The Contribution of Bank and Surface Sediments to 
Fluvial Sediment Transport of the Pra River
J. M. Kusimi
Dept. of Geography and Resource Development, University of Ghana, Legon
Email: jmkusimi@ug.edu.gh
Abstract
Sediment source studies involving a simple mixing model was undertaken in the Pra River Basin in Ghana 
using a single tracer 210Pb to determine the relative contribution of surface and bank sediments to the fluvial 
sediment transport. Sediment source tracing was performed on the basis of sub-basins by comparing the 
concentration of 210Pb in fluvial sediments to the bank sediments and potential surface sediment sources. The 
potential sediment source types sampled for analysis included surface sediments from arable top soils, illegal 
mining sites, path/untarred roads leading to rivers, gullies and gutters from settlements and farms. For bank 
erosion, river channel bank materials were sampled. Lead-210 fallout was determined by alpha spectrometry 
using the low background Gas-less Automatic Alpha counting system (Canberra iMatic™).  Results showed 
that bank material was the dominant sediments and accounted for over 60% of suspended sediment loads in all 
tributaries. Measures should be put in place to control the entrainment of bank materials since bank sediments 
constitute a larger proportion of the fluvial sediments. High fluvial sediment load is known to have 
geomorphological, hydrological, water resource management and ecological implications. 
Introduction USA, and several attempts have been 
High fluvial sediment transport has been made to quantify this source (Collins et al. 
recognised as a major setback to the 1997; Collins et al., 2001; Nagle et al., 
sustainability of reservoirs/dams for 2007; Gellis, 2010). In response to the 
hydro-power generation, potable water problems associated with traditional 
supply, and irrigation (e.g. Alam et al., monitoring and measurement techniques, 
2007; Hazarika and Honda, 2001; Peng et the fingerprinting approach has been 
al., 2008; Schwartz & Greenbaum, 2009). increasingly employed as a means of 
Also, watershed sediment transport can establishing the relative importance of 
lead to a number of environmental potential catchment sediment sources 
problems, including decreases in (Caitcheon et al. 2012; Collins et al., 2001; 
ecological diversity, and decreases in Collins et al., 2013a; Collins et al., 2013b; 
aesthetic properties of rivers and streams, Walling & Collins, 2000; Wilkinson et al., 
impeding navigation and river channel 2013).
morphology and stability (Davis & Fox,    The most commonly used tracers 
2009). 137 210include; radionuclides ( Cs, Pb) (Nagle 
These sediments are derived from bank et al. 2007; Walling, 2004), and 
7
and surface erosion processes. Eroding cosmogenic isotopes ( Be) (Schuller et al. 
channel banks are thought to be a major 2006; Walling, 2004). 
137Cs has a core 
source of sediment in some regions of the depth profile not greater than 20 cm and is 
West African Journal of Applied Ecology, vol. 25(1), 2017: 69–85.
70 West African Journal of Applied Ecology, vol. 25(1), 2017
anthropogenically introduced into the atmosphere is currently near zero or near 
environment through fallouts of nuclear non-detectable limits (Matisoff & 
activities of bombs of 1950s and 1960s and Whiting, 2011) in most parts of the world 
reactors such as the 1986 Chernobyl except northern Europe where release 
disaster. Lead-210 (t1/2 = 22years, with core from the Chernobyl explosion was higher. 
depth of about 10 cm) and beryllium-7 (t1/2 Also due to the short half life of 
7Be, it is 
= 53days and depth profile not exceeding 3 only suitable for simulating erosion of 
cm) are natural fallouts from the small catchments.
