University of Ghana http://ugspace.ug.edu.gh
 
 
COLLEGE OF BASIC AND APPLIED SCIENCES 
SCHOOL OF BIOLOGICAL SCIENCES 
 
ASSESSING THE EFFICIENCY OF TWO IMPROVED LESSER 
KNOWN KILNS AND THEIR EFFECT ON THE QUALITY AND 
SHELF LIFE OF SMOKED FISH IN GHANA 
 
BY 
EUNICE KONADU ASAMOAH 
(10085118) 
 
THIS THESIS IS SUBMITTED TO UNIVERSITY OF GHANA, LEGON, 
IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE 
AWARD OF PHD IN FISHERIES SCIENCE DEGREE 
 
DEPARTMENT OF MARINE AND FISHERIES SCIENCES 
 
JULY, 2019
University of Ghana http://ugspace.ug.edu.gh
DECLARATION 
This thesis is the result of original research work undertaken by Eunice Konadu Asamoah, of 
the Department of Marine and Fisheries Sciences, University of Ghana, under the supervision 
of Prof. Francis K. E. Nunoo, Dr. Samuel Addo and Prof. Grethe Hyldig. This research has not 
been included in any thesis or dissertation submitted to any other institution for a degree or any 
other qualifications. Authors whose works were used have been duly referenced/recognised.  
 
31st July 2019
…………………………………………….…   Date: ………………………. 
Eunice Konadu Asamoah 
(PhD Student) 
 
31st July 2019
………………………………………….    Date: ………………………. 
Prof. Francis. K. E. Nunoo 
 (Principal Supervisor) 
Department of Marine and Fisheries Sciences, University of Ghana, Legon  
 
……………………………….….……….    Date: ………………………. 
Dr. Samuel Addo 
(Supervisor) 
Department of Marine and Fisheries Sciences, University of Ghana, Legon  
 
 
…………………………….…….…………    Date: ………………………. 
Prof. Grethe Hyldig 
(Supervisor) 
National Food Institute, Technical University of Denmark 
 
i 
University of Ghana http://ugspace.ug.edu.gh
ABSTRACT  
Fish smoking is a traditional fish preservation method which is affordable, and hence employed 
in most developing countries with logistical challenges in preserving fresh fish for marketing. 
Smoked fish is a major source of protein in the diets of Ghanaians. Often traditional kilns, that 
rely on firewood as a source of fuel, are used. These kilns have been shown to be less fuel 
efficient and the smoked products have high levels of polycyclic aromatic hydrocarbons 
(PAHs), which are of public health concern. Additionally, poor storage of the products leads to 
quality losses. This study, which contributes to the search for a more efficient and safe smoking 
oven, therefore sought to test and compare the efficiency of two improved, but lesser known 
kilns, the Cabin and Abuesi gas fish smoker (AGFS) to the traditional Chorkor smoker by 
investigating the physicochemical, microbial and sensory qualities of the smoked products. 
Finally, the effect of irradiation and different storage conditions on the shelf life of the smoked 
products was studied. The research was undertaken in Abuesi, in the Western Region, using 
two marine fish species, the chub mackerel and barracuda.  
 
The results showed that the AGFS had 12% lower yield but 86% and 60% higher processing 
rate than the cabin and Chorkor kilns respectively. Again, the fuel consumption was 68% and 
54% better than the Cabin and Chorkor respectively, while the Cabin also saved 29% more fuel 
than the Chorkor. In terms of the fuel costs, the Cabin was 38% and 54% lower than the Chorkor 
and AGFS respectively (owing to the lower cost of firewood, compared to LPG). The cost of 
construction was however extremely high in the AGFS than the Cabin and Chorkor kilns, 
however, its industrial size (500 kg capacity), faster smoking time and lower fuel consumption 
make it good alternative to consider. 
 
ii 
University of Ghana http://ugspace.ug.edu.gh
Smoking improved the physical, chemical, microbiological and sensory quality of mackerel 
and barracuda. These qualities, except for colour and sensory analysis, could not be statistically 
differentiated between the products from the AGFS and Cabin kiln. The Cabin-smoked 
products had the more traditional qualities of smoked fish (appearance, odour and flavour), 
while the gas-smoked products had a pronounced fried appearance and taste. 
 
The AGFS produced smoked products with mean benzo(a)pyrene and PAH4 concentrations 
below the EU MLs (2 and 12 µg/kg respectively). Depending on the type of firewood used, the 
Cabin also produced benzo(a)pyrene below the MLs when C. mildbraedii (Esa) was used, while 
the Chorkor had levels 3 to 8 times higher than the MLS. The PAH4 levels in the Cabin and 
Chorkor products were all above the MLs (4 and 8 times higher respectively). Based on the 
frequency and quantities of smoked mackerel and barracuda consumed by an average Ghanaian 
adult (with a life expectancy of 63 years), the potential carcinogenic risks were of least concern 
in the gas smoked and all barracuda samples (about 1 in 100,000 adults), moderate in the Cabin 
smoked mackerel (3 and 6 in 100,000 adults) and high in the Chorkor smoked mackerel (7 and 
17 in 100,000 adults). Heavy metal (Hg, Pb and Cd) contamination was negligible in fresh and 
smoked mackerel and barracuda.  
 
The effect of irradiation and storage temperature on the quality and shelf life of smoked 
mackerel showed that irradiation did not affect the nutritional quality (protein, fat, moisture and 
ash contents) after 65 days of refrigerated storage. The fatty and amino acid compositions were 
also unaffected by irradiation. Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) 
respectively constituting about 8% and 18% of the total fatty acids. The essential amino acids 
also contributed about 59% of the total amino acid concentrations. The non-irradiated and 
irradiated smoked mackerel, stored at refrigerated temperature, were of good microbial and 
iii 
University of Ghana http://ugspace.ug.edu.gh
chemical quality by Day 65 of storage, even though lipid oxidation and hydrolysis were 
affected. Keeping the non-irradiated and irradiated smoked fish at room temperature were 
rejected by Day 5 of storage due to insect infestation and visible mouldiness (even though most 
microbial and chemical qualities were good). 
 
The results, therefore, indicates that The AGFS performed better, overall, followed by the Cabin 
and then the Chorkor. The kilns produced fish with good nutritional qualities. Irradiation did 
not negatively impact on the quality of the smoked products during the 65 days of storage, but 
refrigerated storage is key to maintaining quality of irradiated fish. 
 
  
iv 
University of Ghana http://ugspace.ug.edu.gh
DEDICATION 
I dedicate this thesis to my dear husband, Dr. Richard Asamoah, our sons (Jayden, Joel and 
Jeremy), extended family and to all smoked fish processors in Ghana and other developing 
countries. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
v 
University of Ghana http://ugspace.ug.edu.gh
ACKNOWLEDGEMENTS 
To God be the glory for the great things he continues to accomplish in my life. 
 
I acknowledge the contributions of all individuals and parties who made this research work 
possible. I am most indebted to DANIDA, through the Building Stronger Universities Phase II 
Scholarship for funding my entire PhD Programme. My appreciation goes to the Director and 
all coordinators of the scholarship programme at the Office of Research, Innovation and 
Development, University of Ghana. I am further grateful to the United Nations University-
Fisheries Training Program, Iceland, for providing me with the schematics and funding for the 
construction of the Cabin kiln. 
 
My profound appreciation goes to my supervisors, Professor Francis K. E. Nunoo, Dr. Samuel 
Addo, both of the University of Ghana Legon, and Prof. Grethe Hyldig of the National Food 
Institute, Danish Technical University for their guidance, inspiration, support and outstanding 
intellect during my study and completion of this work. Also, to the Faculty, Staff and the Fish 
Processing Research Group of the Department of Marine and Fisheries Sciences, University of 
Ghana, I say a big thank you for all your encouragement and support throughout this journey. 
 
I am most grateful to CEO of the Abuesi Fish Processing Association, Mr. George Aidoo-
Abban, and all the fish processors for allowing me to carry out this research at their facility. I 
further wish to thank Engineers George Essandoh and Charles Mensah for constructing the 
Cabin kiln. My special thanks go to my Special Assistants, Ms. Josephine Nyarko and Ms. 
Adwoa Konadu-Twum for their support during my field work. Many thanks go to the staff of 
the chemical laboratory of FRI-CSIR, Envaserv Research Consult, Radiation Technology 
Centre (GAEC), microbiology and sensory laboratories (Department of Nutrition and Food 
vi 
University of Ghana http://ugspace.ug.edu.gh
Science), Ms. Rie Sørensen and Ms. Inge Holmberg, both of Danish Technical University for 
their immense help with analysing my samples.  
 
To my friends, the Kwansa’s, Houphouet’s, Mrs. Audrey Opoku-Acheampong, Dr. Ebenezer 
Nyadjro, Mr. Charles Adjei Ampong and Ms. Edna Quansah, thank you for all your wonderful 
support. Finally, to my husband, children, parents, siblings and in-laws, I would not be here 
without your unwavering support, love, prayers and encouragement.  
 
God bless you all. 
 
  
vii 
University of Ghana http://ugspace.ug.edu.gh
TABLE OF CONTENTS 
DECLARATION ........................................................................................................................ i 
ABSTRACT .............................................................................................................................. ii 
DEDICATION ........................................................................................................................... v 
ACKNOWLEDGEMENTS ...................................................................................................... vi 
LIST OF TABLES .................................................................................................................. xiv 
LIST OF FIGURES ................................................................................................................ xvi 
LIST OF PLATES ............................................................................................................... xviii 
LIST OF ABBREVIATIONS AND ACRONYMS ............................................................... xix 
CHAPTER ONE ........................................................................................................................ 1 
1.0 GENERAL INTRODUCTION ...................................................................................... 1 
1.1 Background ............................................................................................................. 1 
1.2 Aim and Objectives ................................................................................................. 8 
1.3 Organisation of study .............................................................................................. 9 
CHAPTER TWO ..................................................................................................................... 10 
2.0 LITERATURE REVIEW ............................................................................................ 10 
2.1 Fisheries sector in Ghana ...................................................................................... 10 
2.2 Species of interest ................................................................................................. 11 
2.2.1 Atlantic chub mackerel ..................................................................................... 11 
2.2.2 European Barracuda .......................................................................................... 12 
2.3 Fish Preservation Methods .................................................................................... 13 
2.3.1 Fish smoking ..................................................................................................... 14 
2.3.1.1 Changes in fish muscle during smoking .................................................. 16 
viii 
University of Ghana http://ugspace.ug.edu.gh
2.3.2 Fish smoking in Ghana ..................................................................................... 17 
2.3.2.1 Fish smoking kilns ................................................................................... 19 
2.3.3 Chemical composition of smoke and smoked foods ......................................... 20 
2.3.3.1 Polycyclic aromatic hydrocarbons (PAHs) contamination in fresh and 
smoked fish .............................................................................................................. 20 
2.3.3.2 Heavy metal contamination in fresh and smoked fish ............................. 25 
2.3.4 Quality degradation of smoked fish .................................................................. 26 
2.3.4.1 Microbial spoilage ................................................................................... 26 
2.3.4.2 Chemical spoilage .................................................................................... 30 
2.3.5 Strategies for shelf life extension of smoked fish ............................................. 33 
2.3.5.1 Packaging and storage of smoked products ............................................. 33 
2.3.5.2 Irradiation ................................................................................................. 34 
CHAPTER THREE ................................................................................................................. 38 
3.0 COMPARISON OF THE PERFORMANCE AND EFFICIENCY OF THE 
IMPROVED AND TRADITIONAL SMOKING KILNS ...................................................... 38 
3.1 Introduction ........................................................................................................... 38 
3.2 Materials and methods .......................................................................................... 40 
3.2.1 Controlled cooking test (CCT) protocol ........................................................... 40 
3.2.1.1 Test location ............................................................................................. 40 
3.2.1.2 Description of fish smoking kilns tested .................................................. 41 
3.2.1.3 Fuel used .................................................................................................. 44 
3.2.2 Atmospheric conditions .................................................................................... 44 
3.2.3 Experimental procedures .................................................................................. 45 
3.2.3.1 Fish samples preparation .......................................................................... 45 
3.2.3.2 Stack Emissions Monitoring .................................................................... 46 
ix 
University of Ghana http://ugspace.ug.edu.gh
3.2.4 Smoked products measurements ....................................................................... 48 
3.2.5 Data and Statistical Analysis ............................................................................ 48 
3.3 Results ................................................................................................................... 50 
3.3.1 Efficiency of the smoking kilns ........................................................................ 50 
3.3.2 Emissions test ................................................................................................... 52 
3.4 Discussion ............................................................................................................. 53 
3.5 Summary of findings ............................................................................................. 57 
CHAPTER FOUR .................................................................................................................... 59 
4.0 ASSESSMENT OF THE QUALITY OF TWO COMMERCIALLY IMPORTANT 
FISH SPECIES SMOKED USING TWO DIFFERENT KILNS IN GHANA ....................... 59 
4.1 Introduction ........................................................................................................... 59 
4.2 Materials and Methods .......................................................................................... 61 
4.2.1 Description of study area .................................................................................. 61 
4.2.2 Fish smoking kilns ............................................................................................ 61 
4.2.3 Fish sample acquisition and processing ............................................................ 61 
4.2.4 Analyses ............................................................................................................ 62 
4.2.4.1 Physical analysis ...................................................................................... 62 
4.2.4.2 Chemical analyses .................................................................................... 63 
4.2.4.3 Microbiological analyses ......................................................................... 66 
4.2.4.4 Sensory evaluation ................................................................................... 68 
4.2.5 Statistical analysis ............................................................................................. 69 
4.3 Results ................................................................................................................... 72 
4.3.1 Physicochemical quality of fresh and smoked mackerel and barracuda ........... 72 
4.3.2 Chemical composition ...................................................................................... 72 
4.3.3 Colour analysis .................................................................................................. 73 
x 
University of Ghana http://ugspace.ug.edu.gh
4.3.4 Microbiological analyses .................................................................................. 74 
4.3.5 Sensory evaluation ............................................................................................ 76 
4.4 Discussion ............................................................................................................. 80 
4.5 Summary of findings. ............................................................................................ 87 
CHAPTER FIVE ..................................................................................................................... 88 
5.0 POLYCYCLIC AROMATIC HYDROCARBONS (PAHs) AND HEAVY 
METALS IN FRESH AND SMOKED FISH USING THREE DIFFERENT KILNS: 
LEVELS AND HUMAN HEALTH RISK IMPLICATIONS THROUGH DIETARY 
EXPOSURE IN GHANA ........................................................................................................ 88 
5.1 Introduction ........................................................................................................... 88 
5.2 Materials and Methods .......................................................................................... 92 
5.2.1 Fish sampling and preparation .......................................................................... 92 
5.2.2 Fish smoking process ........................................................................................ 92 
5.2.3 Consumer survey .............................................................................................. 93 
5.2.4 Analytical methods ........................................................................................... 93 
5.2.4.1 Moisture and fat content analyses ............................................................ 93 
5.2.4.2 Heavy metal analysis ............................................................................... 93 
5.2.4.3 PAH analysis ............................................................................................ 94 
5.2.5 Human health risk assessment .......................................................................... 95 
5.2.5.1 Health risk associated with heavy metals ................................................ 95 
5.2.5.2 Toxicological risk associated with PAH concentrations ......................... 96 
5.2.5.3 Carcinogenic risk (CR) assessment ......................................................... 96 
5.2.5.4 Non-carcinogenic risk (non-CR) estimates .............................................. 97 
5.2.6 Data analysis ..................................................................................................... 98 
5.3 Results ................................................................................................................... 98 
xi 
University of Ghana http://ugspace.ug.edu.gh
5.3.1 Quality of fresh and smoked fish ...................................................................... 98 
5.3.2 PAHs and heavy metal levels in fresh fish ....................................................... 99 
5.3.3 PAHs and heavy metal levels in smoked fish samples ................................... 102 
5.3.4 Human health risk assessment and dietary exposure of PAHs ....................... 105 
5.3.4.1 Carcinogenic health risk assessment ...................................................... 105 
5.3.4.2 Non-carcinogenic risk assessment ......................................................... 106 
5.3.5 Health risk associated with heavy metals ....................................................... 106 
5.4 Discussion ........................................................................................................... 109 
5.5 Summary of findings ........................................................................................... 116 
CHAPTER SIX ...................................................................................................................... 117 
6.0 THE INFLUENCE OF IRRADIATION AND STORAGE TEMPERATURE ON 
THE QUALITY AND SHELF LIFE OF SMOKED MACKEREL ...................................... 117 
6.1 Introduction ......................................................................................................... 117 
6.2 Materials and methods ........................................................................................ 119 
6.2.1 Gamma Irradiation .......................................................................................... 119 
6.2.2 Analytical methods ......................................................................................... 120 
6.2.2.1 Colour analysis ....................................................................................... 120 
6.2.2.2 Chemical analyses .................................................................................. 120 
6.2.2.3 Microbiological analyses ....................................................................... 123 
6.2.2.4 Sensory evaluation ................................................................................. 123 
6.2.2.5 Insect infestation .................................................................................... 124 
6.2.3 Data analysis ................................................................................................... 124 
6.3 Results ................................................................................................................. 124 
6.3.1 Chemical composition .................................................................................... 125 
6.3.1.1 Proximate composition .......................................................................... 125 
xii 
University of Ghana http://ugspace.ug.edu.gh
6.3.1.2 Fatty acid composition ........................................................................... 126 
6.3.1.3 Amino acid composition ........................................................................ 129 
6.3.1.4 pH ........................................................................................................... 131 
6.3.1.5 Total volatile base (TVB) ...................................................................... 131 
6.3.1.6 Peroxide value (PV) ............................................................................... 132 
6.3.1.7 Free fatty acid (FFA) ............................................................................. 133 
6.3.2 Microbial analyses .......................................................................................... 134 
6.3.2.1 Total mesophilic count (TVC) ............................................................... 134 
6.3.2.2 Faecal coliform and E. coli counts ......................................................... 136 
6.3.2.3 Staphylococcus aureus ........................................................................... 136 
6.3.2.4 Clostridium perfringens ......................................................................... 136 
6.3.2.5 Yeast and moulds ................................................................................... 137 
6.3.3 Insect infestation ............................................................................................. 137 
6.3.4 Colour analysis ................................................................................................ 137 
6.3.5 Sensory analysis .............................................................................................. 138 
6.4 Discussion ........................................................................................................... 140 
6.5 Summary of findings ........................................................................................... 147 
7.0 CONCLUSION AND RECOMMENDATIONS ...................................................... 149 
7.1 Conclusion .......................................................................................................... 149 
7.2 Recommendations ............................................................................................... 151 
7.2.1 Policy .............................................................................................................. 151 
7.2.2 Research .......................................................................................................... 151 
REFERENCES ...................................................................................................................... 153 
 
  
xiii 
University of Ghana http://ugspace.ug.edu.gh
LIST OF TABLES 
Table 2.1: PAH concentrations in smoked fish from different kilns in Ghana ......................... 24 
Table 3.1: Comparison of efficiency of Chorkor, Cabin and AGFS kilns (Mean ± SD) ......... 51 
Table 3.2: Comparison of emissions (mg/m3) from the Chorkor and Cabin kilns (Mean ± SD)
................................................................................................................................................... 52 
Table 4.1: Sensory descriptors developed for smoked mackerel .............................................. 70 
Table 4.2: Sensory descriptors for smoked barracuda .............................................................. 71 
Table 4.3: Chemical quality (Mean ± SD) characteristics of fresh (F), cabin smoked (CS) and 
gas smoked (GS) mackerel (M) and barracuda(B) (n = 4) ....................................................... 73 
Table 4.4: Skin colour (Mean ± SD) attributes of cabin smoked (CS) and gas smoked (GS) 
mackerel (M) and barracuda(B) (n = 5) .................................................................................... 74 
Table 4.5: Muscle colour (Mean ± SD) attributes of cabin smoked (CS) and gas smoked (GS) 
mackerel (M) and barracuda(B) (n = 5) .................................................................................... 74 
Table 4.6: Prevalence of microorganisms in fresh (F), cabin smoked (CS) and gas smoked (GS) 
mackerel (M) and barracuda(B) (n = 4) .................................................................................... 75 
Table 4.7: Microbiological quality (Mean ± SD) of fresh (F), cabin smoked (CS) and gas 
smoked (GS) mackerel (M) and barracuda(B) (n = 4) .............................................................. 76 
Table 5.1: Oral ingestion reference dose (RfDo) and potency equivalency factor (PEFs) used in 
human intake model for estimating cancer risk and hazard index ............................................ 97 
Table 5.2: Chemical quality (Mean ± SD) characteristics of fresh (F), gas smoked (GS) and 
Afena (A) and Esa (E) cabin smoked (CS) and Chorkor smoked (ChS) mackerel (M) ........... 99 
Table 5.3: Chemical quality (Mean ± SD) characteristics of fresh (F), gas smoked (GS) and 
Afena (A) and Esa (E) cabin smoked (CS) barracuda(B) (n = 5) ............................................. 99 
xiv 
University of Ghana http://ugspace.ug.edu.gh
Table 5.4: Concentrations (ug/kg wet weight) of individual PAHs in fresh (F), gas smoked (GS) 
and Afena (A) and Esa (E) cabin smoked (CS) and Chorkor smoked (ChS) mackerel (M). (Mean 
± SD) ....................................................................................................................................... 101 
Table 5.5: Concentrations (ug/kg wet weight) of individual PAHs in fresh (F), gas smoked (GS) 
and Afena (A) and Esa (E) cabin smoked (CS) barracuda (B) (n = 5) (Mean ± SD) ............. 102 
Table 5.6: Estimated carcinogenic risks associated with the consumption of fresh (F), gas 
smoked (GS) and Afena (A) and Esa (E) cabin smoked (CS) and Chorkor smoked (ChS) 
mackerel (M) and barracuda (B). ............................................................................................ 107 
Table 5.7: Estimated non-carcinogenic risks associated with the consumption of fresh (F), gas 
smoked (GS) and Afena (A) and Esa (E) cabin smoked (CS) and Chorkor smoked (ChS) 
mackerel (M) and barracuda (B). ............................................................................................ 108 
Table 6.1: Proximate composition of fresh (F), smoked (S) and gamma irradiated (g) mackerel 
(M) for Day 1 and Day 65 of refrigerated storage .................................................................. 125 
Table 6.2: Fatty acid composition of fresh (F), smoked (S) and gamma irradiated (g) mackerel 
(M) (g fatty acid/100 g oil sample) ......................................................................................... 128 
Table 6.3: Amino acid composition of fresh (F), smoked (S) and gamma irradiated (g) mackerel 
(M) (mg/g) .............................................................................................................................. 130 
Table 6.4: Skin and muscle colour characteristics of smoked (S) and gamma irradiated (g) 
mackerel (M) ........................................................................................................................... 138 
 
 
  
xv 
University of Ghana http://ugspace.ug.edu.gh
LIST OF FIGURES 
Figure 1.1: An overview of the study design for this thesis ....................................................... 9 
Figure 2.1: Catch trends of Atlantic chub mackerel and European barracuda in Ghana (Source: 
FAO FishStat) ........................................................................................................................... 12 
Figure 2.2: The effects of microbial activities on food quality (modified from Cyprian, 2015)
................................................................................................................................................... 29 
Figure 2.3: Autoxidation of polyunsaturated lipid (Adopted from Huss, 1995) ...................... 31 
Figure 2.4: The effects of lipid oxidation on food quality (modified from Kolakowska, 2003)
................................................................................................................................................... 32 
Figure 2.5: Hydrolytic reactions of triglycerides and phospholipids: PL1 & PL2 phospholipases; 
TL, triglyceride lipase (Adopted from Huss, 1995) .................................................................. 33 
Figure 4.1: Spider web plot of sensory profile of cabin smoked (CS) and gas smoked (GS) 
mackerel (M) (AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = 
Mouthfeel; AF- = Aftereffect) .................................................................................................. 77 
Figure 4.2: Spider web plot of sensory profile of cabin smoked (CS) and gas smoked (GS) 
barracuda (B) (AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = 
Mouthfeel; AF- = Aftereffect) .................................................................................................. 78 
Figure 4.3: Product map showing product and descriptor loading of cabin smoked (CS) and gas 
smoked (GS) mackerel (M). (AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; 
MF- = Mouthfeel; AF- = Aftereffect) ....................................................................................... 79 
Figure 4.4: Product map showing product and descriptor loading for smoked barracuda. 
Abbreviations: AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = 
Mouthfeel; AF- = Aftereffect; GSB for gas smoked barracuda; CSB for cabin smoked barracuda
................................................................................................................................................... 80 
xvi 
University of Ghana http://ugspace.ug.edu.gh
Figure 6.1: pH of fresh (F), smoked (S) and gamma irradiated (g) mackerel (M) stored at 
ambient (A) and refrigerated (R) temperatures for 65 days. Refrigerated (R = 2-4°C) and 
ambient temperatures (A = 27-30°C) ...................................................................................... 131 
Figure 6.2: Total volatile base (TVB) content of fresh (F), smoked (S) and gamma irradiated 
(g) mackerel (M) stored at ambient (A) and refrigerated (R) temperatures for 65 days. 
Refrigerated (R = 2-4 °C) and ambient temperatures (A = 27-30°C) ...................................... 132 
Figure 6.3: Peroxide value (PV) of fresh (F), smoked (S) and gamma irradiated (g) mackerel 
(M) .......................................................................................................................................... 133 
Figure 6.4: Free fatty acid (FFA) composition of fresh (F), smoked (S) and gamma irradiated 
(g) mackerel (M) stored at ambient (A) and refrigerated (R) temperatures for 65 days. 
Refrigerated (R = 2-4 °C) and ambient temperatures (A = 27-30°C) ...................................... 134 
Figure 6.5: Total mesophilic (a), S. aureus (b), faecal coliform (c), E. coli (d), C. perfringens 
(e) and yeast and mould (f) counts in fresh (F), smoked (S) and gamma irradiated (g) mackerel 
(M) stored at ambient (A) and refrigerated (R) temperatures for 65 days. Refrigerated (R = 2-4 
°C) and ambient temperatures (A = 27-30°C) ......................................................................... 135 
Figure 6.6: Mean intensity scores for difference-from-control test (Scale: 0 = No difference; 
1= Very slight difference; 2 = Slight difference; 3 = Moderate; 4 = Very different and 5 = 
Extreme difference) ................................................................................................................. 139 
Figure 6.7: Frequency of responses for CATA attributes indicating perceived differences .. 139 
 
 
  
xvii 
University of Ghana http://ugspace.ug.edu.gh
LIST OF PLATES 
Plate 2.1: The Atlantic chub mackerel (S. colias) ..................................................................... 11 
Plate 2.2: The European barracuda (S. sphyraena) ................................................................... 13 
Plate 3.1: Chorkor smoking kiln ............................................................................................... 41 
Plate 3.2: Cabin kiln .................................................................................................................. 42 
Plate 3.3: The Abuesi gas fish smoker (AGFS) ........................................................................ 43 
Plate 3.4: ‘Kontan’ (Uapaca guineensis) used in smoking trials (a) and measuring wood 
moisture (b) for the CCT trials .................................................................................................. 44 
Plate 3.5: Stack emissions monitoring of (a) Cabin and (b) Chorkor kilns; (c) E9000 emission 
system ....................................................................................................................................... 46 
Plate 3.6: PM monitoring (a) in Cabin kiln and filter paper showing PM residue (b) for analysis
................................................................................................................................................... 47 
Plate 3.7: Fish sample from the (a) Cabin, (b) Chorkor and (c) AGFS smoking kilns ............. 48 
 
 
  
xviii 
University of Ghana http://ugspace.ug.edu.gh
LIST OF ABBREVIATIONS AND ACRONYMS  
AGFS:  Abuesi Gas Fish Smoker 
ANOVA: Analysis of Variance 
AOAC: Association of Official Analytical Chemists 
AOCS: American Oil Chemists’ Society 
APHA: American Public Health Association  
AWEP: Association of Women for The Preservation of The Environment 
CCT:  Controlled Cooking Test 
Cd:  Cadmium 
CO:  Carbon Monoxide 
CPAH: Carcinogenic Polycyclic Aromatic Compounds 
CSIR:  Council of Scientific and Industrial Research  
DHA:  Docosahexaenoic Acid 
DMA:  Dimethylamine 
EnDev: Energising Development 
EPA:  Environmental Protection Agency 
EPA:  Eicosapentaenoic Acid 
EU:  European Union 
FAME: Fatty Acid Methyl Esters 
FAO:  Food and Agriculture Organization 
FDA:  Food and Drug Authority 
FFA:  Free fatty acid 
FRI:  Food Research Institute  
FTT:  FAO-Thiaroye processing technique 
GRATIS: Ghana Regional Appropriate Technology Industrial Service 
xix 
University of Ghana http://ugspace.ug.edu.gh
GSA:  Ghana Standards Authority 
Hg:  Mercury 
HMW:  High molecular weight 
HPLC:  High performance liquid chromatography 
IARC:  International Agency for Research on Cancer 
IAEA:  International Atomic Energy Agency 
ICMSF: International Commission for Microbiological Specifications for Foods 
ILCR:  Incremental lifetime cancer risk 
ISO:  International Standards Organisation 
LMW:  Lower molecular weight 
LOD:  Limit of detection 
LOQ:  Limit of quantification 
LSD:  Least significant difference 
MLs:  Maximum limits 
MMW: Medium molecular weight 
MOFA: Ministry of Food and Agriculture 
MOFAD: Ministry of Fisheries and Aquaculture Development 
MUFA: Monounsaturated fatty acids 
NAFPTA: National Association of Fish Processors and Traders 
NOx:  Oxides of Nitrogen 
PAH:  Polycyclic Aromatic Hydrocarbons  
Pb:  Lead 
PCA:  Principal Component Analysis  
PM:  Particulate matter  
PUFA:  Polyunsaturated fatty acid 
xx 
University of Ghana http://ugspace.ug.edu.gh
PV:  Peroxide value  
QDA:  Quantitative descriptive analysis 
SFA:  Saturated fatty acid 
SO2:  Sulphur dioxide   
THQ:  Target hazard quotient 
TVB:  Total volatile base 
TVC:  Total viable mesophilic count 
UNDP/TCDC: United Nations Development Program/ Technical Cooperation among  
Developing Countries 
UNU-FTP: United Nations University-Fisheries Training Programme 
USA:  United States of America 
USEPA: United States Environmental Protection Agency 
WARFP: West Africa Regional Fisheries Programme 
WHO:  World Health Organization 
xxi 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER ONE 
 
1.0 GENERAL INTRODUCTION 
1.1 Background  
Fish and fisheries products serve as significant sources of food and incomes for numerous 
people, particularly those from developing countries (FAO, 2018). Fish is highly nutritious, 
containing approximately 70-84% water; 15-24% protein; 0.1-22% fat; 1-2%, minerals; 0.5% 
calcium; 0.25% phosphorus; and 0.1% vitamins A, D, B and C (Abraha, et al., 2018). 
Consumption of fish and fisheries products has been known to reduce the risks of 
cardiovascular diseases, type 2 diabetes, inflammatory diseases, certain cancers, dementia and 
psychological problems (Hosomi, Yoshida, & Fukunaga, 2015). The fishing industry further 
serves as an important economic activity, especially in Africa, where about 10% of the 
population are engaged in fisheries and aquaculture for their livelihoods (FAO, 2018).  
 
In Ghana, fish and fish products are most important non-traditional export commodities, 
accounting for about 50% of all revenue from non-traditional exports (FAO, 2016-2019). The 
industry also contributes about 3% to the country’s gross domestic product, employs about 
10% of the population and generates an estimated USD 1 billion in total revenue each year 
(MOFAD, 2016; GIPC, 2019). Again, fish contributes significantly to the food security in the 
country. About 50% of the animal protein intake in the diet of Ghanaians (i.e. annual per capita 
consumption of about 19.8 kg; lower than the world, but higher than African average of 20.2 
kg and 9.9 kg respectively) is derived from consumption of fish and fisheries products (FAO, 
2018; MOFAD, 2018).  
 
 1 
University of Ghana http://ugspace.ug.edu.gh
Ghana’s fishery comprises the inland (freshwater and aquaculture) and marine subsectors. The 
marine subsector accounts for about 85% of the total fish catches, exploiting about 347 fish 
species, 17 cephalopod and 25 crustacean species (Nunoo, Asiedu, Kombat, & Samey, 2015). 
The small-scale (artisanal and subsistence), semi-industrial and industrial fishing fleets exploit 
the marine fisheries resources. The species mostly exploited are the small pelagics (making up 
a total of about 70% of total catches), like the round sardinella (Sardinella aurita), flat 
sardinella (S. maderensis), European anchovy (Engraulis encrasicolus), Atlantic horse 
mackerel (Trachurus trachurus), chub mackerel (Scomber colias), round scad (Decapterus 
punctatus) and other species such as the snappers, tunas and barracudas (Sphyraena sphyraena) 
(Nunoo, Asiedu, Kombat, & Samey, 2015; FAO, 2016-2019). 
 
Fresh fish starts deteriorating soon after harvest, due to its almost neutral pH, high water 
activity, fast onset of rigor mortis and high amounts of nutrients that promote development of 
microorganisms (Belusso, Nogueira, Breda, & Mitterer-Daltoe, 2016; de Alba, et al., 2019). It 
therefore requires a degree of processing to preserve and extend its shelf life for extended 
distribution and marketing. Preservation and processing techniques involves a reduction in 
temperature (e.g. freezing and chilling), heat treatment (e.g. smoking, boiling, frying and 
canning), decreasing available water (e.g. salting, drying and smoking) and altering the storage 
environment (packaging and refrigeration) may therefore be employed to ensure a safe and 
stable product (FAO, 2016a). Traditional preservation methods like salting, fermenting, drying 
and smoking account for 12% of all fish destined for human consumption, especially in Africa 
and Asia (FAO, 2018).  
 
Smoking is an ancient and affordable preservation method that enhances the flavour, colour 
and texture of smoked products, while prolonging the shelf life of the product (Arason, 
 2 
University of Ghana http://ugspace.ug.edu.gh
Nguyen, Thorarinsdottir, & Thorkelsson, 2014). In Ghana, many preservation methods are 
employed, with smoking being the most widely used (Ayinsa & Maalekuu, 2013). Smoking is 
used to preserve both marine (mainly common small pelagic species like sardines, anchovies, 
chub and horse mackerels and other species like the barracuda) and freshwater fish species 
(mostly tilapia and catfish). It has been estimated that about 70-80% of fish consumed in Ghana 
is smoked (Nunoo, Asiedu, Kombat, & Samey, 2015), with the products traditionally used in 
soups and sauces (UNDP/TCDC, 2001). Smoked chub mackerel is considered relatively cheap, 
whereas smoked barracuda is highly valued (Nunoo, Asiedu, Kombat, & Samey, 2015). Both 
are consumed locally and also exported, usually to the EU and USA, with smoked barracuda 
estimated to fetch up to EUR 8/kg in some European countries (Asiedu, Failer, & Beygens, 
2018). These species were therefore selected based on the reasons enumerated. 
  
Fish smoking in Ghana is categorized into two, small scale (who make up about 98%) and 
industrial (Asiedu, Failer, & Beygens, 2018). The small scale processors are mainly women, 
living in close proximity to coastal communities and rivers, who engage in both traditional 
smoking and trading of smoked fish products. They operate on small (individual or household 
levels with less than 10 kilns), medium (11-25 kiln) and large (more than 25 kilns, belonging 
to fish processors associations or co-operative) scale basis (Gordon, Pulis, & Owusu-Adjei, 
2011; Nunoo, Asiedu, Kombat, & Samey, 2015). The industrial sector is mostly export-driven, 
with four companies currently active and certified by the Ghana Standards Authority to export 
smoked fish (Asiedu, Failer, & Beygens, 2018). Smoked fish is however mostly consumed 
locally but there is a well-developed marketing system extending into neighbouring countries 
like Togo, Nigeria, Benin, Burkina Faso and a high demand for these products in the EU and 
USA (Gordon, Pulis, & Owusu-Adjei, 2011; Nunoo, Asiedu, Kombat, & Samey, 2015). It has 
 3 
University of Ghana http://ugspace.ug.edu.gh
been estimated that, between 2007 and 2016, about 20 tonnes of smoked fish valued at €79,000, 
on average, was exported from Ghana to the EU every year (Asiedu, Failer, & Beygens, 2018). 
 
Fish in Ghana is mainly smoked using traditional kilns, with the most common one being the 
‘Chorkor Smoker’. This kiln was developed by the Food and Agriculture Organisation of the 
United Nations (FAO) and the Food Research Institute of the Council for Scientific and 
Industrial Research (CSIR) and introduced in Ghana in 1969 (UNDP/TCDC, 2001). It replaced 
traditional smoking kilns (described in UNDP/TCDC, 2001), which were deemed 
environmentally and economically inefficient and affected the health of processors, most of 
whom are women. As of 2015, there were about 120,000 Chorkor and traditional smoking kilns 
in operation in Ghana (Okyere-Nyarko, Aziebor, & Robinson, 2015).  
 
The introduction of the Chorkor smoker has however not solved these problems as it is only 
moderately fuel efficient (resulting in increased fuel use and therefore leading to deforestation 
of especially mangrove forests). It has not met the Energising Development (EnDev) 
requirement of 40% fuel saving potential and also offers moderate emission gains (Okyere-
Nyarko, Aziebor, & Robinson, 2015). Furthermore, the fish is exposed to direct smoke from 
smoldering wood and this can result in the deposition of harmful substances like polycyclic 
aromatic compounds (PAH), dioxins, formaldehyde, nitrogen and sulphur oxides, as well as 
some heavy metals (Codex Alimentarius Commission, 2009). Some potentially toxic heavy 
metals detected in fresh and smoked fish are Lead, Cadmium, Mercury, Chromium, Nickel and 
Arsenic (Daniel, Ugwueze, & Igbegu, 2013; Bandowe, et al., 2014). Again, some PAHs are of 
human health concerns since they are carcinogenic, teratogenic and mutagenic to humans 
(Kim, Jahan, Kabir, & Brown, 2013). Also, smoked fish processors usually stay relatively long 
 4 
University of Ghana http://ugspace.ug.edu.gh
in smoke-filled huts and thus develop eye, skin, lung and other respiratory problems 
(Flintwood–Brace, 2016).  
 
These identified problems with smoking kilns necessitated the development of improved 
smoking technologies in Ghana, such as the Morrison, Ahotor and FAO-Thiaroye processing 
technique (FTT), among others (Entee, 2015a, b; IRI-CSIR, GSA, Kwarteng, 2016). A 
comparison of these kilns with the Chorkor smoker showed that none of the kilns met the 
Energising Development (EnDev) requirement of 40% fuel or energy saving efficiency and 
produced high levels of carbon monoxide (CO) and particulate matter (PM). With the exception 
of the FTT, all the smoked fish from the other kilns had PAH levels that were above the 
maximum limits for benzo(a)pyrene and the sum of four carcinogenic PAHs (i.e. 2.0μg/kg and 
12.0μg/kg respectively) set by the EU (European Commission, 2011). This could be hazardous 
to both the processors and patrons of smoked fish products in Ghana (Aheto, et al., 2017).  
  
These enumerated problems, therefore, make it imperative to search for other cost effective, 
fuel and energy efficient smoking kilns that offer lower emission gains. These interventions 
can help produce safe and good quality fish, with minimal deforestation, while at the same time 
impacting less on the health of fish processors. The need for improved smoking technologies 
was captured under the value chain development component of the West Africa Regional 
Fisheries Programme (WARFP), a World Bank project, which ended in 2018 for Ghana. The 
programme aimed at improving fish smoking technologies that reduce the levels of PAH (to 
conform to international standards) in smoked fish thereby making the product safe and 
reducing the impacts on women fish processors. It is also hoped that this will increase the 
marketability of smoked fish products and contribute to the country’s economic growth (The 
World Bank, 2011).  
 5 
University of Ghana http://ugspace.ug.edu.gh
Two technologies that are in line with the WARFP objectives will be evaluated in this study 
and compared with the Chorkor smoker. They are both existing technologies: the Abuesi Gas 
Fish Smoker (AGFS) and the Cabin smoker. The AGFS was designed by the Ghana Regional 
Appropriate Technology Industrial Service (GRATIS) Foundation, in consultation with 
smoked fish processors, to produce hygienic products that would meet both local and 
international safety requirements (Kleter, 2004). It uses agricultural wastes like sugarcane 
bagasse and coconut husks to impart the smoky flavour to the fish, while liquified petroleum 
gas (LPG) aids in the cooking and drying of the fish. This kiln has a closed design that 
safeguards processors from direct contact with fire or smoke, thereby minimising health 
problems, prevents excessive deforestation and produces better and more hygienic fish (Nunoo, 
Asiedu, Kombat, & Samey, 2015). Also, fish smoking proceeds faster, compared to traditional 
kilns and processors can perform other activities while fish is in the kiln (Nunoo, Tornyeviadzi, 
Asamoah, & Addo, 2019). In spite of these benefits, the kiln is not very popular in fish smoking 
ventures (Kleter, 2004; Nunoo, Asiedu, Kombat, & Samey, 2015). 
 
The Cabin kiln was developed by the United Nations University-Fisheries Training Programme 
(UNU-FTP) and Matis Ltd. (an Icelandic Food and Biotech Research and Development 
company) in Iceland. It is intended for use in rural communities in Africa and is currently in 
use in Tanzania and Sierra Leone. It is estimated to use less firewood compared to traditional 
methods, with a much faster smoking time. It is an enclosed unit that ensures minimal exposure 
of processors to smoke and heat, while offering fish of good nutritional value (UNU-FTP, 
2017). 
 
Adoption of these technologies in Ghana could, therefore, be of benefit to the fishing industry 
(both for small scale and industrial processors). In lieu to this adoption, it is essential to evaluate 
 6 
University of Ghana http://ugspace.ug.edu.gh
the quality of smoked fish produced using these smoking kilns and also, its market and 
economic potentials. Also, information obtained on the costs and performance of the kilns, will 
be very useful for consumers and businesses that will be interested in these technologies. 
 
Apart from the problems with the smoking kilns, some problems relating to quality have also 
been identified throughout the fish value chain in Ghana. Loss of quality in fresh fish 
postharvest, resulting from microbial pathogenic contamination and spoilage, has been 
estimated to be between 11-17% (Akande & Diei-Ouadi, 2010). This loss was encountered by 
smoke fish processors immediately after purchase to just before smoking and they attributed 
the loss to poor or no icing of fish and long bargaining at the landing beaches and markets. 
During smoking with Chorkor and other traditional kilns, some fish also get burnt or drop into 
the fire. In addition, during packaging, storage and marketing of smoked fish, there is the 
development of rancidity in the product and insect infestation (Kleter, 2004; Akande & Diei-
Ouadi, 2010). These losses in smoked fish were estimated at USD 60 million between 2006 to 
mid-2008 (Akande & Diei-Ouadi, 2010), a significantly high loss in monetary terms for the 
processors and the country as a whole.  
 
To limit these postharvest losses, an investigation of the quality of fish after capture and 
through processing, storage and marketing is vital. Again, the use of proper packaging 
materials and subsequent storage at the right temperature can ensure that a safe, attractive and 
nutrient-rich product can be delivered to the consumer (Cyprian, et al., 2015). The use of 
low/medium dose irradiation (<1-10 kGy) can also be employed to lessen these losses and 
prolong the shelf life of fish (Arvanitoyannis & Tserkezou, 2014; Ehlermann, 2016). The 
reference dose for irradiation of fish, according to EU (2009) is 3 kGy, and this can guarantee 
a safe product with an increased shelf life (Arvanitoyannis & Tserkezou, 2014). Irradiation has 
 7 
University of Ghana http://ugspace.ug.edu.gh
also been observed to reduce the levels of PAHs in some foods like wheat grains (Khalil, 
Albachir, & Odeh, 2016).  
 
Based on the background provided above, this project was therefore designed to answer the 
following research questions:  
• How will the improved kilns perform in terms of efficiency and cost effectiveness 
compared to the traditional Chorkor kiln? 
• How will the type of smoking kiln used affect the physicochemical, microbial and 
sensory quality of smoked chub mackerel and European barracuda? 
• How do the levels of PAHs and heavy metals in the smoked products compared with 
the EU standards? 
• How will irradiation and different storage temperatures impact the shelf life of smoked 
chub mackerel?  
 
 
1.2 Aim and Objectives  
The main aim of this study was to assess the efficiency of the Cabin and AGFS kilns and 
determine their effects on the quality and shelf life of the Atlantic chub mackerel and European 
barracuda. 
 
The specific objectives were to: 
1. compare the efficiency of the improved smoking kilns to the traditional Chorkor 
smoker;  
2. investigate the effect of the smoking kilns on the physicochemical, microbial and 
sensory quality of smoked chub mackerel and European barracuda; 
 8 
University of Ghana http://ugspace.ug.edu.gh
3. assess the levels of PAH and heavy metals in fresh and smoked chub mackerel and 
European barracuda to evaluate the safety of the products for consumption and 
4. determine the influence of irradiation and storage conditions on the quality and shelf 
life of mackerel smoked with the AGFS. 
 
1.3 Organisation of study  
This thesis is divided into seven main chapters. Chapter One gives a general introduction to 
the thesis. Chapter Two details a review of relevant literature. Chapters Three to Six, which 
tackle the stated objectives, are presented in Figure 1.1. Chapter Seven presents the conclusions 
and recommendations.  
Chapter Three: Chapter Four: Chapter Five: Chapter Six: Influence 
The efficiency of Effect of smoking Levels of PAH and of irradiation and 
the improved technologies on heavy metals in storage conditions on 
smoking kilns smoked fish quality fresh and smoked smoked fish quality fish and stability
Mackerel Mackerel & Mackerel & Barracuda Barracuda Mackerel
Smoked (AGFS); irradiation 
Smoking (Cabin, Fresh & smoked (Cabin Fresh and smoked at 1.5 & 3kGy; storage at 
Chorkor & AGFS) & AGFS) (Cabin, AGFS & 
Chorkor) refrigerated & room temperatures for 65 days
Measurements Measurements Measurements
•Throughput capacity •Length-weight •PAH Measurements
•Specific fuel •Proximate composition •Heavy metals
•Proximate composition
consumption rate
•Salt content •Moisture, protein & 
•pH
•Carbon monoxide lipid contents •Total volatile base (TVB)
(CO) •pH •Free fatty acid (FFA)
•Particulate matter •Total volatile base 
(TVB) •Lipid oxidation(PM)
•Histamine •Fatty acid profile
•Colour •Amino acid composition
•Microbial •Microbial 
•Sensory evaluation •Colour 
•Sensory evaluation
 
Figure 1.1: An overview of the study design for this thesis 
 9 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER TWO 
2.0 LITERATURE REVIEW 
2.1 Fisheries sector in Ghana 
Ghana’s fishing industry is categorised into the inland (freshwater and aquaculture) and marine 
sectors. The marine sector is contributes the most to local fish production, accounting for about 
85% of the total catches, and exploiting about 347 fish species, 17 cephalopod and 25 
crustacean species (Nunoo, Asiedu, Kombat, & Samey, 2015). The resources are exploited by 
the small-scale (artisanal and subsistence), semi-industrial and industrial fishing fleets (FAO, 
2016-2019). The artisanal fleets consist of wooden canoes (about 12,847 canoes) that are either 
motorised or not; the inshore boats are mainly wooden with an inboard engine, while the 
industrial vessels are generally over 25 m long, steel-hulled and can operate in jurisdictions 
outside Ghana (MOFAD, 2016). Overall, the catch from all fleets averages 300,000 metric 
tonnes annually (FAO, 2016-2019), but this has been decreasing since 2001, while the fishing 
effort has been increasing (Nunoo, Asiedu, Amador, Belhabib, & Pauly, 2014).  
 