atmosphere (Blake et al., 2002; Matisoff The fingerprinting technique could 
and Whiting, 2011), hence are ubiquitous either be a simple mixing model using only 
on the earth’s surface and thus are suitable one diagnostic tracer or a composite 
as environmental radionuclides tracers mixing model involving a combination of 
everywhere. Lead-210 is a product of two or more tracers. Some researchers 
atmospheric decay of 222Rn gas (fallout (Walling et al., 1993, Yu & Oldfield, 1989; 
210Pb) and in situ decay of 226Ra. In most Molinaroli et al., 1991) have, however, 
210 226
soils Pb formed by in situ decay of Ra argued that no single diagnostic property 
will be in approximate equilibrium with of sediment can reliably distinguish 
226Ra, and usually is defined as ‘supported different sources, because individual 
210 210
Pb’. Fallout Pb in a soil or sediment tracers may be subject to physical and 
sample is the excess of 210Pb activity over chemical changes, which limit their use, 
the 226Ra supported component. This is e.g. particle size sorting, organic matter 
210
known as ‘unsupported’ or ‘excess’ Pb s e l e c t i v i t y ,  a n d  g e o c h e m i c a l  
(210Pbex) (Wallbrink and Murray, 1996; transformation during fluvial erosion and 
Motha et al. 2002; Wallbring et al., 1998; transportation (Collins et al., 1997). Also, 
individual properties may be unreliable 
Walling, 2004). 
137 7 210 because of spurious source-sediment 
Cs, Be, and Pb are each suitable as 
matches (Yu & Oldfield, 1989; Molinaroli 
particle tracers because they have a global 
et al., 1991; Walling et al., 1993). 
distribution, adsorb efficiently to soil 
However, Nagle et al. (2007) have 
particles and thus move with soil, and are effectively used a simple mixing model of 
relatively easily measured (Matisoff & 137Cs to distinguish between sediment from 
Whiting, 2011). Also these environmental surface sources and gullies. Lead-210 
radionuclide tracers are effective for (Brigham et al., 2001; Motha et al., 2002) 
distinguishing between surface-derived and Berryllium-7 (Schuller et al., 2006) 
sediments from sheet and shallow rill have also been used singularly in sediment 
erosion and sediments from gullies and source tracing. Sediment source tracing 
stream channel walls, because channel and has also been performed successfully in a 
gully walls deeper than their profile depths subset of intermittent streams using 
of between 3 and 30 cm usually contain amorphous to crystalline ratios of iron to 
little or no traces of the radionuclides estimate the fraction of sediment coming 
(Nagle et al., 2007; Matisoff & Whiting, from in-stream vs. landscape sources 
137
2011). However, Cs fallout from the (Schoonover et al., 2007). 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 71
There are three basic approaches to the required to design effective sediment and 
measurement of 210Pb activity: (i) by non-point pollution control strategies. 
gamma spectrometry, measuring the Sediment source tracing studies will 
gamma radiation from 210Pb directly; (ii) by provide an improved understanding of 
beta spectrometry either counting the 210Pb erosion and suspended sediment transport 
activity directly or after in-growth of its within a basin which is an essential 
daughter 210Bi; and (iii) by alpha precursor to establish sediment budgets, 
spectrometry of the granddaughter 210Po, develop distributed sediment yield models, 
assuming radioactive equilibrium exists and interpret sediment yields in terms of 
between the two nuclides (Johansson, landscape evolution (Walling et al., 1993). 
2008). In sediments, 210Pb and 210P are often However, sediment provenance data are 
found in equilibrium and the activity of lacking for many areas of the world, 
210 210
Pb can be estimated from that of Po. including Africa, because of the spatial and 
The 210Po is typically analysed by alpha temporal sampling constraints and the 
spectrometry following pre-concentration operational difficulties associated with 
and spontaneous deposition onto silver most traditional measurement techniques 
discs (Johansson, 2008). (Peart & Walling, 1988). Consequently, 
The Pra River Basin in Ghana has been very little studies of this kind have been 
engulfed by certain anthropogenic undertaken in Ghana. Most studies have 
activities such as illicit logging, farming, been centred on fluvial sediment transport 
urbanization and illegal small scale mining (Akrasi, 2005; Akrasi and Ansa-Asare, 
(Mensah, 2012). Serious concerns have 2008; Akrasi, 2011; Amisigo and Akrasi, 
been raised by stakeholders such as Water 1998; Boateng et al. 2012; Kusimi, 2008).  