The artisanal sector is the most significant, employing about 250,000 fishermen (about 92% of 
the total fishers) and accounts for 70% of total landings (FAO, 2016-2019). The sector also 
employs about 60% of the women involved in the fishery value chain and operates from 334 
landing sites in 195 fishing towns across four coastal regions (Nunoo, Asiedu, Amador, 
Belhabib & Pauly, 2014; MOFAD, 2016). The artisanal sector employs multiple gears like 
beach seine, set net, hook and line, drift gill net, and purse seine (‘ali’, ‘poli’ and ‘watsa’) nets. 
The small pelagic species i.e. sardines, anchovies and mackerels (approximately 85% of canoe 
catch) are exploited with the large pelagic fish, mostly tuna and demersal stocks i.e. croakers, 
red snapper, sea breams and red mullet (Nunoo, Asiedu, Kombat, & Samey, 2015). 
 10 
University of Ghana http://ugspace.ug.edu.gh
2.2 Species of interest  
The species of interest are the Atlantic chub mackerel (Scomber colias, Gmelin, 1789) and the 
European barracuda (Sphyraena sphyraena, Linnaeus, 1758). These are important pelagic fish 
species that are mainly smoked and consumed locally by Ghanaians (Heinbuch, 1994). They 
are sold mainly unpackaged in open markets and are used in most soups and sauces (Abboah-
Offei, 2016).  
 
2.2.1 Atlantic chub mackerel  
The Atlantic chub mackerel (S. colias) (Plate 2.1) belongs to the Family Scombridae. It is 
coastal pelagic and occur over the continental slope (Collette, et al., 2011). The Atlantic chub 
mackerel occurs in subtropical regions of the Atlantic Ocean, Mediterranean and Black Seas. 
It occurs at depths of 0-300 m (Riede, 2004), can grow to a length of 18 cm and may live for 
up to 13 years (Collette, et al., 2011). The species feeds on small pelagic fishes like anchovies 
and sardines, as well as pelagic invertebrates.  
 
 
Plate 2.1: The Atlantic chub mackerel (S. colias) 
The Atlantic mackerel is very important in the commercial fisheries sector throughout its range 
(Collette, et al., 2011). In Ghana, it is mainly exploited by the artisanal fleets, and catches 
fluctuate annually (Figure 2.1). Fresh mackerel has dark red flesh colour, moist and flaky 
 11 
University of Ghana http://ugspace.ug.edu.gh
texture and rich pronounced flavour. It is a fatty fish, rich in omega-3 and unsaturated fatty 
acids and is also an excellent source of riboflavin, vitamins B6 and B12, protein, selenium and 
niacin B12 (Hyldig, Larsen, & Green-Petersen, 2007; Stadlmayr, et al., 2010; Nogueira, 
Cordeiro, & Aveiro, 2013; NOAA Fishwatch, 2017). It is a relatively cheap fish that is 
consumed by majority of Ghanaians, mainly smoked but sometimes fresh.  
 
30000 Barracudas Chub mackerel
25000
20000
15000
10000
5000
0
70 72 74 76 78 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 109 9 9 9 9 9 9 9 9 9 9 9 9 9 9 0 0 0 0 0 0 01
2 14
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 20
Year 
 
Figure 2.1: Catch trends of Atlantic chub mackerel and European barracuda in Ghana 
(Source: FAO FishStat) 
 
2.2.2 European Barracuda  
The European barracuda (S. sphyraena) (Plate 2. 2) belongs to the Family Sphyraenidae. It 
occurs in the epipelagic zones of coastal and offshore waters of the Eastern Atlantic (de Morais, 
et al., 2015). The species may be found at depths of 100 m and it is a specialized piscivore, 
feeding normally on pelagic and supra-benthic fishes (make up more than 99% of diet), and 
less frequently on cephalopods and crustaceans (Kalogirou, Mittermayer, Pihl, & Wennhage, 
2012). The barracuda can grow a maximum length of 150 m and can attain a maximum age of 
8 years (de Morais, et al., 2015). 
 
 12 
Catch (tonnes)
University of Ghana http://ugspace.ug.edu.gh
The European barracuda is a commercially important fish species and it is mainly caught by 
artisanal and recreational fishers (Nunoo, Asiedu, Kombat, & Samey, 2015). The annual 
catches are generally increasing (Figure 2.1). The barracuda is a semi-fatty fish and it is rich in 
protein, vitamins, minerals and omega-3 and polyunsaturated fatty acids (Hyldig, Larsen, & 
Green-Petersen, 2007; FEP, 2007-2013; Özogul, Özogul, Çi˙çek, Polat, & Kuley, 2009; 
Stadlmayr, et al., 2010). The fresh barracuda has a light colour and tender, tasty meat (Grygus, 
2015). In Ghana, it is a highly valued fish and is consumed mostly in the smoked form (Nunoo, 
Asiedu, Kombat, & Samey, 2015). It is also exported, usually to the EU and USA, where it can 
fetch up to EUR 8/kg in some European countries (Asiedu, Failer, & Beygens, 2018). 
 
 
Plate 2.2: The European barracuda (S. sphyraena) 
 
2.3 Fish Preservation Methods 
Fish is extremely perishable and therefore requires a degree of processing to preserve and 
prolong its shelf life and thus allow for extended distribution and marketing opportunities 
(FAO, 2016a). The preservation techniques may: reduce the temperature (e.g. freezing and 
chilling); use heat treatment (e.g. smoking, boiling, frying and canning); decrease the water 
 13 
University of Ghana http://ugspace.ug.edu.gh
available for microbial attack (e.g. salting, drying and smoking) and alter the storage 
environment (packaging and refrigeration) (FAO, 2016a). Globally, cured fish (dried, smoked, 
salted and fermented) makes up about 12% of fish destined for human consumption, and this 
contribution is mainly from Africa and Asia (FAO, 2018). 
 
In Ghana, fish is preserved primarily by traditional (such as salting, drying, smoking and 
fermenting or a combination of these) and modern (freezing and canning) methods (Asiedu, 
Failer, & Beygens, 2018). Traditionally processed fish is widely patronized by many 
Ghanaians, as they are very affordable and possess good taste (Asiedu, Failer, & Beygens, 
2018). Smoked fish is however the most preferred, accounting for an estimated 70-80% of all 
domestic catch consumed (Nunoo, Asiedu, Kombat, & Samey, 2015; Asamoah, 2018). 
 
2.3.1 Fish smoking 
Fish smoking is a process of preserving fish by subjecting it to smoke from biomass 
combustion, and typically includes an integration of salting, drying, heating and smoking steps 
in a smoking chamber (Codex Alimentarius Commission, 2013). The smoke generated adds 
flavour, taste, colour (sensory qualities that are desired by consumers) and preservative agents 
to the fish (Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). Smoking can increase the 
shelf life of a product due to the synergistic effects of salting, drying and smoke generation 
(Asamoah, 2018). Salting lowers the water activity of the products, thus, inhibiting microbial 
growth; drying (using high temperatures) ensures a physical surface barrier is created, that 
inhibits the passage of microorganisms into the fish; and the smoke generated deposits 
compounds like aldehydes, carboxylic acid and phenols, which slows down growth of 
microorganisms and rancidity development (Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 
2014; Cieślik, Migdał, Topolska, Mickowska, & Cieślik, 2018). Raw materials of good quality 
 14 
University of Ghana http://ugspace.ug.edu.gh
are necessary in producing high quality smoked products, that can ensure a steady market 
demand and profit for processors (Asamoah, 2018). 
 
Fish products can be cold or hot smoked (or a combination of both), depending on the smoke 
delivery mechanism (Codex Alimentarius Commission, 2013). These methods are used both 
in developed and developing countries around the world, even though they are carried out 
mostly under controlled conditions in the developed than the developing countries (Cyprian, et 
al., 2015). Cold smoking occurs at temperatures generally below 30°C for about 20 hours (Doe, 
1998), which leaves the fish proteins mostly intact, while at the same time, reducing the water 
activity (Codex Alimentarius Commission, 2013). The smoke imparts the characteristic aroma 
and flavour, but the final product is not cooked and thereby requires further cooking before 
consumption, with the exception of fish such as salmon, trout and arctic char (Arason, Nguyen, 
Thorarinsdottir, & Thorkelsson, 2014). 
 
Hot smoking, according to Codex Alimentarius Commission (2013) exposes fish to an 
appropriate combination of temperature (usually 70-80 °C) and time that can cause complete 
protein coagulation in muscle. Hot smoking can kill parasites, impair non-sporulated bacterial 
pathogens and damage spores that pose health threats to humans (Codex Alimentarius 
Commission, 2013). The final products are completely cooked and are considered ready-to-eat 
(Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). A drying step could be added to 
further reduce the water activity to about 0.75 or less so products can be stored, transported 
and marketed without refrigeration (Akande, Ayinla, Adeyemi, Olusola, & Salaudeen, 2012). 
This method is preferred in most developing countries like Ghana where refrigeration and other 
logistics might be a problem (Akande, Ayinla, Adeyemi, Olusola, & Salaudeen, 2012). 
 
 15 
University of Ghana http://ugspace.ug.edu.gh
2.3.1.1 Changes in fish muscle during smoking  
Smoking can affect the weight, pH, protein structure, flavour and texture of fish muscle 
(Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). Weight loss results from dehydration 
and the leaching of lipids in fish muscle (Løje, 2007; Asamoah, 2018). This loss relates directly 
to the type of raw material used (whether fatty or lean fish), the final product characteristics, 
the dimensions of the fish and the processing parameters like times and temperature during 
smoking (Sigurgisladottir, Sigurdardottir, Torrissen, Vallet, & Hafsteinsson, 2000). This loss 
has been estimated between 10-25% and this directly affects the yield of the final product, 
important economic consideration that can affect the profitability of the business (Løje, 2007). 
Fatty fish have decreased dehydration and as such, a higher yield than lean fish (Cardinal, et 
al., 2004).  
  
During smoking, the pH of the fish muscle may decrease owing to an uptake of acids from the 
smoke, dehydration, smoking duration and protein constituents reacting with the phenols, 
polyphenols and carbonyl compounds within smoke (Hassan, 1988; Arason, Nguyen, 
Thorarinsdottir, & Thorkelsson, 2014; Lira, Silva, Figueirêdo, & Bragagnolo, 2014). The 
decrease in pH could result in muscle proteins having reduced net surface charge, thereby, 
causing proteins to partially denature, decreasing water-holding capacity and influencing the 
microbiota the muscle (Teixeira, et al., 2014; Bowker & Zhuang, 2015). This can further have 
negative effect on the texture of fish muscle, with muscle toughness increasing with decreasing 
pH (Huss, 1995). Espe, Nortvedt, Lie, & Hafsteinsson (2002) also reported a reciprocal relation 
between the smoking temperature and pH in fish muscle. 
 
Changes in protein structure can result during the salting step of the smoking process. A low 
salt concentration (< 1 M) can cause a swelling of the filament lattice and depolymerization of 
 16 
University of Ghana http://ugspace.ug.edu.gh
myosin (Cyprian, Oduor-Odote, & Arason, 2019), whereas a higher salt concentration (4 M) 
causes protein aggregation in the muscles (Nguyen, Thorarinsdottir, Gudmundsdottir, 
Thorkelsson, & Arason, 2011). These two phenomena can lead to variations in yield, water-
holding capacity, texture and microstructure of the fish muscle (Arason, Nguyen, 
Thorarinsdottir & Thorkelsson, 2014). Cyprian, Oduor-Odote, & Arason (2019) further report 
that these conformational changes can affect the nutritional quality (i.e. loss of thermolabile 
compounds like amino acids) in the final product. The extent of these changes is however 
dependent on the food composition and the type of smoking method used; with textural changes 
in hot smoked fish mainly due to protein denaturation, whereas that in liquid smoking is as a 
result of protease enzyme activities (Arason, Nguyen, Thorarinsdottir & Thorkelsson, 2014). 
 
The characteristic flavour, colour and odour of smoked fish results from the Maillard 
(browning) reaction between the carbonyl amino group, caramelization of fish flesh and lipid 
oxidation during the smoking process (Jaeger, Janositz, & Knorr, 2010; Leksono, Suprayitno, 
& Hardoko, 2014). The type of fuel used can affect these attributes, especially the colour. In 
Ghana for example, mangrove is mostly for smoking because it produces the characteristic 
golden or dark brown colour, that consumers desire (Obodai, Muhammad, Obodai, & Opoku, 
2009). Smoked flavour and odour is further developed from absorption of phenolic compounds 
(guaiacol and syringol) produced from pyrolysis of lignin (Jónsdóttir, Ólafsdóttir, Chanie, & 
Haugen, 2008). An increase in temperature and dryness during fish smoking can enhance these 
reactions (Doe, 1998). 
 
2.3.2 Fish smoking in Ghana 
In Africa countries like Ghana, Nigeria, Cote d’Ivoire, Togo, Benin, Sierra Leone, Liberia, 
Kenya, Uganda, Tanzania, fish smoking is predominantly practiced by women within and 
 17 
University of Ghana http://ugspace.ug.edu.gh
around coastal areas and along the fringes of inland water bodies (Adeyeye & Oyewole, 2016). 
Both fresh and frozen (local and imported) fish can be used in smoking, with the frozen fish 
used mostly in the lean fishing season (Entee, 2015c). The smoking process is usually in the 
following steps: 
• preparation of fish: thawing in the case of frozen fish; then washing and sorting of the 
fish. Seawater is mostly used in the washing step; 
• loading of fish onto smoking racks and allowing it to dry for at least 15 minutes 
(depending on the fish type and size) in open air environment;  
• cooking and smoking of fish in/on the smoking kiln. The cooking step requires heat, 
and once cooked, smoke is added, usually by covering the racks and reducing the heat. 
The duration of these steps relies largely on the type/size/quantity of species, the 
smoking kiln used, the experience of the processor and the consumer preference 
(Asamoah, 2018). 
 
Fish can be soft or hard smoked, based on the type and size of fish, product form, the consumer 
preference and the desired shelf life (Pemberton-Pigott, Robinson, Kwarteng, & Boateng, 
2016). Soft smoking usually takes between 1-3 hours and yield products with moisture contents 
of about 40-50% (Kwarteng, Nsiah, Samey, Boateng, & Aziebor, 2016). The mackerel and 
barracuda are typically soft - to medium-smoked (Abboah-Offei, 2016). Hard smoking is 
preceded by soft smoking and the smoked products are dried afterwards for between 10-18 
hours, sometimes days, depending on the weather (Kwarteng, Nsiah, Samey, Boateng, & 
Aziebor, 2016). The fish produced have moisture contents of between 10-15% and sometimes, 
lower than 10% (Pemberton-Pigott, Robinson, Kwarteng, & Boateng, 2016). The small 
pelagics such as sardines and anchovies are usually smoked using this method and remain 
readily available in many markets in Ghana (Alhassan, Boateng, & Ndaigo, 2012). 
 18 
University of Ghana http://ugspace.ug.edu.gh
2.3.2.1 Fish smoking kilns 
Fish are usually smoked using traditional kilns such as the metal drum and cylindrical or 
rectangular kilns fashioned from mud. These kilns however had a low throughput capacity and 
were fuel inefficient (Avega & Tibu, 2017). Again, the mud kilns easily wore out and the metal 
drums could become extremely hot, and processors suffered burns and complications from 
smoke exposure (Asamoah, 2018; Bomfeh, et al., 2019). These challenges led to the 
development of the Chorkor kiln in 1969 by the Food and Agriculture Organization of the 
United Nations (FAO) and the Food Research Institute of the Council of Scientific and 
Industrial Research (CSIR, Ghana) (UNDP/TCDC, 2001). This kiln can be constructed using 
either cement blocks or mud and has a better product throughput and fuel efficiency, lasted 
longer, with less labour input required (Bomfeh, 2016). These traditional kilns, especially the 
Chorkor, are also used in most parts of Africa where fishsmoking is predominant (Adeyeye & 
Oyewole, 2016). 
 
Other improved kilns have been developed over the years. Some of these are the Morrison, 
Ahotor and FAO Thiaroye Technology (FTT) (Entee, 2015c). Okyere-Nyarko, Aziebor, & 
Robinson (2015) reported an estimated 120,000 traditional and Chorkor kilns in operation 
Ghana. These kilns rely on biomass fuels (mainly firewood, charcoal and agricultural by-
products like sugarcane bagasse and coconut husks) as a source combustion. Entee (2015b) 
reported that a substantial amount of biomass fuel is used by these kilns and the temperature, 
humidity and smoke generation cannot be controlled (except in the case of the FTT that uses 
external smoke generators). Processors using mostly the Chorkor and other traditional kilns are 
thus exposed to smoke, and smoked products are generally of poor quality (Kleter, 2004; 
Antwi-Boasiako, 2017). There is therefore the need to search for or develop efficient and safe 
fish smoking kilns in Ghana. 
 19 
University of Ghana http://ugspace.ug.edu.gh
2.3.3 Chemical composition of smoke and smoked foods  
Smoke is made up of about 400 identified compounds, which include polycyclic aromatic 
hydrocarbons (PAHs), carbonyls, acids, alcohols, esters, aldehydes, phenols, furans, lactones, 
dioxins, and heavy metals (Mejlholm, Andersen, & Dalgaard, 2007; Codex Alimentarius 
Commission, 2009). The phenolic compounds, carbonyls, aldehydes, among others give the 
characteristic flavour of smoked products, as well as possessing antimicrobial and antioxidant 
properties, as discussed above (Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). The 
PAHs and heavy metals are however of human health importance, as such, their levels in 
smoked fish require special attention (Codex Alimentarius Commission, 2009). 
 
2.3.3.1 Polycyclic aromatic hydrocarbons (PAHs) contamination in fresh and smoked 
fish 
 PAHs are pervasive carbon-based ecological contaminants, that are semi-volatile and made up 
of fused benzene rings in linear, cluster or angular arrangements (Abdel-Shafy & Mansour, 
2016). They can dissolve in fish lipids and can be obtained from natural or anthropogenic 
sources (Stogiannidis & Laane, 2015). The natural sources are petrogenic (i.e. petroleum and 
coal) and biogenic (i.e. transformation of natural organic matter), whereas the anthorpogenic, 
the major environmental contributors are pyrogenic (i.e. incomplete biomass or fossil fuel 
combustion) (Bandowe, et al., 2014).  
 
There are about 660 parent PAH (Stogiannidis & Laane, 2015), and they have been classified 
based on the number of ring as: 
• low molecular weight (LMW): two to three rings, with molecular weights (MWs) from 
152 to 178 g/mol, e.g. acenaphthene, fluorene, acenaphthylene, anthracene and 
phenanthrene;  
 20 
University of Ghana http://ugspace.ug.edu.gh
• medium molecular weights (MMW): four rings, with MW of 202 g/mol, e.g. 
fluoranthene and pyrene; and 
• high molecular weight (HMW): five to seven rings with MWs from 208 to 278 g/mol, 
e.g. benzo(a)anthracene, benzo(b)fluoranthene, benzo(a)pyrene, chrysene (ATSDR, 
1995). 
 
Different organisations and regulatory bodies have compiled lists of prority PAHs 
(Stogiannidis & Laane, 2015), with the United States’ Environmental Protection Agency 
(USEPA) list of 16 PAHs widely used. These PAHs have been also categorised by the 
International Agency for Research on Cancer as follows:  
• Group 1 i.e. definite carcinogenic: benzo(a)pyrene;  
• Group 2A i.e. probably carcinogenic: dibenzo (a,- h)anthracene; 
• Group 2B i.e. possibly carcinogenic: benzo(k)fluoranthene, benzo(b)fluoranthene, 
chrysene, naphthalene indeno(1,2,3-cd)pyrene, and benzo(a)anthracene;  
• Group 3 i.e. not classified as to its carcinogenicity to humans: acenaphthalene, 
acenaphthene, fluorene, phenanthrene, anthracene, pyrene, fluoranthene and 
benzo(g,h,i)perylene (IARC, 2018).  
Groups 1 to 2B are mostly HMW PAHs, whereas Group 3 are the LMW PAHs.  
 
Humans can inhale PAHs, ingest them or get if from direct diffusion into the skin (Li, et al., 
2016). PAHs are carcinogenic, mutagenic and teratogenic and pose significant health risks to 
humans (Codex Alimentarius Commission, 2009; Xia, et al., 2010; Kim, Jahan, Kabir, & 
Brown, 2013; Essumang, Dodoo, & Adjei, 2012, 2014; Bandowe, et al., 2014; Ncube, et al., 
2017). With prolonged exposures to sufficiently high concentrations, these risks intensify 
especially in adults than in children (with a ratio of 3:1), as reported by Alomirah, et al. (2011). 
 21 
University of Ghana http://ugspace.ug.edu.gh
PAHs can occur in the marine environment and as such in fish. Bandowe, et al. (2014) and 
Nyarko, Botwe, & Klubi (2011) found PAHs in fresh Cynoglossus senegalensis, Pomadasys 
peroteti, Drapane africana, Sardinella maderensis and Galeoides decadactylus from Ghana’s 
coastal waters. According to European Commission (2011), however, these PAHs do not 
accumulate in fresh fish muscle, as they are rapidly oxidised and metabolied. No limits have 
therefore been set for fresh fish, however, the levels in fish can signify environmental 
contamination (Bouloubassi, et al., 2006).  
 
During smoking, PAHs adsorb to or condense on the surface of the products, with the amounts 
dependent on the:  
• type of fuel (i.e. biomass, fossil fuel, liquid or solid waste and others); 
• direct or indirect smoking or drying method; 
• process of smoke generation, either from smoke generator, or from using smoke 
condensates; 
• the distance and position of the fish relative to the heat source;  
• fish lipid content and how it changes during smoking;  
• time and temperature of smoking and direct drying;  
• hygiene and maintenance of equipment;  
• the density of the smoke in the smoking chamber (Codex Alimentarius Commission, 
2009) 
 
The importance of PAHs in muscle meat of smoked fish products has been recognised by the 
European Commission, which has prescribed maximim limits (MLs). These MLs pertain to 
benzo(a)pyrene and the sum of four carcinogenic PAHs (benzo(a)pyrene, chrysene, 
benzo(a)anthracene and benzo(b)flouranthene) have been set at 2.0 and 12.0 μg/kg respectively 
 22 
University of Ghana http://ugspace.ug.edu.gh
in muscle meat of smoked fish (European Commission, 2011). There have been various studies 
relating to PAHs in smoked fish in Ghana, with most of the levels higher than that of the EU 
MLs (Table 2.1). The results from studies indicated that the type of smoking kiln, smoking 
method (hard/soft smoked), fuel and fish species could affect the levels of PAHs in smoked 
fish. 
 
The health risks resulting from consumption of fish contaminated with PAH can be assessed 
using the benzo[a]pyrene equivalent (B(A)Peq), incremental cancer risk and target hazard 
quotients (THQ) (Bandowe, et al., 2014; Xia, et al., 2010; Li, et al., 2016). The B(A)Peq 
estimates the overall toxicity of the PAH compounds in food based on their individual 
concentrations and potency equivalency factors (PEFs). The cancer risk (CR) is then calculated 
using the B(A)Peq, average fish ingestion rates, life expectancy and body weight of the target 
population, and carcinogenic potency of benzo[a]pyrene (Bandowe, et al., 2014; Xia, et al., 
2010; Li, et al., 2016). THQ is also calculated as the ratio of estimated dose to oral reference 
dose (RfDo) and measures the non-carcinogenic risk from PAH exposure. A CR of 1 x 10-5 is 
considered the carcinogenesis threshold, with those ³ 1 x 10−4 deemed serious and therefore 
requiring attention whereas; THQ < 1 signifies negligible health risk (USEPA, 2004; Xia, et 
al., 2010; Essumang, Dodoo, & Adjei, 2013; Bandowe, et al., 2014). 
 
 
 
 
 
 
 
 23 
University of Ghana http://ugspace.ug.edu.gh
Table 2.1: PAH concentrations in smoked fish from different kilns in Ghana 
Fish type Kiln type Levels (μg/kg) Reference 
Benzo(a)pyrene  PAH4 
Mackerel Chorkor 41.3  452.3 Essumang, Dodoo, & Adjei, 
 1.3** 168.4 (2013) 
 15.2*** 225.8  
Cigar minnows 0.5 50.7 
 3.0** 57.5** 
 2.9*** 6.8*** 
Tuna 5.9 22.9 
 0.6** 24.7** 
 0.6*** 35.3*** 
Sardinella 16.6 40.7 
1.4** 17.5** 
1.6*** 43.0*** 
Sardinella Chorkor 22 84 IRI-CSIR, GSA, & Kwarteng 
Morrison 30 110 (2016) 
Ahotor 5.9 53.1 
Barracuda (soft- *FAO FTT 0.6-1.8 3.6-7.6  Bomfeh, et al., 2019 
hard smoked) Chorkor 50.3-61.1 270.3-
Metal drum 37.4- 69.8 360.3 
167.5-
327.1  
Sardinella (soft- *FAO FTT 0.2-0.3 1.5-2.2 
hard smoked) FAO FTT 1.9 & 7.7 37.0 & 28.9 
Chorkor 26.4-60.3 166.9-
*Chorkor 10.2 394.5 
Metal drum 11.1- 25.6 39.4 
58.1-135.7 
Tuna Chorkor 26.7 195.5 Nunoo, Tornyeviadzi, 
Barracuda  71.1 865.7 Asamoah, & Addo (2019) 
Grouper  50.3 365.8 
    
Tuna Abuesi gas fish 1.4 8.2 
Barracuda smoker (AGFS) ND 4.7 
Grouper ND 0.3 
* = charcoal; ** = sugarcane bagasse; *** = mangrove; ND = not detected 
 
 
 24 
University of Ghana http://ugspace.ug.edu.gh
2.3.3.2 Heavy metal contamination in fresh and smoked fish 
Heavy metals are naturally occurring elements of the earth’s crust or can be introduced into the 
environment via anthropogenic means, such as biomass and fossil fuel combustion, urban 
runoffs, agricultural activities, among others (Bandowe, et al., 2014). Some potentially toxic 
heavy metals detected in fresh and smoked fish are Lead (Pb), Cadmium (Cd), Mercury (Hg), 
Chromium (Cr), Nickel (Ni) and Arsenic (As) (Doe, 1998; Daniel, Ugwueze, & Igbegu, 2013; 
Bandowe, et al., 2014). IARC (2018) has classifed Cd, Pb and Hg as Group 1, 2B and 3 
carcinogens, respectively. 
 
Heavy metals cannot be metabolised by the human body and as such bio-accumulate in the 
liver, muscles and bones (Ekpo, Asia, Amayo, & Jegede, 2007). Bioaccumulation of these 
metals in humans can result in difficulties in swallowing, muscle cramps, hypoxia, diarrhea, and in 
the worst event, death (Abboah-Offei, 2016). The EU has thus set the limits of Cd in muscle meat 
of mackerel at 0.10 mg/kg and other fish at 0.05 mg/kg; that for Pb and Hg are 0.30 and 0.50 
mg/kg respectively (European Commission, 2015). Again, the carcinogenic risks are estimated 
using target hazard quotients (THQ) as described above (Bandowe, et al., 2014). According to 
Abboah-Offei (2016), As and Pb were not detected market samples of smoked tilapia, mudfish, 
tuna and mackerel, whereas Hg and Ni were detected at levels above the EU MLs. Bomfeh (2016) 
also reported levels above the EU MLs for Hg and Cd in market samples of smoked sardines. These 
results from market samples could howver be as a result of contributions from other sources,in 
addition to those from fish smoking, as suggested by Abboah-Offei (2016). Bandowe, et al. (2014) 
however reported THQs < 1 (negligible risks associated with consumption) for three fresh fish 
species found in Ghanaian waters. 
 
 
 25 
University of Ghana http://ugspace.ug.edu.gh
2.3.4 Quality degradation of smoked fish 
During the storage life of smoked products, the problems of microbial growth and lipid 
oxidation arise. This can happen irrespective of the storage conditions and packaging used, 
although the rates may differ (Cyprian, et al., 2015). Akande & Diei-Ouadi (2010) have 
estimated quality losses in the smoked fish industry in Ghana as 37.5% of the total production.  
 
2.3.4.1 Microbial spoilage 
Healthy, live, fish have considerably sterile muscles, but the surfaces of the skin, gills and 
gastrointestinal tracts could harbour microorganisms (Gram, 2009). Upon death, the fish’s 
body resistance is impaired, and this causes opportunistic microorganisms to flourish, with the 
enzymes they secrete further degrading the quality of the fish (Bataringaya, 2007). Processing 
methods like smoking, drying and salting can inhibit but not completely eliminate the effects 
of some of these microbes (Asamoah, 2018).  
 
Smoked products are stored at ambient temperatures and sold unpackaged in open air markets 
(in most developing countries like Ghana), making them susceptible to microbial infestation, 
as well contamination from dust and insects (Kleter, 2004). The shelf life of soft and hard 
smoked fish has been estimated at 1-3 days and 6-9 months (with periodic re-smoking) 
respectively (Pemberton-Pigott, Robinson, Kwarteng, & Boateng, 2016).  
 
The actions of microorganisms can result in undesirable effects on food (Figure 2.2). Bacteria 
such as Aeromonas, Alteromonas, most Enterobacteriaceae, Shewanella, Vibrio and 
Photobacterium are capable of decomposing proteins and other nitrogenous compounds 
(trimethyl amine oxide, TMAO) present in fish (Ryder, Karunasagar, & Ababouch, 2014). This 
breakdown produces trimethylamine (TMA), dimethylamine (DMA) and ammonia, which give 
 26 
University of Ghana http://ugspace.ug.edu.gh
off the typical off-flavours and odours of spoiled fish (Bataringaya, 2007). The levels of TMA, 
DMA and ammonia can be measured using the total volatile base (TVB) as a qualtity index 
(Bataringaya, 2007). The temperature, time, species, hygienic condition, microbial action, 
among other factors, can influence the development of TVB in fish (Oehlenschläger, 2014). 
The nutritional quality of fish can be damaged by biogenic amine (BA)-producing bacteria 
such as Staphylococcus spp., Morganella morganii, Klebsiella pneumoniae, Proteus vulgaris, 
Proteus mirabilis, Enterobacter aerogenes, Clostridium spp., Vibrio alginolyticus, 
Pseudomonas putida, Aeromonas spp., among others (FAO/WHO, 2013). BAs are low MW 
organic bases, formed in fish (usually upon death) by either the removal of the carboxyl group 
of corresponding amino acids by decarboxylase enzyme or the by the transamination of 
aldehydes and ketones (Zhai, et al., 2012; Huang, Wang, Wang, Hu, & Chui, 2019). The BAs 
of significance are histamine, cadaverine, tryptamine, tyramine and putrescine, which are 
derived from the amino acids, histidine, lysine, tryptophan, tyrosine and ornithine, respectively. 
 
Histamine is produced by bacteria usually present in the marine environment; and inhabit the 
gills, skin, and gut of live fish, without causing harm to the fish (Visciano, Schirone, Tofalo, 
& Suzzi, 2012). However, once the fish dies the bacteria can grow rapidly, particularly 
following time/temperature abuse, especially in most developing countries where the cold 
chain after fish harvest is often poor (FAO/WHO, 2013; Asamoah, 2018). Processing methods 
like freezing, canning, smoking, among others, can deactivate the bacteria responsible for 
histamine formation, but the enzyme, histidine decarboxylase, already present in fish cannot 
be eliminated by these methods (FDA, 2011). Quantifying histamine levels in fish therefore 
provides a quality index that that can be used for sanitary surveillance (Zhai, et al., 2012; 
Gopalapura, Gowda, Narayan, & Gopal, 2016). Species belonging to the family Scombridae 
(mostly tuna and mackerel) are most susceptible, with other non-scombroid fish species such 
 27 
University of Ghana http://ugspace.ug.edu.gh
as those from the Clupeidae, Engraulidae, Coryfenidae, Pomatomidae, Scombresosidae 
families also at risk (FAO/WHO, 2013). Histamine poisoning can occur from within minutes 
to an hour, following consumption and can last from 12 hours some days (FDA, 2011). The 
main symptoms may include peppery or metallic taste, numbness of the mouth, headache, 
vertigo, palpitations, drop in blood pressure, swallowing difficulties and dehydration 
(Visciano, Schirone, Tofalo, & Suzzi, 2012).  
 
Certain pathogenic bacteria have been implicated in seafood-related illness in humans. These 
microbes can cause food-borne infections (from directly consuming live pathogens in food) 
and food-borne intoxications (resulting from ingestion of toxins produced by bacteria in food), 
depending on levels of pathogens and food matrix (Morris & Potter, 2013). Examples of 
pathogens causing infections are Listeria monocytogenes, Salmonella sp., Escherichia coli, 
Vibrio vulnificus and Shigella sp. whiles those that cause intoxication include Staphylococcus 
aureus, Clostridium botulinum, C. perfringens (Ryder, Karunasagar, & Ababouch, 2014). S. 
aureus (the second most commmon cause of food poisoning) and C. perfringens can produce 
heat- reistant toxins that can cause nausea, vomiting, abdominal cramps and, sometimes, 
diarrhoea (FDA, 2011; Fletcher, Boonwaat, Moore, Chavada, & Conaty, 2015).  
 
Fungal (yeast and mould) contamination in fishery products is also a matter of concern. Species 
from the genera Aspergillus, Penicillium and Achlya have been isolated from smoked, dried 
and salted fish (da Silva, et al., 2008; Adeyeye, 2016; Osibona, Ogunyebi, & Samuel, 2018). 
These organisms are thermo-tolerant and can produce mycotoxins (e.g. aflatoxins by 
Aspergillus), which are low weight metabolites, that have public health implications for 
humans (Gram, 2009; Marc, et al., 2014; Adeyeye, 2016). Aflatoxins are potent carcinogens 
that have the potential to cause hepatoma, acute hepatitis, anemia and reduced white blood 
 28 
University of Ghana http://ugspace.ug.edu.gh
cells, that can impair the immune system (Samaha, Amer, Abd-Elshahid, & El-bialy, 2015). 
Penicillium species can also produce mycotoxins (i.e. ochratoxin A, citrinin, patulin and 
penicillanic acid), that have cytotoxic effects on mammalian cell lines causing reduced cell 
viability (Egbuta, Mwanza, & Babalolo, 2017). These risks are however dependent on the 
frequency and amount (concentration) of contaminated fish consumed and the duration of 
exposure. 
 
In most developing countries like Ghana, where smoked fish is sold unpacked in open-air 
markets, the threat of microbial contamination is very high. The unsanitary conditions of most 
markets, poor handling of the products by processors, traders and sometimes consumers further 
increase this risk. Bīlgīn, Ünlüsayin, Izci, & Günlü (2008) therefore suggest that smoked 
products should be stored at refrigerated temperatures (i.e. at or below 3 oC) to reduce these 
hazards. 
 
Undesirable effects of microbial activity in foods
Sensory quality Nutritional quality Safety issues
Damage to:
Development of: Foodborne infection and toxicity 
- proteins (TVBN)
- rancid flavour - pathogenic microbes 
- amino acids 
- acrid flavour (e.g. Listeria monocytogenes)
(TMA, biogenic 
- slime - toxin formation 
amines formation)
- gas (neurotoxin, botulism 
- discolouration and mycotoxin e.g. 
- undesirable testure aflatoxin)
(polymer degradation)
 
 
Figure 2.2: The effects of microbial activities on food quality (modified from Cyprian, 
2015) 
 29 
University of Ghana http://ugspace.ug.edu.gh
2.3.4.2 Chemical spoilage 
One important constituent of muscle of teleosts is lipid, which are grouped as phospholipids 
and triglycerides (Huss, 1995). The phospholipids make up the structural lipids as they are 
integral in forming membranes in the cells; whereas the triglycerides are termed depot fat as 
they store energy, usually within special fat cells encased in phospholipid membranes (Huss, 
1995). Fish lipids contain approximately 40% of long-chain fatty acids (14-22 carbon atoms), 
that are significanlty higher than in mammalian lipids (Huss, 1995). Fish lipids are highly 
unsaturated and as such, can easily be oxidised or hydrolysed, throughout the processing, 
distribution, storage and marketing of fish (Cyprian, 2015). . In terms of human nutrition, these 
polyunsaturated fatty acids (PUFAs), such as linoleic and linolenic acid are very essential, as 
they cannot be synthesised by humans (Huss, 1995).  
Oxidation can occur through non-enzymatic (autoxidation), photogenic (photooxidaton) and/or 
enzymatic (liposygenase) reactions (Ghaly, Dave, Budge, & Brooks, 2010; Cyprian, 2015). 
Autoxidation of PUFA (i.e the primary cause of lipid oxidaion in post mortem fish), invloves 
free radical reactions in three stages (Figure 2.3): 
• initiation: where unsaturated lipids lose a hydrogen atom to form a lipid (alkyl) radical 
(L.), in the presence of an intiator; 
• propagation: the alkyl radical reacts (L.) with molecular oxygen to from peroxyl radical 
(LOO.), which then abstracts hydrogen from another molecule of unsaturated lipid to 
form hydroxyperoxides (LOOH) and a new lipid radical (L). The hydroxyperoxides 
undergoes further reaction in the presence of metal catalysts to produce secondary 
intermediates such as aldahydes, ketones, alcohol, small carboxylic acids and alkanes 
that cause off-odour, off-flavour and yellow colour development during lipid oxidation; 
and 
 30 
University of Ghana http://ugspace.ug.edu.gh
• termination: the peroxyl radicals react with each other to form non-radical products or 
the lipid radical reacts with an antioxidant molecule (AH) to produce inert antioxidant 
radicals (Huss, 1995; Kolakowska, 2003). 
 
Figure 2.3: Autoxidation of polyunsaturated lipid (Adopted from Huss, 1995) 
 
Photooxidation, on the other hand, results from a direct reaction of singlet oxygen addition to 
unsaturated lipids, that forms hydroperoxides but not radicals (Kolakowska, 2003). This 
reaction can be caused by UV light, electric shining light or irradiation (Cyprian, 2015). 
Lipoxygenases (enzymatic-initiate lipid oxidation) leads to the formation of hydroperoxides in 
a defined position of the fatty acid chain, contrary to free-radical reactions (Kołakowska & 
Bartosz, 2014). At the initial stage of lipid oxidation, a sudden rise in peroxide value (PV) is 
observed (Okpala, 2016). The extent of oxidation, however, depends on the fish species, fatty 
acid composition, content and action of anti- and pro-oxidants, temperature, irradiation, oxygen 
pressure, surface area exposed to oxygen, storage temperature and water activity (Kolakowska, 
2003).  
 
Lipid oxidation reduces the sensory and nutritive quality of fish, poses health hazards, as well 
as processing challenges, as summarized in Figure 2.4. Kolakowska (2003) reported that 
 31 
University of Ghana http://ugspace.ug.edu.gh
consumption of foods rich in oxidized lipids increased thiobarbituric acid reactive substances 
(TBARS) in plasma and tissue of humans and animals. Aldehydes (especially 
malondialdehyde) can also induce intracellular oxidative stress and they are also genotoxic and 
react with DNA to form highly-mutagenic adducts in human cells (Reitznerová, et al., 2017). 
 
Undesirable effects of lipid oxidation
Sensory quality Nutritional quality Toxic effects
Technological 
suitability
Damage of: Generation of:
- hydroperoxides
Development Losses in: - proteins Decrease in:
of: - PUFA - essential amino - aldehydes - emulsifying 
- off odour - vitamins A, D, E acids - epoxides activity of 
- off flavour - carotenoids - amino acids - dimers proteins
- discolouration - phytosterols (oxidation) - oxycholesterols - protein 
- undesirable - other - S-S bonding - trans fatty acids solubility
texture antioxidants - lipid-protein - Mailard type 
interaction products
 
Figure 2.4: The effects of lipid oxidation on food quality (modified from Kolakowska, 
2003) 
 
Fish lipids can also be hydrolysed as a result of enzymatic activities, mainly triacyl lipase, 
phospholipase A2 and phospholipase B (Shah, Tokunaga, Kurihara, & Takahashi, 2009). 
Hydrolysis leads to the release of free fatty acids (FFAs) (Figure 2.5), which can accumulate 
and thus reduce the quality of the fish muscle (Kaneniwa, Yokoyama, Murata, & Kuwahara, 
2004). The rate of hydrolysis depends on the fish species, temperature and also processing method, 
and it mostly proceeds faster in un-gutted than gutted fish perhaps due to the involvement of 
digestive enzymes (Huss, 1995).  
 
 32 
University of Ghana http://ugspace.ug.edu.gh
 
 
Figure 2.5: Hydrolytic reactions of triglycerides and phospholipids: PL1 & 
PL2 phospholipases; TL, triglyceride lipase (Adopted from Huss, 1995)	
 
2.3.5 Strategies for shelf life extension of smoked fish 
Fish is very sensitive to deterioration during processing, packaging, storage and marketing 
resulting from microbial growth. To ensure safety of the product, proper packaging and storage 
at temperatures that inhibit microbial proliferation are essential. Again, the products can be 
gamma irradiated to render them safe and shelf stable (Arvanitoyannis & Tserkezou, 2014). 
These strategies have been discussed below.  
 
2.3.5.1 Packaging and storage of smoked products 
Smoked fish products can be contaminated during packaging, storage and marketing (IAEA, 
2001), especially where these products are sold on the open market, as happens in Ghana 
(Abboah-Offei, 2016). The extent of this contamination depends on the degree of smoking and 
drying, packaging method and temperature at which they are stored (Cyprian, 2015). It has 
therefore been suggested that smoked fish products must be cooled promptly at ambient or chill 
temperatures (immediately after smoking), and packaged ensure good quality throughout 
marketing (Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). In Ghana, smoked fish is 
 33 
University of Ghana http://ugspace.ug.edu.gh
usually packaged using paper materials and the products are stored at ambient temperatures, 
making the products susceptible to insect attacks and other forms of contamination (Kleter, 
2004). Asamoah (2018) found that the use of plastic packaging materials and subsequent 
storage at refrigerated temperature could extend the shelf life of soft-smoked mackerel from 5 
days to more than 35 days.  
 
 
2.3.5.2 Irradiation 
Food irradiation involves using ionising radiation, from sources such as x-rays, electron beams 
(e-beams) and high-energy gamma rays (cobalt-60, 60Co and cesium-137, 137Cs) to preserve 
foods (Codex Alimentarius Commission, 2003). Cobalt-60 is widely used as it offers high 
penetrability, permanent radiating source and high efficiency, though source replenishment is 
needed and low throughput is achieved (Stewart, 2004; Farkas, 2006; Arvanitoyannis, 2010). 
Ionising radiation is absorbed as it passes through food (in the form of invisible waves) and 
can reduce the incidence of food-borne illness and make food more shelf stable (WHO, 2000; 
Fan & Sommers, 2013). Irradiation doses are classified as low (< 1 kGy), medium (1 – 10 kGy) 
and high (> 10 kGy) (Arvanitoyannis, 2010). Low-dose irradiation have been shown to aid in 
insect/parasite disinfection, inhibit sprouting in crops such as potatoes and yam and delay 
ripening of fresh fruits and vegetables; medium-doses can inactivate spoilage and pathogenic 
microorganisms in foods and extends the shelf life and high-doses are used for industrial 
sterilization and decontamination of some food additives (WHO, 2000; Loaharanu, 2003; 
Mostafavi, Fathollahi, Motamedi, & Mirmajlessi, 2010). 
 
To decrease/inactivate the microbiota in foods, a photon of energy or an electron collides 
directly with a critical element in the cell (most often DNA and RNA) or ionises an adjacent 
molecule (mostly water), which then reacts with the genetic material (Molins, 2001; Farkas, 
 34 
University of Ghana http://ugspace.ug.edu.gh
2006; Mostafavi, Fathollahi, Motamedi, & Mirmajlessi, 2010). These reactions therefore 
prevent the multiplication of microorganisms and randomly terminates most cell functions, 
which improves the storability of the product (Farkas & Mohácsi-Farkas, 2011). The 
differences in the chemical and physical structure of microorganisms, their population and 
ability to recover from radiation injury will determine the success of irradiation (Farkas, 2006). 
Other external factors, such as moisture content, temperature during irradiation, presence or 
absence of oxygen, among others, also can affect the irradiation sensitivity of mostly vegetative 
cells (Farkas, 2006). 
 
Irradiation can also accelerate lipid oxidation, especially in foods with high water content, since 
it generates hydroxyl radicals (free radicals) that are strong initiators of lipid oxidation, (Farkas, 
2006; Fan & Sommers, 2013). Irradiation-induced oxidative changes are dose-dependent and 
can be further hastened under aerobic storage conditions (Fan & Sommers, 2013). This can 
affect the proteins, carbohydrates and lipids in foods; however, these are usually minor changes 
that are comparable to those produced from other processing technologies like pasteurisation 
(Ortega-Rivas, 2012). Before irradition, the product must be properly packaged and the use of 
polyethylene and polypropylene bags have been suggested by (Duah, Emi-Reynolds, Kumah, 
& Larbi, 2018).  
 
Fresh fish and shellfish spoil easily, as such, irradiation can be employed to improve the 
microbiological quality. Various studies have been evaluated the effect of ionizing radiation 
on the physicochemical, microbiological and sensory qualities of different seafood. Su, Duan, 
& Morrissey (2004) reported a log reduction of 2.5 log10 cfu/g of Listeria monocytogenes 
counts in cold-smoked salmon e-beam-irradiated at 1 kGy, with a dose of 2 kGy completely 
eliminating the pathogen. Medina, et al. (2009) also stated that cold-smoked salmon stored at 
 35 
University of Ghana http://ugspace.ug.edu.gh
5°C and 5°C + 8°C (temperature abuse) would require e-beam-irradiation doses of 1.5 and 3 
kGy respectively to attain the Food Safety Objective (FSO) of 2 log10 cfu/g for L. 
monocytogenes for a 35-day shelf-life, without affecting the visual aspect (colour) negatively. 
Irradiating fish at doses of 2-7 kGy and storage at refrigerated temperature reduced pathogens 
such as Salmonella, Listeria, and Vibrio spp., and many specific spoilage organisms such as 
Pseudomonaceae and Enterobacteriaceae (Arvanitoyannis, 2010). Al-Kahtani, et al. (1998) 
found no changes in amino acid in Spanish mackerel irradiated at 1.5 to 10 kGy and stored 
under refrigeration.  
 
The total volatile base nitrogen levels in irradiated sea bass, tilapia and mackerel decreased 
during refrigerated storage, while the moisture, protein and lipid content remained unchanged 
for 20 days of storage (Cozzo-Siqueira, Oetterer, & Gallo, 2003; Ozden, Inugur, & Erkan, 
2007). Aerobically packaged ready-to-eat smoked sardines, irradiated at 7 and 11 kGy in 
Ghana, were studied and the results indicated that microbial activity (i.e. aerobic counts, mould 
and coliforms) was controlled over 12 weeks of ambient storage, though the colour, flavour 
and texture were unacceptable (Nketsia-Tabiri, Adu-Gyamfi, Montford, Gbedemah, & Sefa-
Dedeh, 2003). Akuamoa, Odamtten, & Kortei (2018) irradiated dried and smoked anchovies 
at doses ranging from 2.5 to 10 kGy and reported shelf live increase of four weeks in irradiated 
samples, relative to the control group. Erkan & Ozden (2007) and Fan & Sommers (2013) 
however stated that the chemical properties of irradiated fish may change, depending on the 
species used, the initial quality at the time of capture and good management practices relating 
to on-board handling, icing, packaging and storage. 
 