Resources Commission of Ghana, NGOs, In view of the necessity for accurate and 
District and Municipal Assemblies and reliable sediment source data in 
chiefs of the level of pollution due to the developing countries and the success of the 
release of chemicals and sediments into fingerprinting approach in providing such 
water bodies by these human activities. data in other environments (Collins et al., 
The high concentration levels of fine 2001), this paper reports the application of 
sediments are causing the breakdown of sediment mixing fingerprinting technique 
filters in treatment plants of Ghana Water to identify sediment sources in the Pra 
Company Limited (GWCL) in most urban Bas in  for  e ffec t ive  r iver  bas in  
water supply systems within the basin. management. The paper is just an aspect of 
In order to solve this problem, there is a study that investigated the sediment 
the need to identify the sources of these yield, bank erosion and sediment sources 
sediments because elevated sediment of the basin in order to ascertain sediment 
volumes have an array of detrimental provenance in the basin. When the major 
impacts which threaten sustainable sources of sediments in the basin are 
ecosystem functioning, increasing water known, it will facilitate the development 
treatment costs, flood risk problems etc and implementation of appropriate control 
(Copper et al., 2015). Such information is measures towards protecting water 
72 West African Journal of Applied Ecology, vol. 25(1), 2017
resources. This paper presents only rainfall amount is between 125 and 200cm. 
findings on the sediment source tracking The dry season spans from November to 
aspect of the study; the sediment yield March and the dominant winds during this 
component has been published by Kusimi period are the dry harmattan winds. 
et al., 2014.  Temperatures are high throughout the year 
with  the highest  mean monthly 
Materials and methods temperature being 30 °C occurring 
Physical setting of the study area between March and April and the lowest is 
The Pra River Basin is located between about 26 °C which occurs in August 
Ð Ð
latitudes 5º00 N and 7º15 N and (Dickson & Benneh, 1995). 
longitudes 0º03 ÐW and 2º80 ÐW (Fig.1) in The vegetative cover is that of the moist 
south central Ghana. The major tributaries semi-deciduous forest which is heavily 
of the basin include the Ofin, Oda, Anum logged. Trees grow to heights of about 
and Birim Rivers which drain from the 35–45 m or more. The forest consists of 
Mampong-Kwahu and Atewa Mountain trees, lianas, climbers and shrubs/bushes. 
Ranges. The drainage basin area is 23,188 Due to the rapid expansion of cocoa 
km2 with a mean annual discharge of 214 plantations and the shifting cultivation 
m3s-1 (Akrasi and Ansa-Asare, 2008). The system of farming in this zone, very little of 
basin is generally of low relief the original forest remains and most of 
characterised by undulating topography what is left is secondary growth. The size 
with an average elevation of about 450 m of trees in this belt therefore depends on 
above sea level. The Pra River drains into how long the forest has been allowed to 
the Gulf of Guinea. regenerate (Dickson & Benneh, 1995). 
The main soil type of the catchment is However there are still a number of 
forest ochrosols which are alkaline. The protected Forest Reserves within the basin 
soils are weathered from the Tarkwaian which give good protection to the 
geological formations composing of landscape. 
sandstones and granitoids and metamor- T h e  P r a  B a s i n  t r a n s c e n d s  4  
phosed rocks such as phyllites and schists. administrative regions of Ghana namely; 
The soils are clayey and not well leached; the Ashanti, Eastern, Central and Western 
hence have the capacity to retain more Regions. Consequently, it serves as the 
moisture and are very cohesive (Dickson source of water supply for both 
& Benneh, 1995). commercial and domestic activities for 
   The climate of the basin is the wet most urban centres and other settlements 
semi-equitorial and is characterized by within it. For instance, the Ofin sub-basin 
two rainfall maxima. The first rainy season is the main source of water supply to 
occurs between May–June with the Kumasi, the administrative capital of the 
heaviest rainfall falling in June and the Ashanti Region and its environs from two 
second season is from September– reservoirs, Barekese and Owabi Dams 
October. The rains are brought by the (Fig.1). Also the Birim River supplies 
south-west monsoons and the annual water to towns such as Kibi, Kade, Akim 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 7 3  
Fig.1. Map of the Pra River Basin
74 West African Journal of Applied Ecology, vol. 25(1), 2017
Oda among others. However, human sediment sources were identified through 
activities such as small scale mining and field investigations and interviews with the 
illicit logging are degrading the surface local communities to identify areas of 
water resources of the basin (Water active sediment erosion (Davis & Fox, 
Resources Commission, 2011). 2009). 