Studies have been conducted on the potential of gamma irradiation in improving the safety and 
quality of vegetables like garden eggs, mushrooms, poultry, fermented maize and cassava 
 36 
University of Ghana http://ugspace.ug.edu.gh
products and seafood in Ghana like sardines, anchovies and shrimps (Nketsia-Tabiri, Adu-
Gyamfi, Montford, Gbedemah, & Sefa-Dedeh, 2003; Adu-Gyamfi & Appiah, 2012; Adu-
Gyamfi, Torgby-Tetteh, & Appiah, 2012; Adu-Gyamfi, Riverson, Afful, & Appiah, 2014; 
Akuamoa, Odamtten, & Kortei, 2018; Duah, Emi-Reynolds, Kumah, & Larbi, 2018). Ghana 
has the requisite capacity at the Radiation Technology Centre of Ghana Atomic Energy 
Commission, therefore, to use this technology to ensure safe and shelf-stable fish products, not 
only for the domestic market, but also for export (Adu-Gyamfi, Banson, & Nketsia-Tabiri, 
2011). 
 
The volumes of irradiated food products are increasing (with more than 60 countries regulating 
the practice), and so the future of irradiated foods looks promising (Badr, 2012; Ortega-Rivas, 
2012). The FAO, World Health Organization (WHO) and International Atomic Energy Agency 
(IAEA) have recommended the use irradiation technology in food processing (Banson, 2015). 
Although there have been more than hundred years of research on irradiation, consumer 
perceptions have hindered its development and commmercialisation (Maherani, et al., 2016). 
The reasons given included antinuclear activism, industry hesitation, time-consuming approval 
processes, and insufficient consumer education (Ehlermann, 2016). Consumer perceptions and 
decision to buy irradiated foods were however positively altered after they were made aware 
and understood the benefits of food irradiation (Nayga, Aiew, & Nichols, 2005; Mostafavi, 
Fathollahi, Motamedi, & Mirmajlessi, 2010) .  
 
 
 37 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER THREE 
 
3.0 COMPARISON OF THE PERFORMANCE AND EFFICIENCY OF 
THE IMPROVED AND TRADITIONAL SMOKING KILNS 
 
3.1 Introduction 
Fish smoking and drying are ancient, traditional fish processing methods that are mostly 
employed to preserve fish in most developing countries of the world (FAO, 2018). Smoked 
and dried fish products have longer shelf lives and serve as the main source of fish to rural 
areas that are far from fishing sites (Ndiaye, Komivi, & Diei-Ouadi, 2015). In Ghana, both 
fresh and frozen (mostly imported) fish are smoked (Entee, 2015a). Again, the majority of fish 
consumed is in the smoked form (Nunoo, Asiedu, Kombat, & Samey, 2015).  
 
Fish smoking is usually undertaken using traditional fish smoking kilns (like the metal drum 
and round mud kiln) and improved ones like the Chorkor, Banda, Altona, among others 
(Ndiaye, Komivi, & Diei-Ouadi, 2015; Bomfeh, et al., 2019). In Ghana, the most popular kiln 
is the Chorkor, which was developed in 1969 through the joint collaborative efforts of the Food 
and Agricultal Organisatoin (FAO) and the Food Research Institute, Ghana (Asamoah, 2018). 
This kiln has the advantage of being relatively more energy efficient, has greater throughput 
capacity, poses lower risks to processors in terms of smoke exposure and produce better quality 
products, compared to the other traditional kilns (Ndiaye, Komivi, & Diei-Ouadi, 2015; Entee, 
2015b). 
 
 38 
University of Ghana http://ugspace.ug.edu.gh
The kilns rely on solid/biomass (firewood mainly) combustion to cook, smoke and dry the fish, 
and this imparts the characteristic colour, odour and taste that is favoured by consumers 
(Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014; Ndiaye, Komivi, & Diei-Ouadi, 2015; 
Bomfeh, et al., 2019). Biomass resources are readily accessible and provide a major source of 
renewable energy in most households in developing countries (Global Alliance for Clean 
Cookstoves, 2016). The combustion of these fuels are usually less efficient and release a variety 
of incomplete combustion pollutants, like carbon monoxide (CO), particulate matter (PM), 
sulphur dioxide (SO2) and oxides of nitrogen (NOx) among others, which affect human health 
and local/regional climate change significantly (Roy & Corscadden, 2012; Shen & Xue, 2014; 
Global Alliance for Clean Cookstoves, 2016). Several studies in Ghana have reported the 
potential public health and climate implications of using the Chorkor and other traditional fish 
and improved smoking kilns (Entee, 2015a, b; Flintwood-Brace, 2016; Antwi-Boasiako, 2017; 
Obeng, 2018; Bomfeh, et al., 2019). This calls for the development and testing of new 
improved and efficient fish smoking kilns.  
 
Two improved kilns were therefore evaluated in this study and compared with the Chorkor 
smoker. They are both existing technologies: the Abuesi Gas Fish Smoker (AGFS) and the 
Cabin smoker, developed by GRATIS and UNU-FTP and Matis, respectively. The possibility 
of adopting these kilns will be dependent on an assessment of their performance and costs, 
relative to the Chorkor. The information obtained will be very useful for consumers and 
businesses that will be interested in these technologies. 
 
The aim of the study was therefore to assess the performance and efficiency of the Cabin, 
Chorkor and AGFS smoking kilns. 
 39 
University of Ghana http://ugspace.ug.edu.gh
3.2 Materials and methods 
The efficiencies of the Cabin, Chorkor and AGFS kilns with respect to fuel usage, quantities 
of fish smoked, among others, were tested using the controlled cooking test (CCT) protocol. 
The emissions from the Chorkor and Cabin were also tested. 
 
3.2.1 Controlled cooking test (CCT) protocol 
The controlled cooking test (CCT) procedure, as described in (Bailis, 2004; Entee, 2015b; IRI-
CSIR, GSA, Kwarteng, 2016) was employed to assess the efficency of the three smoking kilns. 
The CCT was developed by the Global Alliance for Clean Cookstoves (Bott, 2014) to compare 
the performance of improved and traditional stoves, with the aim of replacing the traditional 
models. 
 
3.2.1.1 Test location 
The smoking experiments were carried out at the Fish Processing Facility, which belongs to 
the Abuesi Fish Processing Association (AFPA), at Abuesi in the Western Region of Ghana 
from 12th to 14th July 2018. Abuesi is about 20 km from Takoradi, the Western Regional capital. 
Abuesi has a population of about 10,000, with about 52% being females. The main economic 
activities are fishing and trading of fish, particularly, smoked fish (Ghana Statistical Service, 
2014). Fish smoking is mainly carried out on small scale basis in traditional smokehouses, 
which could be enclosed or open (Flintwood–Brace, 2016). This site was selected because of 
the availability of the Abuesi Gas Fish Smoker and a prototype of the Cabin kiln. Again, Abuesi 
was selected because of the high usage of the traditional fish smoking kilns and the reported 
respiratory risks observed in fish processors (Flintwood–Brace, 2016). 
 
 40 
University of Ghana http://ugspace.ug.edu.gh
3.2.1.2 Description of fish smoking kilns tested 
The Chorkor, Cabin and AGFS kilns, representing uncontrolled and semi-controlled and 
controlled smoking technologies respectively, were used in the smoking experiment. The 
Chorkor and Cabin kilns use firewood to smoke the fish, whiles the AGFS used LPG gas.  
 
Chorkor Kiln  
The Chorkor kiln (Plate 3.1) is built with blocks. For the purpose of this CCT, only one side of 
the kiln was used and described accordingly. The kiln dimensions are 120 x 150 x 59 cm (L x 
B x H) with wall width of 13 cm. The kiln can hold between eight (8) to twelve (12) wooden 
trays, with an estimated maximum capacity of 120 kg. The tray measurements are 109 x 98 x 
5 cm (L x B x H), with a wire mesh area of 9,476 cm2 (0.95 m2). There is one (1) front loading 
firepot, with dimensions of 94 x 78 x 80 cm (L x B x H). The trays are loaded atop the fireplace 
with no locking mechanism. The ends of the trays are fashioned into handles for easy handling. 
 
 
Plate 3.1: Chorkor smoking kiln 
 
Cabin kiln 
The Cabin (Plate 3.2) is built with fired bricks. It has an overall dimension of 140 x 138 x 187 
cm (L x B x H), with a brick wall width of 12.5 cm. The kiln has seven (7) wooden trays that 
 41 
University of Ghana http://ugspace.ug.edu.gh
are arranged on wooden rails inside the smoking chamber, with an estimated maximum 
capacity of 90 kg of fish. Each tray has dimensions of 112 x 104 x 7 cm (L x B x H), with a 
wire mesh surface area of 10,388 cm2 (1.04 m2). The kiln has a front loading firepot with 
dimension 77 x 65 x 65 cm (L x B x H). This firepot is covered by a curved metal plate of 
dimensions 100 x 90 cm (L x H). There is a metal door of dimension 73 x 70 cm (L x H) that 
seals the firepot. The distance between the steel plate and the first tray is 38 cm. There are two 
ventilation holes on either side of the kiln with dimensions, 26 x 12.5 x 10 cm (L x B x H). The 
kiln has an opening at the top that serves as a conduit for smoke release. 
 
 
Plate 3.2: Cabin kiln  
 
Abuesi gas fish smoker (AGFS) 
The AGFS (Plate 3.3) is an industrial, double-chamber fish smoking oven built from stainless 
steel. It is a double-chamber oven but for the purposes of the study, only one chamber will be 
 42 
University of Ghana http://ugspace.ug.edu.gh
described. The dimensions of one chamber measure 120 x 180 x 240 cm (L x B x H). Twenty 
five metal trays arranged on rails can be used per smoking section but for proper aeration, the 
recommended number of trays is 15 to 18. The estimated maximum capacity is 500 kg of fresh 
fish. Each tray has dimensions of 116 x 168 cm (L x B), with a wire mesh surface area of 
18,800 cm2 (1.88 m2). Below the bottom tray the perforated metal burner, which is connected 
via hose to an LPG gas cylinder. The chamber has a suction fan (operated by electricity) that 
extracts moisture from the oven during the drying stage of the smoking process. There is a 
chimney that goes through the roof of the building. There is a smoked generator (50 x 10 x 70 
cm) attached to the side of the chamber with a hole connecting the two. Agricultural by-
products e.g. sugarcane bagasse and coconut husk are burnt in this unit to provide the smoked 
flavour.  
 
 
Plate 3.3: The Abuesi gas fish smoker (AGFS)  
   
 43 
University of Ghana http://ugspace.ug.edu.gh
3.2.1.3 Fuel used 
Bundles of the firewood (Plate 3.4), ‘Kontan’ (Uapaca guineensis), were obtained from local 
sellers at Abuesi in the Western Region. This is a hardwood with a calorific value of 4,126 
Kcal/kg (Last, Richards, & Fyfe, 1987). The mean length and width of 54 x 11.6 cm was 
estimated from measuring 16 pieces of firewood sampled from the bundles used. The moisture 
content of the wood was also measured at three different locations in each wood, for 20 samples 
using the General Tools MMD4E Moisture Meter. The mean moisture content was then 
estimated as 19.2%. 
 
 
  (a)      (b) 
Plate 3.4: ‘Kontan’ (Uapaca guineensis) used in smoking trials (a) and measuring wood 
moisture (b) for the CCT trials  
 
In the AGFS, the cooking and smoking step was accomplished using LPG gas and sugarcane 
and then the dryer (which uses electricity) was turned afterwards. 
 
3.2.2 Atmospheric conditions 
The temperature and humidity were measured during the testing period. The mean temperature 
and relative humidity were 25.7°C and 85.7% respectively. There was a light breeze during that 
time. 
 44 
University of Ghana http://ugspace.ug.edu.gh
3.2.3 Experimental procedures 
Two local fish processors, each with with more than 10 years fish smoking experience, who 
use smoking kilns were recruited for the CCT, under the supervision of the researcher. Each 
processor tested each kiln once daily, over the three-day period. Each processor was given 60 
kg of fish per kiln (Chorkor and Cabin) per day and 60 kg of firewood per kiln per day(which 
was about twice the quantity of the fuel that they estimated would be enough to smoke all the 
fish per day). The smoking was accomplished in two steps: a cooking and smoking step and a 
drying step. The weights of fish and wood were taken using a KERN CH50K100 electronic, 
hanging balance (KERN & Sohn GmbH, Germany), with a maximum capacity of 50 kg and a 
resolution of 0.10 kg. For the AFGS, 100 kg of fish was used for each smoking session due to 
its industrial size and also to avoid excessive fuel wastage. During the drying step, the oven 
fan was turned on for about 30 minutes to circulate the hot air, after which it was turned off 
and the residual heat used for further drying. The processor was given a 30 kg LPG cylinder 
per smoking session. The temperature inside the fish was taken at 30-minute intervals with a 
Thermco Precision Handheld Pt100 Digital Thermometer (resolution of 0.1 °C). Once the 
temperature inside the fish reached 70 °C (temperature threshold to kill bacteria on the skin and 
in the muscle), the fish were allowed to be in the kiln for an extra 30 minutes, as described in 
Asamoah (2018), after which they were considered smoked and taken out.  
 
3.2.3.1 Fish samples preparation 
Frozen chub mackerel (Scomber colias), procured from cold stores, was used in the smoking 
trials. The fish was thawed at ambient temperature for about 2 hours and then weighed. After 
weighing, the fish was washed and then brined in an 8% NaCl solution for 30 minutes. They 
were taken out of the brine solution, arranged on the smoking racks and allowed to dry for 
about 30 minutes before smoking started.  
 45 
University of Ghana http://ugspace.ug.edu.gh
3.2.3.2 Stack Emissions Monitoring 
The carbon monoxide (CO), sulphur dioxide (SO2) oxides of nitrogen (NOx) and particulate 
matter (PM) emissions were measured for the Cabin and Chorkor kilns using the E9000 
Portable Stack Emissions Analyser. The probe of the analyser was inserted into the smoke 
‘escape’ hole on top of the Cabin and in the fireplace of the Chorkor kiln, as shown in Plate 
3.5. The gas sample was drawn through the probe, with a diaphragm suction pump. The gas 
sample was then cleaned of humidity and impurities by a condensate trap and filter located 
inside the instrument. The gas components were analysed using the electrochemical and 
infrared sensors. The CO, SO2 and NOx were tested following the BS EN 15058, TGN M21 
and BS EN14794 methods.  
 
 
(a)                                                      (b) 
 
                   (c) 
 
Plate 3.5: Stack emissions monitoring of (a) Cabin and (b) Chorkor kilns; (c) E9000 
emission system 
 46 
University of Ghana http://ugspace.ug.edu.gh
The particulate matter measurement was done using the RESS stack pump, a component of the 
E9000 emission system (Plate 3.6). PM was measured isokinetically (BS EN 13284-1 method) 
by placing a previously conditioned glass micro fibre filter in the RESS stack pump. The pump 
was then inserted into the same place as used for the gas emissions. The flue gas was drawn at 
a flow rate of 1.9 LPM, in ten uniform pulls per kiln. After this, the filter paper was removed 
and kept in the zip lock bag for laboratory analysis. In the laboratory, the mass of the deposited 
particles was analysed to obtain the mass per unit volume of PM particles present in the 
emission. The emission tests were done for 1 hour. For quality assurance and control purposes 
for the emissions monitoring, all equipment were calibrated with span gases, prior to the 
measurements, to authenticate the precision of the equipment and data generated. On the site 
the auto zero calibration phase was also activated, then compared with the programmed values 
and compensated for standard deviations. All filters for particulate matter where dried, weighed 
and kept in silica gel to keep the integrity of the filter before measurement. 
 
 
(a)                                                                  (b) 
Plate 3.6: PM monitoring (a) in Cabin kiln and filter paper showing PM residue (b) for 
analysis 
 
 47 
University of Ghana http://ugspace.ug.edu.gh
3.2.4 Smoked products measurements 
After smoking, the fish from each kiln was cooled at room temperature (about 26°C) and then 
weighed. Plate 3.7 depicts the smoked products from the three kilns. To determine the moisture 
contents of the smoked fish, 5 g of the fish sample was dehydrated at 105°C for 4 hours to 
constant weight (AOAC, 2005). The loss in weight during drying was expressed as g water/100 
g sample.  
 
 
(a)                                          (b)    (c) 
Plate 3.7: Fish sample from the (a) Cabin, (b) Chorkor and (c) AGFS smoking kilns  
 
3.2.5 Data and Statistical Analysis 
Statistical analysis was performed using Microsoft Excel 2016 and XLSTAT (Addinsoft, New 
York, USA). The data from the CCTs were inputted into the CCT 2.0 Spreadsheet software 
(Global Alliance for Clean Cookstoves, 2018). The software estimated the equivalent dry wood 
consumed (kg), specific fuel consumption (kg of firewood used per kg of fish smoked) and 
total cooking time (min), for the individual tests, based on the equations below: 
 48 
University of Ghana http://ugspace.ug.edu.gh
!"#$%&'()*	,-.	/00,	10)2#3(,	(5!) = 85"	 −	5$: ∗ 81 − (1.12 ∗ 3): − 1.5 ∗ 	∆A% 
        Equation 3.1 
 
BC(1$5$1	5#('	10)2#3C*$0)	(BA) = 	 $!&  * 1000 Equation 3.2 "
 
D0*&'	230E$)F	*$3(	(∆*) = 	 *$ −	*"  Equation 3.3 
 
Where: fi and ff are the initial and final weights of firewood respectively (g); m is the wood 
moisture content (%); ∆A% is the weight of char remaining (g); Wf is the weight of smoked fish 
(g); and ti and tf are the start and finish times of smoking (min) (Bailis, 2004). 
 
The mean of the three tests performed per stove was then estimated. The results were tested for 
statistical significance using one-way ANOVA followed by Tukey's HSD.  
 
The smoking yield (%) was also calculated according to (Cyprian, et al., 2015) as: 
B30E$)F	.$(',	(%) = 	 HI'JI K 	L	100   Equation 3.4 (
 
Where Ws if the weight of smoked fish and Wr is the weight of raw fish.  
 
Also, the processing rate (g/min), cost of firewood used per kiln (GHS) and the cost of firewood 
per kg of smoked fish (GHS/kg) were estimated. The conversion rate, as of 3rd June 2019 was 
USD 1 = GHS 5.20 (Bank of Ghana, 2019). For the AGFS, similar estimates were made but 
some costs like that for electricity could not be estimated. Results are presented as mean ± 
standard deviation.  
 
 
 49 
University of Ghana http://ugspace.ug.edu.gh
3.3 Results  
 
3.3.1 Efficiency of the smoking kilns 
The moisture contents of the smoked mackerel were 38.87, 33.71 and 25.38 g/100g for the 
Chorkor, Cabin and AGFS kilns respectively. The results from the CCT are presented in Table 
3.1. The mean total weights of smoked fish were 34.6, 34.1 and 50.2 kg, which corresponded 
to yields of 57.6, 56.9% and 50.2% for the Chorkor, Cabin and AGFS kilns respectively. The 
AGFS had significantly lower (p < 0.05) yields compared to the Chorkor and Cabin. The 
equivalent dry wood consumed in the Chorkor averaged 33.2 kg compared to the Cabin which 
used 23.3 kg. This represented a 30% lower usage in the Cabin compared to the Chorkor, and 
this was statistically significant (p < 0.05). The mean quantity of gas and sugarcane consumed 
were 15.3 and 0.3 kg respectively per smoking trial, which in total was significantly lower (p 
< 0.05) compared to the Chorkor and cabin kilns. The mean specific fuel consumption per kg 
of smoked fish in the Chorkor, Cabin and AGFS were 0.96, 0.67 and 0.31, which were all 
statistically different (p < 0.05). The amount of char remaining in the Chorkor averaged 0.35 
kg, which was significantly lower (p < 0.05) than in the Cabin, which produced 0.99 kg of char. 
 
The total smoking time was recorded when the racks were placed on/in the kilns till they were 
taken out for weighing. The Chorkor used a mean of 256 minutes while the Cabin used 218 
minutes, representing a 15% less time to smoke that was statistically significant (p < 0.05). The 
AGFS also used 200 minutes, which was 22% significantly lower (p < 0.05) than the Chorkor. 
The processing rate of the Chorkor was 135.2 g/min, which was about 16% significantly lower 
than (p < 0.05) the Cabin, with 156.8 g/min rate. That for AGFS was 251 g/min, which was 
86% and 60% significantly higher (p < 0.05) than the Chorkor and Cabin respectively. 
 
 50 
University of Ghana http://ugspace.ug.edu.gh
The cost of firewood per kg was estimated at GHS 0.84. The mean cost of firewood by the 
Chorkor and Cabin kilns were GHS 36.09 and 28.76 respectively per smoking session. There 
was a fuel saving of GHS 7.34, on average when the Cabin was used, compared to the Chorkor 
and this was statistically significant (p < 0.05). The cost of firewood/kg of smoked fish was 
also significantly higher (p < 0.05) in the Chorkor compared to the Cabin (i.e. GHS1.05 and 
0.78/kg for Chorkor and Cabin respectively). The mean cost of gas used was GHS 76.33 and 
it cost GHS 1.59 to produce 1 kg of smoked fish, which was significantly higher (p < 0.05) 
than when using firewood. The cost of electricity used for drying the fish could however not 
be determined, and this could increase the cost. The cost of sugarcane used for the smoked 
flavoring was about GHS 0.10, which was very low. 
 
Table 3.1: Comparison of efficiency of Chorkor, Cabin and AGFS kilns (Mean ± SD) 
Parameter Chorkor Cabin AGFS 
Weight of fresh fish (kg) 60.00±0.00 60.00±0.00 100.0±0.0 
Weight of smoked fish (kg) 34.57±1.07 34.13±0.49 50.17±0.76 
Yield (%) 57.61±1.80a 56.89±0.80a 50.17±0.76b 
Weight of char remaining (kg) 0.35±0.35a 0.99±0.16b - 
Equivalent fuel consumed (kg) 33.20±3.39a 23.29±0.46b 15.60±0.92c 
Specific fuel consumption /kg of smoked fish 0.96±0.12a 0.67±0.01b 0.31±0.02c 
Total smoking time (min) 256±15a 218±9b 200±10b 
Processing rate (g/min) 135.20±5.60a 156.80±8.30b 251.18±10.65c 
Cost of fuel/kg of smoked fish (GHS/kg)  1.05±0.12a 0.78±0.01a 1.69±0.17b 
Maximum capacity (kg)  120.0 90.0 500 
Construction cost (GHS) 950.00 4,350.00 65,000.00 
a, b,c Means with different superscripts within a row are significantly different at a < 0.05. 
 
The cost of construction was GHS 4,350.00, GHS 950.00 and GHS 60,000 for the Cabin, 
Chorkor and AGFS respectively. The major costs items for the Cabin were those of the steel 
 51 
University of Ghana http://ugspace.ug.edu.gh
door and plate (28%) and bricks (25%), whereas that for the AGFS were the aluminum frame 
and the capacity. The estimated lifespan of the Cabin and Chorkor is about 10 years, but the 
smoking racks may last for between 2 to 4 years. The AGFS can also last for about 15 years. 
 
3.3.2 Emissions test 
The results from the stack emissions tests are presented in Table 3.2. The Chorkor kiln recorded 
mean emission values of 2058.5, 33.5, 19.25 and 400 mg/m3 for CO, NO2, SO2 and PM 
respectively. The cabin kiln also recorded 1792.4, 30.0, 4.5 and 398.2 mg/m3 for CO, NO2, 
SO2 and PM respectively. These emissions were not statistically different (p > 0.05), with the 
exception of SO2. The flue temperatures for the Chorkor and Cabin were 136.6 and 131.1°C 
respectively.  
 
During the smoking trials, it was observed that there were periods of high smoke production in 
the Cabin at the start of the smoking trial and during addition of extra wood. Once the wood 
burnt to embers, smoke production was considerably reduced. The Chorkor behaved similarly, 
except that there was continual but reduced smoked production as the test progressed. The 
continued smoking was as a result of the fish oils dropping into the fire directly.  
 
Table 3.2: Comparison of emissions (mg/m3) from the Chorkor and Cabin kilns (Mean ± 
SD) 
Parameters Chorkor Cabin 
CO 2058.5±180.3 1792.3±13.1 
NOx 33.5±2.1 30.0±7.1 
SO2 19.3±7.4a 4.5±6.36b 
PM 400.0±10.0 398.2±1.9 
CO is carbon monoxide, NOx is oxides of nitrogen, SO2 is sulphur dioxide and PM is particulate 
matter  
 52 
University of Ghana http://ugspace.ug.edu.gh
3.4 Discussion 
The efficiency of a kiln is measured by the temperature distribution within the kiln, the specific 
fuel consumption, the throughput capacity, the length of time spent smoking and the quality of 
the final product, in comparison with another (Hilderbrand, 1992; Bailis, 2004; Entee, 2015b). 
With respect to the throughput capacity, it was estimated that the Cabin and Chorkor could 
each load approximately 90 and 120 kg respectively per smoking session. From the CCT, the 
Cabin and Chorkor kilns used 683.4 g and 955.1 g of firewood respectively per kg of smoked 
fish. The Cabin was thus more fuel efficient, saving about 29% more firewood, which was 
significantly lower compared to the Chorkor. This saving potential did not however meet the 
40% fuel saving requirement of Energising Development (EnDev) programme (Entee, 2015b). 
According to the processors, the remaining charcoal (char) was not used in subsequent fish 
smoking but used to fuel their local stoves in the preparation of household meals. This was an 
added benefit to them. 
 
A study by IRI-CSIR, GSA, Kwarteng (2016), which compared the Chorkor and Ahotor kilns 
in Ghana, gave the Ahotor kiln a fuel saving potential of 32% over the Chorkor kiln. Ahotor 
recorded higher fuel saving than that recorded by the Cabin in this study. The higher potential 
of the Ahotor could however be attributed to the fact that the moisture content of the firewood 
was high (61.1%) compared to that from this study which was 19.2%. Taylor (2010) found that 
firewood with moisture content of 60% and above burn slowly and can lead to incomplete 
combustion, that produces dirty smoke, which is not healthy. The best firewood for effective 
burning must therefore be properly seasoned/dried and must have a moisture content below 
20% (Taylor, 2010). Dry firewood has been shown to burn hotter and completely, as less heat 
is used up evaporating water, guaranteeing a cleaner and safer fire (Taylor, 2010).  
 53 
University of Ghana http://ugspace.ug.edu.gh
The time spent smoking in the Cabin was 15% less and this corresponded to a higher processing 
rate (16%) compared to the Chorkor. The closed design of the Cabin and the control of air 
entering the fireplace may have contributed to its fuel and time savings as well as higher 
processing rate, as most of the heat was kept in the smoke chamber and the fire burned in a 
controlled way. The smoking time was less than that in a similar study by Entee (2015b), who 
reported smoking times of 285 to 709 minutes (corresponding to processing rates of 190.4 and 
27.5 g/min) in the Morrison and Association of Women for the Preservation of the 
Environment (AWEP) kilns respectively. That for Chorkor was higher than in the present study 
but the FAO Thiaroye Technique (FTT) had a rate of 44.2 g/min.  
 
A comparison of the Chorkor and Cabin kilns to the AGFS showed that the latter performed 
best, followed by the Cabin and then the Chorkor kilns. In terms throughput capacity, the latter 
could hold more fish per smoking cycle (maximum capacity of 500 kg) and smoked it faster 
than the former. Thus, the processing rate was much faster. This could be very useful, 
especially during the bumper fish season, in reducing the postharvest losses that are 
encountered (Akande & Diei-Ouadi, 2010). The AGFS had a fuel saving potential of 68% and 
54% higher than the Chorkor and Cabin respectively, which met the 40% fuel saving 
requirement of EnDev (Entee, 2015b). The yield was however significantly lower in the AGFS 
and was as a result of the superior drying achieved. This, combined with the reduced moisture 
content, could make products more shelf stable, relative to those from the Chorkor and Cabin 
(Asamoah, 2018). Yield, on the other hand, is also of high economic importance as the price 
of smoked fish mostly depends on the weight (Entee, 2015a) so this might affect the 
profitability of the operation. The cost of fuel was higher than when firewood was used, and 
this is as a result of the high price of LPG generally (about GHS 5 per kg compared to wood 
which was GHS 0.84 per kg in this study). However, with about 51% Ghanaians dependent on 
 54 
University of Ghana http://ugspace.ug.edu.gh
firewood and another 35% relying on charcoal (derived from wood), as their primary fuel 
source, coupled with the fact that about 72% of the country is vulnerable to desertification, 
firewood is considered an unsustainable choice (Global Alliance for Clean Cookstoves, 2016). 
 
In terms of construction costs of the kilns, the Chorkor was the lowest with the AGFS being 
the highest. A study comparing the costs of fish smoking kiln in Ghana estimated the costs of 
approximately GHS 1,200.00 and 5,600.00 for the Chorkor and FTT kilns respectively (Entee, 
2015b). The AGFS is an industrial kiln but the cost might be a deterrent for small scale 
processors. There are however other sizes and specifications that costs less, and processors can 
explore these options. Further, the fear of accidental explosions of the gas plus sometimes 
unreliable supply of gas in the country might be further barriers for its adoption (Nunoo, 
Asiedu, Kombat, & Samey, 2015). 
 
From the results, with the exception of SO2, the concentrations of CO, NOx and PM were 
similar between the two kilns. The incomplete combustion of biomass fuels like firewood as a 
result of low temperature of combustion, inadequate oxygen and poor mixing of fuel and 
combustion air may have resulted in the high concentrations of CO released, as suggested by 
Xiu, et al. (2018). A comparison of different biomass fuels in domestic wood stoves produced 
between 750 to above 3000 mg/m3 and this was attributed to limited mixing of air with volatiles 
in the biomass fuel (Roy & Corscadden, 2012). 
 
Oxides of nitrogen (NOx) may be formed from a mixture of atmospheric nitrogen (that forms 
thermal NOx above 1300°C temperatures), prompt NOx (formed at the flame front) and 
elemental nitrogen content in the fuel (which forms fuel-NOx on combustion) (Roy & 
Corscadden, 2012; Russell, 2013). With biomass burning at temperatures below 1300oC, the 
 55 
University of Ghana http://ugspace.ug.edu.gh
level of NOx is usually low as it is formed from the nitrogen constituents in the fuel (Roy & 
Corscadden, 2012; Trozzi, 2013; Rabaçal, 2013; Xiu, et al., 2018). This may have accounted 
for the low levels obtained in this study. The elemental sulphur in wood during combustion is 
mostly converted to sulphur dioxide (SO2), with the levels directly correlating to the amount 
of fuel used (Zandaryaa & Buekens, 2009; Russell, 2013). This assertion agrees with the results 
from this study that showed a significantly lower SO2 level in the Cabin which correlated with 
its lower fuel consumption, relative to the Chorkor kiln.  
 
The combustion of biomass fuels can further result in the formation of particulate matter (PM), 
whose main constituents are soot and condensed and adsorbed organic vapour (Sippula, 2009; 
Garcia, Borrega, & Luís, 2018). Roy & Corscadden (2012) reported PM emission was 
dependent on the type of fuel used, with hardwood having lower emissions than softwood. For 
environmental and human health protections, Ghana’s Environmental Protection Agency 
(EPA) has set standards for point source/stack emissions. Per the guidelines, the limits for CO, 
SO2, NOx and PM are 100, 200, 200 and 50 mg/m3 respectively (EPA, 2019). Comparing the 
results obtained in this study to the EPA limits indicated that SO2 and NOx were below, whereas 
CO and PM were about eighteen and eight times higher than the set limits respectively. The 
high levels of CO and PM could be because there were no pollution reduction mechanisms 
incorporated into the design of the kiln, which the EPA (2019) expected. 
The high levels of CO and PM could negatively impact the health of processors and the 
community as a whole, since emissions can travel long distances in the atmosphere and 
sometimes form secondary products in the atmosphere (Edwards, 2014). Particulate matter 
emissions can negatively affect the respiratory systems, damage lung tissue, cause cancer, and 
ultimately premature deaths in especially children, elderly and asthmatics (Global Alliance for 
Clean Cookstoves, 2016; Garcia, Borrega, & Luís, 2018). Some studies have suggested the 
 56 
University of Ghana http://ugspace.ug.edu.gh
adoption of fish smoking kilns that utilise charcoal since they posed little environmental risk 
compared to the Chorkor and other kilns that use firewood (Entee, 2015a, b; Ndiaye, Komivi, 
& Diei Ouadi, 2015; Bomfeh, et al., 2019). It was however reported that the production of 
charcoal alone had an estimated 2.4 times PM formation impact than firewood and LPG used 
in cookstoves in Ghana (Global Alliance for Clean Cookstoves, 2016).  
 
It was assumed that closed nature of the Cabin and AGFS (with vents on top to direct smoke) 
as well as the shorter time spent smoking, as compared to the Chorkor, might reduce the 
emissions that the processors will be exposed to. There have been some recent studies in Ghana 
that have investigated personal exposure of processors and non-processor to CO and PM 
emissions in fish smokng areas (Antwi-Boasiako, 2017; Obeng, 2018). These studies reported 
PM emissions higher than the EPA standards, even in control locations that were further away 
from area where fish smoking took place. This suggested there was some communal pollution 
from fish smoking that was attributed to the siting of the facilities in the study areas. CO on the 
other hand was below the guideline values prescribed by the EPA. Carbon monoxide is 
however lethal, even at minimal exposure levels, and so continuous exposure may result in 
headaches, nausea, convulsions, increased heart rate, unconsciousness and death (at high 
concentration), especially to vulnerable population groups (Goldstein, 2008; Lucarelli, Wyne, 
& Svenson, 2018; Bilsback, et al., 2019). 
 
3.5 Summary of findings 
Three kilns, the Chorkor, Cabin and AGFS kilns representing uncontrolled, semi-controlled 
and controlled smoking technologies, were investigated based on their efficiency and amount 
of flue gases produced. The yield after smoking followed the trend of the AGFS < Cabin < 
Chorkor and conversely, the specific fuel consumption/kg of smoked fish and processing rate 
 57 
University of Ghana http://ugspace.ug.edu.gh
followed the trend of Chorkor < Cabin < AGFS. These findings were most likely generated 
because of the higher heat loss to the surroundings in the case of the Chorkor kiln compared to 
the two. This led to an unevn distribution of heat in the Chorkor klin which resulted in higher 
consumption of fuel, longer processing times and higher yield. Furthermore, the kilns which 
used firewood as fuel (Chorkor and Cabin) produced flue gases containing CO and PMwere 
that well above Ghana’s EPA emmisions standard. For SO2, NOx however, the amount 
produced was below the emission standard and therefore it stands to reason that with better 
heat management, the firewood consumption, CO and PM emission could be decreased. This 
would help safeguard the health of the predominately female population who use these kilns 
and their environments. To conclude, the AGFS kiln showed a better performance than the 
Chorkor and Cabin but due to fuel and cost consideration, the Cabin kiln has the flexibility to 
be suited to the fishing communities in Ghana in the short run. The environmental benefits 
associated with the AGFS and its industrial operations however make it a preferable choice, in 
the long run. 
 
 
 58 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR 
4.0 ASSESSMENT OF THE QUALITY OF TWO COMMERCIALLY 
IMPORTANT FISH SPECIES SMOKED USING TWO DIFFERENT 
KILNS IN GHANA 
4.1 Introduction 
Fish and fisheries products provide essential amino acids, highly polyunsaturated fatty acids 
(PUFAs) like omega-3 and omega-6, vitamins (A, B2, and D) and minerals (calcium, 
phosphorus, iodine, iron, zinc, magnesium, selenium and potassium), all of which are essential 
for healthy living (Hyldig, Larsen, & Green-Petersen, 2007; Stadlmayr, et al., 2010; Pal et al., 
2018). In Ghana, about 75% of the annual national fish production is locally consumed, which 
accounts for about 50% of the animal protein intake (MOFA, 2017; FAO, 2018). As a highly 
perishable commodity, fish requires some degree of processing. This is because it begins to 
deteriorate immediately after harvest. According to FAO (2016a), the processing of fish 
preserves and extends its shelf life and also allows for extended distribution and marketing 
opportunities. 
 
Fish smoking is an ancient, traditional and affordable preservation method used in most parts 
of the world, especially in areas where there are logistical challenges in cold storage (Asamoah, 
2018). Smoking enhances qualities of great consumer demand like flavour, colour and texture 
(Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). It also prolongs the shelf life of the 
product due to the combined effects of salting, dehydration and antimicrobial and antioxidant 
activity of some smoke constituents (Doe, 1998; Goulas & Kontominas, 2005; Codex 
Alimentarius, 2013). In Ghana, it has been estimated that about 70-80% of fish consumed is 
smoked (Nunoo, Asiedu, Kombat, & Samey, 2015), with the products used in soups and sauces 
 59 
University of Ghana http://ugspace.ug.edu.gh
(UNDP/TCDC, 2001). The mackerel, sardines, anchovies and barracuda are the main smoked 
fish products of commercial importance in Ghana (Nunoo, Asiedu, Kombat, & Samey, 2015).  
 
There have been concerns about the quality and safety of fresh and smoked fish in the country. 
Healthy, freshly caught fish have sterile muscles but the surfaces of the skin, gills and 
gastrointestinal tracts could harbour microorganisms (Gram, 2009). These microorganisms 
could cause spoilage during processing and subsequent storage and display at the various 
markets (Plahar, Nerquaye-Tetteh, & Annan, 1999; Edris, Hasanen, Khater, & Lela, 2012; 
Bomfeh, 2016; Aheto, et al., 2017). In terms of safety, microbial contamination has been 
strongly linked to cases of food-borne infections like cholera, salmonellosis, staphylococcal 
intoxication, histamine poisoning, among others (ICMSF, 1986; FDA, 2011; Visciano, 
Schirone, Tofalo, & Suzzi, 2012; Berjia & Brimer, 2013; Costa, 2013). In addition, there are 
chemical and enzymatic changes, and all these factors can affect the colour, flavour, texture 
and odour, which are important sensory attributes that can influence consumer acceptance 
(Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). They can also affect the nutritional 
status of the processed fish by damaging proteins and amino acids (Cyprian, 2015). 
 
The raw material quality, type of smoking technology used, composition of smoke, among 
others, can affect the quality of smoked fish (Doe, 1998; Onyia, Sogbesan, Milam, & Joseph, 
2011; Sogbesan & Ibrahim, 2017). This study therefore sought to investigate the quality of 
fresh Atlantic chub mackerel (S. colias) and the European barracuda (S. sphyraena) and the 
possible influence of smoking on the quality of the smoked products. It is also aimed at 
assessing the physicochemical, microbial and sensory qualities of the smoked products from 
two different fish smoking kilns. 
 
 60 
University of Ghana http://ugspace.ug.edu.gh
4.2 Materials and Methods 
 
4.2.1 Description of study area 
The smoking experiments were carried out, on fish purchased in January and July 2018, at the 
Fish Processing Facility, which belongs to the Abuesi Fish Processing Association (AFPA), at 
Abuesi in the Western Region of Ghana (described in Chapter Three). 
 
4.2.2 Fish smoking kilns 
Smoking was carried out using two different smoking kilns, i.e. the Abuesi Gas Fish Smoker 
(AGFS) and the Cabin kiln (Chapter Three). The AGFS and Cabin represented controlled and 
semi-controlled smoking technologies. The AGFS operates on liquefied petroleum gas (LPG), 
for heating and cooking and agricultural products like sugarcane (cut pieces or bagasse) for 
smoke generation. It is constructed from steel with removable metal smoking racks. The Cabin 
on the other hand relies on firewood as the smoking fuel and it is constructed from burnt bricks 
with wooden doors and removable wooden smoking racks.  
 
4.2.3 Fish sample acquisition and processing  
Two batches of Atlantic chub mackerel (Scomber colias, Gmelin, 1789) and the European 
barracuda (Sphyraena sphyraena, Linnaeus, 1758), landed on 19th January 2018 and 13th July 
2018, were purchased from artisanal fishers in Sekondi (in the Western Region) for the 
assessment. Twenty kilograms (20 kg) each of mackerel and barracuda were gutted and 
subjected to brining, by immersion in an 8% brine solution (fish: brine ratio of 1: 2 w/v) for 30 
minutes. After brining, the samples were placed on racks to drain for 30 minutes before 
smoking. The mackerel was smoked whole, whereas the barracuda was smoked coiled as 
pertains to how the fish is prepared by local fish processors. The fish samples were smoked for 
 61 
University of Ghana http://ugspace.ug.edu.gh
approximately 4 hours in both kilns. Fish samples were taken out and turned upside down 
midway during the smoking period to ensure even smoking. After smoking, the fish were 
cooled at ambient temperature for about an hour, after which they were packed into ziplock 
bags, iced and transported to the laboratory for analysis. 
 
4.2.4 Analyses 
The physicochemical, microbial and sensory quality of the samples were assessed, in duplicate, 
for fresh and smoked fish. Laboratory analyses were conducted at the Food Microbiology and 
Sensory Laboratories of the Department of Nutrition and Food Science, University of Ghana 
and the Food Research Institute of the Centre for Scientific and Industrial Research (FRI-
CSIR).  
 
4.2.4.1 Physical analysis  
Condition factor 
The total length and weight measurements of 30 individual fresh fish samples of each species 
were taken. The condition factor (K; g/cm3) was calculated for each fish species using the 
Fulton’s index (Froese, 2006) as: 
K= 100W/L3      Equation 4.1 
 
where W is the ungutted weight of fish (g) and L is the length of fish (cm). 
 
Colour Measurements 
The skin and muscle colour of smoked mackerel and barracuda were measured with a Minolta 
CR-310 chromameter (Minolta Camera Co., Ltd; Osaka, Japan). The chromameter was 
calibrated with a reference white porcelain tile (N = 97.57, & = +2.29, and O = 1.88) before 
 62 
University of Ghana http://ugspace.ug.edu.gh
measurements. The colour intensity was described in N∗, &∗, and O∗ notation on the CIE LAB 
colour scale, according to (CIE, 1979). The N∗ measured lightness (L* = 0 for black and L* = 
100 for white); &∗ defined components on the red-green axis (a* > 0 for red and a* < 0 for 
green), and O∗ defined components on the yellow-blue axis (b* > 0 for yellow and b* < 0 blue) 
(Cardinal, et al., 2004). Measurements were made at three locations from posterior to anterior, 
on the skin surface and muscle of the smoked fish samples and the mean and standard deviation 
calculated. The hue angle (Hoab) and chromaticity (c*) were then calculated, according to (Rørå, 
et al., 1998) as: 
P+ = &-1*&) O∗⁄&∗)* 	     Equation 4.2 
A∗ =	R)∗
$
*∗$      Equation 4.3 
 
Hoab ranges from total redness (Hoab = 0) to total yellowness (Hoab = 90), whereas the more 
intense the colour the higher the c* value (Rørå, et al., 1998).  
 
4.2.4.2 Chemical analyses 
Proximate Analysis 
The fresh and smoked fish samples were analysed for moisture, protein, fat, salt and ash 
contents using the methods described in AOAC (2005). Samples were deboned and the heads 
removed, after which the muscle was minced thoroughly before analysis. 
 
The moisture content of the samples was determined using the air-oven method (AOAC, 
32.1.03). Five grams of the fish sample was dehydrated at 105°C for 4 hours to constant weight. 
The loss in weight during drying was expressed as g water/100 g sample. 
 
 63 
University of Ghana http://ugspace.ug.edu.gh
The dry ashing method by AOAC (32.1.05) was used to analyse the ash content. About 5 g of 
the fish sample was weighed into a crucible and incinerated at 550°C for 8 hours in a muffle 
furnace. Results were expressed as g ash/100 g sample. 
 
Protein was analysed using the Kjeldahl method (AOAC 4.2.09). About 2 g of minced fish was 
digested by sulphuric acid in the presence of a catalyst. The digest was rendered alkaline, then 
the liberated ammonia was distilled and titrated with hydrochloric acids. The crude protein was 
obtained by multiplying the nitrogen content by 6.25. The results were expressed as g 
protein/100 g sample. 
 
The Soxhlet method (AOAC 4.5.01) was used to evaluate the fat content. Fat was extracted 
from 5 g of the minced sample with petroleum ether (boiling point of 40-60°C) for 8 hours. The 
solvent was removed from the extract by evaporation (in a fume chamber) and the residue was 
weighed and reported as fat. The fat content was expressed as g fat/100 g sample.  
 
Sodium chloride (Salt) content  
The salt content was determined by weighing 5 g of sample into an extraction bottle, 200 ml 
of deionised water was added and shaken using a shaker for 50 minutes. 20 ml of nitric acid 
was then added to 20 ml of the supernatant and titrated with silver nitrate (AOAC, 2000). 
Results were presented as g NaCl/100 g sample. The salt content in the water phase (Z-value) 
for smoked mackerel was calculated as: 
S − %&'#( = 	 -..	×	%	1%2	3%	1    Equation 4.4 
where: %S is percent salt content and %M is percent water (moisture) content in final product. 
 
 
 64 
University of Ghana http://ugspace.ug.edu.gh
pH 
The pH was determined following a modified AOAC (2.8.01). About 5 g sample from the 
fish was homogenised with distilled water. The pH of the homogenate was measured with a 
glass electrode pH meter. The pH was expressed in 10 g/100 ml water. 
 
Total Volatile Base 
The steam distillation method of Pearson (1970) was used to analyse the total volatile base 
(TVB) content of the samples. 10 g minced fish sample was added to the distilling flask, into 
which 2 g magnesium oxide and 300 ml distilled water were added. 25 ml of 2% boric acid 
and a few drops of screened methyl red indicator were added to a 500 ml receiving flask. The 
macro-Kjeldahl apparatus containing the mixture was heated for 10 minutes and using the same 
rate of heating, distillation was carried out for 25 minutes. The distillate was then titrated with 
0.1 N sulphuric acid. The titre (less blank) was multiplied by 14 to obtain the TVB content as 
mg N/100g flesh. 
 
Histamine analysis  
The levels of histamine were measured following the method by Hardy & Smith (1979). 10 g 
of minced flesh (without skin, scales, bones or other undesirable parts) was put into a blender, 
into which 100 ml of freshly prepared 2.5% trichloroacetic acid (TCA) was added. The mixture 
was homogenised for 3 minutes. The solution was filtered, and the volume recorded. The TCA 
sample solution was then neutralised to pH 7 with 1 N KOH and 0.2 N HCl. The new volume 
was then recorded. A narrow chromatography column was packed with 1 g Amberlite CG-50 
resin and washed with 150 ml acetate buffer, then the eluent was drained so that the surface of 
the liquid aligned the surface of the resin. 75 ml of the neutralised TCA sample solution was 
then applied to the prepared column and the flow adjusted to approximately 1.5ml/min. The 
 65 
University of Ghana http://ugspace.ug.edu.gh
surface of the Amberlite CG-50 was drained and 100 ml acetate buffer was applied to remove 
interfering substances. The histamine was eluted with exactly 25 ml of 0.2 N HCl and the eluate 
was collected in a 50 ml beaker. 1 ml of the HCl eluate was added to 15 ml 5% Na2CO3 in a 
stoppered test tube previously chilled in an ice water bath. 2 ml of chilled diazo reagent was 
added to the mixture and allowed to stand at 0°C for 15 minutes prior to absorbance 
measurement. A blank determination was performed using similar volume of 2.5% TCA. The 
absorbance was measured at 495 nm using distilled water as a reference. A standard curve was 
prepared by using 1 ml aliquots of a standard histamine solution (0-80 μg histamine/ml 0.2 N 
HCl). 80 μg/ml (2 mg/25 ml) in the acid eluent. The results were expressed as mg/kg. 
 
4.2.4.3 Microbiological analyses 
Fish samples (fresh and smoked) were analysed for total count of aerobic mesophiles (TVC), 
coliforms (faecal coliforms and Escherichia coli), staphylococci, Clostridium perfringens and 
yeasts and moulds as per the Ghana Standards Authority quality criteria (Failler, Yolaine, & 
Asiedu, 2014).  
 
The rinse method (APHA, 2015) was used in sample preparation. 25 g of each sample 
(including head, skin, bones and muscle) was aseptically weighed into a sterile stomacher bag, 
into which 225 ml of 0.1% peptone water was added. The bag was then massaged by hand for 
about 1 minute. Serial dilutions were performed on the rinse fluids and homogenised tissue 
samples for the various microbial count tests, per the methods described by the International 
Commission for Microbiological Specifications for Foods (ICMSF, 1986).  
 
Total mesophilic count (TVC) was determined by plating aliquots on plate count agar (PCA) 
using pour plate technique. Enumeration of TVC was performed after 24 hours incubation 
 66 
University of Ghana http://ugspace.ug.edu.gh
under aerobic conditions at 37°C. Two replicates of at least two dilutions with 25-250 discrete 
colonies were enumerated following incubation.  
 
Fecal coliform and E. coli were enumerated and detected by spread plating three serial decimal 
dilutions of the rinse fluid on Levine Eosin Methylene Blue (EMB) agar (Oxoid CM0069) and 
incubated at 37˚C and 44˚C for 24 hours respectively. Plates with 20-200 metallic green 
colonies were counted and sub-cultured on MacConkey agar (Oxoid). Faecal coliforms 
fermented lactose and appeared as pale pink colour on MacConkey, whereas E. coli colonies 
appeared red. Five E. coli colonies were sub-cultured on Sorbitol MacConkey (SMAC) agar 
(Merck), where they appeared as pinpoint, pale pink colonies. These suspected colonies were 
then purified on nutrient agar and then transferred unto Triple sugar iron agar (TSI) slants, 
Simon’s citrate slants and Sulphur-Indole-Motility (SIM) agar. E. coli colonies were indole 
positive, gas positive, H2S negative, citrate negative and could ferment glucose and sucrose.  
 
Staphylococci were determined by spread plating three serial decimal dilutions of the rinse 
fluid on Baird-Parker agar (Oxoid CM275) supplemented with egg yolk tellurite emulsion 
(Oxoid). After incubation at 37°C for up to 24 hours, plates containing 20-200 typical 
staphylococcal colonies (black, circular) were counted. Up to 5 colonies on each plate were 
sub-cultured on nutrient agar (Oxoid CM003), gram stained and tested for catalase activity. All 
typical staphylococci were catalase positive and were gram positive stacked cocci. S. aureus 
colonies were circular, convex, grey-black to jet-black with an off-white margin. They were 
also coagulase positive (APHA, 2015). 
 
The pour plate technique was used to enumerate C. perfringens. Serial dilutions were made 
using 0.1 ml aliquots on TSC (tryptose-sulfite-cycloserine) agar (Oxoid CM0587) 
 67 
University of Ghana http://ugspace.ug.edu.gh
supplemented with egg-yolk and TSC supplement (Oxoid). After the agar had dried slightly, 
the surface was overlaid with 5 ml of TSC agar and incubated upright in an anaerobic jar 
containing an aerobic gas generating kit (Oxoid anaerogen) and incubated at 37 ̊ C for 24 hours. 
Plates containing 20-200 black colonies with opaque halos were selected and counted. To 
confirm presumptive positive C. perfringens colonies, 5 black colonies were selected and tested 
for motility in SIM agar and for nitrate reduction in nitrate broth (Fluka 72548). C. perfringens 
reduced nitrate and was non motile (APHA, 2015). 
 
Yeast and moulds were analysed from the aliquots by pour plating onto Potato Dextrose Agar 
(Oxoid, CM0139) and incubated at 25 ˚C for 3-5 days. Plates containing 20-200 colonies were 
counted. All microbial counts were expressed as log CFU/g sample. 
 
4.2.4.4 Sensory evaluation 
A Quantitative Descriptive Analysis (QDA®) method was carried out to compare the sensory 
descriptive profiles of smoked mackerel and barracuda from the two different kilns (Lawless 
& Heymann, 2010). Two different descriptive tests were carried out for the mackerel and 
barracuda. An eleven-member panel with prior training in Quantitative Descriptive test were 
recruited for the test. The assessors were trained, in three sessions, to describe, assess and score 
sensory attributes of smoked fish i.e. the appearance, aroma, flavour, texture (in hand), 
mouthfeel and aftereffect of samples. The assessors developed two different lists of attributes 
to describe the smoked mackerel and barracuda (Table 4.1 and 4.2). Food and non-food 
reference materials suggested by the panel were used to further comprehend the descriptive list 
of attributes to ensure that all descriptors were used in the same way by all panelists. Further 
training involved a ranking test, where assessors were given coded products and asked to rank 
 68 
University of Ghana http://ugspace.ug.edu.gh
based on intensities for the generated attributes. This ensured that the panelists understood and 
used the attributes generated in the same way.  
 
For assessments and evaluations, samples of fish were taken out from each batch of mackerel 
and barracuda and cut into pieces while frozen to enable easy and uniform cuts and prevent 
any possible structural disintegration due to flaking. The head and tail fin portions of the 
samples were cut off and discarded. The body was then cut transversely into two equal parts. 
Each part was then split along the backbone into two symmetrical halves, de-boned and gutted, 
and further divided to obtain four equal portions. The prepared samples were transferred into 
appropriately labelled 80 cc plastic containers with lids and left out at ambient temperature to 
thaw to about 18±2°C before serving to assessors. Samples were presented to assessors, in 
triplicate, in a randomised balanced order using the William’s design in Compusensecloud 
(Compusensecloud®, Guelph, Ontario, Canada). Samples were served in a monadic sequential 
order. The assessors used Compusense Cloud® (Compusensecloud ®, Guelph, Ontario, 
Canada) to score the intensities of the different attributes defined on a 15 cm line scale (Lawless 
& Heymann, 2010). A total of 26 and 24 sensory descriptors were generated to describe the 
smoked mackerel and barracuda respectively (Table 4.1 and 4.2). 
 
4.2.5 Statistical analysis 
For statistical purposes all microbiological data that were below the limit of detection (LOD) 
were assumed to be half of the respective LOD. Statistical analysis was performed using 
Microsoft Excel 2016 and XLSTAT (Addinsoft, New York, USA). Analysis of variance 
(ANOVA), with stepwise comparison was performed using Tukey’s HSD test at the 5% level 
of significance. Multivariate comparison of different variables and samples was performed 
 69 
University of Ghana http://ugspace.ug.edu.gh
using Principal Component Analysis (PCA) on means to identify differences and similarities 
between the samples.  
 
Table 4.1: Sensory descriptors developed for smoked mackerel 
Sensory descriptor Term Definition Anchor 
 Glossy Having a shiny surface Dull / Shiny 
Appearance (skin of Oily  Amount of oil on the surface of the sample Not / Very 
fish) Wrinkle  Depth of folds on the skin Not / Very 
Iridescent  Typical gold, silver, brown and black Not / Very 
appearance of traditionally smoked fish 
 Flaky Sample appears it will crumble when Not / Very 
 touched  
Appearance (muscle Cream colour  Not / Very 
of fish) Dry Absence of moisture Not / Very 
Brown colour  Not / Very 
 Smoked fish Characteristic aroma of traditionally smoked Not / Very 
Aroma fish  
Fresh tuna-like Characteristic aroma of fresh tuna fish Not / Very 
Dry herring- Characteristic aroma of dried herrings Not / Very 
like  
 Umami Basic taste  Not / Very 
Flavour Smoked fish Characteristic flavour of smoked fish Not / Very 
Salty  Basic taste Not / Very 
 Tough Not easily compressed with the hand  Not / Very 
Texture-in-hand Flaky Ease with which the fillet crumbles into Not / Very 
pieces 
 Astringent Dry feel in the mouth Not / Very 
 Juicy How much fluid oozes out during Not / Very 
 mastication  
Mouthfeel Chewy Needs to be crushed a lot before it is Not / Very 
swallowed 
Tough Not easily compressed in the mouth Not / Very 
Fibrous Forms a fibrous mass in the mouth when Not / Very 
chewed 
 Umami Basic taste  Not / Very 
 Salty Basic taste Not / Very 
Aftereffect Salivation Production of saliva in mouth Not / Very 
Residue Presence of fibrous pieces of fish in mouth Not / Very 
Smoked fish Lingering smoked fish taste Not / Very 
 70 
University of Ghana http://ugspace.ug.edu.gh
Table 4.2: Sensory descriptors for smoked barracuda 
Sensory descriptor Term Definition Anchor 
 Soft Sample gives the impression that it will be Not / Very 
Appearance (skin of easily compressed when touched 
fish) Oily  Presence of oil on the surface of the sample Not / Very 
Wrinkle  Having folds on the skin Not / Very 
Iridescent  Typical gold, silver, brown and black Not / Very 
appearance of traditionally smoke fish 
Cream colour  Not / Very 
Appearance (muscle Dry Absence of moisture Not / Very 
of fish) Soft Sample gives the impression that it will be Not / Very 
easily compressed when touched 
 Smoked Characteristic aroma of smoked barracuda Not / Very 
Aroma barracuda fish 
Fried Characteristic aroma like that of fried fish Not / Very 
Smoky Characteristic aroma of burnt palm nut chaff Not / Very 
 Umami Basic taste  Not / Very 
 Smoked fish Characteristic flavour of traditionally Not / Very 
Flavour smoked fish 
Fried Characteristic aroma like that of fried fish Not / Very 
Salty  Basic taste Not / Very 
Soft Easily compressed with the hand  Not / Very 
Texture-in-hand Flaky Ease with which the fillet crumbles into Not / Very 
pieces 
 Juicy Fluid oozes out during mastication  Not / Very 
 Chewy Needed to be crushed a lot before it is Not / Very 
 swallowed 
Mouthfeel Soft Easily compressed in the mouth Not / Very 
Fibrous Forms a fibrous mass in the mouth when Not / Very 
chewed 
 Umami Basic taste Not / Very 
 Salivation Production of saliva in mouth Not / Very 
Aftereffect Residue Presence of fibrous pieces of fish in mouth Not / Very 
Smoked fish Lingering smoked fish taste Not / Very 
 
 
 71 
University of Ghana http://ugspace.ug.edu.gh
4.3 Results 
 
4.3.1 Physicochemical quality of fresh and smoked mackerel and barracuda 
Fresh mackerel and barracuda samples of mean total lengths 25.2 ± 3.1 cm and 40.7 ± 4.1 cm 
and corresponding mean weight 225.46 ± 53.39 g and 474.67 ± 182.05 g respectively, were 
used in the smoking trials. The mean condition factors (K) were 1.49 ± 0.54 and 0.74 ± 0.33 
for mackerel and barracuda respectively. 
 
4.3.2 Chemical composition 
The mean moisture, fat, protein, ash, sodium chloride, pH, TVB and histamine contents of the 
fresh and smoked fish samples are presented in Table 4.3. The moisture, protein, fat and ash 
contents were 71.97, 24.33, 5.51, 1.41 g/100g and 74.72, 21.76, 2.27 and 1.42 in fresh mackerel 
and barracuda respectively. After smoking, the moisture content of smoked mackerel 
significantly decreased (p < 0.05) by about 53% and 56% in CSM and GSM respectively. The 
protein and TVB were significantly higher (p < 0.05) in smoked samples (i.e. 111% and 282% 
in CSM and 97% and 238% in GSM respectively) as compared to FM samples. Fat, ash, salt, 
pH and histamine contents showed no significant differences (p > 0.05) between the fresh and 
smoked samples. For barracuda, the moisture content was also significantly lower (p < 0.05) 
in smoked samples (34% and 40% in CSB and GSB respectively) than fresh ones. The protein, 
ash and TVB contents were significantly higher (p < 0.05) in the smoked samples (102%, 105% 
and 163% in CSB and 101%, 159% and 197% in GSB respectively) compared to the fresh 
ones. The fat and salt contents were significantly higher (p < 0.05) in GSB and CSB 
respectively as compared to the FB. There was no significant difference (p > 0.05) between the 
samples in terms of pH and histamine. There were no significant differences (p > 0.05) between 
the smoked samples from the two kilns, with respect to the chemical composition. 
 72 
University of Ghana http://ugspace.ug.edu.gh
Table 4.3: Chemical quality (Mean ± SD) characteristics of fresh (F), cabin smoked (CS) 
and gas smoked (GS) mackerel (M) and barracuda(B) (n = 4) 
Parameter Mackerel Barracuda 
FM CSM GSM FB CSB GSB 
Moisture 
71.97±3.10a 34.18±9.68b 32.00±10.13b 74.72±2.85a 49.15±4.26b 44.75±5.30b 
(g/100g) 
Fat 
5.51±5.56a 11.23±2.99a 16.06±7.04a 2.27±3.30a 6.94±2.02ab 10.5±5.37b 
(g/100g) 
Protein 
24.33±1.37a 51.35±12.47b 47.95±4.39b 21.76±0.82a 43.92±2.87b 43.75±3.62b 
(g/100g) 
Ash 
1.41±0.10a 3.67±0.54a 4.09±2.32a 1.42±0.16a 2.91±0.28b 3.69±0.99b 
(g/100g) 
Salt 
0.30±0.12a 0.44±0.18a 0.51±0.25a 0.18±0.02a 0.49±0.15b 0.40±0.13ab 
(g/100g) 
pH 5.98±0.10a 6.00±0.08a 6.03±0.10a 6.60±0.18a 6.65±0.13a 6.58±0.10a 
TVB (mg 
26.32±2.92a 100.62±25.50b 89.05±29.92b 30.70±6.99a 80.85±31.50b 91.23±10.83b 
N/100g) 
Histamine 
11.83±7.79a 7.23±6.63a 10.60±7.64a 4.31±5.47a 6.85±4.72a 5.19±2.82a 
(mg/kg) 
a, b Means of each species with different superscripts within a row are significantly different at 
a < 0.05. 
 
4.3.3 Colour analysis  
The skin and muscle colour attributes are presented in Table 4.4 and 4.5 for smoked mackerel 
and barracuda respectively. The skin redness (a*), yellowness (b*) and chromaticity (c*) were 
significantly higher (p < 0.05) in the CSM compared to GSM (a*, b* and c* of 0.97, 7.56, 7.62 
and 0.01, 5.29, 5.29 in CSM and GSM respectively). The hue angle (Hoab) of 88.83° in the 
GSM was significantly higher (p < 0.05) than that in the CSM (82.70o). There was no 
significant difference (p > 0.05) between the L* for CSM and GSM. Muscle colour showed no 
statistical differences between mackerel from the two kilns. For the barracuda, only the skin a* 
was statistically different (p < 0.05) between the kilns (i.e. 0.95 and -1.78 in CSB and GSB 
respectively). Muscle a* and c* were significantly higher (p < 0.05) in the GSB than in CSB 
(a*, c* of 8.49, 11.94 and 2.62, 4.72 in GSB and CSB respectively). 
 73 
University of Ghana http://ugspace.ug.edu.gh
Table 4.4: Skin colour (Mean ± SD) attributes of cabin smoked (CS) and gas smoked 
(GS) mackerel (M) and barracuda(B) (n = 5) 
Mackerel Barracuda 
Colour attribute CSM GSM CSB GSB 
Lightness (L*) 41.74±2.90a 39.34±1.87a 49.73±0.89a 55.21±5.50a 
Redness (a*) 0.97±0.33a 0.01±0.13b 0.95±0.55a -1.78±0.78b 
Yellowness (b*) 7.56±2.00a 5.29±0.37b 9.80±1.10a 11.31±2.15a 
Hue angle (Hoab) 82.70±1.54a 88.83±0.47b 84.67±2.82a 81.43±3.00a 
Chromaticity (c*) 7.62±2.02a 5.29±0.37b 9.86±1.13a 11.46±2.23a 
 
Table 4.5: Muscle colour (Mean ± SD) attributes of cabin smoked (CS) and gas smoked 
(GS) mackerel (M) and barracuda(B) (n = 5) 
Colour Mackerel Barracuda 
attribute CSM GSM CSB GSB 
Lightness (L*) 10.87±2.95a 15.08±7.40a 15.52±6.00a 14.64±5.32a 
Redness (a*) 1.69±2.50a 0.83±0.52a 2.62±1.32a 8.49±4.54b 
Yellowness (b*)  1.73±1.46a 3.17±2.28a 3.87±2.37a 8.05±4.02a 
Hue angle (Hoab) 56.09±23.93a 72.28±4.63a 52.98±13.93a 44.64±11.83a 
Chromaticity 
2.57±2.75a 3.31±2.30a 4.72±2.59a 11.94±5.48b 
(c*) 
a, b Means of each species with different superscripts within a row are significantly different at 
a < 0.05 
 
4.3.4 Microbiological analyses 
Faecal coliform was observed in all fresh fish samples. This decreased in smoked samples with 
CSB having the least (25%). The percent prevalence of E. coli was the same (50%) in all 
samples, except for CSM, which was higher (75%). C. perfringens decreased in the smoked 
mackerel to 25%, when compared to the fresh ones and were absent in smoked barracuda. 
Yeast and mould were in all samples of FM, GSM and FB, and decreased to 75% in CSM, 
 74 
University of Ghana http://ugspace.ug.edu.gh
CSB and GSB. S. aureus was in 50% of FM and FB, but decreased to 25% in CSM and CSB 
and were absent in GSM and GSB, compared to the fresh samples (Table 4.6).  
 
Table 4.6: Prevalence of microorganisms in fresh (F), cabin smoked (CS) and gas smoked 
(GS) mackerel (M) and barracuda(B) (n = 4) 
% Prevalence (n/N) 
Mackerel Barracuda 
Microorganism FM CSM GSM FB CSB GSB 
Faecal coliform 100 50 75 100 25 50 
E. coli 50 75 50 50 50 50 
C. perfringens 75 25 25 75 0 0 
Yeast and mould 100 75 100 100 75 75 
S. aureus 50 25 0 50 25 0 
 
The mean counts of microorganisms in fresh and smoked samples is presented in Table 4.7. 
FM had the higher total counts of 7.5 log CFU/g compared to that of FB which was 5.8 log 
CFU/g. The faecal coliform, E. coli, C. perfringens, yeast and mould and S. aureus were 3.1, 
2.4, 1.9. 3.9, 1,2 and 4.6, 2.2, 1.7, 4.8 and 1.5 log CFU/g in F< and FB respectively. The 
microbial counts decreased in all smoked samples as compared to the fresh samples. C. 
perfringens was statistically lower in CSM (0.7 log CFU/g) and GSM (0.8 log CFU/g) and 
below the detection limit in the smoked barracuda samples from both kilns. S. aureus was not 
detected in smoked mackerel and barracuda from the AGFS.  
 
 
 
 
 
 75 
University of Ghana http://ugspace.ug.edu.gh
Table 4.7: Microbiological quality (Mean ± SD) of fresh (F), cabin smoked (CS) and gas 
smoked (GS) mackerel (M) and barracuda(B) (n = 4) 
Microorganism Mackerel Barracuda 
(log CFU/g) FM CSM GSM FB CSB GSB 
Total mesophilic 
count 7.5±1.0a 3.0±2.3a 4.0±4.2a 5.8±2.0a 3.8±2.8a 4.3±4.4a 
Faecal coliform 3.1±0.9a 2.2±1.8a 1.6±1.3a 4.6±1.5a 2.4±3.3a 2.4±2.7a 
E. coli 2.4±2.0a 1.4±0.5a 1.6±1.2a 2.2±2.0a 1.8±1.3a 1.4±1.0a 
C. perfringens 1.9±0.8a 0.7±0.2b 0.8±0.2b 1.7±0.7a ND ND 
Yeast and mould 3.9±0.4a 3.3±2.3a 3.8±0.6a 4.8±0.7a 3.0±2.6a 3.0±2.6a 
S. aureus 1.2±0.6a 0.8±0.2a ND 1.5±1.1a 1.5±1.6a ND 
a, b Means of each species with different superscripts within a row are significantly different at 
a < 0.05; ND = not detected  
 
4.3.5 Sensory evaluation 
The spider web plots in Figures 4.1 and 4.2 describe the sensory attributes of smoked mackerel 
and barracuda from the two kilns. Of the 26 sensory descriptors, 13 significantly discriminating 
between the two samples (p < 0.05). GSM scored significantly higher (p < 0.05) in terms of 
glossy, iridescent skin appearance (i.e. 6.1 and 11.4 respectively) compared to CSM that scored 
0.2 and 4.0 respectively. CSM on the other hand had significantly dry and brown muscles (p < 
0.05), scoring 12.6 and 9.0 respectively. CSM scored 13.3 compared to a score of 0.0 for fresh 
tuna-like aroma whereas GSM had a decidedly dry herring-like aroma (scoring 13.1 to 0.0 
scored by CSM). CSM had a significantly higher (p < 0.05) umami flavour and flaky texture 
(scoring 10.2 and 8.4 respectively, compared to 3.9 and 5.3 for GSM). CSM had an astringent, 
tough and fibrous mouthfeel (scoring 7.4, 6.6 and 8.6, compared to GSM scores of 4.0, 4.2 and 
3.5 respectively). Finally, CSM left and umami after taste and residue in the mouth (4.0 and 
4.5 for CSM compared to 0.7 and 2.0 for GSM respectively).  
 76 
University of Ghana http://ugspace.ug.edu.gh
GSM CSM
AP-Glossy (Skin)
AF-Residue14.0 AP-Oily (Skin)
AF-Smoked fish AP-Wrinkles (Skin)
12.0
AF-Salivation AP-Iridiscent (Skin)
10.0
AF-Salty 8.0 AP-Cream colour…
AF-Umami 6.0 AP-Dry (Muscle)
4.0
MF-Juicy 2.0 AP-Brown (Muscle)
0.0
MF-Fibrous AP-Flaky (Muscle)
MF-Tough AR-Tuna
MF-Chewy AR_Smoked fish
MF-Astringent AR-Herrings
FL-Smoked fish TX-Hard
FL-Salty TX-Flaky
FL-Umami
 
Figure 4.1: Spider web plot of sensory profile of cabin smoked (CS) and gas smoked (GS) 
mackerel (M) (AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = 
Mouthfeel; AF- = Aftereffect) 
 
Smoked barracuda similarly had 13 descriptors that were significantly discriminated between 
the two samples from the two kilns (Figure 4.2). GSB had significantly higher (p < 0.05) score 
for oily skin (3.2) and cream muscle colour (8.4) as compared to CSB (with 0.6 and 6.0 
respectively). CSB on the other hand was more wrinkled than GSB (scoring 10.0 and 4.1 
respectively). The CSB samples had a generally smoky aroma and flavour (12.4 and 10.3 
respectively) whereas GSB had a fried fish aroma and flavour (12.3 and 10.9 respectively). 
The texture of GSB was flakier (6.3 to 4.4 for CSB). CSB was chewier and more fibrous (9.0 
and 7.7 respectively) and left a more smoked fish and residue aftereffect in the mouth (3.3 and 
3.4 respectively).  
 
 77 
University of Ghana http://ugspace.ug.edu.gh
GSB CSB
AP-Soft (Skin)
AE-Residue14.0 AP-Oily (Skin)
AE-Smoked fish AP-Wrinkles (Skin)
12.0
AE-Salivation AP-Iridiscent (Skin)
10.0
AE-Umami 8.0 AP-Cream colour…
6.0
MF-Juicy AP-Dry (Muscle)
4.0
MF-Fibrous 2.0 AP-Soft (Muscle)
0.0
MF-Soft AP-Oil (Muscle)
MF-Chewy AR_Smoked Barracuda
FL-Fried fish AR-Fried fish
FL-Smoked_fish AR-Smoky
FL-Salty TX-Soft
FL-Umami TX-Flaky
TX-Oily
 
Figure 4.2: Spider web plot of sensory profile of cabin smoked (CS) and gas smoked (GS) 
barracuda (B) (AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = 
Mouthfeel; AF- = Aftereffect) 
 
To understand the similarities or differences between sensory attributes of the smoked products 
from the two kilns, a principal component analysis was carried out. From Figure 4.3, the first 
principal component (F1) explained 48.77% of the total variation of 66.33% in all the sensory 
attributes. GSM samples were located to the left and characterised by a glossy appearance, an 
iridescent colour and a herring aroma. CSM was on the right and described to have brown 
muscle, a tuna aroma with an intense umami taste. The sample was also flaky, fibrous and left 
the mouth feeling dry (astringent). 
 
 78 
University of Ghana http://ugspace.ug.edu.gh
Sensory profiles (axes F1 and F2: 66.33%)
1
0.75 MF-Fibrous
0.5 AF-Residue AR-Tuna
AF-Umami
AP-GlossyOuter
0.25 CSMGSM
FL-Umami
0
-0.25 AP-DryInner
AP-IridiscentOuter
AP-BrownInner
AR-Herrings
-0.5 MF-Astringent
TX-Flaky
MF-Tough
-0.75
-1
-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1
F1 (48.77%)  
Figure 4.3: Product map showing product and descriptor loading of cabin smoked (CS) 
and gas smoked (GS) mackerel (M). (AP- = Appearance; AR- = Aroma; TX- = Texture; 
FL- = Flavour; MF- = Mouthfeel; AF- = Aftereffect)  
 
Based on descriptor and product loading in the product space (Figure 4.4), F1 accounted for 
52.95%, out of a total of 68.56%, of the variation in the smoked barracuda samples. These were 
grouped to the right of the product space. The GSB was best described as having an oily outer 
appearance, creamy muscle, flaky texture, with a fried fish aroma and flavour. On the other 
hand, the CSB looked more wrinkled, with an intense smoked barracuda aroma, flavour and 
left a smoky aftertaste in the mouth. It was also chewier and more fibrous and left fish residues 
in the mouth. 
 
 79 
F2 (17.55%)
University of Ghana http://ugspace.ug.edu.gh
Sensory profiles (axes F1 and F2: 68.56%)
1
TX-Flaky
0.75 MF-Fibrous
AP-Cream colour (Inner)
AF-Residue
MF-Chewy
0.5
GSB
CSB
FL-Fried f0is.h25
AR-Fried fish AP-Wrinkles (Outer)
AP-Oily (Outer)
AF-Smoked_fish
FL-Smoked fish
0
AR-Smoked Barracuda
AR-Smoky
-0.25
-0.5
-0.75
-1
-1 -0.75 -0.5 -0.25 0 0.25 0.5 0.75 1
F1 (52.95%)  
Figure 4.4: Product map showing product and descriptor loading for smoked barracuda. 
Abbreviations: AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = 
Mouthfeel; AF- = Aftereffect; GSB for gas smoked barracuda; CSB for cabin smoked 
barracuda 
 
4.4 Discussion 
The nutritional content of fish may differ according to species, body size, season, 
environmental factors and nutritional status (Al-Reza, et al., 2015). The moisture, fat, protein 
and ash contents obtained for the fresh mackerel and barracuda in the present study agreed with 
findings from other studies (Stamatis & Arkoudelos, 2007; Mbarki, Miloud, Selmi, Dhib, & 
Sadok, 2009; Nogueira, Cordeiro, & Aveiro, 2013; Adeyeye, Oyewole, Obadina, Omemu, & 
Omoniyi, 2017).  
 
Smoking caused a significant reduction in moisture content (greater than 50% and 30% in 
smoked mackerel and barracuda respectively), relative to the raw materials. Reports by Goulas 
 80 
F2 (15.61%)
University of Ghana http://ugspace.ug.edu.gh
& Kontominas (2005) and Ljubojević, et al. (2016) stated that smoking significantly reduced 
the moisture content, whih supports the findings of this study. A product with a moisture 
content of less than 65% is considered to meet industrial specification of ‘‘smoked finished 
products’’ (Cardinal et al., 2001). From the results, both the smoked mackerel and barracuda 
met this requirement. To inhibit the proliferation of spoilage microorganisms, such as mould, 
and therefore preserve smoked fish for a longer period, a moisture content of 25% or less (wet 
weight) has been recommended (Idah & Nwankwo, 2013). This requirement was not met by 
the smoked products from both kilns and this would suggest that further drying time is needed 
to make the products more shelf stable. 
 
Smoking resulted in a general increase in the protein, fat and ash contents, which corresponded 
with an increase in the dry matter content, resulting from dehydration. Idah & Nwankwo, 2013; 
Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014; Cyprian, Oduor-Odote, & Arason, 
2019; Ljubojević, et al., 2016; Adeyeye, Oyewole, Obadina, Omemu, & Omoniyi, 2017 
reported similar findings. Protein constituted the greatest percentage of the dry matter in the 
smoked samples and this agrees with a study by Daramola, Fasakin, & Adeparusi (2007). A 
study by Suryanti & Suryaningrum (2017) however found significantly lower protein content 
in smoked tilapia fillets, as compared to fresh ones and attibuted this proteins denaturation and 
subsequent uncoiling of polypeptide chains resulting during smoking. Matsuura, et al. (2015) 
reported that temperatures of 100°C caused protein denaturation and the loss of essential amino 
acids. This suggests that the two smoking kilns used possibly operated at a temperature low 
enough to prevent protein denaturation and keep the quality of the smoked fish. The higher fat 
content in the smoked fish, as compared to the fresh ones might cause an increase in the energy 
value of the smoked fish, as suggested by Adeyeye, Oyewole, Obadina, Omemu, & Omoniyi, 
2017). Lipid oxidation can however occur, which may affect the flavour, odour and general 
 81 
University of Ghana http://ugspace.ug.edu.gh
quaility of the smoked fish (Arason, Nguyen, Thorarinsdottir, & Thorkelsson, 2014). The 
increase in the ash content may be as a result of the deposition of mineral elements present in 
the salt during brining process (Ikasari, Suryanti, & Suryaningrum, 2017).  
 
The nutritional quality of the smoked products did not differ significantly between the AGFS 
and the Cabin. This could be as a result of the similar moisture losses observed in smoked fish 
from both kilns, which allowed the concentration of nutrients not to significantly differ, as 
observed by Tiwo, Tchoumbougnang, Nganou, Pankaj, & Nayak (2019). This was in line with 
a study by Odoi (2014) who found no significant difference in proximate composition between 
open fire drum and Cabin kiln smoked mackerel and herring. Chukwu & Shaba (2009), on the 
other hand, found siginificant differences in proximate composition of kiln-dried and electric 
oven-dried catfish in Nigeria. Bouriga, Ben Ismail, Gammoudi, Faure, & Trabelsi (2012) and 
Sogbesan & Ibrahim (2017) also reported similar findings in nutritional composition of fish 
smoked using two smoking kilns.  
 
To preserve fish effectively, a suitable salt concentration, in addition to the use of smoke and 
heat produced during smoking is required. According to Hilderbrand (1992), an appropriate 
salt concentration of 3.5%, in the water phase, can reduce water activity of smoked products to 
0.97 or less that can retard (but not stop) bacteria growth. From the results, all the smoked 
samples had a salt content less than 3.5% in the water phase. This might imply the products 
would have a shorter shelf life.  
 
The acidity of fish muscles generally decreases with smoking, as a result of high temperature, 
absorption of acid from smoke, dehydration and reaction of phenols, polyphenols and carbonyl 
compounds with protein and protein constituents (Arason, Nguyen, Thorarinsdottir, & 
 82 
University of Ghana http://ugspace.ug.edu.gh
Thorkelsson, 2014). The pH values of smoked samples in this study however increased slightly 
(but not significantly) from that in the raw samples. Similar findings were made by Bīlgīn, 
Ünlüsayin, Izci, & Günlü (2008) and da Silva, et al. (2008). pH is one of the most important 
factors that can influence microbial growth and cause spoilage of seafood (Adeyeye, Oyewole, 
& Obadina, 2016). A pH value of 7 presents optimal condition for microbial growth, as most 
tolerate pH between 5 to 8 (Nester, Anderson, Roberts, & Nester, 2009). The pH of the smoked 
mackerel and barracuda in the present study were all within this range and so their shelf life 
could be affected, thus proper storage is required.  
 
The fitness of fishery products for human consumption is usually determined, chemically, by 
the levels of total volatile base (TVB). The TVB is the sum of ammonia, dimethylamine, 
trimethylamine (TMA) and other basic volatile nitrogenous compounds (Koral, Tufan, 
Başçınar, & Köse, 2015). TVB limit of 25-35 mg N/100 g in the muscle has been specified for 
unprocessed products from various fish species (EU Directive, 2008). However, according to 
industry specifications, a TVBN level of < 15 mg N/100g represents fresh fish whereas levels 
> 35 mg N/100g represents stale fish (Tobin & Gormley, 2016). From the results, both the fresh 
mackerel and barracuda were below the limits and thus were fit for smoking. The significant 
increase in the smoked products could be due to the partial dehydration of smoked fish resulting 
in concentration of the TVBN constituents (Goulas & Kontominas, 2005). The TVB limit in 
dried fish products is estimated at 100-200 mg N/100g (Özoğul & Özoğul, 2000). The levels 
of TVB in smoked mackerel and barracuda, in this study, were below this limit. In a related 
study, Plahar, Nerquaye-Tetteh, & Annan (1999) also found TVB levels of 120.6 and 133.8 
mg/100g in freshly smoked Sardinella and anchovy respectively in Ghana. These observations 
implied that the smoked fish were fit for consumption. 
 
 83 
University of Ghana http://ugspace.ug.edu.gh
Consumers usually use the colour of dehydrated foods as a quality indicator, that determines 
their purchase decision (Cyprian, et al., 2015). In fish smoking, the colour of the final product 
is dependent, to a large extent, on the method and the fuel used (Arason, Nguyen, 
Thorarinsdottir, & Thorkelsson, 2014). In Ghana, the skin colour of smoked fish ranges from 
black, dark brown, golden brown or light brown to dirty white, however, consumers prefer 
golden brown or dark brown colour (Obodai, Muhammad, Obodai, & Opoku, 2009; Asamoah, 
2018). From the results, the type of kiln used had a significant impact on the redness, 
yellowness, hue angle and chromaticity of the smoked mackerel, but only redness in smoked 
barracuda. This was consistent with results obtained by Asamoah (2018) indicating a 
significant difference in Cabin and Bradley smoked mackerel. The hue angles for all samples 
was close to 90o, which correspond to total yellowness, with the cabin smoked mackerel being 
more intense. This could imply a general acceptability of the products. Yellowness of dried 
fish muscle gives an indication of lipid oxidation and long storage (Cyprian, et al., 2015). From 
the results, the muscle colour attributes were not significantly different between the two kilns, 
except for the redness and chromaticity of the smoked barracuda. Yellowness values were low, 
and this was in line with the sensory evaluation results.  
 
Fresh or processed seafood are excellent substrates for the growth of most common bacterial 
agents of food-borne diseases, especially when held at improper temperatures (Asamoah, 
2018). In fish, the proposed limit of acceptance for human consumption of total mesophilic 
counts is 7 log CFU/g (ICMSF, 1986; GSA, 2019). The initial quality of the raw mackerel, in 
this study, had counts exceeding this limit (7.5 log CFU/g), which might suggest poor handling, 
whereas that of the barracuda was below the limit (5.8 log CFU/g). The Ghana Standards 
Authority (GSA) has set limits for E. coli and S. aureus in fresh and smoked fish at 3 and 4 log 
CFU/g respectively ; and that for faecal coliform and yeast and mould 2 and 4 log CFU/g 
 84 
University of Ghana http://ugspace.ug.edu.gh
respectively (GSA, 2013; GSA, 2019). The results obtained showed faecal coliform counts 
above the limits in all samples. The E. coli and S. aureus counts were below the limits in all 
samples. Yeast and mould counts were high in the raw materials, with fresh barracuda 
exceeding the limits (i.e. 4.8 log CFU/g), but smoked samples were within the limits. The 
results suggest that smoking led to the final products having lower microbial counts than the 
fresh products. This could have resulted from the combined effects of salting, which lowered 
the water activity; high temperature drying, which provided a physical surface barrier to the 
passage of microorganisms, and deposition of antimicrobial compounds, that delayed 
microbial growth (Arason, Nguyen, Thorarinsdottir & Thorkelsson, 2014). 
 
Two kilns performed similarly, with respect to their smoking capabilities, as the differences 
were not statistically significant. Smoking also caused a general decrease in prevalence of 
microorganisms except in the case of E. coli, which increased in CSM or remained the same. 
Faecal coliform and E. coli are indicator organisms and their presence could indicate faecal 
contamination and possible presence of pathogenic bacteria. (Fernandes, 2009; Kombat, 
Nunoo, Ampofo, & Addo, 2013). The presence of faecal coliforms and yeast and moulds may 
be a result of their high thermal tolerance, making smoking less effective, or poor adherence 
to good management practices after smoking (Koral, Tufan, Başçınar, & Köse, 2015). This is 
consistent with observations by Aheto, et al. (2017) that smoking usually decreases total viable 
bacteria counts, while other microorganisms like moulds and yeast, may still persist due to 
their resistance to heat. A study by Marc, et al. (2014) also found high counts of thermo-tolerant 
coliforms and yeast and moulds in smoked mackerel in Benin. C. perfringens have the ability 
to produce heat-resistant spores that can cause food poisoning (Cortés-Sánchez, 2018). The 
results from the present study however show counts below the borderline limit of 10 - < 104 
(Center for Food Safety, 2014) in both the fresh and smoked samples, an indication of their 
 85 
University of Ghana http://ugspace.ug.edu.gh
good quality. Contrary to this result, Sabry, El-Moein, Hamza, & Kader (2016) found high 
prevalence of C. perfringens in fresh fish, and this posed a public health hazard for both 
processors and consumers in Egypt.  
 
Good quality fish must have a histamine level below 100 mg/kg (but not exceedig 200 mg/kg 
in any of the samples) in raw and smoked fish respectively (European Commission, 2005; 
Codex Alimentarius, 2013), while the US-FDA has set a defect limit of 50 mg/kg (FDA, 2011). 
Fish samples containing levels of 200-500 mg/100g have been known to produce poisoning 
when consumed (FDA, 2011; Food Safety Authority of Ireland, 2018). From the results, levels 
of histamine in both fresh and smoked samples were well below these limits, implying that 
they were safe for consumption. From the microbial results, E. coli and C. perfringens, 
microorganisms implicated in histamine production (FDA, 2011), had low counts and this 
might be the reason for the low histamine levels recorded. A study by Bomfeh (2016) found 
histamine levels in smoked mackerel and barracuda ranging from <10-26 mg/kg for freshly 
smoked fish and 11-450 mg/kg in market samples from Ghana. The levels in the market 
samples suggested post-processing contamination and probable temperature abuse. Amponsah, 
et al. (2017) found histamine levels of 63.0 mg/kg in smoked mackerels in Ghana, which was 
above the US FDA defect limit. Yesudhason, et al. (2013) also found high concentrations in 
frozen mackerel and barracuda as compared to fresh fish in Oman and suggested that handling 
and temperature abuse may have been the cause. Smoking can inactivate the enzyme and 
microorganisms from continuing to produce histamine but that already formed cannnot be 
eliminated since its heat stable (FDA, 2011) and this might account for the high levels in the 
study by Amponsah, et al. (2017). 
 
 86 
University of Ghana http://ugspace.ug.edu.gh
To ensure that a product does not experience market failure, sensory analysis is required. 
Consumers often demand products that are safe and of good nutritional and sensory quality 
(Duxbury, 2005). The results indicated that overall, the cabin smoked products had the more 
traditional smoky appearance, flavour and aroma, as opposed to the gas smoked samples which 
had a more fried aroma, flavour and appearance. The lower smoky notes in the gas smoked 
samples may have been as a result of insufficient smoke generation and time. This could affect 
the acceptability of the product by consumers who are used to the smoked colour, flavour and 
aroma of traditionally smoked fish in Ghana. The TVBN and sensory scores of the smoked 
product were in agreement since neither of them indicated spoilage characteristics in the 
products. The sensory analysis was able to statistically differentiate between the smoked 
products from the two kilns, which the chemical analysis or colour analysis (of the smoked 
barracuda) could not achieve. 
 
4.5 Summary of findings 
The results of the study indicated smoking improved the physical, chemical, microbiological 
and sensory quality of smoked mackerel and barracuda. These qualities, with the exception of 
colour and sensory analysis, could not be statistically differentiated between the products from 
the two different smoking kilns. The Cabin-smoked products had the more traditional qualities 
of smoked fish (appearance, odour and flavour) that the gas-smoked products lacked. This was 
probably due to the indirect smoke generation and/or short contact between the fish and the 
smoke in the AGFS. Based on these factors, it can be concluded that the Cabin kiln which uses 
semi-controlled conditions might combine the fuel flexibility of the Chorkor kiln and the 
efficiency of the gas-fired AGFS kiln.  
 87 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FIVE 
 
5.0 POLYCYCLIC AROMATIC HYDROCARBONS (PAHS) AND 
HEAVY METALS IN FRESH AND SMOKED FISH USING THREE 
DIFFERENT KILNS: LEVELS AND HUMAN HEALTH RISK 
IMPLICATIONS THROUGH DIETARY EXPOSURE IN GHANA 
 
5.1 Introduction  
Smoking preserves fish, or other food products, by exposing it to smoke from smouldering 
wood or plant materials (FAO/WHO, 2012). In many developing countries, including Ghana, 
smoking is the most popular way of preserving fish, using traditional fish smoking kilns 
(Adeyeye & Oyewole, 2016). The most common fuel used by these kilns is firewood, which is 
burnt at temperatures of 300-700°C (Essumang, Dodoo, & Adjei, 2013). 
 
The smoking process can result in the deposition of beneficial substances like antimicrobial 
and antioxidant compounds (Arason, Nguyen, Thorarinsdottir & Thorkelsson, 2014). At the 
same time chemical contaminants like polycyclic aromatic hydrocarbons (PAH), dioxins, 
formaldehyde, nitrogen, sulphur oxides and heavy metals can be deposited on smoked products 
(Codex Alimentarius Commission, 2009).  
 
PAHs are a class of persistent organic ecological toxicants which are of health concerns, as 
they have been observed to be carcinogenic, teratogenic and mutagenic to humans (Codex 
Alimentarius Commission, 2009; Xia, et al., 2010; Kim, Jahan, Kabir, & Brown, 2013; 
Essumang, Dodoo, & Adjei, 2012, 2014; Bandowe, et al., 2014; Li, et al., 2016; Ncube, et al., 
 88 
University of Ghana http://ugspace.ug.edu.gh
2017). PAHs may be present in raw food such as fish, owing to environmental contamination 
of the water or as a result of lignin pyrolysis in biomass fuels, such as firewood, during fish 
smoking (Codex Alimentarius Commission, 2009). Other factors that determine the degree of 
PAH contamination in smoked fish include the type and composition of fuel used (e.g. wood 
and other plant materials, gas among others), smoking method (direct or indirect), distance 
between fish and heat source, fat content of fish, duration of smoking and the design of smoking 
chamber products (Codex Alimentarius Commission, 2009).  
 
The USEPA has identified sixteen priority PAHs and these have been classified as follows: 
benzo(a)pyrene is a definite carcinogenic (Group 1); dibenzo (a,- h)anthracene is probably 
carcinogenic (Group 2A); and benzo(k)fluoranthene, benzo(b)fluoranthene, chrysene, 
naphthalene, indeno(1,2,3-cd)pyrene, and benzo(a)anthracene are possibly carcinogenic 
(Group 2B). The other PAHs, acenaphthalene, acenaphthene, fluorene, phenanthrene, 
anthracene, pyrene, fluoranthene and benzo(g,h,i)perylene, have however not been classified 
as to their carcinogenicity to humans (Group 3) (IARC, 2018). The EU has set maximim limits 
(MLs) for benzo(a)pyrene and the sum of four carcinogenic PAHs at 2.0 and 12.0 μg/kg 
respectively (European Commission, 2011). 
 
Apart from the consumption of smoked fish products, emissions from the combustion of 
biomass fuels (mainly wood) during the smoking process have been identified as a potential 
human exposure pathway to carcinogenic PAHs and other substances (Group 2A) (IARC, 
2018). Smoked fish processors who usually stay in smoke-filled huts for long periods are 
particularly at a greater risk of exposure to these carcinogenic emissions, as well as developing 
other health problems related to the eye, skin and lung (Flintwood–Brace, 2016).  
 
 89 
University of Ghana http://ugspace.ug.edu.gh
Some potentially toxic heavy metals detected in fresh and smoked fish are Lead (Pb), Cadmium 
(Cd), Mercury (Hg), Chromium (Cr), Nickel (Ni) and Arsenic (As) (Daniel, Ugwueze, & 
Igbegu, 2013; Bandowe, et al., 2014). Cd, Pb and Hg have been classified as Group 1, 2B and 
3 carcinogens, respectively (IARC, 2018).  
 
In Ghana, traditional fish smoking kilns commonly used include the “Chorkor” kiln (also 
known as the “Chorkor” smoker), which was introduced in Ghana in 1969 (UNDP/TCDC, 
2001) and was an improvement over the existing kilns due to its higher product throughput and 
fuel efficiency, longer lifespan, shorter operation time and lower labour input. The main 
drawback of the Chorkor kiln is in relation to the high concentrations of PAHs in the smoked 
products, resulting from using firewood (Essumang, Dodoo, & Adjei, 2013, 2014; IRI-CSIR, 
GSA, & Kwarteng, 2016; Bomfeh, et al., 2019). This reduces the quality of fish products and 
poses health risks to consumers, which often leads to the rejection of smoked products on 
international markets (Bomfeh, et al., 2019). To address these concerns, the value chain 
development component of the West Africa Regional Fisheries Programme (WARFP), a World 
Bank project that ended in 2018, aimed at developing fish smoking technologies that reduce 
the levels of PAH in smoked fish to conform to international standards and thereby making the 
product safe for consumption, while at the same time being fuel and cost efficient with minimal 
adverse health impacts on fish processors. Such technology will increase the marketability of 
smoked fish products and contribute to the country’s economic growth (The World Bank, 
2011). 
 
Currently, two improved kilns, the Ahotor and FAO-Thiaroye Technique (FTT), have been 
introduced in Ghana (FAO, 2016b; IRI-CSIR, GSA, & Kwarteng, 2016). An assessment of the 
two kilns showed that the FTT produced good quality fish with PAHs below the EU maximum 
 90 
University of Ghana http://ugspace.ug.edu.gh
limits (MLs), whereas the Ahotor still produced levels higher than the MLs (IRI-CSIR, GSA, 
& Kwarteng, 2016; Bomfeh, et al., 2019). The improved performance of the FTT could be 
attributed to the use of charcoal and an external smoke generator (to provide smoke flavoring) 
as compared to the use of firewood in the Ahotor kiln. The relatively higher cost of constrution 
of the FTT (≈ USD 1,300) (Entee, 2015a), however, might be a deterrent for its adoption in 
Ghana. The need to explore other kilns is therefore paramount. 
 
Two improved but lesser known kilns, the Abuesi Gas Fish Smoker (AGFS) and the Cabin kiln 
were therefore explored in this study and compared with the Chorkor kiln. The AGFS is an 
indirect smoking kiln that relies on liquefied petroleum gas (LPG) for cooking and drying the 
fish and agricultural wastes like sugarcane bagasse to impart the smoky flavour to the fish 
(Nunoo, Asiedu, Kombat, & Samey, 2015; Omri, et al., 2019). The Cabin kiln relies on 
firewood and it is semi-controlled. Both kilns are enclosed units and are expected to ensure 
minimal exposure of processors to smoke and heat, while offering fish of good nutritional 
value.  
 
It is therefore anticipated that the adoption of these technologies in Ghana will be of benefit to 
the fishing industry (for both small scale and industrial processors) and the country as a whole. 
Prior to their adoption in Ghana, however, it is imperative to assess the quality of smoked 
products from these technologies in order to protect human health.  
 
This study therefore aimed to: assess the PAH and heavy metal levels in fresh Atlantic chub 
mackerel (S. colias) and European barracuda (S. sphyraena) in Ghana’s coastal waters; 
compare the PAH and heavy metal levels in fish smoked using the AGFS and Cabin kilns to 
 91 
University of Ghana http://ugspace.ug.edu.gh
the existing Chorkor kiln using different fuel sources; and determine the possible carcinogenic 
health risks associated with smoked fish consumption from these kilns. 
 
5.2 Materials and Methods  
 
5.2.1 Fish sampling and preparation 
A total of 10 kg each of fresh Atlantic chub mackerel (S. colias) and European barracuda (S. 
sphyraena) were purchased, for smoking, from fishers at the Sekondi fish landing site in the 
Western Region of Ghana in December 2017 and January and July 2018. These species were 
selected because they are commercially important pelagic fish species that are (1) mainly 
smoked and consumed locally by a large section of the Ghanaian population, and (2) exported 
to the EU and the USA (Entee, 2015c). 
 
The fish were then degutted, washed, brined, drained and smoked using the Abuesi Gas Fish 
Smoker (AGFS), Cabin Smoker (CS) and Chorkor Smoker (ChS) (described in Chapter Three). 
Two readily available and common hardwood species used for smoking in coastal communities 
i.e. ‘Afena’ (Strombosia glaucescens) and ‘Esa’ (Celtis mildbraedii) were chosen as fuel 
sources in the CS and ChS.  
 
5.2.2 Fish smoking process 
The fish was smoked for approximately 4 hours. The smoking was in two phases: cooking and 
smoking of the fish for about two hours and drying of the samples for another two hours. For 
the AGFS, pieces of sugarcane were placed in the smoke chamber and lighted during the 
smoking phase. The AGFS is equipped with a fan that helps with the heat distribution and the 
drying of the fish. After smoking, the samples were cooled, wrapped in aluminium foil and 
 92 
University of Ghana http://ugspace.ug.edu.gh
transported to the laboratory for heavy metal and PAH analyses. For the moisture and fat 
content analyses, samples were packed in plastic ziplock bags and transported to the laboratory. 
 
5.2.3 Consumer survey 
A survey was undertaken in three communities in the Western region (Abuesi, Sekondi and 
Aboadze). A simple semi-structured questionnaire was administered to for hundred consumers, 
both male and female, to obtain information about their fish consumption patterns. There 
questions were asked:  
• Do you consume mackerel/barracuda? 
• How much (quantity) do you consume per meal?  
• How often do you consume this fish, daily, weekly or monthly?  
 
5.2.4 Analytical methods 
5.2.4.1 Moisture and fat content analyses 
The moisture and fat contents were analysed following the descriptions in Chapter Four. 
 
5.2.4.2 Heavy metal analysis 
The levels of three heavy metals, lead (Pb), mercury (Hg) and cadmium (Cd), in fresh and 
smoked fish samples (n = 4) were determined for each species. The samples were deboned, de-
headed and minced thoroughly (using a blender) before analysis. They were then bagged, 
labelled and stored in the freezer until analysis. Prior to the analysis, the minced fish samples 
were thawed and air-dried to constant weight. About 1.0±0.01 g portions of air-dried fish 
samples were weighed and transferred into 250 ml conical flasks and acid digested with a di-
acid mixture of 10 ml concentrated HNO3 and 10 ml concentrated HCl. This was then refluxed 
at 95oC±5oC in a fume chamber until the digestion was completed. Samples were allowed to 
 93 
University of Ghana http://ugspace.ug.edu.gh
cool and diluted with deionised water to the 50 ml mark, then filtered through a Whatman No. 
1 filter paper into a 50 ml volumetric flask. About 10 ml aliquots of each digested sample was 
taken for analysis of Cd, Pb and Hg concentrations by the Anodic Stripping Voltammetry 
method (ASV) along with standard solutions. Blank samples were also prepared and analysed 
in the same manner as the samples but without fish samples. The target metals were identified, 
quantified and their concentrations expressed in parts per billion (ppb). The limits of 
quantification (LOQ) for Cd, Pb and Hg were 0.001, 0.001 and 0.005 ppb, respectively. 
 
5.2.4.3 PAH analysis 
Fish samples were stored at -80°C prior to analyses at the Ghana Standards Authority. Prior to 
the analysis, the heads and bones of the fish samples were removed. The samples, with their 
skins on, were then minced thoroughly to achieve sample homogeneity, bagged, labelled and 
stored at -20°C ready for extraction. The sixteen (16) US-EPA priority PAHs were targeted 
(Ncube, et al., 2017). All reagents and solvents for the analysis were of HPLC or Ultra-pure 
grade. This PAH stock solution (10 μg/ml of 18 polyaromatic hydrocarbons) was diluted in 
acetonitrile to produce a spiking solution of 1 ppm (μg/ml) (Aheto, et al., 2017). The spiking 
solution was serially diluted to produce concentrations of 5, 10, 20, 50, 100 and 200 ppb, which 
were then used to generate a 6-point multi-level calibration curve. The Agilent Bond Elut 
QuEChERS procedure (Brondi, De MacEdo, Vicente, & Nogueira, 2011) followed by dSPE 
clean-up technique (Aheto, et al., 2017) were used for PAH extraction. 3.0 g (± 0.05 g) of each 
minced sample was weighed into a 50 ml centrifuge tube. 12 ml of de-ionised water and 15 ml 
of acetonitrile were added to the sample, which was then macerated for 2 min. The QuEChERS 
extraction salt containing 6 g MgSO4 and 1.5 g NaCl was then added to the samples in 
centrifuge tubes. The tubes were capped and vortexed for 1 min at 1500 rpm for liquid-liquid 
partitioning and then centrifuged for 3 min at 3000 rpm. 6 ml of the acetonitrile (ACN) layer 
 94 
University of Ghana http://ugspace.ug.edu.gh
was transferred into a 15 ml centrifuge tube with dSPE clean-up agents containing 150 mg 
PSA, 300 mg C18 and 900 mg MgSO4. The mixture was vortexed for 1 min and then 
centrifuged for 3 min at 3000 rpm. 4 ml of the upper ACN layer was transferred to a 50 ml 
pear-shaped flask and concentrated to near-dryness using rotary evaporator at below 40 oC. 
The near-dry extract was then re-dissolved in 1 ml ethyl acetate, and then transferred into a 2 
ml autosampler vial, ready for PAH quantification by gas chromatography-mass spectrometry 
(GC/MS). The limit of quantification (LOQ) for the various PAHs was 0.1 µg/kg (Aheto, et 
al., 2017). 
5.2.5 Human health risk assessment 
5.2.5.1 Health risk associated with heavy metals 
The health risk associated to heavy metal ingestion was evaluated by comparing the target 
hazard quotients (THQ) with reference doses given by the US-EPA as:  
DPT =	 45	6	47	6	859	6	: ?@9$7;	6	<&	6	=> 	L	10    Equation 5.1 
where C is the metal concentration (mg/kg); ED is the exposure duration/life expectancy of 63 
years (UNDP, 2018); EF is the exposure frequency of 365 days/year (Bandowe, et al., 2014); 
BW is the adult body weight of 64.5 kg (WorldData.info, 2019); AT is the average lifespan for 
carcinogens (365 days/year x exposure duration); RfDo is the oral reference dose (mg/kg/day) 
and IFR is the fish ingestion rate in Ghana (g/day). RfDo values for lead, mercury and cadmium 
are 4 x 10-3, 3 x 10-4 and 1 x 10-3 mg/kg/day respectively (USEPA, 2000). IFR was calculated 
based on consumer survey as 28.6 g/day, 42.9 g/day and 8.7 g/day for fresh and smoked 
mackerel and barracuda respectively. The IFR was estimated based on a mean consumption of 
100 and 130 g per meal; and frequency of two and three times a week and two times a month 
for fresh and smoked mackerel and barracuda respectively.  
 
 95 
University of Ghana http://ugspace.ug.edu.gh
5.2.5.2 Toxicological risk associated with PAH concentrations 
The observed concentrations of PAHs in the fish samples were compared to regulatory limits 
and guidelines to determine their toxicity. Individual PAH concentrations, the sum of all 
measured PAHs concentrations (Σ16 PAHs) and total carcinogenic PAHs i.e. PAH4 (sum of 
benzo(a)anthracene, chrysene, benzo(a)pyrene and benzo(b)fluoranthene) were assessed. Also, 
the ratio of the sum of all lower molecular weight (LMW) PAHs (2-3 rings) to that of higher 
molecular weight (HMW) PAHs (4-6 rings) were used to assess the source of the PAH in the 
fresh fish samples, whether petrogenic or pyrogenic (Nyarko, Botwe, & Klubi, 2011).  
 
5.2.5.3 Carcinogenic risk (CR) assessment  
The CR was calculated using concentrations of the seven carcinogenic PAHs (CPAHs) i.e. 
benzo(a)anthracene, chrysene, benzo(a)pyrene, benzo(b)fluoranthene, indeno(1,2,3-
cd)pyrene, dibenzo(a,- h)anthracene and benzo(k)fluoranthene. The dietary daily PAH 
exposure level (ED) was assessed for the adult population using Eq. (1) (Bandowe, et al., 2014; 
Xia, et al., 2010; Li, et al., 2016): 
!7 = U!A	L	VWX"       Equation 5.2 
U!A = 	∑A" A" 	L	Z!W"      Equation 5.3 
where Ci is the concentration of PAHs (µg/kg) in the fish tissue; BEC is the converted sum of 
seven carcinogenic PAHs based on the potency equivalency factors (PEFs) of BAPeq (µg/kg) 
(Table 5.1).  
 
The incremental lifetime cancer risk (ILCR) caused by dietary exposure to PAH from daily 
intake was then calculated using Eq. (3) (Xia, et al., 2010; Li, et al., 2016): 
VNAX = 	 4%	6	45	6	47	6	15	6	:5<&	6	=> 	    Equation 5.4 
 96 
University of Ghana http://ugspace.ug.edu.gh
where ED is the daily PAH exposure level (µg/day); SF is the oral cancer slope factor for 
benzo(a)pyrene [(with geographic mean of 7.3 mg/kg/day)-1] and CF is the conversion factor 
(10-6 kg/mg) (Li, et al., 2016).  
 
5.2.5.4 Non-carcinogenic risk (non-CR) estimates 
The non-CR estimates were based on the Hazard Index (HI) relating to eight non-carcinogenic 
PAHs (naphthalene, acenaphthalene, acenaphthene, fluorene, phenanthrene, anthracene, 
pyrene and fluoranthene). The Hazard Index (HI) was calculated using Eq. (4) (Li, et al., 2016). 
PV = ∑B :&	6	859	6	:5	6	47	6	45" <&	6	=>	6	9$7;      Equation 5.5 
where RfDo is the oral ingestion reference dose (mg/kg/day)-1 (Table 5.1) and Ci is the PAH 
concentration (µg/kg). Values for estimating CR and HI are presented in Table 5.1. 
 
Table 5.1: Oral ingestion reference dose (RfDo) and potency equivalency factor (PEFs) 
used in human intake model for estimating cancer risk and hazard index  
 
PAHs PEF RfDo (mg/kg/day) 
(USEPA, 2013) (Li, et al., 2016) 
Naphthalene - 0.04 
Acenaphthalene - 0.06 
Acenaphthene - 0.06 
Fluorene - 0.04 
Phenanthrene - 0.03 
Anthracene - 0.03 
Fluoranthene - 0.04 
Pyrene - 0.03 
Benzo(a)anthracene 0.1 - 
Chrysene 0.001 - 
Benzo(b)fluoranthene 0.1 - 
Benzo(k)fluoranthene 0.01 - 
Benzo(a)pyrene 1.0 - 
Indeno(1,2,3-cd)pyrene 0.1 - 
Dibenzo (a,- h)anthracene 1.0 - 
 
 97 
University of Ghana http://ugspace.ug.edu.gh
5.2.6 Data analysis 
Statistical analysis was performed using XLSTAT (Addinsoft, New York, USA) and IBM 
SPSS version 23. Comparisons of means of moisture and fat contents, across kilns and species 
were tested by two-way ANOVA followed by Tukey's HSD test at the 5% significance level. 
All PAH and heavy metal concentrations were expressed as μg/kg and mg/kg wet weight 
respectively. For statistical purposes, individual PAH and heavy metal concentrations that were 
below the limit of detection (LOD) or limit of quantification (LOQ) were assumed to be half 
of the respective LOD or LOQ. Normality of the PAH data was assessed using the Shapiro-
Wilk test. The data were not normally distributed, even after log transformation; therefore, only 
nonparametric tests were used. The Kruskal-Wallis with Dunn’s multiple comparison tests 
were performed to detect significant differences between raw and smoked samples from the 
different kilns. Results are presented as mean±standard deviation. 
 
5.3 Results  
 
5.3.1 Quality of fresh and smoked fish 
Fresh mackerel and barracuda measuring 25.2 cm and 40.7 cm in total length and weighing 
225.46 and 474.67 g respectively were smoked using the Cabin (C), Chorkor (Ch) and Gas (G) 
smoking kilns. The moisture and fat contents of fresh mackerel and barracuda are presented in 
Tables 5.2 and 5.3. Fresh mackerel had moisture and fat contents of 71.31 and 7.12 g/100g 
respectively (Table 5.2). The smoked mackerel samples had a moisture content significantly 
lower (p < 0.05) than in the fresh fish. The fat contents were higher in the smoked samples but 
only the GSM was significantly higher (21.53 g/100g) compared to the fresh samples.  
 
 98 
University of Ghana http://ugspace.ug.edu.gh
Table 5.2: Chemical quality (Mean ± SD) characteristics of fresh (F), gas smoked (GS) 
and Afena (A) and Esa (E) cabin smoked (CS) and Chorkor smoked (ChS) mackerel 
(M)  
Smoked (n = 6) 
Fresh (n = 5) 
Chorkor Cabin AGFS 
Parameter FM A-ChSM E-ChSM A-CSM E-CSM GSM 
Moisture (g/100g) 71.31±2.24a 35.45±5.07b 35.70±1.71b 27.19±7.60b 34.18±8.96b 32.00±9.38b 
Fat (g/100g) 7.12±3.93a 15.20±2.39a 10.75±4.33a 13.95±2.81a 11.23±2.78a 21.53±3.01b 
a, b Means of with different superscripts within a row are significantly different at a < 0.05. 
 
The fresh barracuda on the other hand had moisture and fat contents of 75.52 and 2.32 g/100g 
respectively (Table 5.3). The moisture and fat contents in the smoked barracuda behaved 
similarly like in the smoked mackerel. The GSB had a significantly lower (p < 0.05) moisture 
content than the A-CSB (43.14 and 54.11 g/100g respectively). The fat content was also 
significantly higher (p < 0.05) in the GSB (9.66 g/100g). 
 
Table 5.3: Chemical quality (Mean ± SD) characteristics of fresh (F), gas smoked (GS) 
and Afena (A) and Esa (E) cabin smoked (CS) barracuda(B) (n = 5) 
Fresh Smoked 
Parameter FB A-CSB E-CSB GSB 
Moisture (g/100g) 75.52±0.94a 54.11±1.10bc 49.32±3.93bc 43.14±5.23bd 
Fat (g/100g) 2.32±1.79a 6.16±2.87a 6.69±1.86a 9.66±2.57b 
 
a, b Means of with different superscripts within a row are significantly different at a < 0.05 
 
 
5.3.2 PAHs and heavy metal levels in fresh fish 
The total PAH in FM ranged between 0.87 to 23.72 µg/kg, with a mean of 11.27 µg/kg. For 
individual PAH compounds, the concentrations ranged from below detection limits (chrysene 
and benzo(b)fluoranthene) to 2.5 µg/kg (anthracene) in FM. The benzo(a)pyrene in fresh 
 99 
University of Ghana http://ugspace.ug.edu.gh
mackerel ranged from below detection limits to 0.80 ug/kg, with a mean of 0.20 ug/kg. The 
sum of four carcinogenic PAHs (PAH4) was 0.40 ug/kg. 
 
FB had total PAH concentrations ranging from 0.80 to 30.95 µg/kg, with a mean of 13.56 
µg/kg. Chrysene, benzo(a)pyrene and benzo(b)fluoranthene were below detection limits, with 
phenanthrene having the highest concentration of 3.04 µg/kg. The PAH4 averaged 0.45 ug/kg.  
 
The lower molecular weight (LMW) PAHs accounted for 74 and 83% of the total PAHs in FM 
and FB respectively. The ∑LMW-PAH/∑HMW-PAH ratios were 2.96 and 5.21 for FM and 
FB respectively.  
 
The concentrations of Cd, Pb and Hg were below their detection limits of 0.001, 0.001 and 
0.005 mg/kg, respectively, in both FM and FB. 
 100 
University of Ghana http://ugspace.ug.edu.gh
Table 5.4: Concentrations (ug/kg wet weight) of individual PAHs in fresh (F), gas smoked (GS) and Afena (A) and Esa (E) cabin smoked 
(CS) and Chorkor smoked (ChS) mackerel (M). (Mean ± SD) 
Smoked (n = 6) 
Fresh (n = 5) 
Chorkor Cabin AGFS 
PAH  FM A-ChSM E-ChSM A-CSM E-CSM GSM 
Naphthalene 1.69±2.32  28.95±15.60 26.91±21.22 28.95±22.13 23.26±21.23 15.17±12.59 
Acenaphthalene 0.22±0.21 23.27±18.56 17.99±26.53 15.68±23.00 15.33±21.75 0.95±1.17 
Acenaphthene 0.14±0.20 1.32±1.22 1.58±1.92 1.37±1.32 5.14±7.09 0.17±0.19 
Fluorene 1.81±2.17 34.77±19.18 23.29±25.09 20.99±16.22 23.11±20.98 2.71±3.71 
Phenanthrene 1.97±3.14 38.26±17.94 27.75±26.06 45.15±27.07 53.24±54.73 1.22±1.49 
Anthracene 2.50±5.34 131.08±120.79 81.99±97.77 61.58±81.32 108.13±98.85 1.09±2.46 
Fluoranthene 0.06±0.02 73.76±48.45 49.59±52.10 39.76±42.72 50.87±46.99 3.83±4.74 
Pyrene 0.06±0.02 73.8O±48.40 49.44±52.20 39.81±42.67 50.87±46.99 3.83±4.74 
*Benzo(a)anthracene 0.10±0.11 27.55±20.17 22.68±19.46 18.56±18.83 39.15±13.26 0.81±1.15 
*Chrysene ND  40.30±34.59 25.59±25.39 16.46±22.33 24.01±26.54 1.01±1.26 
*Benzo(b)fluoranthene ND  38.25±48.13 11.33±19.18 11.11±20.61 1.42±2.16 ND 
Benzo(k)fluoranthene 0.25±0.36 27.38±19.85 17.17±19.63 10.60±9.51 9.09±10.93 0.53±0.42 
*Benzo(a)pyrene 0.20±0.34 15.51±16.63 5.68±7.36 3.59±3.24 1.29±1.67 0.66±0.61 
Indeno(1,2,3-cd)pyrene 0.66±1.36 11.82±10.24 4.90±8.47 5.30±6.49 0.32±0.54 2.53±3.89 
Dibenzo(a,-h)anthracene 0.78±1.63 11.61±10.34 4.79±8.32 4.61±5.63 0.33±0.42 2.17±3.32 
Benzo(g,h,i)perylene 0.72±1.50 11.98±9.90 4.78±8.40 4.93±5.97 ND 2.42±3.68 
∑PAHs 11.27±10.98 589.59±378.34 375.42±347.09 328.42±231.71 405.58±273.52 39.13±27.74 
∑PAH4 0.40±0.45 121.60±98.88 65.27±65.78 49.71±45.56 65.87±35.09 2.52±2.55 
* denotes PAHs used in ∑PAH4 calculation; ND denotes not detected (i.e. below the detection limit of 0.1 μg/kg); ∑PAHs denotes total PAH 
concentration derived from the sum of individual mass concentrations of all 16 PAH congeners measured; ∑PAH4 denotes sum of 
benzo(a)anthracene, chrysene, benzo(a)pyrene and benzo(b)fluoranthene.
 101 
University of Ghana http://ugspace.ug.edu.gh
Table 5.5: Concentrations (ug/kg wet weight) of individual PAHs in fresh (F), gas smoked 
(GS) and Afena (A) and Esa (E) cabin smoked (CS) barracuda (B) (n = 5) (Mean ± SD) 
Fresh Smoked 
PAH  FB A-CSB E-CSB GSB 
Naphthalene 2.67 ± 2.53 12.36±10.54 17.50±10.41 6.61±2.63 
Acenaphthalene 0.72 ± 0.70 26.51±32.84 17.51±11.31 0.44±0.41 
Acenaphthene 0.92 ± 1.26 11.59±24.37 0.08±0.07 0.17±0.17 
Fluorene 1.65 ± 2.91 17.07±29.47 20.16±12.33 2.05±2.08 
Phenanthrene 3.04 ± 4.52 14.52±19.17 30.87±22.93 14.96±19.40 
Anthracene 2.19 ± 4.54 32.16±53.51 75.30±64.80 9.80±13.69 
Fluoranthene 0.11 ± 0.08 34.00±51.14 58.25±40.24 2.29±1.63 
Pyrene 0.11 ± 0.08 33.99±51.14 58.25±40.22 2.29±1.63 
*Benzo(a)anthracene 0.30 ± 0.56 13.48±11.14 29.11±16.47 0.44±0.55 
*Chrysene ND 13.47±12.25 35.73±22.78 0.47±0.66 
*Benzo(b)fluoranthene ND 4.78±7.01 1.83±3.23 0.35±0.66 
Benzo(k)fluoranthene 0.30 ± 0.48 3.68±3.04 8.41±10.63 0.21±0.24 
*Benzo(a)pyrene ND 2.57±3.54 1.32±2.09 0.32±0.28 
Indeno(1,2,3-cd)pyrene 0.45 ± 0.76 3.10±5.24 2.54±5.57 3.56±3.86 
Dibenzo (a,- h)anthracene 0.53 ± 0.89 4.76±6.46 2.18±4.76 2.02±2.04  
Benzo(g,h,i)perylene 0.42 ± 0.83 5.14±6.98 2.34±5.12 2.56±2.49 
∑PAHs 13.56 ± 12.58 233.19±290.49 361.39±155.58 48.53±34.86 
∑PAH4 0.45 ± 0.56 34.30±30.51 67.99±32.63 1.57±1.44 
* denotes PAHs used in ∑PAH4 calculation; ND denotes not detected (i.e. below the detection 
limit of 0.1 μg/kg); ∑ denotes total PAH concentration derived from the sum of individual mass 
concentrations of all 16 PAH congeners measured; ∑PAH4 denotes sum of 
benzo(a)anthracene, chrysene, benzo(a)pyrene and benzo(b)fluoranthene. 
 
5.3.3 PAHs and heavy metal levels in smoked fish samples 
Smoking caused an increase in the concentrations of the individual PAHs with the exception 
of acenaphthene which was significantly lower (p < 0.05) in GSM compared to FM (0.17 and 
 102 
University of Ghana http://ugspace.ug.edu.gh
0.23 ug/kg respectively) (Table 5.4). The mean concentrations of individual PAHs in smoked 
mackerel ranged from below detection limit of 0.1 ug/kg (benzo(g,h,i)perylene and 
benzo(b)fluoranthene in E-CSM and GSM respectively) to 131.08 ug/kg (i.e. anthracene in A-
ChSM). Naphthalene, acenaphthalene, fluorene, fluoranthene, pyrene and chrysene were 
significantly higher (p < 0.05) in A-ChSM compared to FM. The concentrations of fluorene, 
phenanthrene, anthracene, benzo(a)anthracene were also significantly lower (p < 0.05) in FM 
compared to E-CSM and A-CSM. Fluoranthene and benzo(g,h,i)perylene concentrations were 
significantly different (p < 0.05) between E-ChSM and E-CSM and A-ChSM and E-CSM 
respectively. GSM samples significantly differed from A-ChSM, E-CSM, A-CSM with respect 
to fluorene, phenanthrene, anthracene and benzo(b)fluoranthene concentrations. There were no 
significant differences (p > 0.05) in indeno(1,2,3-cd)pyrene, dibenzo (a,- h)anthracene and 
benzo(k)fluoranthene between all the smoked samples and FM. The concentration of 
benzo(a)pyrene were 0.66, 1.29, 3.59, 5.68 and 15.51 ug/kg for GSM, E-CSM, A-CSM, E-
ChSM and A-ChSM respectively. The Chorkor smoked samples generally had higher 
benzo(a)pyrene concentrations than the Cabin smoked samples, irrespective of the firewood 
type used. There were however no significant differences (p > 0.05) in PAH concentrations 
between the samples smoked with the different kiln and also FM. 
 
The mean total PAH concentrations in smoked mackerel ranged from 39.13 ug/kg in GSM to 
589.59 ug/kg in A-ChSM (Table 5.4). The total concentrations in FM and GSM were 
significantly lower (p < 0.05) than in E-CSM and A-ChSM. The PAH4 concentrations also 
ranged from 2.52 to 121.60 ug/kg in GSM and A-ChSM respectively. A-CSM, E-CSM and E-
ChSM recorded PAH4 concentrations of 49.71, 65.87 and 65.27 ug/kg respectively. PAH4 was 
significantly lower (p < 0.05) in FM and GSM with respect to E-CSM and A-ChSM. The A-
CSM had about 80% less total PAHs compared to the A-ChSM, whereas the E-ChSM had 8% 
 103 
University of Ghana http://ugspace.ug.edu.gh
less than the E-CSM (the main contributor being anthracene in the E-CSM). The GSM had on 
average, 91 and 89% less PAH than the Chorkor and Cabin smoked mackerel respectively. 
Also, the GSB had 87% lower PAH than the Cabin smoked barracuda. The LMW PAHs 
accounted for between 53 to 56% of the total PAHS of A-CSM, GSM and E-CSM and 44 to 
48% in A-ChSM and E-ChSM. 
 
The individual PAH concentrations in smoked barracuda ranged from a mean of 0.08 to 75.30 
ug/kg for acenaphthene and anthracene respectively. There were no significant differences (p 
< 0.05) in individual PAH concentrations between the smoked barracuda samples from the 
different kilns and also the FB for naphthalene, acenaphthalene, acenaphthene, 
benzo(b)fluoranthene, benzo(k)fluoranthene, indeno(1,2,3-cd)pyrene, dibenzo (a,- 
h)anthracene and benzo(g,h,i)perylene. The concentrations of fluoranthene, phenanthrene, 
anthracene, fluorene, pyrene and chrysene were significantly higher (p < 0.05) in E-CSB 
compared to FB. Benzo(a)anthracene had a significantly higher (p < 0.05) concentration in E-
CSB compared to FB and GSB. Benzo(a)pyrene concentrations were 0.32, 1.32and 2.57 ug/kg 
in GSB, E-CSB and A-CSB respectively. 
 
The mean total PAH concentrations ranged from 48.53 to 361.39 ug/kg in GSB and E-CSB 
respectively (Table 5.5). The Cabin smoked samples (A-CSB and E-CSB) recorded total PAH 
concentrations that were significantly higher (p < 0.05) than in the FB. The PAH4 
concentration were 1.57, 34.30 and 67.99 ug/kg in GSB, A-CSB and E-CSB respectively. Only 
E-CSB had significantly higher (p < 0.05) PAH4 concentration than FB. The HMW PAHs 
contributed 51 and 55% to the total PAHs in CSB and 30% to the total PAHs in GSB. 
 
 104 
University of Ghana http://ugspace.ug.edu.gh
The concentrations of Cd and Hg were below detection limits (< 0.001 and < 0.005 mg/kg 
respectively) in smoked mackerel and barracuda from the Cabin, Chorkor and Gas fish 
smoking kilns, irrespective of firewood used. Lead was present in only one sample each of A-
CSM and A-CSB and in two samples of GSM, with mean concentrations of 0.16, 0.23 and 0.59 
mg/kg respectively, but absent in A-ChSM, E-ChSM and GSB.  
 
5.3.4 Human health risk assessment and dietary exposure of PAHs 
5.3.4.1 Carcinogenic health risk assessment 
The carcinogenic risks associated with the consumption of fresh and smoked mackerel and 
barracuda is presented in Table 5.6. From the results, the toxicity equivalencies based on the 
seven US-EPA priority carcinogenic PAHs, namely benzo(a)anthracene, chrysene, 
benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, indeno(1,2,3-cd)pyrene and 
dibenzo (a,- h)anthracene) in fresh mackerel and barracuda were 1.06 and 0.66 ug/kg 
corresponding to BaPEQ daily dose of 30.39 and 5.75 ug/day. For the smoked mackerel, the 
equivalencies ranged from 3.17 ug/kg in GSM to 35.19 ug/kg in A-ChSM, with BaPEQ daily 
doses of 135.86 and 1508.21 ug/day. That for smoked barracuda ranged from 2.78 ug/kg in 
GSB to 9.52 ug/kg in A-CSB. An adult Ghanaian (with a life expectancy of 63 years and 
consumption of 28.6, 42.9 and 8.7 g/day for FM, SM and barracuda respectively) would have 
an estimated cancer risk of 3.44 x10-6, 1.54 x10-5, 2.82 x 10-5, 5.73 x 10-5, 7.06 x 10-5 and 1.71 
x 10-4 associated with FM, GSM, E-CSM, A-CSM, E-ChSM and A-ChSM consumption 
respectively. The estimates for barracuda were 6.50 x10-7, 2.72 x 10-6, 6.83 x 10-6 and 9.34 x 
10-6, corresponding to FB, GSB, E-CSB and A-CSB respectively.  
 
 105 
University of Ghana http://ugspace.ug.edu.gh
5.3.4.2 Non-carcinogenic risk assessment  
The results of non-carcinogenic risks associated with consumption of fresh and smoked 
mackerel and barracuda are presented in Table 5.7. These estimates were based on the eight 
non-carcinogenic PAHs (i.e. naphthalene, acenaphthalene, acenaphthene, fluorene, 
phenanthrene, anthracene, pyrene and fluoranthene). The hazard indices estimated for fresh 
mackerel and barracuda were 1.12 x 10-4 and 4.38 x 10-5 respectively. That for smoked 
mackerel were 5.86 x 10-4, 6.12 x 10-3, 4.29 x 10-3, 5.38 x 10-3 and 6.08 x 10-3 for GSM, E-
CSM, A-CSM, E-ChSM and A-ChSM respectively. Smoked barracuda recorded estimates of 
1.64 x 10-4, 1.13 x 10-3 and 6.08 x 10-3 for GSB, E-CSB and A-CSB respectively. 
 
5.3.5 Health risk associated with heavy metals 
The health risk associated with heavy metal ingestion was evaluated by comparing the target 
hazard quotients (THQ) with reference doses given by the US-EPA. From the results, lead was 
present in only A-CSM, A-CSB and GSM and these corresponded to THQs of 2.67 x 10-2, 7.58 
x 10-3 and 9.85 x 10-2 respectively. 
 
 106 
University of Ghana http://ugspace.ug.edu.gh
Table 5.6: Estimated carcinogenic risks associated with the consumption of fresh (F), gas smoked (GS) and Afena (A) and Esa (E) cabin 
smoked (CS) and Chorkor smoked (ChS) mackerel (M) and barracuda (B). 
Mackerel  Barracuda (n = 5) 
Carcinogenic equivalency Chorkor (n = 6) Cabin (n = 6) Cabin 
FM (n = 5) GSM (n = 6) FB GSB 
A-ChSM E-ChSM A-CSM E-CSM A-CSB E-CSB 
Benzo(a)anthracene 0.01 2.75 0.08 1.86 3.92 0.08 0.03 1.35 2.91 0.04 
Chrysene 0.00 0.04 0.00 0.02 0.02 0.00 0.00 0.01 0.04 0.00 
Benzo(b)fluoranthene 0.01 3.82 0.01 1.11 0.14 0.01 0.01 0.48 0.18 0.03 
Benzo(k)fluoranthene 0.00 0.27 0.01 0.11 0.09 0.01 0.00 0.04 0.08 0.00 
Benzo(a)pyrene 0.20 15.51 0.66 3.59 1.29 0.66 0.05 2.57 1.32 0.32 
Indeno(1,2,3-cd)pyrene 0.07 1.18 0.25 0.53 0.03 0.25 0.05 0.31 0.25 0.36 
Dibenzo (a,- h)anthracene 0.78 11.61 2.17 4.61 0.33 2.17 0.53 4.76 2.18 2.02 
BEC (ug/kg)  1.06 35.19 3.17 11.81 5.82 3.17 0.66 9.52 6.97 2.78 
ED (µg/day) 30.39 1508.21 623.63 506.32 249.29 135.86 5.75 82.53 60.37 24.07 
Carcinogenic risk 3.44E-06 1.71E-04 7.06E-05 5.73E-05 2.82E-05 1.54E-05 6.50E-07 9.34E-06 6.83E-06 2.72E-06 
BEC denotes the converted sum of seven carcinogenic PAHs based on the potency equivalency factors (PEFs) of BAPeq (µg/kg); ED denotes the 
daily PAH exposure level (µg/day). 
 
 
 
 107 
University of Ghana http://ugspace.ug.edu.gh
Table 5.7: Estimated non-carcinogenic risks associated with the consumption of fresh (F), gas smoked (GS) and Afena (A) and Esa (E) 
cabin smoked (CS) and Chorkor smoked (ChS) mackerel (M) and barracuda (B). 
Mackerel Barracuda 
PAH  Chorkor (n = 6) Cabin (n = 6) Cabin 
FM (n = 5) GSM (n = 6) FB GSB 
A-ChSM E-ChSM A-CSM E-CSM A-CSB E-CSB 
Naphthalene 1.93E-05 3.81E-04 2.64E-04 4.42E-04 6.40E-04 3.00E-04 9.23E-06 4.28E-05 6.06E-05 2.29E-05 
Acenaphthalene 1.64E-06 1.38E-04 1.13E-04 1.24E-04 1.93E-04 1.29E-05 1.66E-06 6.12E-05 4.04E-05 1.02E-06 
Acenaphthene 1.07E-06 1.01E-05 5.88E-05 1.22E-05 2.41E-05 2.17E-06 2.12E-06 2.68E-05 1.85E-07 3.92E-07 
Fluorene 2.07E-05 2.95E-04 3.01E-04 4.04E-04 4.36E-04 5.37E-05 5.70E-06 5.91E-05 6.98E-05 7.09E-06 
Phenanthrene 3.00E-05 9.86E-04 1.15E-03 6.24E-04 7.92E-04 3.25E-05 1.41E-05 6.71E-05 1.43E-04 6.91E-05 
Anthracene 3.81E-05 1.49E-03 2.74E-03 2.53E-03 9.99E-04 1.92E-06 1.01E-05 1.48E-04 3.48E-04 4.53E-05 
Fluoranthene 6.85E-07 4.25E-04 6.40E-04 8.33E-04 9.85E-04 7.80E-05 3.81E-07 1.18E-04 2.02E-04 7.92E-06 
Pyrene 9.13E-07 5.68E-04 8.53E-04 1.11E-03 1.31E-03 1.04E-04 5.08E-07 1.57E-04 2.69E-04 1.06E-05 
Hazard index 1.12E-04 4.29E-03 6.12E-03 6.08E-03 5.38E-03 5.86E-04 4.38E-05 6.80E-04 1.13E-03 1.64E-04 
 108 
University of Ghana http://ugspace.ug.edu.gh
5.4 Discussion 
The moisture content of the smoked fish was significantly reduced so as to preserve the 
products, while at the same time keeping their sensory properties (Cardinal, et al., 2004; 
Essumang, Dodoo, & Adjei, 2014). The fat contents in the smoked mackerel and barracuda 
were higher than in the fresh samples, with only GSM and GSB samples being significanlty 
higher than the fresh samples.  
 
The PAH concentrations in the fresh mackerel and barrcuda were all low, compared to that in 
the smoked samples. Essumang, Dodoo, & Adjei (2013) obtained higher concentrations of 
PAH in fresh mackerel, compared to the present study, with benzo[b]fluoranthene contributing 
the highest percentage. This indicates that these fish accumulate low levels of PAHs in the 
marine environment, which is consistent with findings by Stołyhwo & Sikorski (2005). The 
maximum levels of benzo(a)pyrene and PAH4 concentration in muscle meat of fish other than 
smoked fish are used as indicators of potential environmental pollution. However, the current 
EU regulations has established that PAHs are rapidly oxidised and metabolised in fresh fish 
and hence do not accumulate in the muscles, thus no maximum level is prescribed (European 
Commission, 2011). The lower molecular weight (LMW) PAHs constituted the highest 
percentage of the total PAHs implying that the source of PAH was petrogenic (åLMW/åHMW 
> 1.0) for both the FM and FB. This could be due the higher solubility of LMW-PAHs in water, 
which makes them more bioavailable, and the ease of metabolism and removal of higher 
molecular weight PAHs (European Commission, 2011; Bandowe, et al., 2014). Again, the site 
is within the main fishing harbour and its close proximity to vehicular activities, fuel discharge 
points, among other activities could have caused petrogenic PAH contamination of the fish 
(Essumang, Dodoo, & Adjei, 2012). 
 
 109 
University of Ghana http://ugspace.ug.edu.gh
The total PAH content in the smoked mackerel and barracuda, based on fuel type was of the 
order Gas < Esa < Afena for Gas-Chorkor and Gas < Afena < Esa in Gas-Cabin combinations. 
The gas smoked fish samples had up to 83% less PAHs than the Chorkor and Cabin smoked 
fish samples. This implies that in terms of PAH content, the gas kiln had a better product than 
the other two kilns. The high PAH content in wood smoked fish could be because of high lignin 
content of both Esa and Afena (25% and 34.5% respectively) (Oteng-Amoako, 2012; Brauns 
& Brauns, 2013), which causes them to burn hot (averagely between 345.9-465.8°C) 
(Essumang, Dodoo, & Adjei, 2013). This may have caused increased production and deposition 
of PAHs in the exposed fish samples than in the gas smoked samples. Again, the high PAH 
content in smoked mackerel compared to barracuda might be related to the high fat content in 
the mackerel samples, which has also been observed by several authors (Codex Alimentarius 
Commission, 2009; Bandowe, et al., 2014; Essumang, Dodoo, & Adjei, 2013, 2014; Aheto, et 
al., 2017). 
 
With respect to kiln type, the levels were of the order GS < CS < CHS. The Cabin kiln produced 
mackerel with lower total PAH than the Chorkor (about 80%), when Afena was used. When 
Esa was however used for the smoking, the Chorkor had a lower total PAH than the Cabin and 
this was as a result of the higher percentage of lower molecular weight PAHs in the E-CSM 
compared to the E-ChSM. The high levels of PAH obtained in the Chorkor and Cabin kilns 
could be due of the of stacking of several trays of fish and covering the topmost tray in the 
former, and the closed nature of the latter. These could lead to a buildup of smoke around the 
fish (Bomfeh, et al., 2019). Again, the high levels from both kilns could be as a result of testing 
for PAHs from a mince of both the skin and muscles in the fish samples tested. This was 
because both species are usually consumed with skin on. Stołyhwo & Sikorski (2005) in their 
 110 
University of Ghana http://ugspace.ug.edu.gh
study reported that the traditional kilns mostly produce fish that are heavily smoked, with 
benzo(a)pyrene concentrations up to about 50 ug/kg, especially on their surfaces.  
 
Essumang, Dodoo, & Adjei (2013) found higher levels of total PAHs in Chorkor smoked 
sardine, cigar minnows, tuna and mackerel using three different fuel sources (acacia, sugarcane 
bagasse and mangrove). Aheto, et al. (2017) also found higher levels (ranging between 1156.60 
to 4443.40 ug/kg) of total PAHs in market samples of sardines, mackerel and anchoives. Their 
study however did not differentiate between the types of fuel wood and kiln used. Another 
study by Essumang, Dodoo, & Adjei (2014) using modified traditional kilns with or without 
activated charcoal filters found an average of more than 40% reduction in the PAH content of 
fish smoked with filters, which was less than concentrations obtained in the AGFS in the 
present study. 
 
The maximum limits (MLs) of benzo(a)pyrene and PAH4 in the muscle meat of smoked fish 
have been set by the EU at 2 and 12 ug/kg (European Commission, 2011). The lowest 
concentration of benzo(a) pyrene was measured in GSB (0.32 ug/kg), with the highest in A-
ChSM (15.51 ug/kg). From the results, benzo(a)pyrene levels in GSM, GSB, E-CSM, E-CSB 
were all below the EU recommended MLs. The A-CSB, A-CSM, E-ChSM and A-ChSM were 
1.3, 1.8, 2.8, times 7.8 times higher than the recommended limits.  
 
The PAH4 concentration was proposed as the most suitable indicator of PAH contamination 
in foods (European Commission, 2011). The estimated PAH4 ranged between 1.57 to 121.60 
ug/kg for GSB and A-ChSM respectively). In PAH4 levels in only the gas smoked mackerel 
and barracuda were below the EU limits. For Cabin and Chorkor smoked fish, chrysene and 
 111 
University of Ghana http://ugspace.ug.edu.gh
benzo(a)anthracene concentrations contributed the most to the total PAH4. The Chorkor 
smoked mackerel 5.4 and 10.1 times higher (for E-ChSM and A-ChSM respectively).  
 
Several studies have compared the PAHs in smoked fish using different technologies in Ghana 
and beyond. IRI-CSIR, GSA, & Kwarteng (2016) compared the PAH concentrations of 
smoked fish from two traditional smoking kilns (i.e. the Chorkor, Morrison) to the Ahotor (an 
improved kiln) in Ghana. The Chorkor, Morrison and Ahotor recorded benzo(a)pyrene and 
PAH4 concentrations of 22 and 84 ug/kg; 30 and 110 ug/kg and 5.9 and 53.1 ug/kg 
respectively. Aidoo (2017) compared the PAH in mackerel (whole and fillet) from two fish 
smoking kilns (Cabin and Bradley) in Iceland. The PAH and PAH4 concentrations were 2.8 
and 25.85 ug/kg in the smoked fillets and 1.05 and 8.3 ug/kg in the whole smoked mackerel 
from the Cabin kiln. The Bradley kiln gave concentrations below the EU MLs. These estimates 
were lower than in the present study and this could be attributed to the type of wood used and 
the shorter length of time used for smoking. 
 
Essumang, Dodoo, & Adjei (2013) reported benzo(a)pyrene and PAH4 concentrations of 41.27 
and 452.33; 1.26 and 169.4 and 15.22 and 225.75 ug/kg respetively in acacia, sugarcane 
bagasse and mangrove smoked mackerel from the Chorkor smoker in Ghana. This supported 
the fact that different firewood sources influenced the level of PAH in smoked fish. The authors 
also concluded that the elevated levels of benzo(a)pyrene and PAH4 in smoked mackerel may 
have been due, in part to its high lipid contents. 
 
Bomfeh, et al. (2019) also compared barracuda smoked with the FAO FTT, Chorkor and metal 
drum kilns. Soft and dry-smoked barracuda recorded benzo(a)pyrene concentrations of 0.6, 
50.3 and 37.4 ug/kg and 1.8, 61.1 and 69.8 ug/kg respectively in FTT, Chorkor and metal drum 
 112 
University of Ghana http://ugspace.ug.edu.gh
smoked fish. These results were however from the use of different sources of fuel. The FTT 
used charcoal for cooking and sugarcane bagasse for smoke flavouring, while the Chorkor and 
metal drum used firewood. Also, the FTT represented indirect smoking while the other two 
were direct smoking methods and this could account for the different levels of PAH in fish 
products. This was comparable to results from the present study where the gas smoked 
barracuda had lower PAH concentrations compared to the Cabin smoked ones that used 
firewood. Another comparison was made between the FTT and the Chorkor smoker using the 
same types of fuel for smoking sardines. The FTT recorded benzo(a)pyrene and PAH4 
concnetrations of 0.2 and 1.5; 1.9 and 37.0 and 7.7 and 28.9 ug/kg for charcoal, Pterocarpus 
erinaceus and Azadirachta indica respectively. The Chorkor smoked fish recorded 
benzo(a)pyrene and PAH4 concentrations that were 51 and 26; 32 and 10 and 8 and 7 times 
higher than the FTT using charcoal, Pterocarpus erinaceus and Azadirachta indica 
respectively as fuels. This also indicates that the type of kiln and fuel used can have an impact 
on the PAH outcomes, which agreed with the findings from the Cabin and Chorkor smoked 
mackerel using the Afena and Esa wood for smoking in this study. 
 
Another study by Diei-Ouadi (2013) to test different smoking kilns and fuel sources found high 
levels of benzo(a)pyrene and PAH4 (4.6 and 7.2 ug/kg) in smoked fish when the FTT was used 
with LPG gas as the fuel source. This was higher than findings from the recent study, implying 
that the gas smoker used here performed better than the FTT, when LPG gas was used. 
 
An advantage of the AGFS over the Chorkor, Cabin, FTT and Ahotor kilns, among other 
traditional fish smoking kilns in Ghana is that it does not rely on firewood or charcoal, derived 
from burning firewood, as source of fuel. This implies deforestation and its attendant problems 
could be largely eliminated in the fish smoking industry. 
 113 
University of Ghana http://ugspace.ug.edu.gh
The carcinogenic risk associated with the consumption of fresh mackerel and barracuda for an 
adult Ghanaian with a life expectancy of 63 years were estimated to be between 3.44 x10-6 and 
6.50 x 10-6. This implied that an estimated 3 in 1,000,000 and 7 in 10,000,000 adults 
respectively were potentially at risk of suffering from cancer in their lifetime. The gas smoked 
mackerel and barracuda estimated 2 in 100,000 and 3 in 1,000,000 adults were likely to suffer 
from cancer. Likewise, Cabin smoked fish had estimates between 3 to 6 in 100,000 whereas 
Chorkor smoked mackerel fell between 7 per 100,000 and 2 per 10,000 adults (for Esa and 
Afena respectively in each kiln). The US-EPA has estimated a 1 in 1,000,000 (ILCR of 1 x 10-
6) chance of additional human cancer given a 70-year exposure time (USEPA, 2004). This level 
is considered acceptable or inconsequential and compares with risks resulting from ‘normal’ 
human activities like diagnostic x-rays, fishing, among others (Xia, et al., 2010). A level of risk 
of 1 in 100,000 (ILCR of 1 x 10-5) is considered the carcinogenesis threshold, with 1 in 10,000 
or greater (ILCR ³ 1 x 10−4) deemed serious and therefore requiring attention (USEPA, 2004; 
Xia, et al., 2010; Essumang, Dodoo, & Adjei, 2013). From the results, fresh mackerel and all 
barracuda samples were within the acceptable limits and may therefore pose a very low risk. 
This could be due to their lower frequency of consumption. All the smoked mackerel samples 
were above the limits, with the gas smoked being lowest, followed by the Cabin smoked and 
the Chorkor smoked the highest potential risks. Based on the risks with respect to the fuel type, 
it could be inferred that LPG was most suitable, followed by Esa, with Afena being very 
unfavorable, especially when used in the Chorkor kiln.   
 
Bandowe, et al. (2014) estimated risks of magnitude between 7 x 10-7 to 4 x 10-4 in fresh 
Cynoglossus senegalensis, Pomadasys peroteti and Drepane africana from different coastal 
marine areas in Ghana. A report by Essumang, Dodoo, & Adjei (2013) found 5, 15 and 29 out 
of 100,000 adults were at risk of developing cancer in their lifetime from consuming Chorkor 
 114 
University of Ghana http://ugspace.ug.edu.gh
smoked mackerel using sugarcane bagasse, mangrove and acacia wood in Ghana. Other 
smoked fish samples (tuna, sardines and cigar minnow) assessed posed low to moderate risks. 
The authors concluded that the use of sugarcane bagasse was more suitable for fish smoking 
compared to the use of hardwood.  
 
The non-carcinogenic risks (hazard index, HI) from the results were all less than one. It could 
therefore be inferred that consumers of the smoked products were unlikely to experience non-
carcinogenic effects, as was also reported in a study by Li, et al. (2016). 
 
The concentrations of Cd, Pb and Hg were below detection limits (< 0.001, <0.001 and < 0.005 
mg/kg respectively) in both FM and FB. This was consistent with a study by Bandowe, et al. 
(2014), which found low levels of these metals in fresh fish, with the exception of 2 samples. 
Smoking had no significant impact on the levels of Cd and Hg (remained below detection 
limits) in mackerel and barracuda from the Cabin, Chorkor and Gas fish smoking kilns. Lead 
was present in A-CSM, GSM and A-CSB (0.16±0.32; 0.59±0.69 and 0.23±0.45 mg/kg 
respectively) but absent in ChSM and GSB. The EU has set the limits of Cd in muscle meat of 
mackerel at 0.10 mg/kg and other fish at 0.05 mg/kg; that for Pb and mercury are 0.30 and 0.50 
mg/kg respectively (European Commission, 2015). From the results, only GSM had Pb levels 
exceeding the recommended limits. The levels in the gas smoked mackerel could be linked to 
the fact that liquified petroleum gas naturally contain Pb. Adeyeye, Oyewole, Obadina, 
Omemu, & Omoniyi (2017) and Bandowe, et al. (2014) found metal levels below the EU limits 
in smoked barracuda and fresh fish respcetively.  
 
The target hazard quotients (THQs) calculated for each metal and the sum of all metals for all 
the fresh and smoked samples were however below 1. This implies that possible health risks 
 115 
University of Ghana http://ugspace.ug.edu.gh
of heavy metals via the consumption of fresh and smoked mackerel and barracuda may be 
negligible, similar to findings by Bandowe, et al. (2014). However, increased consumption of 
these smoked fish products would likely lead to increased risks (Bandowe, et al. 2014). 
 
5.5 Summary of findings  
The smoking process, while aiding in the preservation of fish, also produces some carcinogenic 
substances such as PAHs and heavy metals in the processed fish. The fresh mackerel and 
barracuda were of good quality for smoking and posed minimal carcinogenic risks to 
consumers. With respect to the kilns, AGFS performed best, by producing smoked products 
with benzo(a)pyrene and PAH4 concentrations below the EU MLs (2 and 12 µg/kg 
respectively). The Cabin also produced smoked mackerel with 77% and 59% lower 
benzo(a)pyrene and PAH4 (only in Esa smoked fish) than the Chorkor. The levels of 
benzo(a)pyrene and PAH4 were however greater than the EU’s MLs in all Chorkor and Cabin 
smoked samples (except for benzo(a)pyrene in E-CSM and E-CSB). Based on the frequency 
and quantities of smoked mackerel and barracuda consumed by an average Ghanaian adult 
(with a life expectancy of 63 years), the potential carcinogenic risks were of least concern in 
the gas smoked and all barracuda samples, moderate in the Cabin smoked mackerel and high 
in the Chorkor smoked mackerel. It could therefore be inferred that the presence of PAHs in 
the smoked fish was due to the type of kiln, smoking method (direct and indirect) and fuel used 
(LPG and firewood) for the treatment. Again, the magnitude of the carcinogenic risks depended 
largely on the fish ingestion rate, with higher benzo(a)pyrene and PAH4 levels not always 
corresponding to increased risks (as shown in the barracuda). Heavy metal (Hg, Pb and Cd) 
contamination was negligible in fresh and smoked mackerel and barracuda. This indicated that 
the smoke emitted might not have contained significant amounts of heavy metals and thus 
contamination was avoided.  
 116 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER SIX 
 
6.0 THE INFLUENCE OF IRRADIATION AND STORAGE 
TEMPERATURE ON THE QUALITY AND SHELF LIFE OF 
SMOKED MACKEREL 
 
6.1 Introduction 
Smoked fish is a delicacy enjoyed by a majority of the population in most developed and 
developing nations, including Ghana. It is as such one of the most traded primary fisheries 
products (FAO, 2018; Bomfeh, et al., 2019). The value chain of smoked fish, from capture to 
distribution and marketing, however, faces a number of losses (Akande & Diei-Ouadi, 2010). 
In Ghana, smoked fish processors encountered quality losses between 11-17% immediately 
after purchase to just before smoking resulting from poor or no icing of fish and lengthy 
bargaining (Akande & Diei-Ouadi, 2010). During smoking with Chorkor and other traditional 
kilns, physical losses (about 3-17%) occurred as a result of some fish getting burnt or dropping 
into the fire. Finally, after smoking and during subsequent packaging, storage and marketing 
of smoked fish, quality losses estimated at 37.5% occurred via the development of rancidity in 
the product, microbial contamination and insect infestation (Kleter, 2004; Akande & Diei-
Ouadi, 2010). These losses in smoked fish processing have been valued at USD 60 million 
annually (Akande & Diei-Ouadi, 2010), a significantly high loss in monetary terms for the 
processors and the country as a whole.  
 
To reduce quality losses encountered after smoking and extend the shelf life of the products, a 
number of interventions can be employed. In Ghana and most developing nations, smoked fish 
 117 
University of Ghana http://ugspace.ug.edu.gh
is mostly stored at ambient temperatures and sold unpackaged, and this can result in 
contamination from microorganisms, mainly air-borne moulds and insect attacks (Akande & 
Ajayi, 2005; Akande & Diei-Ouadi, 2010). For soft smoked mackerel, the shelf life is limited 
to between 1-3 days when stored unpackaged at ambient temperatures (Kwarteng, Nsiah, 
Samey, Boateng, & Aziebor, 2016; Asamoah, 2018). The use of effective packaging and the 
subsequent storage at the right temperature could ensure that a safe and attractive product is 
delivered to the consumer (Cyprian, et al., 2015). 
 
The use gamma irradiation (low to medium dose i.e. <1-10 kGy) can be employed to both 
inactivate pathogenic microorganisms (thereby rendering the product safe), and extend the 
shelf life of a product (Badr, 2012; Arvanitoyannis & Tserkezou, 2014; Ehlermann, 2014). 
Foods, depending on the type, can be irradiated from below 1 kGy to more than 10 kGy, but 
the acceptable dose for fish 3 kGy (EU, 2009). With the development of national and 
international standards following research outcomes, more than 60 countries allow food 
irradiation of at least one food product (Badr, 2012). Studies have been conducted on the 
potential of gamma irradiation in improving the safety and quality of vegetables like garden 
eggs, mushrooms, poultry, fermented maize and cassava products and seafood in Ghana like 
sardines, anchovies and shrimps (Nketsia-Tabiri, Adu-Gyamfi, Montford, Gbedemah, & Sefa-
Dedeh, 2003; Adu-Gyamfi & Appiah, 2012; Adu-Gyamfi, Torgby-Tetteh, & Appiah, 2012; 
Adu-Gyamfi, Riverson, Afful, & Appiah, 2014; Akuamoa, Odamtten, & Kortei, 2018; Duah, 
Emi-Reynolds, Kumah, & Larbi, 2018). Ghana has the requisite capacity, therefore, to use this 
technology to ensure safe and shelf-stable fish products, not only for the domestic market, but 
also for export. 
 118 
University of Ghana http://ugspace.ug.edu.gh
The objective of the study was to evaluate the influence of gamma irradiation and different 
storage temperatures on the physicochemical, microbial and sensory quality of smoked Atlantic 
chub mackerel (Scomber colias).  
 
6.2 Materials and methods 
Two batches of frozen Atlantic chub mackerel (200 kg), purchased on February 2019, was used 
for the assessment. The fish was gutted and subjected to brining, by immersion in an 8% brine 
solution (fish: brine ratio of 1: 2 w/v) for 30 minutes. After brining, the samples were placed 
on racks to drain for 30 minutes before smoking. The mackerel were smoked whole for 
approximately 4 hours in the Abuesi gas fish smoker (AGFS) (Chapter Three). After smoking, 
the samples were cooled at room temperature for about an hour, after which they were packed 
into sterile ziplock bags and transported to the laboratory for analysis. The samples were 
divided into 3 lots of approximately 38 kg each. One lot served as the Control (no irradiation) 
whiles the other two were irradiated at 1.5 and 3 kGy each. The samples were further divided 
into two batches each for storage at refrigerated (2-4 °C) and ambient temperatures (26-30°C). 
The samples for fatty acid and amino acid composition were frozen at -20°C until they were 
analysed. 
 
6.2.1 Gamma Irradiation 
Gamma Irradiation of smoked mackerel was carried out in a category (IV) wet storage cobalt 
60 multipurpose gamma irradiator facility (Type CoS43HH) at the Radiation Technology 
Centre of Ghana Atomic Energy Commission, Accra, Ghana. The smoked mackerel samples 
were packed in zip lock bags and placed in cardboard boxes for irradiation. Each box with 
smoked fish weighed 5 kg. Two doses, 1.5 and 3 kGy were targeted and for each dose, about 
44 kg of smoked fish was used. The irradiation was performed at a dose rate of 0.495 kGy/hr 
 119 
University of Ghana http://ugspace.ug.edu.gh
and cartons were turned 180° halfway into the processing time to ensure homogenous 
distribution of the dose delivered under the same conditions. Doses delivered were confirmed 
using Ethanol-chlorobenzene dosimeters placed inside and outside the cartons. The delivered 
doses were 1.65±0.11 and 3.35±0.13 kGy. 
 
6.2.2 Analytical methods 
Analyses were conducted on fresh/raw, smoked control and irradiated mackerel samples, in 
duplicate. Laboratory analyses were conducted at the Food Microbiology and Sensory 
Laboratories of the Department of Nutrition and Food Science, University of Ghana, FRI-CSIR 
and the National Food Institute of the Danish Technical University. For shelf life studies, 
sampling was undertaken on Days 1, 5, 10, 15, 25, 45 and 65. 
 
6.2.2.1 Colour analysis 
The skin and muscle colour of smoked mackerel (irradiated and non-irradiated) were measured 
with a Minolta CR-310 chromameter (Minolta Camera Co., Ltd; Osaka, Japan) (Chapter Four). 
The colour intensity was described in !∗, "∗, and #∗ notation on the CIE LAB colour scale, 
according to (CIE, 1979). 
 
6.2.2.2 Chemical analyses 
The proximate composition (moisture, protein, fat and ash), total volatile base (TVB) and pH 
of the fresh, smoked and irradiated mackerel were determined using the methods described in 
Chapter Four.  
 
 
 
 120 
University of Ghana http://ugspace.ug.edu.gh
Peroxide Value (PV)  
The peroxide value was analysed using the AOAC method 965.33 (AOAC, 2005). Extracted 
oil sample (5 g) was dissolved in 30 ml of acetic acid-chloroform (2:1 v/v). Saturated potassium 
iodide (0.5 ml) was added and the sample was kept in the dark for 1 min, after which 30 ml of 
water was added. The mixture was slowly titrated with 0.1 M sodium thiosulphate (Na2S2O3), 
while vigorously shaking until the yellow colour disappeared. After this 0.5 ml of 1% starch 
solution was added, and the resulting mixture was continuously titrated with Na2S2O3 until 
colourless. A blank (T) titration was run alongside the samples. The peroxide value (expressed 
in milliequivalent peroxide/kg fat, mEq O2/kg) was calculated from: 
$% = !	#	$	#	%&&&'    Equation 6.1 
Where: S is ml of Na2S2O3 (blank corrected); M is the molarity of Na2S2O3 solution and W is 
the weight of the test sample. 
 
Free fatty acid (FFA) 
The free fatty acids content was determined, according to (ISO 660:2000) method, for the fat 
extracted by the AOAC 4.5.01 method (described in Chapter Four). Sample (2 g) was weighed 
and dissolved in 40 ml of 95% neutralized ethanol. The mixture was heated on a hot plate till 
it boiled, and titrated with 0.1 N sodium hydroxide (NaOH). The FFA was calculated as a 
percentage mass fraction of the oleic acid as: 
''( = 	 (	#	)	#	$	#	%&&%&&&	#	*    Equation 6.2 
Where: V is the volume (ml) of the standard volumetric NaOH solution used; c is the 
concentration (mol/l) of the standard volumetric sodium or potassium hydroxide solution used; 
M is the molar mass of oleic acid (282 g/mol) and m is the mass (g) of the test portion. The 
free fatty acid (FFA) was expressed as % oleic acid. 
 121 
University of Ghana http://ugspace.ug.edu.gh
Fatty acid composition 
The fatty acid composition of the lipid extracts (Bligh & Dyer, 1959) was analysed by gas 
chromatography according to AOCS (1998). Toluene (100 μL), 200 μL heptane with 0.01% 
(v/v) butylated hydroxytoluene (BHT) and 100 μL internal standard (C23:0) (2% w/v) were 
added to about 30 mg of lipid extract. One millilitre of borontriflouride (BF3) in methanol was 
added to the lipid extract mixture and the lipids were methylated in a one-step procedure using 
a microwave oven (Multiwave 3000 SOLV, Anton Paar, Graz, Austria) with a 64MG5 rotor. 
The settings for the microwave were 5 min at 500 Watt followed by 10 min cooling. The fatty 
acid methyl esters (FAMEs) were washed with 1 mL saturated NaCl and 0.7 mL heptane with 
0.01% (v/v) BHT. The heptane phase was analysed by gas chromatography (Agilent 
Technology Model 7890A series GC, China) fitted with automatic sampler (Model 7693, 
Agilent Technology), fused silica capillary column (HP-88, 100 m x 0.25 mm x 0.20 μm film 
thickness; Agilent Technology), split injector, and flame ionization detector (FID). The carrier 
gas was helium with a flow rate of 0.38 mL/min and an inlet pressure of 51 psi. The oven 
temperature program for separation was from 160 to 200°C, 200 to 220°C, and 220 to 240°C at 
10.6°C /min. For separation, DB127-7012 column (10 m × ID 0.1 mm × 0.1 μm film thickness, 
Agilent Technologies, Palo Alto, CA, USA) was used. Injection volume was 0.2 μL in split 
mode (1:50). All analyses were carried out in duplicate. The result of each fatty acid was 
expressed as g fatty acid/100 g lipid. 
 
The lipid quality indices, polyene index (PI), atherogenic index (AI) and thrombogenic index 
(TI) were calculated using the following equations (Rosa & Nunes, 2003; Chaula, et al., 2019):  
* = 	 +,&:./+,,:0+%0:&         Equation 6.3 
(* = +%,:&/(2	#	+%2:&)/+%0:&∑$567/	∑8567!"#/	∑8567 	     Equation 6.4 !"$
 122 
University of Ghana http://ugspace.ug.edu.gh
 
+* = +%2:&/+%0:&/+%9:&(:..	#	∑$567)/(&..	∑8567 )/	(<	#	∑8567 )	/(=><? )	 Equation 6.5 !"# !"$ =>0
where: MUFA is monounsaturated fatty acids and PUFA is polyunsaturated fatty acids.  
 
Amino acid composition 
The amino acid composition was analysed using a Phenomenex EZ:faast amino acid analysis 
kit (California, USA). About 30 mg of the sample was hydrolysed in 6 M HCl in a microwave 
oven (Microwave 3000 SOLV, Anton Paar, Austria) for 60 min at 110°C. The hydrolysate was 
filtered into 1.5 mL screw-cap vial through a cellulose acetate 0.22 μm Q-Max RR syringe 
filter using a 1 mL syringe. The amino acid composition was determined by liquid 
chromatography with a mass spectrometry detector (Agilent 1100 series, LC/MSD Trap, 
Agilent Technologies, Denmark) using a Phenomenex Z:faast 4u AAA-MS column (250 × 3.0 
mm, California, USA). The total protein concentration in the samples was calculated by 
summarizing all the amino acids and subtracting the water incorporated during hydrolysis (18 
g H2O mol
-1 amino acid) (Mols-Mortensen, Ortind, Jacobsen, & Holdt, 2017).  
 
6.2.2.3 Microbiological analyses 
Fish samples (fresh and smoked) were analysed for the total count of aerobic mesophiles 
(TVC), coliforms (faecal coliforms and E. coli), Staphylococci aureus, Clostridium perfringens 
and yeasts and moulds based on the methods described in Chapter Four. 
 
6.2.2.4 Sensory evaluation 
The difference-from-control test (Lawless & Heymann, 2010) was used to determine if there 
were any differences between non-irradiated (control) and irradiated smoked mackerel 
immediately after smoking and irradiation. 15 panelists were screened to be able to detect and 
 123 
University of Ghana http://ugspace.ug.edu.gh
discriminate between basic tastes. Assessors first tasted a labelled control sample (R) and then 
proceeded to taste three test samples; Blind control (non-irradiated), Irradiated I (1.5 kGy) and 
Irradiated II (3 kGy) in a pre-determined randomized order. For each test sample, assessors 
determined overall how different that test sample was from the labelled control sample on a 6-
point labelled category scale with the following categories: 0 = No difference; 1= Very slight 
difference; 2 = Slight difference; 3 = Moderate; 4 = Very different and 5 = Extreme difference. 
Overall product differences were evaluated and not attribute differences. Assessments were 
done in duplicates to give 30 responses for each sample. Check-all-that-apply (CATA) list was 
provided for panelists to indicate where differences (if any) could be perceived. 
 
6.2.2.5 Insect infestation 
The samples were visually inspected for insects and pest attacks during storage. This was done 
prior to samples been sent to the laboratory for microbial and chemical quality analyses. 
 
6.2.3 Data analysis 
Statistical analysis was performed using Microsoft Excel 2016 and XLSTAT (Addinsoft, New 
York, USA). For statistical purposes all microbiological data that were below the limit of 
detection (LOD) were assumed to be half of the respective LOD. Analysis of variance 
(ANOVA) was performed, with stepwise comparison, using Fisher’s least significant 
difference (LSD) test at the 5% significance level. Results are presented as mean ± standard 
deviation (n = 2). 
 
6.3 Results 
Results from the physicochemical, microbiological and sensory analyses are presented below. 
 124 
University of Ghana http://ugspace.ug.edu.gh
6.3.1 Chemical composition 
6.3.1.1 Proximate composition 
The mean moisture, protein, fat and ash content of the fresh and smoked mackerel are presented 
in Table 6.1. FM had the highest moisture (64.93 g/100g) and lowest protein (25.30 g/100g), 
fat (10.15 g/100g) and ash (1.45 g/100g) contents relative to the smoked samples. Freshly 
smoked and irradiated products (Day 1) were statistically different (p < 0.05) from the FM in 
terms of the moisture and protein contents. Comparing the proximate composition between 
Day 1 and Day 65 (refrigerated storage) showed a general decrease moisture content by Day 
65, with the lowest percentage reduction (16%) in the SM and SMg-3.0 kGy samples (39.33 
and 36.30 g/100g respectively by Day 65). The ash, fat and protein contents in SM increased 
by 22%, 45% and 12% respectively by Day 65. The SMg-1.5 kGy samples remained relatively 
unchanged during the 65 days, except for the fat content that increased from 11.57 g/100g to 
20.46 g/100g (77%). The ash and fat content of SMg-3.0 kGy increased from 1.97 and 11.55 
g/100g to 2.49 and 20.19 g/100g (27% and 75%) respectively, whereas the protein content 
decreased from 45.91 g/100g to 37.56 g/100g by Day 65. The changes between Day 1 and Day 
65 were however not statistically different (p > 0.05) between the SM and SMg products. 
 
Table 6.1: Proximate composition of fresh (F), smoked (S) and gamma irradiated (g) 
mackerel (M) for Day 1 and Day 65 of refrigerated storage 
Parameter SM SMg-1.5 kGy SMg - 3.0 kGy 
(g/100g) FM Day 1 Day 65 Day 1 Day 65 Day 1 Day 65 
Moisture  64.93±4.91 46.65±5.80 39.33±9.49 45.98±1.17 45.75±7.29 43.13±2.30 36.30±1.71 
Protein  25.30±0.11 37.21±4.10 45.91±3.13 37.56±7.37 38.07±5.19 43.29±0.94 39.39±1.51 
Fat  10.15±5.89 12.03±9.01 17.50±2.81 11.57±2.91 20.46±0.58 11.545±2.49 20.19±2.42 
 
Ash  1.45±0.13 1.98± 0.11 2.42±0.52 2.15±0.46 2.34±0.46 1.965±0.32 2.49±0.11 
 
 
 125 
University of Ghana http://ugspace.ug.edu.gh
6.3.1.2 Fatty acid composition 
A total of 33 fatty acids were identified and quantified in the fresh and smoked mackerel 
samples (Table 6.2), with the unsaturated fatty acids being relatively dominant (27), compared 
to the saturated ones (6). Of the unsaturated fatty acids identified, 18 were polyunsaturated 
fatty acids (PUFAs), with nine being monounsaturated fatty acids (MUFAs). Among the 
PUFAs, there were 9 omega-3 fatty acids (ω-3) and 5 omega-6 (ω-6) fatty acids. 
 
The total saturated fatty acids (SFAs) of the irradiated and non-irradiated smoked samples were 
significantly higher (p < 0.05) than in the FM (i.e. 23.01, 35.39, 34.13 and 35.65 g fatty 
acid/100 g oil sample in FM, SM, SMg-1.5 kGy and SMg-3.0 kGy respectively). Irradiation 
however did not cause a significant increase (p > 0.05) in SFAs, compared to SM. Palmitic, 
myristic and stearic acid constituted 13.53 to 22.61, 5.07 to 5.70 and 2.79 to 6.23 g fatty 
acid/100 g of the total lipid content, respectively. Stearic acid was significantly higher (p < 
0.05) in smoked samples. Arachidic acid was statistically different (p < 0.05) in all the fresh 
and smoked samples, but not detected in SMg-3.0 kGy. The total MUFA was significantly 
reduced (p < 0.05) in the smoked samples (ranging from 20.68 g fatty acid/100 g in SM to 
45.29 g fatty acid/100 g in FM). Oleic acid accounted for a higher percentage of the MUFAs, 
ranging from 9.15 to 15.06 g fatty acid/100 g in SM and FM respectively.  
 
The PUFAs also significantly increased (p < 0.05) in the smoked samples, relative to the FM 
(26.22, 34.59, 35.04 and 36.08 g fatty acid/100 g for FM, SMg-3.0 kGy, SMg-1.5 kGy and SM 
respectively). Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) accounted for 
the highest percentages (ranging from 6.91 to 7.36 g fatty acid/100 g and 8.78 to 17.02 g fatty 
acid/100 g respectively) of the ω-3 PUFAs, in both the fresh and smoked mackerels. There 
were no significant differences (p > 0.05) in the EPA for all samples, but DHA was 
 126 
University of Ghana http://ugspace.ug.edu.gh
significantly higher (p < 0.05) in the smoked samples, compared to the FM. Again, irradiation 
at 1.5 and 3.0 kGy had no significant effect on the ω-3 fatty acids, compared to the non-
irradiated samples. The total ω-6 PUFAs ranged from 3.54 g fatty acid/100 g in FM to 4.78 g 
fatty acid/100 g in SMg-1.5 kGy. Linoleic and γ-linolenic acids were only significantly lower (p 
< 0.05) in SM and SMg-3.0 kGy respectively, compared to the FM. Dihomo-γ-linolenic was 
not detected in FM. The total ω-3 and ω-6 PUFAs was also significantly higher (p < 0.05) in 
the smoked than fresh mackerel.  
 
The ratios of ω-3:ω-6 PUFAs were 5.07, 7.72, 5.55 and 5.49 in FM, SM, SMg-1.5 kGy and 
SMg-3.0 kGy respectively. The PUFA to SFA ratios were also 1.14, 1.02, 1.03 and 0.97 in FM, 
SM, SMg-1.5 kGy and SMg-3.0 kGy respectively. The polyene index (PI) ranged from and 
1.04 to 1.17 in SMg-3.0 kGy and SMg-1.5 kGy respectively. There were no significant 
differences (p > 0.05) in these ratios for all samples. Atherogenic index (AI) was significantly 
lower (p < 0.05) in FM (0.54) compared to SM, SMg-1.5 kGy and SMg-3.0 kGy (0. 79, 0.76 
and 0.82 respectively). The thrombogenic index (TI) also ranged between 0.26 to 0.35 in FM 
and SMg-3.0 kGy respectively, and these were not statistically different (p > 0.05) from the 
other samples.  
 
  
 127 
University of Ghana http://ugspace.ug.edu.gh
Table 6.2: Fatty acid composition of fresh (F), smoked (S) and gamma irradiated (g) mackerel (M) (g fatty 
acid/100 g oil sample)  
Fatty acid FM SM  SMg-1.5 kGy SMg-3.0 kGy 
Myristic 14:0 5.70 ± 0.07a 5.09 ± 0.40b 5.07 ± 0.12b 5.12 ± 0.04a 
Pentadecanoic 15:0 0.40 ± 0.02a 1.48 ± 0.16b 1.28 ± 0.11b 1.41 ± 0.05b 
Palmitic 16:0 13.53 ± 0.18a 21.68 ± 1.70b 21.06 ± 0.44b 22.61 ± 0.22b 
Margaric 17:0 0.43 ± 0.05 0.47 ± 0.38 0.65 ± 0.07 0.63 ± 0.03 
Stearic 18:0 2.79 ± 0.05a 6.23 ± 0.24bc 5.56 ± 0.05bd 5.89 ± 0.20b 
Arachidic 20:0 0.16 ± 0.03a 0.44 ± 0.03b 0.51 ± 0.01c NDd 
å SFA 23.01 ± 0.19a 35.39 ± 2.06b 34.13 ± 0.57b 35.65 ± 0.07b  
• Myristovaccenic 14:1 0.21 ± 0.02 0.21 ± 0.04 0.26 ± 0.01 0.28 ± 0.04 
• Palmitoleic  16:1 (n-7) 4.10 ± 0.08a 5.89 ± 0.49b 6.08 ± 0.08b 6.07 ± 0.24b 
Oleic  18:1 (n-9) 15.06 ± 2.23a 9.15 ± 0.74b 9.89 ± 0.17b 10.21 ± 0.39b 
• Vaccenic 18:1 (n-7) 3.02 ± 0.11 2.68 ± 0.52 2.73 ± 0.00 2.74 ± 0.03 
• Gondoic 20:1 (n-9,11) 9.43 ± 1.96a 1.15 ± 0.06b 1.43 ± 0.75b 0.99 ± 0.02b 
Paullinic 20:1 (n-7) 0.31 ± 0.02 0.38 ± 0.03 0.38 ± 0.01 0.35 ± 0.00 
Setoleic 22:1 (n-11) 12.35 ± 0.08a 0.37 ± 0. 10b 1.04 ± 1.01b 0.38 ± 0.02b 
• Erucic 22:1 (n-9) NDa 0.23 ± 0.11a 0.35 ± 0.14b 0.17 ± 0.07a 
• Nervonic  24:1 (n-9) 0.81 ± 0.14 0.62 ± 0.07 0.72 ± 0.10 0.75 ± 0.04 
å MUFAs 45.29 ± 0.92a 20.68 ± 1.93b 22.88 ± 1.90b 21.95 ± 0.15b 
• 9,12-Hexadecadienoic 16:2 (n-4) 0.37 ± 0.01a 0.75 ± 0.18b 0.62 ± 0.10a 0.75 ± 0.10b 
6,9,12-Hexadecatrienoic  16:3 (n-4) 0.40 ± 0.02a 1.59 ± 0.00b 1.47 ± 0.13b 1.57 ± 0.10b 
11,14-Octodecadienoic 18:2(n-4) 0.09 ± 0.00a 0.47 ± 0.02b 0.45 ± 0.12b 0.50 ± 0.00b 
8,11,14-
Octodecatrienoic 18:3 (n-4) 3.86 ± 0.04
a 0.92 ± 0.12b 1.19 ± 0.16b 1.08 ± 0.08b 
å (n-4) 4.73 ± 0.05a 3.73 ± 0.28b 3.72 ± 0.20b 3.89 ±0.13b 
Palmitidonic 16:4 (ω-3) 0.35 ± 0.00 0.40 ± 0.09 0.47 ± 0.14 0.37 ± 0.02 
• α-Linolenic 18:3 (ω-3) ND 0.02± 0.03 ND ND 
• Stearidonic 18:4 (ω-3) 0.19 ± 0.04 ND  ND ND 
• Dihomo-α-linolenic  20:3 (ω-3) 0.15 ± 0.02 1.25 ± 1.58 0.13± 0.01 0.13 ± 0.00  
• Eicosatetraeonic 20:4 (ω-3) 0.92 ± 0.06a 0.42 ± 0.00b 0.45 ± 0.02b 0.44 ± 0.04b 
Eicosapentaenoic (EPA) 20:5 (ω-3) 6.91 ± 0.19 7.36 ± 0.68 7.32 ± 0.39 6.96 ± 0.08 
Heneicosapentaenoic  21:5 (ω-3) 0.40 ± 0.06 0.47 ± 0.12 0.33 ± 0.03 0.40 ± 0.02 
Docosapentaenoic  
(DPA) 22:5 (ω-3) 0.24 ± 0.09
a 1.42 ± 0.11b 0.58 ± 0.04c 1.04 ± 0.15d 
• Docosahexaenoic 
(DHA) 22:6 (ω-3)  8.78 ± 0.66
a 17.02 ± 1.98b 17.25 ± 0.48b 16.63 ± 0.18b 
å (ω-3) 17.95 ± 0.54a 28.36 ± 4.35b 26.53 ± 0.21b 25.97 ± 0.46b 
• Linoleic 18:2 (ω-6) 1.52 ± 0.01a 1.36 ± 0.00b 1.43 ± 0.01a 1.41 ± 0.11a 
• γ-Linolenic 18:3 (ω-6) 1.35 ± 0.06a 1.05 ± 0.05a 1.08 ± 0.00a 0.83 ± 0.22b 
• Dihomo-linoleic 20:2 (ω-6) 0.25 ± 0.01 0.32 ± 0.07 0.28 ± 0.04 0.27 ± 0.00 
• Dihomo-γ-linolenic 20:3 (ω-6) NDa  0.14 ± 0.04b 0.12 ± 0.01b 0.13 ± 0.01b 
• Arachidonic 20:4 (ω-6) 0.41 ± 0.08 1.11 ± 1.39 1.87 ± 0.12 2.09 ± 0.12 
å (ω-6) 3.54 ± 0.02 3.99 ± 1.33 4.78 ± 0.10 4.74 ± 0.24 
å PUFA 26.22 ± 0.51a 36.08 ± 2.75b 35.04 ± 0.52b 34.59 ± 0.11b 
 ω-3/ω-6 5.07 ± 0.12 7.72 ± 3.36 5.55 ± 0.08 5.49 ± 0.37 
PUFA/SFA 1.14 ± 0.03 1.02 ± 0.14 1.03 ± 0.00 0.97 ± 0.00 
Polyene Index 1.16 ± 0.05 1.13 ± 0.21 1.17 ± 0.02 1.04 ± 0.00 
Atherogenic index (AI) 0.54 ± 0.00a 0.79 ± 0.08b 0.76 ± 0.02b 0.82 ± 0.00b 
Thrombogenic index (TI) 0.26 ± 0.01 0.32 ± 0.06 0.32 ± 0.00 0.35 ± 0.01 
a, b, c, d, e Means of with different superscripts within a row are significantly different at a < 0.05. å = sum; SFA = 
saturated fatty acids; MUFA = monounsaturated fatty acids; PUFA = polyunsaturated fatty acids; ND = not 
detected. 
 128 
University of Ghana http://ugspace.ug.edu.gh
6.3.1.3 Amino acid composition 
The total amino acids (Table 6.3) in the smoked samples were all significantly higher (p < 0.05) 
than that of the FM (i.e. 190.39, 397.36, 408.73 and 449.46 mg/g for FM, SM, SMg-1.5 kGy 
and SMg-3.0 kGy respectively). Irradiation increased the amino acid composition, however 
this was not statistically different (p > 0.05) from the SM samples or between the two doses 
(1.5 and 3.0 kGy). The essential amino acids (EAA) accounted for a greater proportion (about 
55%) of the total, with FM significantly lower than SM, SMg-1.5 kGy and SM-3.0 kGy 
(106.31, 211.50, 228.98 and 252.82 mg/g respectively). With the exception of methionine, all 
other EAAs were significantly higher (p < 0.05) in the smoked samples than FM. Histidine 
concentration was significantly higher (p < 0.05) in SMg-3.0 kGy than SM (i.e. 25.57 and 
15.84 mg/g respectively). The non-essential amino acids (NEAA) followed a similar pattern as 
the EAAs, with SMg-3.0 kGy having the highest proportion (78.57 mg/g) and FM having the 
lowest (33.43 mg/g). Serine and alanine were significantly higher (p < 0.05) in all smoked 
samples, with aspartic acid being significantly higher in the SMg samples relative to both FM 
and SM. The total conditionally indispensable (CI) amino acids were significantly higher in all 
smoked samples than FM, and ranged between 50.65 to 118.06 mg/g for FM and SMg-3.0 kGy 
respectively). Glycine and proline were significantly higher (p < 0.05) in all smoked samples, 
cysteine was only significantly higher (p < 0.05) in the SMg samples, whereas there was no 
significant difference in glutamine and arginine between the FM and smoked samples. The 
ratio of EAA to NEAA was not significantly different (p > 0.05) between all samples, and 
ranged from 1.35 to 1.45. 
 
 
 
 
 129 
University of Ghana http://ugspace.ug.edu.gh
Table 6.3: Amino acid composition of fresh (F), smoked (S) and gamma irradiated (g) 
mackerel (M) (mg/g)  
Amino acids FM SM  SMg-1.5 kGy SMg-3.0 kGy 
Essential/Indispensable (EAA)     
Threonine  9.23 ± 0.89a 17.95 ± 1.08b 18.10 ± 1.03b 18.08 ± 1.90b 
Methionine  4.93 ± 1.10 9.45 ± 0.26 12.07 ± 1.28 11.87 ± 3.99 
Valine  19.80 ± 1.00a 41.67 ± 2.54b 43.84 ± 2.32b 49.38 ± 5.59b 
Histidine  7.06 ± 1.41a 15.84 ± 1.02bc 22.01 ± 2.25b 25.57 ± 1.87bd 
Lysine  25.26 ± 0.36a 46.52 ± 4.29b 48.30 ± 1.11b 51.92 ± 4.28b 
Leucine  13.27 ± 0.49a 29.26 ± 1.10b 27.48 ± 2.01b 30.81 ± 2.52b 
Phenylalanine  7.36 ± 0.07a 16.26 ± 0.49b 14.97 ± 1.11b 16.65 ± 0.19b 
Tryptophan  6.44 ± 0.39a 12.98 ± 0.29b 12.57 ± 0.35b 13.82 ± 1.25b 
Isoleucine  12.96 ± 1.86a 31.58 ± 0.07b 29.65 ± 1.26b 34.74 ± 3.20b 
å EAA 106.31 ± 4.03a 211.50 ± 10.61b 228.98 ± 7.94b 252.82 ± 24.79b 
Non-essential/Dispensable (NEAA)     
Serine  7.58 ± 0.74a 14.12 ± 0.31b 15.40 ± 0.89b 16.70 ± 2.19b 
Alanine  7.97 ± 0.39a 17.99 ± 0.69b 18.11 ± 0.48b 19.67 ± 2.24b 
Aspartic acid  17.89 ± 0.63a 35.78 ± 3.55a 39.38 ± 3.31b 42.20 ± 7.44b 
å NEAA 33.43 ± 1.72a 67.90 ± 4.56b 72.89 ± 4.69b 78.57 ± 11.87b 
Conditionally indispensable (CI)     
Glycine  6.29 ± 0.42a 15.63 ± 1.07b 15.11 ± 0.25b 17.52 ± 2.33b 
Proline  5.21 ± 0.55a 12.63 ± 1.28b 13.12 ± 1.02b 14.18 ± 1.25b 
Glutamine  28.72 ± 2.32 52.60 ± 0.56 52.51 ± 11.80 57.30 ± 9.95 
Cysteine  1.10 ± 0.29a 2.32 ± 0.12a 2.60 ± 0.31b 2.67 ± 0.48b 
Arginine  9.33 ± 0.91 24.77 ± 6.25 23.51 ± 3.77 26.38 ± 5.91 
å CI 50.65 ± 3.91a 107.96 ± 9.28b 106.86 ± 14.50b 118.06 ± 19.92b 
å TAA 190.39 ± 9.66a 397.36 ± 24.45b 408.73 ± 27.12b 449.46 ± 56.59b 
å EAA/å NEAA+CI 1.27 ± 0.04 1.26 ± 0.04 1.28 ± 0.09 1.29 ± 0.08 
a, b, Means of with different superscripts within a row are significantly different at a < 0.05. 
 130 
University of Ghana http://ugspace.ug.edu.gh
6.3.1.4 pH 
The pH increased in all the smoked samples (not significantly though, p > 0.05) relative to FM 
(Figure 6.1). There was a general increase in pH during the refrigerated storage of the smoked 
mackerel, with that in SMg samples higher than SM. Samples stored at ambient temperature 
(SM-A, SMg-1.5-A and SMg-3.0-A) decreased by Day 5 of storage. Again, the changes 
between the different products, storage conditions and days were not statistically significant (p 
> 0.05).  
FM SM-A SM-R SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-3.0-R
7.5
7.0
6.5
6.0
5.5
5.0
0 1 5 10 15 25 45 65
Storage time (days)
 
Figure 6.1: pH of fresh (F), smoked (S) and gamma irradiated (g) mackerel (M) stored at 
ambient (A) and refrigerated (R) temperatures for 65 days. Refrigerated (R = 2-4°C) and 
ambient temperatures (A = 27-30°C) 
 
6.3.1.5 Total volatile base (TVB) 
The total volatile base (TVB) increased significantly (p < 0.05) in all smoked samples (Day 1), 
compared to the FM (Figure 6.2). The SMg samples were however significantly lower (p < 
0.05) than the SM ones on Day 1. By Day 5, the SMg samples stored at ambient temperature 
had increased TVB content (84.70 mg N/100g and 87.15 mg N/100g for SMg-1.5-A and SMg-
3.0-A respectively), which were significantly higher (p < 0.05) than the SM samples. There 
was, however, a general decreasing trend in TVB during the rest of the storage period at 
 131 
pH
University of Ghana http://ugspace.ug.edu.gh
refrigerated temperature. This trend was however not significant different (p > 0.05) in terms 
of irradiation treatment and storage days.  
  
FM SM-A SM-R SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-3.0-R
120
100
80
60
40
20
0
0 1 5 10 15 25 45 65
Storage time (days)
 
Figure 6.2: Total volatile base (TVB) content of fresh (F), smoked (S) and gamma 
irradiated (g) mackerel (M) stored at ambient (A) and refrigerated (R) temperatures for 
65 days. Refrigerated (R = 2-4 °C) and ambient temperatures (A = 27-30°C) 
 
6.3.1.6 Peroxide value (PV) 
The PVs for FM, SM, SMg-1.5 kGy and SM-3.0 kGy were 10.34, 10.43, 16.85 and 9.61 meq 
O2/kg respectively (Figure 6.3), with SMg-1.5 kGy being significant higher (p < 0.05) than all 
the other samples. On Day 5, there was a reduction in PV for SM-A and SM-R and a significant 
increase (p < 0.05) in all SMg samples, relative to SM, irrespective of their storage temperature. 
SMg-1.5 kGy was also significantly higher (p < 0.05) under refrigeration, compared to ambient 
storage. There was a general increase in PV for all samples during storage at refrigerated 
temperature, in the order of SM < SMg-1.5 kGy < SMg-3.0 kGy. On Days 10 and 45, SM was 
significantly lower (p < 0.05) than both SMg samples, whereas SMg-3.0 kGy was significantly 
higher (p < 0.05) than SM (on Day 15) and also SM and SMg-1.5 kGy (on Day 25). On Day 
65 however, there were no significant differences in the PV for all samples. 
 132 
TVB (mg N/100g)
University of Ghana http://ugspace.ug.edu.gh
FM SM-A SM-R SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-3.0-R
240
210
180
150
120
90
60
30
0
0 1 5 10 15 25 45 65
Storage time (days)  
 
Figure 6.3: Peroxide value (PV) of fresh (F), smoked (S) and gamma irradiated (g) 
mackerel (M)  
 
6.3.1.7 Free fatty acid (FFA) 
Fresh mackerel had an FFA of 8.84 g/100g fat, which was significantly reduced (p < 0.05) after 
smoking and irradiation (Figure 6.4). By Day 5, SMg-3.0-A had the highest FFA content (7.8 
g/100g fat), which was significantly higher (p < 0.05) than all the other samples, irrespective 
of their storage temperature. There is a general increase in FFA during storage, with SMg-1.5-
R having a significantly higher (p < 0.05) level (9.15 g/100g) than SM-R and SMg-3.0-R (5.38 
and 4.37 g/100g respectively) on Day 25, after which the levels decreased to 4.01, 4.69 and 
4.07 g/100g (SM-R, SMg-1.5-R and SMg-3.0-R respectively) on Day 65. 
 
 133 
PV  (meq O2/kg fat) 
University of Ghana http://ugspace.ug.edu.gh
FM SM-A SM-R SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-3.0-R
11
9
8
6
5
3
2
0
0 1 5 10 15 25 45 65
Storage time (days)
 
Figure 6.4: Free fatty acid (FFA) composition of fresh (F), smoked (S) and gamma 
irradiated (g) mackerel (M) stored at ambient (A) and refrigerated (R) temperatures for 
65 days. Refrigerated (R = 2-4 °C) and ambient temperatures (A = 27-30°C) 
 
6.3.2 Microbial analyses 
The microbial quality of non-irradiated and irradiated smoked fish stored for 65 days is 
presented in Figure 6.5.  
 
6.3.2.1 Total mesophilic count (TVC) 
The TVC for the FM was 2.5 log CFU/g, and this increased in the irradiated samples on Day 1 
(Figure 6.5a). On Day 5, the counts decreased in all samples, except for SM-A, which was 
significantly higher (p < 0.05). The counts in SMg-1.5-R were further significantly lower (p < 
0.05) than in SM and SMg-3.0-A. From Day 10, there was a general increase in TVC for all 
samples with Day 65 recording counts of 2.2, 3.3 and 1.4 log CFU/g in SM-R, SMg-1.5-R and 
SMg-3.0-R respectively. The samples however did not significantly differ (p > 0.05) during 
this time.  
 134 
FFA (g/100g fat)
University of Ghana http://ugspace.ug.edu.gh
FM SM-A SM-R
FM SM-A SM-R SMg-1.5-A SMg-1.5-R SMg-3.0-A
SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-3.0-R
SMg-3.0-R
7 7
6 6
5 5
4 4
3 3
2 2
1 1
0 0
0 1 5 10 15 25 45 0 1 5 10 15 25 45 65
(a) Storage time (days) (b)
Storage time (days)
   
FM SM-A SM-R FM SM-A SM-R
SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-1.5-A SMg-1.5-R SMg-3.0-A
SMg-3.0-R SMg-3.0-R
7 7
6 6
5
5
4
4
3
2 3
1 2
0 1
0 1 5 10 15 25 45
(C) Storage time (days) 0
(d ) 0 1 5 10 15 25 45 65  
FM SM-A SM-R FM SM-A SM-R
SMg-1.5-A SMg-1.5-R SMg-3.0-A SMg-1.5-A SMg-1.5-R SMg-3.0-A
SMg-3.0-R SMg-3.0-R
7 7
6 6
5 5
4 4
3 3
2 2
1 1
0 0
0 1 5 10 15 25 45 0 1 5 10 15 25 45 65
Storage time (days)
(e) Storage time (days)( f)  
Figure 6.5: Total mesophilic (a), S. aureus (b), faecal coliform (c), E. coli (d), C. 
perfringens (e) and yeast and mould (f) counts in fresh (F), smoked (S) and gamma 
irradiated (g) mackerel (M) stored at ambient (A) and refrigerated (R) temperatures for 
65 days. Refrigerated (R = 2-4 °C) and ambient temperatures (A = 27-30°C)
 
 135 
TVC (log CFU/g)
C. perfringens (log CFU/g) Faecal coliform (log CFU/g)
Yeast and moulds (log CFU/g) E. coli (log CFU/g) S. aureus (log CFU/g)
University of Ghana http://ugspace.ug.edu.gh
6.3.2.2 Faecal coliform and E. coli counts 
The faecal coliform and E. coli counts in FM were 2.2 and 1.4 log CFU/g respectively, which 
increased after smoking to 2.7 and 2.0 log CFU/g in SM on Day 1 (Figure 6.5b, c). Irradiation 
resulted in a significant decrease (p < 0.05) in SMg-1.5-A, relative to SM. The counts generally 
increased in all samples, except SM-A, on Day 5. The changes however were not statistically 
different (p < 0.05). From Day 10 to 65, faecal coliform and E. coli were below detectable 
limits in SM-3.0-R. SMg-1.5-R only recording counts on Day 45 (1.7 and 1.2 log CFU/g for 
faecal coliform and E. coli respectively), while SM-R had counts of 0.9 log CFU/g only on 
Day 65. 
 
6.3.2.3 Staphylococcus aureus  
S. aureus counts reduced after smoking (Figure 6.5d), from 2.3 log CFU/g in FM to 1.6 log 
CFU/g in SMg-3.0-A (which were significantly lower (p < 0.05) than the other smoked 
samples). There were further decreases by Day 5, with the irradiated samples being 
significantly lower than SM, at both refrigerated and ambient storage temperatures. S. aureus 
counts increased in all samples on Day 10, after which they generally increased. The highest 
count occurred on Day 45 in SMg-1.5-R (3.0 log CFU/g), which was significantly higher than 
SM-R and SMg-3.0-R.  
 
6.3.2.4 Clostridium perfringens  
C. perfringens were below detectable limits in FM but increased in SM to 2.3 log CFU/g on 
Day 1 (Figure 6.5e). Irradiation further reduced the counts, with SMg-1.5-A being significantly 
lower. The counts increased in all samples stored at ambient temperature, but the irradiated 
samples were significantly lower than SM, on Day 5. The refrigerated samples had reduced 
counts, with SMg-3.0-R having significantly lower (p < 0.05) counts. The counts generally 
 136 
University of Ghana http://ugspace.ug.edu.gh
decreased during storage, with Days 15, 25 and 65 having counts below detection limits in all 
irradiated samples.  
 
6.3.2.5 Yeast and moulds  
Yeast and moulds were also below detection limits in FM and SM samples (Figure 6.5f). 
Irradiated samples had increased counts, i.e. 1.3 and 2.5 log CFU/g in SMg-1.5-and SMg-3.0 
respectively. Moulds were visually observed in smoked samples stored at ambient temperature 
on Day 5. Laboratory analysis estimated increased counts for SM and SMg-1.5 and decreased 
counts in SMg-3.0, though these were not statistically different (p > 0.05). The counts in SMg-
1.5-R were significantly higher (p < 0.05) on Days 10 and 45 (1.8 and 1.7 log CFU/g 
respectively), whiles SMg-3.0-R had significantly higher counts on Day 25. All samples had 
counts below detectable limits on Day 65. 
  
6.3.3 Insect infestation 
Visual inspection of the samples on Day 5 showed insect larvae in all samples, both irradiated 
and non-irradiated, stored at ambient temperature. 
 
6.3.4 Colour analysis 
The instrumental skin colour measured showed irradiation increased the L* and b* from 34.45 
to 43.91 and 5.93 to 11.66 in SM and SMg-3.0 kGy respectively (Table 6.4). The a* was also 
highest in SMg-1.5 kGy. The Hoab and c* were also highest in the SMg-3.0 kGy samples. 
Similar results were obtained for the muscle colour. These differences in skin and muscle 
colour between the SM and irradiated samples were however not significantly different (p > 
0.05).  
 
 137 
University of Ghana http://ugspace.ug.edu.gh
Table 6.4: Skin and muscle colour characteristics of smoked (S) and gamma irradiated 
(g) mackerel (M)  
Parameter SM  SMg-1.5 kGy SMg-3.0 kGy 
Lightness (L*) 34.45±1.31 39.62±2.62 43.91±10.15 
Redness (a*) -0.62±0.66 0.31±2.16 -1.10±0.19 
Skin Yellowness (b*) 5.93±5.16 5.24±0.88 11.66±8.20 
Hue angle (Hoab) 76.77±17.43 73.94±1.80 82.50±6.15 
Chromaticity (c*) 6.06±5.00 5.46±0.96 11.74±8.13 
  
   
Lightness (L*) 60.56±12.25 66.61±6.75 66.62±0.86 
Redness (a*) 2.66±2.49 3.96±1.24 1.91±0.47 
Muscle Yellowness (b*) 20.03±0.95 17.28±2.13 20.66±1.05 
Hue angle (Hoab) 82.33±7.33 76.83±5.46 84.76±1.03 
Chromaticity (c*) 20.29±0.61 17.77±1.80 20.75±1.10 
 
 
6.3.5 Sensory analysis 
The mean scores for SM, SMg-1.5 kGy and SMg-3.0 kGy were 2.3, 2.3 and 2.6 respectively, 
on the 5-point scale (Figure 6.6). These were not statistical different (p > 0.05). The results 
from the Check-all-that-apply (CATA) (Figure 6.7), showed important attributes that may have 
accounted for the perceived differences between the fish treatments and the control sample 
were juicy texture, umami flavour, smoked flavour and aroma and herring-like aroma. Other 
attributes related to the saltiness (SMg-1.5 kGy and SMg-3.0 kGy saltier and slightly softer 
than SM). 
 138 
University of Ghana http://ugspace.ug.edu.gh
5
4
3
2
1
0
SM SMg-1.5 kGy SMg-3.0 kGy
Samples
 
 
Figure 6.6: Mean intensity scores for difference-from-control test (Scale: 0 = No 
difference; 1= Very slight difference; 2 = Slight difference; 3 = Moderate; 4 = Very different 
and 5 = Extreme difference) 
 
 
SM SMg - 1.5 kGy SMg - 3 kGy
20
18
16
14
12
10
8
6
4
2
0
ing ke ed gh y i d t
s y i r
ok -li ok ou Fl
ak am ok
e eng ro
u
uic mb J a Ot
he
 lo ing Sm X-
T n -
y X
- Um m ri Fi- S t _ F -U
m
Dr- He
rr - T T - s M
- A
R FL L -A F F
P F F M
A
A AR M
Sensory attributes
 
Figure 6.7: Frequency of responses for CATA attributes indicating perceived differences 
(AP- = Appearance; AR- = Aroma; TX- = Texture; FL- = Flavour; MF- = Mouthfeel; AF- = Aftereffect) 
 139 
Frequency of Responses Degree of difference from labelled 
reference sample 
University of Ghana http://ugspace.ug.edu.gh
6.4 Discussion 
The moisture content of SM and SMg samples were significantly reduced (compared to FM) 
and met the industrially specified limit (< 65%) for ‘smoked finished products’, which was 
consistent with Cardinal et al. (2001) and Asamoah (2018). The protein content was also 
significantly increased in the smoked samples relative to FM. However, there were no 
significant differences in the moisture, protein, fat, ash and pH between the SM and SMg 
mackerel. The sensory and colour analyses also found no statistical difference between the SM 
and SMg samples. This was in consonance with a study by Badr (2012) on the effect of 
irradiation on cold-smoked salmon that found that the proximate composition and sensory 
acceptability were not affected by doses up to 3 kGy. A similar study by Dvorak, Kratochv, & 
Grolichov (2005) found no differences in colour and pH of Onchorynchus mykiss irradiated at 
a dose of 3 kGy. The proximate composition of the SM and SMg samples did not significantly 
change during the 65 days of storage. This could have been as a result of the refrigerated 
storage, since Al-Reza, et al. (2015) reported protein and fat degradation in sun-dried fish 
stored at ambient temperature for 60 days. Silva, Mendes, Nunes, & Empis (2006) further 
indicated that irradiation up to 10 kGy had no significant impact on proteins in horse mackerel 
stored in ice. 
 
Smoking resulted in a significant increase in SFA, MUFA and PUFA, which could be due to 
the loss of water via evaporation during the smoking process (Bouriga, Bejaoui, Jemmali, 
Quignard, & Trabelsi, 2020). The proportions of fatty acids in FM was of the order MUFA > 
PUFA > SFA, similar to findings by Nogueira, Cordeiro, & Aveiro (2013). In contrast, the SM 
and SMg samples were of the order PUFA ≈ SFA > MUFA, which was not consistent with 
results from Cyprian (2015), Chaula, et al. (2019) and Erkan & Özden (2007). The individual 
fatty acids that contributed the greatest proportions in all samples were palmitic, myristic and 
 140 
University of Ghana http://ugspace.ug.edu.gh
stearic acids (SFAs), oleic acid (MUFA) and eicosapentaenoic (EPA) and docosahexaenoic 
acids (DHA) (PUFAs). Similar findings were reported for irradiated sea breams (Erkan & 
Özden, 2007), Atlantic chub mackerel and other species captured in north-eastern Altantic 
(Nogueira, Cordeiro, & Aveiro, 2013), golden grey mullet and gold band goatfish 
(Küçükgülmez, Yanar, Çelik, & Ersor, 2018) and tilapia (Surendra, Edirisinghe, & 
Rathnayake, 2018). Oleic acid is important in human nutrition as it stimulates bile secretion, 
which aids in the digestion and absorption of fats (Nogueira, Cordeiro, & Aveiro, 2013). Again, 
EPA and DHA are also hypotriglyceridemic and very important in the prevention of 
cardiovascular and inflammatory diseases in humans (Nogueira, Cordeiro, & Aveiro, 2013; 
Cyprian, 2015). The high levels of EPA and DHA in FM, SM and SMg implied that they were 
of good nutritional quality.  
 
There was a higher proportion of ω-3 PUFAs compared to ω-6, which agrees with the assertion 
that marine fish are richer in ω-3 than ω-6 (Osman, Suriah, & Law, 2001). The ω-3:ω-6 ratio 
is a good index for comparing the relative nutritional value of fish (Küçükgülmez, Yanar, 
Çelik, & Ersor, 2018). The ratios from the present study were between 5.07 and 7.72 (FM and 
SM respectively), and these were within the estimates of 2.67 to 12.61 in marine species (Bayir, 
Haliloglu, Sirkecioglu, & Aras, 2006). Küçükgülmez, Yanar, Çelik, & Ersor (2018) stated that 
higher rations (>1) are of great importance in order as this diminishes the risks of coronary 
heart diseases, plasma lipid levels, and cancer in humans.  
 
Lipid quality was assessed using three indices: the polyene index (PI), index of atherogenicity 
(AI) and index of thrombogenicity (TI). The PI measures oxidative damage to PUFAs 
(Rodríguez, et al., 2007). Consuming foods rich in fatty acids can directly affect the simulation 
or preclusion of atherosclerosis and coronary thrombosis due to their effect on blood 
 141 
University of Ghana http://ugspace.ug.edu.gh
cholesterol and low-density lipoprotein (LDL) cholesterol concentrations (Omri, et al., 2019). 
According to Garaffo, et al. (2011), the index of atherogenicity (AI) and index of 
thrombogenicity (TI) can be used to characterise these effects. The AI denotes the relationship 
between the main pro-atherogenic (saturated fatty acids that favour the adhesion of lipids to 
cells of the immunological and circulatory system) and anti-atherogenic (unsaturated fatty 
acids that inhibit the aggregation of plaque and diminishes the levels of esterified fatty acid, 
cholesterol, and phospholipids, there-by preventing the appearance of micro- and macro- 
coronary diseases) fatty acids. The index of thrombogenicity (TI) defines the relationship 
between the pro-thrombogenetic (saturated) and the anti-thrombogenetic fatty acids (MUFAs, 
ω-6 PUFAs and ω-3 PUFAs) that have the tendency to form clots in blood vessels (Garaffo, et 
al., 2011). According to Łuczyńska, Paszczyk, Nowosad, & Łuczyński (2017), AI and TI 
greater than 1.0 can be detrimental to human health. The results from the present study show 
AI and TI values below 1.0 (with FM having the lowest estimates), indicating there are reduced 
risks of atherosclerosis and coronary thrombosis from consuming these products. These 
estimates were however higher than those for fresh mackerel, shrimp and lobster (Rosa & 
Nunes, 2003), lower than estimates in raw, frozen, solar-dried and oven-dried sardinella 
(Telahigue, Hajji, Rabeh, & El Cafsi, 2013), but similar to those for bluefin tuna roes and salted 
‘Bottarga’ (Garaffo, et al., 2011). Küçükgülmez, Yanar, Çelik, & Ersor (2018) observed 
seasonal variations in AI, TI, and TI of gold band goatfish and golden grey mullet form the 
Mediterranean seas. The PUFA/SFA ratios calculated in the present study were about 1 
(between 0.97 to 1.14, FM and SMg-3.0 kGy respectively). Regulska-Ilow, et al. (2013) stated 
that a ratio ≥ 1 could decrease the possibility of atherosclerosis and coronary heart disease, 
which was consistent with the results of this study. Irradiation at 1.5 and 3.0 kGy did not have 
a significant effect on the fatty acid composition of smoked fish, which agrees with assertions 
by (Erkan & Özden, 2007).  
 142 
University of Ghana http://ugspace.ug.edu.gh
Amino acids form the molecular structure of proteins, responsible for the synthesis of 
hormones, enzymes, body tissues and other metabolic molecules (Salma & Nizar, 2015). They 
are classified as essential (EAA, i.e. indispensable as they cannot be synthesized de novo, and 
therefore must be supplied by diet); non-essential (NEAA, i.e. dispensable as they synthesised 
by the body) and conditionally indispensable (CI, as they are normally synthesised by the body 
but can be limited under special pathophysiological conditions) (Otten, Hellwig, & Meyers, 
2006). A total of 17 amino acids (made up of 9 EAA, 3 NEAA and 5 CI) were identified in this 
study. The presence of the 9 EAA makes fish protein a ‘complete protein’, as was reported by 
Otten, Hellwig, & Meyers (2006). The total amino acid content in FM, SM and SMg-1.5 kGy 
and Smg-3.0 kGy were 190.39, 397.36, 408.73 and 449.46 respectively. The dominant amino 
acids were valine, lysine, leucine and isoleucine (EAA); aspartic acid (NEAA); and glutamine 
(CI), which was comparable with findings by Rosa & Nunes (2003), Erkan & Özden (2007) 
and Atowa, Nwabu, & Ogiedu (2014). The benefits of amino acids in human nutrition have 
been examined by several authors. Mohanty, et al. (2014) reported that aspartic acid, 
methionine arginine and glycine play an improtant role in wound healing and also in 
maintaining the solubility and ionic properties of proteins. Sarma, et al. (2013) further reports 
that methionine, histidine, lysine and tryptophan have antioxidant properties. Again, Kim, et 
al. (1999) stated that glutamine, proline, aspartic acid, glycine and leucine have cytotoxic 
abilities to kill or damage cancer cells. Irradiation at 1.5 and 3.0 kGy did not significantly affect 
these amino acids, meaning that the products i.e. FM, SM and SMg, were all of very good 
nutritional status. 
 
Amino acids can also be used as quality indices in seafood. Histidine, tyrosine, arginine, 
tryptamine and lysine are very important during fish spoilage, as they can produce biogenic 
amines (via decarboxylation by microorganisms), like histamine, tyramine, agmatine, 
 143 
University of Ghana http://ugspace.ug.edu.gh
tryptophan and cadaverine respectively (Biji, Ravishankar, Venkateswarlu, Mohan, & Gopal, 
2016). From the results, the fresh and smoked samples had appreciable levels of these amino 
acids and hence care should be taken to ensure that microbial activities are diminished during 
storage to inhibit biogenic amine formation. Again, glutamic acid, aspartic acid, alanine and 
glycine, are responsible for characteristic flavor and taste of fish Erkan & Özden (2007). The 
levels of glycine and alanine were not statistically different between the SM and SMg samples. 
Only aspartic acid was significantly higher in SMg samples compared to SM but this was 
probably not sufficient to change the flavour and taste of the irradiated samples, as evidenced 
by the results from sensory analysis. Higher irradiation doses however can change the flavour 
and taste in foods, as reported by Erkan & Özden (2007). Smoking has however been shown 
to reduce the lysine content in foods (as they react with carbonyls in smoke during browning 
or Mailard reaction) (Kaya, Turan, & Erdem, 2008). The result from the present study however 
contradicts this assertion, as lysine levels significantly increased in the smoked products.  
 
The ratio of EAA to NEAA highlights the important contributions of each to the diet. From the 
present study, the ratios were 1.27, 1.26, 1.28 and 1.29 for FM, SM, SMg-1.5 kGy and SMg-
3.0 kGy respectively. These ratios were higher than those reported by Rosa & Nunes (2003), 
Salma & Nizar (2015) and Kaya, Erdem, & Turan (2014). It can therefore be inferred that fresh, 
smoked and irradiated (at 1.5 and 3.0 kGy) mackerel are important sources of essential amino 
acids.  
 
The total volatile base (TVB) is a measure of the fitness for consumption of fish products 
(Asamoah, 2018). The limits for fresh and smoked/dried fish are 25-30 mg N/100g and 100-
200 mg N/100g (Özoğul & Özoğul, 2000; EU Directive, 2008). From the results, FM met this 
requiement and as such was fit for smoking and consumption. There was a significant increase 
 144 
University of Ghana http://ugspace.ug.edu.gh
in TVB after smoking, which Goulas & Kontominas (2005) attributed to the partial drying and 
subsequent concentration of TVB contents in the smoked products. The levels of TVB in the 
non-irradiated and irradiated smoked mackerel were below 100 mg N/100g, even by Day 65 
of storage at refrigerated temperature and Day 5 of ambient storage. This gave an indication 
that the products were fit for human consumption. Özoğul & Özoğul (2000) reported that TVB 
was not a reliable indicator of early quality changes as levels usually increase appreciably 
during the later stages of storage as a result of bacterial activity. Plahar, Nerquaye-Tetteh, & 
Annan (1999), in a study in Ghana obtained higher TVB levels of 120.6 and 133.8 mg/100g in 
freshly smoked Sardinella and anchovy respectively and 108.8 mg N/100g and 294.8 mg 
N/100g in the products stored for 6 months at ambient temperatures.  
 
The peroxide value (PV) measures primary oxidation of lipids in foods (Salaudeen, 2013). PV 
values between 20-40 meq O2/kg may indicate oxidation, and thus, developing spoilage 
(Daramola, Fasakin, & Adeparusi, 2007). The PV of FM was below this guideline, indicating 
it was of good quality for smoking. Smoking did not significantly increase the PV, which, could 
mean that the antioxidant properties of smoke may have exceeded the effects of hot smoking 
temperature on lipid oxidation (Marc, Kaakeh, & Mbofung, 1998; Cyprian, 2015). There was 
a general increase in PV during storage, which was more pronounced in irradiated samples 
than in non-irradiated ones, that was in agreement with results from Surendra, Edirisinghe, & 
Rathnayake (2018). Irradiation-induced lipid peroxidation has been shown be dose-dependent, 
and can be sped up as a result of the formation of lipid-free radicals, which under aerobic 
conditions can break down hydroperoxides and destroy antioxidants (Fan & Sommers, 2013). 
Irrespective of the storage temperature, the PVs were however below the suggested guideline, 
up till Day 10, 15 and 25 for SMg-3.0 kGy, SMg-1.5 kGy and SM respectively, implying a 
delay in oxidation, with respect to the dose rate. Gecgel (2013) also obtained a positive 
 145 
University of Ghana http://ugspace.ug.edu.gh
correlation between the increase in PV, the dose applied and storage duration for irradaited 
meatballs. Ahn & Love (1999) obtained an increase in oxidation in irradiated pork products 
(immediately after irradiation), relative to non-irradiated ones but this difference disappeared 
during storage, which was contrary to results from the present study. Cozzo-Siqueira, Oetterer, 
& Gallo (2003) however found that oxidation was higher in irradiated tilapia and Spanish 
mackerel, when doses were higher than 10 kGy and fish was stored under low refrigeration. 
 
Free fatty acid (FFA) is a tertiary product of rancidity and is produced by the hydrolysis of fish 
fats and oils, through lipase action (Cyprian, et al., 2017; Chaula, et al., 2019). According to 
Daramola, Fasakin, & Adeparusi (2007), rancidity of fish oils is usually noticeable when FFA 
is between 0.5-1.5 g/100g. From the results, all samples (fresh, non-irradiated and irradiated) 
had FFA values above this limit (with FM having the highest FFA), which could mean they 
had undergone lipase action. Smoking produced a significant reduction in FFA, which could 
have resulted from the hot smoking temperatures denaturing most enzymes that catalyse lipid 
hydrolysis and thereby liberating FFA (Cyprian, et al., 2017). The FFA levels increased during 
storage, with SM samples generally lower than the SMg samples (significantly so with SMg-
1.5 kGy on Day 25). This could mean that irradiation at the two doses may have accelerated 
the hydrolysis of glycerol-fatty acid esters, which in turn liberated the FFA (Cyprian, 2015) 
 
Fresh or processed seafood can be contaminated by microorganisms during processing or 
through cross-contamination from regular handling and packaging (Ehlermann, 2016). This 
can affect the quality and shelf life of seafood. The recommended levels of total mesophilic 
bacteria, faecal coliform, E. coli, yeast and mould and S. aureus (for human consumption) have 
been established by GSA (2013; 2019) as 7, 2, 3, 4 and 4 log CFU/g respectively for fresh and 
smoked fish in Ghana. That for C. perfringens is 4 log CFU/g (Center for Food Safety, 2014). 
 146 
University of Ghana http://ugspace.ug.edu.gh
From the results, FM and SM, SMg-1.5 kGy and SMg-3.0 kGy (irrespective of their storage 
temperatures) were within these limits during the study period, except for faecal coliform 
counts in FM, SM-A and SM-R (on Day 1) and SMg-3.0 kGy (on Day 5). Smoking alone was 
not enough to reduce the counts of faecal coliforms, as these may be thermo-tolerant (Marc, et 
al., 2014). Irradiation however reduced these counts, but doses of 4.0 kGy and above have been 
known to completely eliminate coliforms in smoked salmon (WHO, 2000). 
 
In terms of shelf life, the ambient-stored SM and SMg samples had a shelf life of 5 days, due 
to insect infestation and visible mouldiness. Similar findings were reported by (Plahar & 
Amevor, 2003) for smoked sardines stored at ambient temperatures. The authors, therefore, 
recommended the use of heat-sealed low-density polythene bags and frozen storage to extend 
shelf life. Non-irradiated and irradiated smoked samples stored at refrigerated storage were of 
good microbiological, chemical (TVB and pH) and nutritional quality by Day 65 (9 weeks), 
even though oxidation was evident (from FFA and PV results). Moini, et al. (2009) reported 
that irradiation at 3 kGy could control the microbial, protein and lipid oxidation in fresh 
rainbow trout stored at refrigerated temperature for up to 4 weeks. Duah, Emi-Reynolds, 
Kumah, & Larbi (2018) reported shelf lives of 5 and 9 weeks for non-irradiated and irradiated 
(2.5-10 kGy) smoke-dried anchovies, stored in polyethylene bags at ambient temperature. The 
authors recorded no insect infestation in the irradiated samples within the 9 weeks of storage. 
Aworh, Okparanta, & Oyedokun (2002) also obtained a shelf life of 4-6 months for irradiated 
(up to 6 kGy) smoke-dried catfish stored at ambient temperatures.  
 
6.5 Summary of findings 
The effect of irradiation at two doses (1.5 and 3 kGy) and storage temperature (ambient, 27-
30°C and refrigerated, 2-4°C) on the quality and shelf life of smoke mackerel was investigated. 
 147 
University of Ghana http://ugspace.ug.edu.gh
Irradiation at 1.5 and 3 kGy had no impact on the nutritional composition (protein, fat, fatty 
acid and amino acid compositions), sensory and colour characteristics of the smoked mackerel. 
With respect to storage, the non-irradiated and irradiated, kept at ambient temperature had a 
shelf life of 5 days, as a result of insect infestation and visible mouldiness. With refrigeration, 
the products were of good quality by Day 65, in terms of their microbiological and nutritional 
status. Lipid oxidation and hydrolysis however increased during storage, and were more 
pronounced in the irradiated mackerel. This study therefore indicates that fish smoking, with 
or without irradiation (at 1.5 and 3 kGy), combined with refrigeration is a very viable way to 
preserve fish even beyond 65 days.  
 
 
 148 
University of Ghana http://ugspace.ug.edu.gh
CHAPTER SEVEN 
7.0 CONCLUSION AND RECOMMENDATIONS 
7.1 Conclusion 
Three kilns, the Chorkor, Cabin and AGFS kilns representing uncontrolled, semi-controlled 
and controlled smoking technologies, were investigated based on their efficiency. Further, the 
Chorkor and Cabin were assessed based on the amount of flue gases produced. The yield was 
better in the Cabin and Chorkor but the fuel consumption/kg of smoked fish and processing 
rate were better in the AGFS. The firewood-operated kilns produced flue gases containing CO 
and PM well above Ghana’s EPA emmisions standard. For SO2, NOx however, the amount 
produced was below the emission standard and therefore it stands to reason that with better 
heat management, the firewood consumption, CO and PM emission could be decreased. This 
would help safeguard the health of the predominately female population who use these kilns 
and their environments. The Cabin could be a better alternative to the Chorkor since it 
performed better and was less expensive than the AGFS. However, the environmental benefits 
of the AGFS (use of LPG which meant deforestation would not be a problem) and high 
throughput capacity meant it could be the best alternative of the three.  
 
Smoking improved the physical, chemical, microbiological and sensory quality of mackerel 
and barracuda. Only colour and sensory analysis could statistically differentiate between the 
products from the two different smoking kilns. The Cabin-smoked products had the more 
traditional qualities of smoked fish (appearance, odour and flavour) that the gas-smoked 
products lacked, probably because of the indirect smoke generation and/or short contact 
between the fish and the smoke in the AGFS.  
 149 
University of Ghana http://ugspace.ug.edu.gh
The smoking process produced some carcinogenic substances such as PAHs and heavy metals 
in the processed fish. The fresh mackerel and barracuda were of good quality for smoking and 
posed minimal carcinogenic risks to consumers. With respect to the kilns, AGFS performed 
best, by producing smoked products with benzo(a)pyrene and PAH4 concentrations below the 
EU MLs (2 and 12 µg/kg respectively). The Cabin also produced smoked mackerel with 77% 
and 59% lower benzo(a)pyrene and PAH4 (only in Esa smoked fish) than the Chorkor. The 
levels of benzo(a)pyrene and PAH4 were however greater than the EU’s MLs in all Chorkor 
and Cabin smoked samples (except for benzo(a)pyrene in E-CSM and E-CSB). The potential 
carcinogenic risks were of least concern in the gas smoked and all barracuda samples, moderate 
in the Cabin smoked mackerel and high in the Chorkor smoked mackerel. It could therefore be 
inferred that the presence of PAHs in the smoked fish was due to the type of kiln, smoking 
method (direct and indirect) and fuel used (LPG and firewood) for the treatments. Again, the 
magnitude of the carcinogenic risks depended largely on the fish ingestion rate, with higher 
benzo(a)pyrene and PAH4 levels not always corresponding to increased risks (as shown in the 
barracuda). Heavy metal (Hg, Pb and Cd) contamination was negligible in fresh and smoked 
mackerel and barracuda. This indicated that the smoke emitted might not have contained 
significant amounts of heavy metals and thus contamination was avoided. With increased 
consumption of smoked products however, these risks are likely to increase thereby 
necessitating the need for the AGFS and the use of alternate sources of fuel in the Cabin. 
 
The effect of irradiation at two doses (1.5 and 3 kGy) and storage temperature (ambient, 27-
30°C and refrigerated, 2-4°C) on the quality and shelf life of smoke mackerel was investigated. 
Irradiation at 1.5 and 3 kGy had no impact on the nutritional composition (protein, fat, fatty 
acid and amino acid compositions), sensory and colour characteristics of the smoked mackerel. 
With respect to storage, the non-irradiated and irradiated samples, kept at ambient temperature 
 150 
University of Ghana http://ugspace.ug.edu.gh
had a shelf life of 5 days, as a result of insect infestation and visible mouldiness. With 
refrigeration, the products were of good quality by Day 65, in terms of their microbiological 
and nutritional status, though lipid oxidation was detectable. This study therefore indicates that 
fish smoking, with or without irradiation (at 1.5 and 3 kGy), combined with refrigeration is a 
very viable way to preserve fish even beyond 65 days.  
 
7.2 Recommendations  
The following recommendations are being made based on the findings from this research: 
 
7.2.1 Policy 
1. The Ministry of Fisheries and Aquaculture Development (MOFAD), National Fish 
Processors and Traders Association (NAFPTA) and non-governmental organisations 
(NGOs) should adopt and popularise the two smoking kilns. The cost of construction 
of the kilns could be subsidised to make them more affordable and attractive to 
processors. 
 
2. Fish processors should be trained in healthy fish handling, processing and storage and 
also on the use and maintenance of the kilns.  
 
7.2.2 Research 
1. To ascertain if the differences between the Cabin smoked and gas smoked samples 
would be acceptable, a consumer study should be undertaken. The results would help 
in the education (where necessary) and commercialisation of the products. 
 
 151 
University of Ghana http://ugspace.ug.edu.gh
2. There should be further PAH studies of smoke-dried fish using other sources of fuel in 
the Cabin and AGFS. There should also be regular data collection and testing of the 
samples from these kilns to ensure that they adhere to good manufacturing practices. 
 
3. The potential of irradiation on smoked fish using different packaging materials and 
longer shelf life study should be explored for fresh, and other processed seafoods in 
Ghana. Consumer education on the benefits of irradiation should also be carried out. 
  
 152 
University of Ghana http://ugspace.ug.edu.gh
REFERENCES 
Abboah-Offei, O. (2016). Analysis of heavy metal concentrations in smoked fish from the major 
food markets in the Accra-Tema Metropolis. Accra: Master Thesis in Occupational 
Hygiene. Univerity of Ghana. 
Abdel-Shafy, H. I., & Mansour, M. S. (2016). A review on polycyclic aromatic 
hydrocarbons:Source, environmental impact, effect on humanhealth and remediation. 
Egyptian Journal of Petroleum, 25, 107-123. 
Abraha, B., Admassu, H., Mahmud, A., Tsighe, N., Shui, X. W., & Fang, Y. (2018). Effect of 
processing methods on nutritional and physicochemical composition of fish: A review. 
MOJ Food Processing & Technology, 6(4), 376-382. 
Adeyeye, S. A. (2016). Fungal mycotoxins in foods: A review. Cogent Food & Agriculture, 2, 
11. 
Adeyeye, S. A., Oyewole, O. B., & Obadina, A. O. (2016). Assessment of quality and safety 
of traditional smoked spotted tilapia fish (Tilapia mariae) from Lagos State, Nigeria. 
Nutrition & Food Science, 46(1), 142-155. 
Adeyeye, S. A., Oyewole, O. B., Obadina, A. O., Omemu, A. M., & Omoniyi, S. A. (2017). 
Quality attributes, polycyclic aromatic hydrocarbon, and heavy metal profile of 
traditional drum-smoked Guinean barracuda fish from Lagos State, Nigeria. Journal of 
Aquatic Food Product Technology, 26(3), 296-307. 
Adeyeye, S. O., & Oyewole, O. B. (2016). An overview of traditional fish smoking In Africa. 
Journal of Culinary Science & Technology, 14(3), 198-215. 
Adu-Gyamfi, A., & Appiah, V. (2012). Enhancing the hygienic quality of some Ghanaian food 
products by gamma irradiation. Food and Nutrition Sciences, 219-223. 
Adu-Gyamfi, A., Banson, K. E., & Nketsia-Tabiri, J. (2011). Challenges to implementation of 
new and emerging technologies: the case of the utilization of ionizing radiation. 
 153 
University of Ghana http://ugspace.ug.edu.gh
International Conference on Creating and Sustaining Globally Competitive 
Businesses: Challenges For Developing Economies. Trinidad: WAITRO. 
Adu-Gyamfi, A., Riverson, N. M., Afful, N., & Appiah, V. (2014). Effect of gamma irradiation, 
packaging and storage on the microbiological quality of garden eggs. International 
Journal of Nutrition and Food Sciences, 3(4), 340-346. 
Adu-Gyamfi, A., Torgby-Tetteh, W., & Appiah, V. (2012). Microbiological quality of chicken 
sold in Accra and determination of D10-value of E. coli. Food and nutrition sciences, 
3, 693-698. 
Aheto, D. W., Adinortey, C. A., Essumang, D. K., Adjei, J., Ahadzi, E., Kwarteng, E., & 
Avega, B. (2017). Microbiological and Polycyclic Aromatic Hydrocarbons (PAHs) 
Analysis of Fish from selected Areas from Central and Western Regions of Ghana. 
Narragansett, RI: Coastal Resources Center, Graduate School of Oceanography, 
University of Rhode Island and SNV Netherlands Development Organisation: The 
USAID/Ghana Sustainable Fisheries Management Project (SFMP). 
Ahn, D. U., & Love, J. (1999). Effect of irradiation on lipid oxidation and off-flavor 
development in cooked pork products with different fatty acids and packaging. Swine 
Research Report, 59. 
Aidoo, B. (2017). Effects of different smoking methods on the formation of polyaromatic 
hydrocarbons (PAHs) and physicochemical quality in smoked Atlantic mackerel 
(Scomber scombrus) products. Master Thesis in Environment and Natural Resources. 
University of Iceland. 
Akande, G. R., & Ajayi, A. A. (2005). Comparative proximate composition, physical and 
sensory evaluation of fish smoked with sawdust and firewood using improved Chorkor 
oven. FAO Fisheries Report No. 712. 
 154 
University of Ghana http://ugspace.ug.edu.gh
Akande, G. R., & Diei-Ouadi, Y. (2010). Post-harvest losses in small-scale fisheries: case 
studies in five sub-Saharan African countries. Rome: FAO Fisheries and Aquaculture 
Technical Paper. 550. 
Akande, G. R., Ayinla, O. A., Adeyemi, R. S., Olusola, A. O., & Salaudeen, M. M. (2012). 
Blueprint for smoked fish products. Lagos: Nigerian Institute for Oceanography and 
Marine Research. 
Akuamoa, F., Odamtten, G. T., & Kortei, N. K. (2018). Nutritional and shelf-life studies of dry 
smoked and gamma irradiated shrimps (Penaeus notialis) from three different water 
sources in Ghana. Cogent Food & Agriculture, 4, 9. 
Al-Kahtani, H. A., Abu-Tarboush, H. M., Atia, M., Bajaber, A. S., Ahmed, M. A., & El-
Mojaddidi, M. A. (1998). Amino acid and protein changes in tilapia and Spanish 
mackerel after irradiation and storage. Radiation Physics and Chemistry, 51(1), 107–
114. 
Al-Reza, S., Karmaker, S., Hasan, M., Roy, S., Hoque, R., & Rahman, N. (2015). Effect of 
traditional fish processing methods on the proximate and microbiological 
characteristics of Laubuka dadiburjori during storage at room temperature. Journal of 
Fisheries and Aquatic Science, 10(4), 232-243. 
Alhassan, E. H., Boateng, V. F., & Ndaigo, C. (2012). Smoked And frozen fish consumption 
and marketing channels in the Tamale Metropolis of Ghana. Ghana Journal of 
Development Studies, 9(1), 21-28. 
Alomirah, H., Al-Zenki, S., Al-Hooti, S., Zaghloul, S., Sawaya, W., Ahmed, N., & Kannan, K. 
(2011). Concentrations and dietary exposure to polycyclic aromatic hydrocarbons 
(PAHs) from grilled and smoked foods. Food Control, 22(12), 2028-2035. 
 155 
University of Ghana http://ugspace.ug.edu.gh
Amponsah, S. K., Nketiah, S., Arthur, W., Tandoh, I., Narh, D., & Baffour, V. (2017). 
Histamine levels in smoked finfishes from Ghanaian coastal waters. Accra, Ghana: 
CSIR Food Research Institute. 
Antwi-Boasiako, A. (2017). Biomass burning among small-scale fish processors in coastal 
Ghana: implications for health and climate. Accra: PhD Thesis. University of Ghana. 
AOAC. (2000). Salt (Chlorine as Sodium Chloride) in seafood. The Journal of AOAC, 17th 
Edn.. 976:18. 
AOAC. (2005). Official method of Analysis. 18th Edn. Washington DC: Association of 
Officiating Analytical Chemists. 
AOCS. (1998). Official method Ce 1b‐89. Fatty acid composition by GLC: marine oils. Official 
methods and recommended practices of the AOCS. 
APHA. (2015). Compendium of methods for the microbilogical examination of foods. (Y. 
Salfinger, & M. L. Tortorello, Eds.) Washington, DC, USA: American Public Health 
Association. 
Arason, S., Nguyen, M. V., Thorarinsdottir, K. A., & Thorkelsson, G. (2014). Preservation of 
fish by curing. Seafood Processing: Technology, Quality and Safety; Boziaris, IS, Ed, 
129-160. 
Arvanitoyannis , I. S., & Tserkezou, P. (2014). Irradiation of Fish and Seafood. In Seafood 
Processing: Technology, Quality and Safety (pp. 83-127). John Wiley & Sons, Ltd. 
Arvanitoyannis, I. S. (2010). Irradiation of food commodities: Techniques, applications, 
detection, legislation, safety and consumer opinion. London: Academic Press. 
Asamoah, E. (2018). The quality and shelf life of Atlantic mackerel (Scomber scombrus) as 
influenced by smoking technology, packaging material and storage temperature. 
Reykjavik: United Nations University Fisheries Training Programme, Iceland [final 
project]. 
 156 
University of Ghana http://ugspace.ug.edu.gh
Asiedu, B., Failer, P., & Beygens, Y. (2018). Ensuring food security: an analysis of the 
industrial smoking fishery sector of Ghana. Agriculture & Food Security, 7(38). 
Atowa, C. O., Nwabu, A. O., & Ogiedu, T. A. (2014). Storage and drying time effects on 
digestibility and amino acid composition of dried smoked horse mackerel (Trachurus 
trachurus) fillets. International Journal of Food Science and Nutrition Engineering, 
4(4), 98-105. 
ATSDR. (1995). Toxicological profile for Polycyclic Aromatic Hydrocarbons(PAHs). 
Department of Health and Human Services. Atlanta, GA: Agency for Toxic Substances 
and Disease Registry. 
Avega, B., & Tibu, G. A. (2017). Performance evaluation survey - Ahotor oven. Narragansett, 
RI: Coastal Resources Center, Graduate School of Oceanography, University of Rhode 
Island and Netherlands Development Organisation. GH2014_ACT099_SNV: The 
USAID/Ghana Sustainable Fisheries Management Project (SFMP). 
Aworh, O. C., Okparanta, R. N., & Oyedokun, E. O. (2002). Effect of irradiation on quality, 
shelf life and consumer acceptance of traditional Nigerian meat and fish products. Study 
of the impact of food irradiation on preventing losses: Experience in Africa. Pretoria, 
South Africa: IAEA. 
Ayinsa, H. K., & Maalekuu, B. K. (2013). Effect of traditional fish processing methods on the 
proximate composition of red fish stored under ambient room conditions. American 
Journal of Food and Nutrition, 3(3), 73-82. 
Badr, H. M. (2012). Control of the potential health hazards of smoked fish by gamma 
irradiation. International Journal of Food Microbiology, 154, 177-186. 
Bailis, R. (2004). Stove perfromance testing protocols, Controlled Cooking Test (CCT) Version 
2. Shell Foundation. Uniersity of Berkeley. 
 157 
University of Ghana http://ugspace.ug.edu.gh
Bandowe, B. A., Bigalke, M., Boamah, L., Nyarko, E., Saalia, F. K., & Wilcke, W. (2014). 
Polycyclic aromatic compounds (PAHs and oxygenated PAHs) and trace metals in fish 
species from Ghana (West Africa): bioaccumulation and health risk assessment. 
Environment International, 65, 135-146. 
Bank of Ghana. (2019). Markets: Daily onterbank FX rates. Retrieved June 03, 2019 from 
https://www.bog.gov.gh/markets/daily-interbank-fx-rates.  
Banson, K. E. (2015). Consumer acceptance of irradiation for sterilisation: a systemic approach 
to the case of medical items and food in Ghana. International Journal of Markets and 
Business Systems, 1(1), 70-92. 
Bataringaya, A. (2007). Analysis of quality deterioration at critical steps/points in the fish 
handling in Uganda and Iceland and suggestions for improvement. Reykjavik: Nations 
University Fisheries Training Programme, Iceland [final project]. 
Bayir, A., Haliloglu, H. I., Sirkecioglu, A. N., & Aras, N. M. (2006). Fatty acid composition 
in some selected marine fish species living in Turkish waters. Journal of the Science of 
Food and Agriculture, 163-168. 
Belusso, A. C., Nogueira, B. A., Breda, L. S., & Mitterer-Daltoe, M. L. (2016). Check all that 
apply (CATA) as an instrument for the development of fish products. Food Science and 
Technology, 36(2), 275-281. 
Berjia, F. L., & Brimer, L. (2013). Determination of Histamine in freshwater fish using ELISA 
Method: A food safety concern. American-Eurasian Journal of Toxicological Sciences, 
5(4), 94-96. 
Biji, K. B., Ravishankar, C. N., Venkateswarlu, R., Mohan, C. O., & Gopal, T. K. (2016). 
Biogenic amines in seafood: a review. Journal of Food Science and technology, 53(5), 
2210-2218. 
 158 
University of Ghana http://ugspace.ug.edu.gh
Bīlgīn, S., Ünlüsayin, M., Izci, M., & Günlü, A. (2008). The determination of the shelf life and 
some nutritional components of gilthead seabream (Sparus aurata L., 1758) after cold 
and hot smoking. Turkish Journal of Veterinary Animal Science, 32(1), 49-56. 
Bilsback, K. R., Dahlke, J., Fedak, K. M., Good, N., Hecobian, A., Herckes, P., . . . Volckens, 
J. (2019). A laboratory assessment of 120 air pollutant emissions from biomass and 
fossil-fuel cookstoves. Environmental Science & Technology, 53(12), 7114-7125. 
Bligh, E. G., & Dyer, W. J. (1959). A rapid method of total lipid extraction and purification. 
Canadian Journal of Biochemistry and Physiology, 37(8), 911-917. 
Bomfeh, K. (2016). Meeting Food Safety Targets in Shifting from Traditional to New Fish 
Smoking Systems: A Case Study in Ghana. Food Safety Technical Network (FSTN) 
Webinar/Seminar. FAO. 
Bomfeh, K., Jacxsens, L., Amoa-Awua, W. K., Tandoh, I., Afoakwa, E. O., Garrido Gamarro, 
E., . . . De Meulenaer, B. (2019). Reducing polycyclic aromatic hydrocarbon (PAH) 
contamination in smoked fish in the Global South: a case study of an improved kiln in 
Ghana. Journal of the Science of Food and Agriculture, 99(12), 5417-5423. 
Bott, L. (2014). Controlled Cooking Test. Vientiane: Renewable Energy and New Materials 
Institute. 
Bouloubassi, I., Méjanelle, L., Pete, R., Fillaux, J., Lorre, A., & Point, V. (2006). Transport by 
sinking particles in the open Mediterranean Sea: a 1 year sediment trap study. Marine 
Pollution Bulletin, 52, 560-571. 
Bouriga, N., Bejaoui, S., Jemmali, B., Quignard, J. P., & Trabelsi, M. (2020). Effects of 
smoking processes on the nutritional value and fatty acid composition of Zander fish 
(Sander lucioperca). Grasas Aceites, 71(1), e340. 
 159 
University of Ghana http://ugspace.ug.edu.gh
Bouriga, N., Ben Ismail, H., Gammoudi, M., Faure, E., & Trabelsi, M. (2012). Effect of 
smoking methos on biochemical and microbiological quality of Nile tilapia 
(Oreochromis niloticus). American Jounal of Food Technology, 7(11), 679-689. 
Bowker, B., & Zhuang, H. (2015). Relationship between water-holding capacity and protein 
denaturation in broiler breast meat. Poultry Science, 94, 1657–1664. 
Brauns, F. E., & Brauns, D. A. (2013). The Chemistry of Lignin: Covering the Literature for 
the Years 1949–1958 . Elsevier. 
Brondi, S. D., De MacEdo, A. N., Vicente, G. H., & Nogueira, A. R. (2011). Evaluation of the 
QuEChERS method and gas chromatography–mass spectrometry for the analysis 
pesticide residues in water and sediment. Bulletin of environmental contamination and 
toxicology, 86(1), 18-22. 
Cardinal, M., Gunnlaugsdottir, H., Bjoernevik, M., Ouisse, A., Vallet, J., & Leroi, F. (2004). 
Sensory characteristics of cold-smoked Atlantic salmon (Salmo salar) from European 
market and relationships with chemical, physical and microbiological measurements. 
Food Research International, 37, 181–193. 
Cardinal, M., Knockaert, C., Torrissen, O., Sigurgisladottir, S., Mørkøre, T., Thomassen, M., 
& Vallet, J. L. (2001). Relation of smoking parameters to the yield, colour and sensory 
quality of smoked Atlantic salmon (Salmo salar). Food Research International, 34(6), 
537-550. 
Center for Food Safety. (2014). Microbiological guidelines for food (for ready-to-eat food in 
general and specific food items). Queensway, Hong Kong: Food and Environmental 
Hygiene Department. 
Chaula, D., Laswai, H., Chove, B., Dalsgaard, A., Mdegela, R., & Hyldig, G. (2019). Fatty 
acid profiles and lipid oxidation status of sun dried, deep fried and smoked sardine 
 160 
University of Ghana http://ugspace.ug.edu.gh
(Rastrineobola argentea) from Lake Victoria, Tanzania. Journal of Aquatic Food 
Product Technology, 1-12. 
Chukwu, O., & Shaba, I. M. (2009). Effects of drying methods on proximate oompositions of 
catfish (Clarias gariepinus). World Journal of Agricultural Sciences, 5(1), 114-116. 
CIE. (1979). Colorimetry. Vienna, Austria: Bureau Central de la CIE, Publ. No. 15. 
Cieślik, I., Migdał, W., Topolska, K., Mickowska, B., & Cieślik, E. (2018). Changes of amino 
acid and fatty acid profile in freshwater fish after smoking. Journal of Food Processing 
and Preservation, 41(1), e13357. 
Codex Alimentarius Commission. (2003). Codex general standards for irradiated foods and 
recommended international code of practice for the operation of radiation facilities 
used for the treatment of foods. Rome: Food and Agriculture Organization of the United 
Nations. 
Codex Alimentarius Commission. (2009). Code of practice for the reduction of contamination 
of food with polycyclic aromatic hydrocarbons (PAH) from smoking and direct drying 
processes. CAC/RCP. 
Codex Alimentarius Commission. (2013). Standard for smoked fish, smoke-flavoured fish and 
smoke-dried fish (CXS 311- 2013). Rome: FAO/WHO. 
Collette, B., Amorim, A. F., Boustany, A., Carpenter, K. E., de Oliveira Leite Jr, N., Di Natale, 
A., . . . T, P. F. (2011). Scomber colias. Retrieved September 05, 2017, from The IUCN 
Red List of Threatened Species 2011: e.T170357A6767497: 
http://dx.doi.org/10.2305/IUCN.UK.2011-2.RLTS.T170357A6767497.en  
Cortés-Sánchez, A. D. (2018). Clostridium perfringens in foods and fish. Regulatory 
Mechanisms in Biosystems, 9(1), 112-117. 
Costa, R. (2013). Escherichia coli in seafood: A brief overview. Advances in Bioscience and 
Biotechnology, 4, 450-454. 
 161 
University of Ghana http://ugspace.ug.edu.gh
Cozzo-Siqueira, A., Oetterer, M., & Gallo, C. R. (2003). Effects of irradiation and refrigeration 
on the nutrients and shelf-life of tilapia (Oreochromis niloticus). Journal of Aquatic 
Food Product Technology, 12(1), 85–101. 
Cyprian, O. O. (2015). Drying and smoking of capelin (Mallotus villosus) and sardine 
(Sardinella gibbosa)–the influence on physicochemical properties and consumers 
acceptance. Reykjavik: PhD Dissertation. University of Iceland. 
Cyprian, O. O., Oduor-Odote, P., & Arason, S. (2019). The influence of lipid content and 
pretreatment methods on protein conformation in fish (capelin, Mallotus villosus) 
during smoking and drying. Food Science & Nutrition, 7(4), 1446-1454. 
Cyprian, O. O., Sveinsdottir, K., Jonsson, A., Nguyen, M. V., Tomasson, T., Thorkelsson, G., 
& Arason, S. (2017). Influence of lipid content and packaging methods on the quality 
of dried capelin (Mallotus villosus) during storage. Journal Food Science and 
Technology, 54(2), 293-302. 
Cyprian, O. O., Van Nguyen, M., Sveinsdottir, K., Jonsson, A., Tomasson, T., Thorkelsson, 
G., & Arason, S. (2015). Influence of smoking and packaging methods on lipid stability 
and microbial quality of Capelin (Mallotus villosus) and Sardine (Sardinella gibossa). 
Food Science & Nutrition, 3(5), 404-414. 
da Silva, L. V., Prinyawiwatkul, W., King, J. M., No, H. K., Bankston, J. D., & Ge, B. (2008). 
Effect of preservatives on microbial safety and quality of smoked blue catfish (Ictalurus 
furcatus) steaks during room-temperature storage. Food Microbiology, 25, 958-963. 
Daniel, E. O., Ugwueze, A. U., & Igbegu, H. E. (2013). Microbiological quality and some 
heavy metals analysis of smoked fish sold in Benin City, Edo State, Nigeria. World J. 
Fish Marine Sci, 5(3), 239-243. 
 162 
University of Ghana http://ugspace.ug.edu.gh
Daramola, J. A., Fasakin, E. A., & Adeparusi, E. O. (2007). Changes in physicochemical and 
sensory characteristics of smoke-dried fish species stored at ambient temperature. 
African Journal of Food, Agriculture, Nutrition and Development, 7(6), 16. 
de Alba, M., Pérez-Andrés, J., Harrison, S. M., Brunton, N. P., Burgess, C. M., & Tiwari, B. 
K. (2019). High pressure processing on microbial inactivation, quality parameters and 
nutritional quality indices of mackerel fillets. Innovative Food Science & Emerging 
Technologies, 55, 80-87. 
de Morais, L., Smith-Vaniz, W. F., Kara, M., Yokes, B., Pollard, D., Carpenter, K. e., & de 
Bruyne, G. (2015). Sphyraena sphyraena. Retrieved July 02, 2017, from The IUCN 
Red List of Threatened Species 2015: e.T198565A15596165: 
http://dx.doi.org/10.2305/IUCN.UK.2015-4.RLTS.T198565A15596165.en  
Diei-Ouadi, Y. (2013). The complaince efforts for the PAHs control in processed fish from 
developing countries. FAO. 
Doe, P. E. (Ed.). (1998). Fish drying and smoking: Production and quality. Lancaster, PA, 
USA: Taylor & Francis Routledge. 
Duah, S. A., Emi-Reynolds, G., Kumah, P., & Larbi, D. A. (2018). Effect of gamma irradiation 
on the microbiological quality and shelf life of Engraulis encrasicolus in Ghana. 
International Journal of Fisheries and Aquaculture, 10(3), 16-21. 
Duxbury, D. (2005). Sensory evaluation provides value. Food Tecnology, 59(5), 68-72. 
Dvorak, P., Kratochv, B., & Grolichov, M. (2005). Changes of colour andpHin fish 
musculature after ionizing radiation exposure. European Food Research and 
Technology, 220, 309-311. 
Edris, A. M., Hasanen, F. S., Khater, D. F., & Lela, R. A. (2012). Microbiological studies of 
some fishery products. Benha Veterinary Medical Journal, 23(1), 185-191. 
 163 
University of Ghana http://ugspace.ug.edu.gh
Edwards, R. K. (2014). WHO indoor air quality guidelines: Household fuel combustion - 
Review 2: emissions of health-damaging pollutants from household stoves. Geneva: 
World Health Organization. 
Egbuta, M. A., Mwanza, M., & Babalolo, O. O. (2017). Health risks associated with exposure 
to filamentous fungi. International Journal of Environmental Research and Public 
Health, 14(7), 719. 
Ehlermann, D. A. (2016). Particular applications of food irradiation: Meat, fish and others. 
Radiation Physics and Chemistry, 129, 53-57. 
Ekpo, K. E., Asia, I. O., Amayo, K. O., & Jegede, D. A. (2007). Determination of lead, 
cadmium and mercury in surrounding water and organs of some species of fish from 
Ikpoba river in Benin city, Nigeria. International Journal of Physical Sciences, 3(11), 
89–292. 
Entee, S. (2015a). Post-Harvest Process Characterization Studies and Needs Assessment 
Report on Fish Smoking and Salting of Fish. Narragansett, RI: The USAID/Ghana 
Sustainable Fisheries Management Project (SFMP). Coastal Resources Center, 
Graduate School of Oceanography, University of Rhode Island and SNV Netherlands 
Development Organization. GH2014_ACT006_SNV. 
Entee, S. (2015b). Post-harvest processing stove performance report. Coastal Resources 
Center, Graduate School of Oceanography, University of Rhode Island and SNV 
Netherlands Development Organization. Narragansett, RI: The USAID/Ghana 
Sustainable Fisheries Management Project (SFMP). GH2014_ACT015_SNV. 
Entee, S. (2015c). Post Harvest Processing Value Chain Literature Review Report. Coastal 
Resources Center, Graduate School of Oceanography, University of Rhode Island and 
SNV Netherlands Development Organization. Narragansett, RI: The USAID/Ghana 
Sustainable Fisheries Management Project (SFMP). GH2014_ACT015_SNV. 
 164 
University of Ghana http://ugspace.ug.edu.gh
EPA. (2019). Ghana standards for environmental and health protection - Requirements for 
ambient air quality and point source/stack emissions (GS 1236, 2019) . Accra, Ghana: 
Environmental Protection Agency. 
Erkan, N., & Özden, Ö. (2007). The changes of fatty acid and amino acid compositions in sea 
bream (Sparus aurata) during irradiation process. Radiation Physics and Chemistry, 
76, 1636–1641. 
Espe, M., Nortvedt, R., Lie, Ø., & Hafsteinsson, H. (2002). Atlantic salmon (Salmo salar) as 
raw material for the smoking industry. II: Effect of different smoking methods on losses 
of nutrients and on the oxidation of lipids. Food Chemistry, 77, 41-46. 
Essumang, D. K., Dodoo, D. K., & Adjei, J. K. (2012). Polycyclic aromatic hydrocarbon (PAH) 
contamination in smoke-cured fish products. Journal of Food Composition and 
Analysis, 27, 128-138. 
Essumang, D. K., Dodoo, D. K., & Adjei, J. K. (2013). Effect of smoke generation sources and 
smoke curing duration on the levels of polycyclic aromatic hydrocarbon (PAH) in 
different suites of fish. Food and Chemical Toxicology, 58, 86-94. 
Essumang, D. K., Dodoo, D. K., & Adjei, J. K. (2014). Effective reduction of PAH 
contamination in smoke cured fish products using charcoal filters in a modified 
traditional kiln. Food Control, 35, 85-93. 
EU Directive. (2008). COMMISSION REGULATION (EC) No 1022/2008 of 17 October 2008 
amending Regulation (EC) No 2074/2005 as regards the total volatile basic nitrogen 
(TVB-N) limits. L277/18 EN. Official Journal the European Union. 
EU. (2009). List of Member States’ authorisations of food and food ingredients which may be 
treated with ionising radiation. Official Journal of the European Union (C 283/02). 
European Commission. (2005). Commission Regulation (EC) No 2073/2005 of 15 November 
2005 on microbiological criteria for foodstuffs. 50, pp. 1-26. 
 165 
University of Ghana http://ugspace.ug.edu.gh
European Commission. (2011). Commission Regulation (EU) No 835/2011 of 19 August 2011 
amending Regulation (EC) No 1881/2006 as regards maximum levels for polycyclic 
aromatic hydrocarbons in foodstuffs. Official Journal of the European Union (L 215), 
4-8. 
European Commission. (2015). Commission Regulation (EC) No 1881/2006 of 19 December 
2006 setting maximum levels for certain contaminants in foodstuffs. Official Journal 
of the European Union. 364. 
Failler, P., Yolaine, B., & Asiedu, B. (2014). Value chain analysis of the fishery sector in 
Ghana with focus in quality, eniveronmental, social, sustainability, food security, 
organic requirements and its compliance infrastructure. Accra: UNIDO/MOTI TCP 
Project. 
Fan, X., & Sommers, C. H. (Eds.). (2013). Food irradiation research and technology. 2nd Edn. 
Wyndmoor, USA: Wiley-Blackwell. 
FAO. (2016-2019). Fishery and Aquaculture Country Profiles. Ghana (2016). Country Profile 
Fact Sheets. FAO Fisheries and Aquaculture Department [online]. 
FAO. (2016a). The State of the World fisheries and Aquaculture 2016 - Contributing to food 
security and nutrition for all. Rome: FAO. 
FAO. (2016b). FTT-Thiaroye Ovens: Clearing the air for women fish processors in Côte 
d'Ivoire, and beyond. Retrieved January 03, 2018, from 
http://www.fao.org/gender/insights/detail/en/c/458484/  
FAO. (2018). The State of the World fisheries and aquaculture 2018 - Meeting the sustainable 
development goals. Rome: FAO. 
FAO/WHO. (2012). Code of practice for fish and fishery products. 2nd Edn. Rome, Italy: 
World Health Organization; Food and Agriculture Organization of the United Nations. 
Food Standards Programme. Codex Alimentarius Commission. 
 166 
University of Ghana http://ugspace.ug.edu.gh
FAO/WHO. (2013). Public health risks of histamine and other biogenic amines from fish and 
fishery products. Meeting report. Rome: Food and Agriculture Organization of the 
United Nations. 
Farkas, J. (2006). Irradiation for better foods. Trends in Food Science and Technology, 17, 
148-152. 
Farkas, J., & Mohácsi-Farkas, C. (2011). History and future of food irradiation. Trends in Food 
Science & Technology, 22, 121-126. 
FEP. (2007-2013). Barracuda and sea pike. Retrieved October 05, 2017, from http://www.fish-
cheap.eu/wp-content/uploads/2016/01/Barracuda.pdf  
Fernandes, R. (Ed.). (2009). Microbiology handbook: fish and seafood. Surrey, UK: 
Leatherhead Food International Ltd. 
Fletcher, S., Boonwaat, L., Moore, T., Chavada, R., & Conaty, S. (2015). Investigating an 
outbreak of staphylococcal food poisoning among travellers across two Australian 
states. Western Pacific Surveillance Response Journal, 6(2), 17-21. 
Flintwood-Brace, A. (2016). Biomass smoke exposure in traditional smokehouses and 
respiratory symptoms among fish smokers at Aboadze/Abuesi in the Western Region of 
Ghana. University of Ghana. Accra: Master of Public Health Dissertation, University 
of Ghana. 
Food and Drug Authority (FDA). (2011). Fish and fishery products hazards and controls 
guidance. 4th Edn. Washington, DC: Department of Health and Human Services, Food 
and Drug Administration, Center for Food Safety and Applied Nutrition . 
Food Safety Authority of Ireland. (2018). Histamine in Fish and Fishery Products. Retrieved 
August 04, 2019, from https://www.fsai.ie/faq/histamine_fish.html  
Froese, R. (2006). Cube law, condition factor and length-weight relationships: history, meta-
analysis and recommendations. Applied Ichthyology, 22, 241-253. 
 167 
University of Ghana http://ugspace.ug.edu.gh
Garaffo, M. A., Vassallo-Agius, R., Nengas, Y., Lembo, E., Rando, R., Maisano, R., . . . 
Giuffrida, D. (2011). Fatty acids profile, atherogenic (IA) and thrombogenic (IT) health 
lipid indices, of raw roe of bluefin tuna (Thunnus thynnus l.) and their salted product 
“bottarga". Food and Nutrition Sciences, 736-743. 
Garcia, J. N., Borrega, J., & Luís, M. R. (2018). Airborne PM impact on health, overview of 
variables, and key factors to decision making in air quality. Air Pollution: monitoring, 
quantification and removal of gases and particles. IntechOpen. Available from: 
https://www.intechopen.com/books/air-pollution-monitoring-quantification-and-
removal-of-gases-and-particles/airborne-pm-impact-on-health-overview-of-variables-
and-key-factors-to-decision-making-in-air-quality  
Gecgel, U. (2013). Changes in some physicochemical properties and fatty acid composition of 
irradiated meatballs during storage. Journal of Food Science and Technology, 50(3), 
505-513. 
Ghaly, A. E., Dave, D., Budge, S., & Brooks, M. S. (2010). Fish spoilage mechanisms and 
preservation techniques: Review. American Journal of Applied Sciences, 7(7), 859-877. 
Ghana Standards Authority (GSA). (2013). GS955:2013: Microbiology analysis of foods – 
sampling plans and microbiological criteria. Accra, Ghana: Ghana Standards 
Authority. 
Ghana Standards Authority (GSA). (2019). GS955:2019: Microbiological analysis of food-
Sampling plans and microbiological criteria. Accra: Ghana Standards Authority. 
Ghana Statistical Service. (2014). 2010 Population and housing census: District analytical 
report. Shama district. Ghana Statistical Service. 
GIPC. (2019). Investing in Ghana's fishing industry - Sector Overview. Retrieved February 21, 
2019, from http://gipcghana.com/21-investment-projects/agriculture-and-
agribusiness/fishing-and-aquaculture/300-investing-in-ghana-s-fishing-industry.html  
 168 
University of Ghana http://ugspace.ug.edu.gh
Global Alliance for Clean Cookstoves. (2016). Comparative analysis of fuels for cooking: Life 
cycle environmental impacts and economic and social considerations. Appendix A: 
Detailed environmental, economic and cocial technical analyses. Global Alliance for 
Clean Cookstoves. 
Global Alliance for Clean Cookstoves. (2018). Protocols. Retrieved April 05, 2019, from 
http://cleancookstoves.org/technology-and-fuels/testing/protocols.html  
Goldstein, M. (2008). Carbon monoxide poisoning. Journal of Emergency Nursing, 34(6), 538-
542. 
Gopalapura, S., Gowda, S., Narayan, B., & Gopal, S. (2016). Bacteriological properties and 
health-related biochemical components of fermented fish sauce: An overview. Food 
Reviews International, 32(2), 203-229. 
Gordon, A., Pulis, A., & Owusu-Adjei, E. (2011). Smoked marine fish from the Western 
Region, Ghana: a value chain assessment. Narragansett, USA: WorldFish Center. 
USAID Integrated Coastal and Fisheries Governance Initiative for the Western Region, 
Ghana. 
Goulas, A. E., & Kontominas, M. G. (2005). Effect of salting and smoking-method on the 
keeping quality of chub mackerel (Scomber japonicus): biochemical and sensory 
attributes. Food Chemistry, 93, 511-520. 
Gram, L. (2009). Microbiological spoilage of fish and seafood products. In Compendium of the 
microbiological spoilage of foods and beverages (pp. 87-119). New York: Springer. 
Grygus, A. (2015). Barracuda. Retrieved September 11, 2017, from 
http://www.clovegarden.com/ingred/sf_baraz.html  
Hardy, R., & Smith, J. (1979). The storage of mackerel (Scomber scombrus). Development of 
histamine and rancidity. Journal of the Science of Food and Agriculture, 27(7), 595-
599. 
 169 
University of Ghana http://ugspace.ug.edu.gh
Hassan, I. M. (1988). Processing of smoked common carp fish and its relation to some 
chemical, physical and organoleptic properties. Food Chemistry, 27(2), 95-106. 
Heinbuch, U. (1994). Animal Protein Sources for Rural and Urban Populations in Ghana. 
IDAF/WP/58. Cotonou: Programme for the Integrated Development of Artisanal 
Fisheries in West Africa. 
Hilderbrand, K. S. (1992). Fish smoking procedures for forced convection smokehouses. 
Newport, Or: Oregon State University, Extension Sea Grant Program, Hatfield Marine 
Science Center. 
Hosomi, R., Yoshida, M., & Fukunaga, K. (2015). Seafood consumption and components for 
health. Global Journal of Health Science, 4(3), 72. 
Huang, C. Y., Wang, Y. X., Wang, X. Z., Hu, C. C., & Chui, T. C. (2019). Analysis of seven 
biogenic amines and two amino acids in wines using micellar electrokinetic 
chromatography. Applied Sciences, 9(6), 1193. 
Huss, H. H. (1995). Quality and quality changes in fresh fish. Rome: Food and Agriculture 
Organization. 
Hyldig, G., Larsen, E., & Green-Petersen, D. (2007). Fish and sensory analysis in the fish 
chain. Handbook of Meat, Poultry and Seafood Quality, 499-510. 
IAEA. (2001). Irradiation to control Vibrio infection from consumption of raw seafood and 
fresh produce. Results of a co-ordinated research project (No. IAEA-TECDOC--1213). 
International Atomic Energy Agency. 
ICMSF. (1986). Microorganisms in Foods 2. Sampling for Microbiological Analysis: 
Principles and Specific Applications. 2nd Edn. London: Blackwell Scientific 
Publications. 
 170 
University of Ghana http://ugspace.ug.edu.gh
Idah, P. A., & Nwankwo, I. (2013). Effects of smoke-drying temperatures and time on physical 
and nutritional quality parameters of Tilapia (Oreochromis niloticus). International 
Journal of Fisheries and Aquaculture, 5(3), 29-34. 
Ikasari, D., Suryanti, S., & Suryaningrum, T. D. (2017). Proximate composition and sensory 
characteristics of traditional and oven-drying smoked tilapia fillets enriched with olive 
oil. Squalen Bulletin of Marine and Fisheries Postharvest and Biotechnology, 12(3), 
127-137. 
International Agency for Research on Cancer, IARC. (2018). Agents Classified by the IARC 
Monographs, Volumes 1–123. Retrieved October 02, 2018, from 
https://monographs.iarc.fr/wp-content/uploads/2018/09/List_of_Classifications.pdf  
IRI-CSIR, GSA, & Kwarteng, E. (2016). Testing of low PAH improved fish smoking stove 
(Ahotor oven). RI: Coastal Resources Center, Graduate School of Oceanography, 
University of Rhode Island. GH2014_ACT067_SNV. Narragansett: The 
USAID/Ghana Sustainable Fisheries Management Project (SFMP). 
ISO. 660 (2000). Animal and vegetable fats and oils - Determination of acid value and acidity. 
International Organization for Standardization. 
Jaeger, H., Janositz, A., & Knorr, D. (2010). The Maillard reaction and its control during food 
processing. The potential of emerging technologies. Pathologie Biologie, 58, 207-213. 
Jónsdóttir, R., Ólafsdóttir, G., Chanie, E., & Haugen, J. E. (2008). Volatile compounds suitable 
for rapid detection as quality indicators of cold smoked salmon (Salmo salar). Food 
Chemistry. 109, 184-195. 
Kalogirou, S., Mittermayer, F., Pihl, L., & Wennhage, H. (2012). Feeding ecology of 
indigenous and non‐indigenous fish species within the family Sphyraenidae. Journal of 
fish biology, 80(7), 2528-2548. 
 171 
University of Ghana http://ugspace.ug.edu.gh
Kaneniwa, M., Yokoyama, M., Murata, Y., & Kuwahara, R. (2004). Enzymatic hydrolysis of 
lipids in muscle of fish and shellfish during cold storage. In Quality of fresh and 
processed foods (pp. 113-119). New York: Springer. 
Kaya, Y., Erdem, M. E., & Turan, H. (2014). Monthly differentiation in meat yield, chemical 
and amino acid composition of wild and cultured brown trout (Salmo trutta forma fario 
Linneaus, 1758). Turkish Journal of Fisheries and Aquatic Sciences, 479-486. 
Kaya, Y., Turan, H., & Erdem, M. E. (2008). Fatty acid and amino acid composition of raw 
and hot smoked sturgeon (Huso huso, L. 1758). International Journal of Food Sciences 
and Nutrition, 59(7), 635-642. 
Khalil, A., Albachir, M., & Odeh, A. (2016). Effect of Gamma Irradiation on Some 
Carcinogenic Polycyclic Aromatic Hydrocarbons (PAHs) in Wheat Grains. Polycyclic 
Aromatic Compounds, 36(5), 873-883. 
Kim, J. Y., Woo, H. J., Ahn, C. W., Nam, H. S., Shin, Z. I., & Lee, H. J. (1999). Cytotoxic 
effects of peptides fractionated from bromelain hydrolyzates of soybean protein. Food 
Science and Biotechnology, 8, 333-337. 
Kim, K. H., Jahan, S. A., Kabir, E., & Brown, R. J. (2013). A review of airborne polycyclic 
aromatic hydrocarbons (PAHs) and their human health effects. Environment 
International(60), 71-80. 
Kleter, G. A. (2004). Control and prevention of contamination and spoilage in the traditional 
production of smoked fish in Ghana. Discussion paper prepared for the Tailor Made 
Biotechnologies project. Wageningen, The Netherlands: RIKILT - Institute of Food 
Safety. Projectnr.: 805.71.456.01. 
Kolakowska, A. (2003). Lipid oxidation in food systems. Chemical and functional properties 
of food lipids, pp. 133-165. 
 172 
University of Ghana http://ugspace.ug.edu.gh
Kołakowska, A., & Bartosz, G. (2014). Oxidation of food components: An introduction. In 
Food oxidants and antioxidants chemical, biological, and functional properties (pp. 1-
20). Boca Raton: CRC Press. 
Kombat, E. O., Nunoo, F. K., Ampofo, J. A., & Addo, P. G. (2013). Effects of environmental 
conditions on the microbiological quality of two small marine pelagic fishes landed in 
Accra and Tema, Ghana. Archives of Applied Science Research, 5(2), 180-188. 
Koral, S., Tufan, B., Başçınar, N., & Köse, S. (2015). Quality Changes and Shelf-Life of 
Cultured and Wild Hot-Smoked Mediterranean Horse Mackerel (Trachurus 
mediterraneus, Steindachner, 1868) at Frozen Storage (-20±2 C). Turkish Journal of 
Fisheries and Aquatic Sciences, 15(3), 495-504. 
Küçükgülmez, A., Yanar, Y., Çelik, M., & Ersor, B. (2018). Fatty acids profile, atherogenic, 
thrombogenic, and polyene lipid indices in golden grey mullet (Liza aurata) and gold 
band goatfish (Upeneus moluccensis) from Mediterranean Sea. Journal of Aquatic 
Food Product Technology, 27(8), 912-918. 
Kwarteng, E., Nsiah, A. A., Samey, B., Boateng, L., & Aziebor, E. (2016). Livelihood Options 
Synthesis Report. The USAID/Ghana Sustainable Fisheries Management Project 
(SFMP). Narragansett, RI: Coastal Resources Center, Graduate School of 
Oceanography, University of Rhode Island. GH2014_ACT072_SNV. 
Last, M., Richards, P., & Fyfe, C. (1987). Sierra Leone, 1787-1987: Two centuries of 
intellectual life. Manchester University Press. 
Lawless, H. T., & Heymann, H. (2010). Sensory evaluation of food: Principles and practices. 
2nd Edn. New York: Springer Science and Business. 
Leksono, T., Suprayitno, E., & Hardoko, H. P. (2014). Profile of traditional smoking striped 
catfish (Pangasius hypophthalmus Sauvage 1878) and firewood used in province of 
riau, Indonesia. Journal of Biodiversity and Environmental Sciences, 5(1), 11. 
 173 
University of Ghana http://ugspace.ug.edu.gh
Li, J., Dong, H., Li, X., Han, B., Zhu, C., & Zhang, D. (2016). Quantitatively assessing the 
health risk of exposure to PAHs from intake of smoked meats. Ecotoxicology and 
Environmental Safety, 124, 91-95. 
Lira, G. M., Silva, K. B., Figueirêdo, B. C., & Bragagnolo, N. (2014). Impact of smoking on 
the lipid fraction and nutritional value of seabob shrimp (Xiphopenaeus kroyeri, Heller, 
1862). Food Science and Technology, 58(1), 183-187. 
Ljubojević, D., Radosavljević, V., Pelić, M., Đorđević, V., Živkov Baloš, M., & Ćirković, M. 
(2016). Fatty acid composition, chemical composition and processing yield of 
traditional hot smoked common carp (Cyprinus carpio, L). Iranian Journal of Fisheries 
Sciences, 15(4), 1293-1306. 
Loaharanu, P. (2003). Irradiated foods. New York: American Council on Science and Health. 
Løje, H. (2007). The quality of cold smoked salmon: Influence of raw material and 
technological parameters. Technical University of Denmark. 
Lucarelli, K., Wyne, K., & Svenson, J. E. (2018). Improved cookstoves and their effect on 
carbon monoxide levels in San Lucas Tolimán, Guatemala. International Journal of 
Environmental Health Research, 28(1), 64-70. 
Łuczyńska, J., Paszczyk, B., Nowosad, J., & Łuczyński, M. (2017). Mercury, fatty acids 
content and lipid quality indexes in muscles of freshwater and marine fish on the polish 
market. Risk assessment of fish consumption. International Journal of Environmental 
Research and Public Health, 14(10), 1120. 
Maherani, B., Hossain, F., Criado, P., Ben-Fadhel, Y., Salmieri, S., & Lacroix, M. (2016). 
World market development and consumer acceptance of irradiation technology. Foods, 
5, 79. 
 174 
University of Ghana http://ugspace.ug.edu.gh
Marc, C., Kaakeh, R., & Mbofung, C. M. (1998). Effect of salting and smoking method on the 
stability of lipid and microbiological quality of Nile perch (Lates niloticus). Journal of 
Food Quality, 22, 517-528. 
Marc, K., Philippe, S., Eustache, H., Boniface, Y., Dominique, S., & Souaibou, F. (2014). 
Microbiological quality of smoked mackerel (Trachurus trachurus), sold in Abomey-
Calavi township markets, Benin. Journal of Microbiology Research, 4(5), 175-179. 
Matsuura, Y., Takehira, M., Joti, Y., Ogasahara, K., Tanako, T., Ono, N., . . . Yutani, K. (2015). 
Thermodynamics of protein denaturation at temperatures over 100 °C: CutA1 mutant 
proteins substituted with hydrophobic and charged residues. Scientific Reports, 
5(15545), 1-9. 
Mbarki, R., Miloud, N. B., Selmi, S., Dhib, S., & Sadok, S. (2009). Effect of vacuum packaging 
and low-dose irradiation on the microbial, chemical and sensory characteristics of chub 
mackerel (Scomber japonicus). Food Microbiology, 26, 821-826. 
Medina, M., Cabeza, M. C., Bravo, D., Cambera, I., Monteil, R., Ordóñez, J. A., . . . Hoz, L. 
(2009). A comparison between e-beam irradiation and high pressure treatment for cold-
smoked salmon sanitation: Microbiological aspects. Food Microbiology, 26, 224–227. 
Mejlholm, O., Andersen, J. E., & Dalgaard, P. (2007). Microbial changes and safety of lightly 
preserved seafood. Denmark: DIFRES, DTU. 
Ministry of Food and Agriculture (MOFA). (2017). Fisheries. Retrieved October 15, 2017, 
from www.mofa.gov.gh/site/?page_id=244  
MOFAD. (2016). Fisheries management plan of Ghana. A national Policy for the management 
of the marine fisheries sector. Accra: MOFAD. 
MOFAD. (2018). Medium term expenditure framework (MTEF) for 2018-2021. Programme 
based budget estimates for 2018. Accra: Ministry of Finance. 
 175 
University of Ghana http://ugspace.ug.edu.gh
Mohanty, B., Mahanty, A., Ganguly, S., Sankar, T. V., Chakraborty, K., Rangasamy, A., . . . 
Sharma1, A. P. (2014). Amino acid compositions of 27 food fishes and their importance 
in clinical nutrition. Journal of Amino Acids, 2014, 7. 
Moini, S., Tahergorabi, R., Hosseini, S. V., Rabbani, M., Tahergorabi, Z., Feas, X., & Aflaki, 
F. (2009). Effect of gamma radiation on the quality and shelf life of refrigerated 
rainbow trout (Oncorhynchus mykiss) fillets. Journal of Food Protection, 72(7), 1419-
1426. 
Molins, R. A. (Ed.). (2001). Food irradiation: Principles and applications. New York: John 
Wiley & Sons Inc. 
Mols-Mortensen, A., Ortind, E. Á., Jacobsen, C., & Holdt, S. L. (2017). Variation in growth, 
yield and protein concentration in Saccharina latissima (Laminariales, Phaeophyceae) 
cultivated with different wave and current exposures in the Faroe Islands. Journal of 
Applied Phycology, 29(5), 2277-2286. 
Morris, J. G., & Potter, M. (2013). Foodborne infections and intoxications. 4th Edn. Academic 
Press. 
Mostafavi, H. A., Fathollahi, H., Motamedi, F., & Mirmajlessi, S. M. (2010). Food irradiation: 
Applications, public acceptance and global trade. African Journal of Biotechnology, 
9(20), 2826-2833. 
Nayga, R. M., Aiew, W., & Nichols, J. P. (2005). Information effects on consumers’ 
willingness to purchase irradiated food products. Review of Agricultural Economics, 
27, 37–48. 
Ncube, S., Kunene, P., Tavengwa, N. T., Tutu, H., Richards, H., Cukrowska, E., & Chimuka, 
L. (2017). Synthesis and characterization of a molecularly imprinted polymer for the 
isolation of the 16 US-EPA priority polycyclic aromatic hydrocarbons (PAHs) in 
solution. ournal of Environmental Management, 199, 192-200. 
 176 
University of Ghana http://ugspace.ug.edu.gh
Ndiaye, O., Komivi, B. S., & Diei-Ouadi, Y. (2015). Guide for developing and using the FAO-
Thiaroye processing technique (FTT-Thiaroye). Rome: Food and Agriculture 
Organization of the United Nations. 
Nester, E. W., Anderson, D. G., Roberts, C. E., & Nester, M. T. (2009). Microbiology: A 
Human Perspective. 6th Edn. New York, NY: McGraw-Hill. 
Nguyen, M. V., Thorarinsdottir, K. A., Gudmundsdottir, A., Thorkelsson, G., & Arason, S. 
(2011). The effects of salt concentration on conformational changes in cod (Gadus 
morhua) proteins during brine salting. Food Chemistry, 125, 1013-1019. 
Nketsia-Tabiri, J., Adu-Gyamfi, A., Montford, K. G., Gbedemah, C. M., & Sefa-Dedeh, S. 
(2003). Optimising processing conditions for irradiated cured fish. Radiation 
processing for safe, shelf-stable and ready-to-eat food. Proceedings of a final research 
co-ordination meeting held in Montreal, Canada, 10–14 July 2000 (pp. 207-216). (No. 
IAEA-TECDOC-1337). Vienna: International Atomic Energy Agency Technical 
Document, 1337. 
NOAA Fishwatch. (2015). Atlantic Mackerel (Scomber scombrus). Retrieved December 02, 
2018, from NOAA FishWatch. U.U. Seafood Facts: 
http://www.fishwatch.gov/profiles/atlantic-mackerel  
Nogueira, N., Cordeiro, N., & Aveiro, M. J. (2013). Chemical composition, fatty acids profile 
and cholesterol content of commercialized marine fishes captured in Northeastern 
Atlantic. Journal of Fisheries Sciences, 7(3), 271-286. 
Nunoo, F. K., Asiedu, B., Amador, K., Belhabib, D., & Pauly, D. (2014). Reconstruction of 
marine fisheries catches for Ghana, 1950-2010. Working Paper Series., 26. 
Nunoo, F. K., Asiedu, B., Kombat, E. O., & Samey, B. (2015). Sardinella and other small 
pelagics value and supply chain of the fishery sector, Ghana. Narragansett, USA: The 
USAID/Ghana Sustainable Fisheries Management Project (SFMP). RI: Coastal 
 177 
University of Ghana http://ugspace.ug.edu.gh
Resources Center, Graduate School of Oceanography, University of Rhode Island and 
Netherlands Development Organisation. GH2014_ACT044_SNV. 
Nunoo, F. K., Tornyeviadzi, E., Asamoah, E. K., & Addo, S. (2019). Effect of two fish smoking 
ovens on the nutritional composition and PAH content of smoked fish. Elixir 
Aquaculture, 129, 53073-53076. 
Nyarko, E., Botwe, B. O., & Klubi, E. (2011). Polycyclic aromatic hydrocarbons (PAHs) levels 
in two commercially important fish species from the coastal waters of Ghana and their 
carcinogenic health risks. West African Journal of Applied Ecology, 19(1), 53-66. 
Obeng, G. M. (2018). Self-reported health status of fish smokers at Abuesi, a fishing community 
in the western region of Ghana. MPhil Dissertation. University of Cape Coast. 
Obodai, E. A., Muhammad, B. A., Obodai, G. A., & Opoku, E. (2009). Effect of fuel wood on 
the quality of smoked freshwater fish species sold in Tamale Central Market, Northern 
Region, Ghana. Ethiopian Journal of Environmental Studies and Management, 2(2), 
27-35. 
Odoi, R. N. (2014). Comparing quality and storage life of hot smoked mackerel and herring 
using two different smoking kilns. Reykjavik: United Nations University Fisheries 
Training Programme, Iceland (final project). 
Oehlenschläger, J. (2014). Seafood quality assessment. In I. S. Boziaris, Seafood Processing: 
Technology, Quality and Safety. 1st Edn. (pp. 362-386). John Wiley & Sons, Ltd. 
Okpala, C. O. (2016). Lipid autoxidation in ozone-processed crustacea under cold storage: A 
treatise. Lipid Technology, 28(5-6), 93-95. 
Okyere-Nyarko, A. A., Aziebor, E., & Robinson, J. (2015). Namhowfo, mangroves and 
woodlots: a more complete approach to reducing deforestation in Ghana. Netherlands: 
SNV Netherlands Development Organisation. Retrieved September 09, 2017, from 
 178 
University of Ghana http://ugspace.ug.edu.gh
http://www.snv.org/public/cms/sites/default/files/explore/download/case_study_-
_mangroves_and_woodlots.pdf  
Omri, B., Chalghoumi, R., Izzo, L., Ritieni, A., Lucarini, M., Durazzo, A., . . . Santini, A. 
(2019). Effect of dietary incorporation of linseed alone or together with tomato-red 
pepper mix on laying hens’ egg yolk fatty acids profile and health lipid indexes. 
Nutrients, 11, 813. 
Onyia, L. U., Sogbesan, A. O., Milam, C., & Joseph, A. I. (2011). Microbial load analysis of 
fresh mud catfish (Clarias anguillaris) and tilapia (Oreochromis niloticus) in Lake 
Gerio, Adamawa State, Nigeria using two different smoking kilns. Journal of 
Environmental Issues and Agriculture in Developing Countries, 3(1), 165-170. 
Ortega-Rivas, E. (2012). Non-thermal food engineering operations. London: Springer Science 
& Business Media. 
Osibona, A. O., Ogunyebi, O. O., & Samuel, T. 0. (2018). Storage fungi and mycotoxins 
associated with stored smoked Catfish (Clarias gariepinus). Journal of Applied 
Science, Environment and Managment, 22(5), 643-646. 
Osman, H., Suriah, A. R., & Law, E. C. (2001). Fatty acid composition and cholesterol content 
of selected marine fish in Malaysian waters. Food Chemistry, 73, 55-60. 
Oteng-Amoako, A. A. (2012). Strombosia pustulata Oliv. In R. H. Lemmens, D. Louppe, & A. 
A. Oteng-Amoako, PROTA (Plant Resources of Tropical Africa / Ressources végétales 
de l’Afrique tropicale). Wageningen, Netherlands. 
Otten, J. J., Hellwig, J. P., & Meyers, J. D. (Eds.). (2006). Dietary reference intakes: the 
essential guide to nutrient requirements. Washington, D.C.: National Academies Press. 
Ozden, O., Inugur, M., & Erkan, N. (2007). Effect of different dose gamma radiation and 
refrigeration on the chemical and sensory properties and microbiological status of 
 179 
University of Ghana http://ugspace.ug.edu.gh
aquacultured sea bass (Dicentrarchus labrax). Radiation Physics and Chemistry, 76, 
1169-1178. 
Özoğul, F., & Özoğul, Y. (2000). Comparision of methods used for determination of total 
volatile basic nitrogen (TVB-N) in rainbow trout (Oncorhynchus mykiss). Turkish 
Journal of Zoology, 24(1), 113-120. 
Özogul, Y., Özogul, F. H., Çi˙ çek, E., Polat, A., & Kuley, E. (2009). Fat content and fatty acid 
compositions of 34 marine water fish species from the Mediterranean Sea. International 
Journal of Food Sciences and Nutrition, 60(6), 464-475. 
Pearson, D. (1970). The chemical analysis of foods. 6th Edn. Longman Group Ltd. 
Pemberton-Pigott, C., Robinson, J., Kwarteng, E., & Boateng, L. (2016). Low PAH Improved 
Fish Smoking Stove Design Development Report. Narragansett, RI: Coastal Resources 
Center, Graduate School of Oceanography, University of Rhode Island and Netherlands 
Development Organisation. GH2014_ACT063_SNV.: The USAID/Ghana Sustainable 
Fisheries Management Project (SFMP). 
Plahar, W. A., & Amevor, W. K. (2003). Effect of storage methods on some quality indices of 
smoked herrings (Sardinella eba). Accra: Food Research Institute. 
Plahar, W. A., Nerquaye-Tetteh, G. A., & Annan, N. T. (1999). Development of an intergrated 
quality assurance system for the traditional Sardinella sp. and anchovy fish smoking 
inductry in Ghana. Food Control, 10, 15-25. 
Rabaçal, M. F. (2013). Combustion and emission characteristics of a domestic boiler fired with 
pellets of pine, industrial wood wastes and peach stones. Renewable Energy, 51, 220-
226. 
Regulska-Ilow, B., Ilow, R., Konikowska, K., Kawicka, A., Róźańska, D., & Bochińska, A. 
(2013). Fatty acid profile of the fat in selected smoked marine fish. Roczniki 
Państwowego Zakładu Higieny, 64(4), 299-307. 
 180 
University of Ghana http://ugspace.ug.edu.gh
Reitznerová, A., Šuleková, M., Nagy, J., Marcinˇcák, S., Semjon, B., Čertík, M., & Klempová, 
T. (2017). Lipid peroxidation process in meat and meat products: a comparison study 
of malondialdehyde determination between modified 2-thiobarbituric acid 
spectrophotometric method and reverse-phase high-performance liquid 
chromatography. Molecules, 22. 
Riede, K. (2004). Global register of migratory species: from global to regional scales: final 
report of the R&D-Projekt 808 05 081. Bonn, Germany: Federal Agency for Nature 
Conservation. 
Rodríguez, A., Losada, V., Larraíne, M. A., Quitral, V., Vinagre, J., & Aubourg, S. P. (2007). 
Development of lipid changes related to quality loss during the frozen storage of farmed 
coho salmon (Oncorhynchus kisutch). Journal of the American Oil Chemists' Society. 
84 (4), 727–734. 
Rørå, A., Kvåle, A., Mørkøre, T., Rørvik, K.-A., Hallbjoørn, S., Thomassen, S., & Magny, S. 
(1998). Process yield, colour and sensory quality of smoked Atlantic salmon (Salmo 
salar) in relation to raw material characteristics. Food Research International, 31 (8), 
601-609. 
Rosa, R., & Nunes, M. L. (2003). Nutritional quality of red shrimp, Aristeus antennatus 
(Risso), pink shrimp, Parapenaeus longirostris (Lucas), and Norway lobster, Nephrops 
norvegicus (Linnaeus). Journal of the Science of Food and Agriculture, 84, 89-94. 
Roy, M. M., & Corscadden, K. W. (2012). An experimental study of combustion and emissions 
of biomass briquettes in a domestic wood stove. Applied Energy, 99, 206-212. 
Russell, A. T. (2013). Combustion emissions. Air Pollution and Cancer, 161. 
Ryder, J., Karunasagar, I., & Ababouch, L. (2014). Assessment and management of seafood 
safety and quality: current practices and emerging issues. Rome: FAO Fisheries and 
Aquaculture Technical Paper No. 574. 
 181 
University of Ghana http://ugspace.ug.edu.gh
Sabry, M., El-Moein, K. A., Hamza, E., & Kader, F. A. (2016). Occurrence of Clostridium 
perfringens Types A, E, and C in fresh fish and its public health significance. Journal 
of Food Protection, 79(6), 994-1000. 
Salaudeen, M. M. (2013). Quality analysis of dried cod (Gadus morhua) heads along the value 
chain from Iceland to Nigeria. Reykjavik: United Nations University Fisheries Training 
Programme, Iceland [final project]. 
Salma, E. G., & Nizar, M. (2015). Atlantic mackerel amino acids and mineral contents from 
the Tunisian middle eastern coast. International Journal of Agricultural Policy and 
Research, 3(2), 77-83. 
Samaha, I. A., Amer, A. A., Abd-Elshahid, Y. S., & El-bialy, S. M. (2015). Mycological 
evaluation of imported frozen Fish. Alexandria Journal of Veterinary Sciences, 45, 127-
131. 
Sarma, D., Akhtar, M. S., Das, P., Das, P., Shahi, N., Ciji, A., . . . Debnath, D. (2013). 
Nutritional quality in terms of amino acid and fatty acid of five coldwater fish species: 
Implications to human health. National Academy Science Letters, 36(4), 385-391. 
Shah, A. A., Tokunaga, C., Kurihara, H., & Takahashi, K. (2009). Changes in lipids and their 
contribution to the taste of migaki-nishin (dried herring fillet) during drying. Food 
Chemistry, 115(3), 1011-1018. 
Shen, G., & Xue, M. (2014). Comparison of carbon monoxide and particulate matter emissions 
from residential burnings of pelletized biofuels and traditional solid fuels. Energy & 
Fuels, 28(6), 3933-3939. 
Sigurgisladottir, S., Sigurdardottir, M. S., Torrissen, O., Vallet, J. L., & Hafsteinsson, H. 
(2000). Effects of different salting and smoking processes on the microstructure, the 
texture and yield of Atlantic salmon (Salmo salar) fillets. Food Reseach International, 
33(10), 847-855. 
 182 
University of Ghana http://ugspace.ug.edu.gh
Silva, H. A., Mendes, R., Nunes, M. L., & Empis, J. (2006). Protein changes after irradiation 
and ice storage of horse mackerel (Trachurus trachurus). European Food Research and 
Technology, 224, 83-90. 
Sippula, O. H.-P. (2009). Comparison of particle emissions from small heavy fuel oil and 
wood-fired boilers. Atmospheric Environment, 43(32), 4855-4864. 
Sogbesan, A. O., & Ibrahim, P. (2017). Assessment of smoked fish quality using two smoking 
kilns and hybrid solar dryer on some commercial fish species in Yola, Nigeria. Journal 
of Animal Research and Nutrition, 2(6), 7. 
Stadlmayr, B., Charrondiere, U. R., Addy, P., Samb, B., Enujiugha, V. N., Bayili, R. G., . . . 
Burlingame, B. (2010). Composition of selected foods from West Africa. Rome: Food 
and Agriculture Organization. 
Stamatis, N., & Arkoudelos, J. (2007). Quality assessment of Scomber colias japonicus under 
modified atmosphere and vacuum packaging. Food Control, 292-300. 
Stewart, E. M. (2004). Food irradiaition: more pros than cons? Biologist, 51(2), 91-96. 
Stogiannidis, E., & Laane, R. (2015). Source characterization of polycyclic aromatic 
hydrocarbons by using their molecular indices: an overview of possibilities. (D. M. 
Whitcare, Ed.) Reviews of Environmental Contamination and Toxicology, 234. 
Stołyhwo, A., & Sikorski, Z. E. (2005). Polycyclic aromatic hydrocarbons in smoked fish – a 
critical review. Food Chemistry, 91, 303-311. 
Su, Y. C., Duan, J., & Morrissey, M. T. (2004). Electron beam irradiation for reducing Listeria 
monocytogenes contamination on cold-smoked salmon. Journal of Aquatic Food 
Product Technology, 13(1), 3-11. 
Surendra, I. H., Edirisinghe, E. M., & Rathnayake, R. M. (2018). Effect of low-dose gamma 
irradiation on the quality of tilapia fish muscle with storage at 0°C. American Journal 
of Food Science and Technology, 6(6), 237-246. 
 183 
University of Ghana http://ugspace.ug.edu.gh
Suryanti, D. I., & Suryaningrum, T. D. (2017). Proximate composition and sensory 
characteristics of traditional and oven-drying smoked tilapia fillets enriched with olive 
oil. Squalen Bulletin of Marine and Fisheries Postharvest and Biotechnology, 12(3), 
127-137. 
Taylor, A. (2010). W179 Wood Products Information - Moisture Content of 'Seasoned' 
Firewood. The University of Tennessee Agricultural Extension Service, W179 08-
0079. 
Teixeira, B., Marques, A., Mendes, R., Gonçalves, A., Fidalgo, L., Oliveira, M., . . . Nunes, M. 
L. (2014). Effects of high-pressure processing on the quality of sea bass (Dicentrarchus 
labrax) fillets during refrigerated storage. Food Bioprocess Technology, 7(5), 1333-
1343. 
Telahigue, K., Hajji, T., Rabeh, I., & El Cafsi, M. (2013). The changes of fatty acid 
composition in sun dried, oven dried and frozen hake (Merluccius merluccius) and 
sardinella (Sardinella aurita). African Journal of Biochemistry Research, 7(8), 158-
164. 
The World Bank. (2011). Project appraisal document on a proposed credit in an amount of 
SDR 31.1 million (US$50.3 million equivalent) and a proposed grant from the Global 
Environment Facility Trust Fund in an amount equal to US$3.5 million to the Republic 
of Ghana for the Ghana . Washinton DC: Report No. 57898-GH. The World Bank 
Group. 
Tiwo, C. T., Tchoumbougnang, F., Nganou, E., Pankaj, K., & Nayak, B. (2019). Effect of 
different smoking processes on the nutritional and polycyclic aromatic hydrocarbons 
composition of smoked Clarias gariepinus and Cyprinus carpio. Food Science & 
Nutrition, 7, 2412– 2418. 
 184 
University of Ghana http://ugspace.ug.edu.gh
Tobin, L., & Gormley, R. (2016). Fish freshness at retail level. UCD Institute of Food & 
Health. Dublin: SeaHealth-UCD. Issue 21. 
Trozzi, C. N. (2013). Small combustion. EEA, Emission Inventory Guidebook. 
UNDP. (2018). Human Development Indices and Indicators: Ghana. United Nations 
Development Programme. 
UNDP/TCDC. (2001). Sharing innovative experiences. Vol. 5, Examples of successful 
initiatives in agriculture and rural development in the South. Chapter 3. Retrieved 
September 2, 2016, from 
http://tcdc2.undp.org/GSSDAcademy/SIE/Docs/Vol5/improved.pdf  
UNU-FTP. (2017). UNU short course on improved fish smoking in Kigoma, Tanzania. 
Retrieved May 15, 2017, from http://www.unuftp.is/en/about-us-
frontpage/frettir/short-course-in-kigoma-tanzania  
USEPA. (2000). Risk based concentration table . Philadelphia, PA, Washington, DC: USEPA. 
USEPA. (2004). Risk assessment guidance for superfund volume I: Human health evaluation 
manual (Part E, Supplemental Guidance for Dermal Risk Assessment). Washington, 
DC: USEPA. 
USEPA. (2013). Regional Screening Levels for Chemical Contaminants at Superfund Sites. 
Washington DC: Environmental Protection Agency. Office of Research and 
Development. 
Visciano, P., Schirone, M., Tofalo, R., & Suzzi, G. (2012). Biogenic amines in raw and 
processed seafood. Frontiers in Microbiology, 3, 10 pp. 
WHO. (2000). Irradiation of fish, shellfish and frog legs. A compilation of technical data for 
authorization and control (IAEA-TECDOC-1158). Vienna: International Atomic 
Energy Agency. 
 185 
University of Ghana http://ugspace.ug.edu.gh
WorldData.info. (2019). WorldData.info. Retrieved June 01, 2019, from Average sizes of men 
and women: https://www.worlddata.info/average-bodyheight.php  
Xia, Z., Duan, X., Qiu, W., Liu, D., Wang, B., Tao, S., . . . Hu, X. (2010). Health risk 
assessment on dietary exposure to polycyclic aromatic hydrocarbons (PAHs) in 
Taiyuan, China. Science of the Total Environment, 408, 5331-5337. 
Xiu, M., Stevanovic, S., Rahman, M. M., Pourkhesalian, A. M., Morawska, L., & Thai, P. K. 
(2018). Emissions of particulate matter, carbon monoxide and nitrogen oxides from the 
residential burning of waste paper briquettes and other fuels. Environmental Research, 
167, 536-543. 
Yesudhason, P., Al-Zidjali, M., Al-Zidjali, A., Al-Busaidi, M., Al-Waili, A., Al-Mazrooei, N., 
& Al-Habsi, S. (2013). Histamine levels in commercially important fresh and processed 
fish of Oman with reference to international standards. Food Chemistry, 140, 777-783. 
Zandaryaa, S., & Buekens, A. (2009). Control of sulphur oxides. Pollution Control 
Technologies, 2, pp. 1-10. 
Zhai, H., Yang, X., Li, L., Xia, G., Cen, J., Huang, H., & Hao, S. (2012). Biogenic amines in 
commercial fish and fish products sold in southern China. Food Control, 25, 303-308. 
 
 
 
 
 
 
 
 
  
 186