The primary land uses within the Pra To accommodate local  spat ia l  
210
Basin are settlement development, variability in Pb in arable top soils and 
farming, logging and mining. The basin illegal mining sites, core samples of depth 
contains the highest density of settlements 10 cm within plot sizes of 15×15 m
2 at grid 
both rural and urban in the country. It also intervals of 3× 3 m were taken at each site 
contains most of the largest cocoa growing with a 5 cm diameter pvc pipe as a corer.  
areas in south central Ghana. Agricultural R e c e n t l y  e r o d e d  p a t c h e s  f r o m  
practices include cash crop such as cocoa, paths/untarred roads, sidewalls of gullies 
oil palm and food crops. It has a high and gutters not beyond 10 cm deep (since 
concentration of mining activities mainly Lead-210 depth profile does not exceeding 
concerned with gold exploitation. Mining 10 cm) were randomly collected using the 
i s  under t aken  by  mul t ina t iona l  scapper. Same samples at each site were 
corporations and small scale miners most composited and one sample taken for 
of them being illegal miners. Some of the analysis (Nagle et al., 2007; Walling, 
large scale mining companies are 2004). Bank sediments were also collected 
AngloGold Ashanti and Perseus Mining by scrapping channel wall sediments not 
Ltd (Water Resources Commission, 2011). below 10 cm (Davis & Fox, 2009). 
The activities of illegal miners are of Suspended sediments were sampled at the 
primary concern as their activities lead to catchment outlets of sub-basins (Fig.2) 
the discharge of mine wastes or sediments using the depth integrated sampler based 
into the rivers. on the equal-width-incremental approach 
across each channel transect (Davis & Fox, 
Data collection andaAnalysis 2009; Edwards & Glysson, 1999; Matisoff 
Sediment source tracing was performed & Whiting, 2011) and composited for 
on the basis of sub-basins using the finger analysis to account for the spatial 
print method (Gruszowski et al., 2003). variability in 
210Pb concentration across the 
The following potential sediment source section. Collected surface sediments were 
types were sampled for analysis; for placed in labelled polyethene bags whiles 
surface erosion (arable top soils, illegal those of suspended were poured into 
mining sites, paths/untarred roads leading plastic bottles and preserved in iced chests 
to rivers, sidewalls of gullies and (Ohio EPA, 2001) to minimize biological 
gutters/ditches that drain into streams; and chemical changes from the time of 
recently eroded sediments at the end of collection to the time of analysis. Samples 
gullies and gutters/ditches) and for bank were collected in September at the peak of 
erosion channel banks were scrapped. the rainy season when flows were high 
because 210Sampling sites (Fig.2) of these potential Pb environmental radionuclides 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 7 5  
Fig. 2. Sampling sites
76 West African Journal of Applied Ecology, vol. 25(1), 2017
are known to fallout highest with high Where Cs (%) is the percentage 
precipitation (Matisoff & Whiting, 2011). contribution from surface sources; Pr is 
Samples were prepared for analysis 210the value of Pb for stream sediments, Ps 
through the following processes. All the concentration value of 210Pb in surface 
source material samples were air-dried, 210sediment sources and Pb the value of Pb 
ground in a mortar and dry-sieved to 63µm for bank materials. The use of the 
because radionuclides tend to have higher simplistic mixing model is acceptable 
sorption to finer soils (Collins et al., 2013a; when smaller samples and or two sources 
Matisoff & Whiting, 2011) and also to are involved. The procedure follows 
facilitate direct comparison of suspended previously published works (Nagle & 
sediment and source samples (Collins et Richie, 2004; Nagle et al., 2007). Where 
al., 2001). Suspended sediments were more sources are being discriminated, 
centrifuged and the supernatant discarded, more complex mixing model approaches 
recovered solids were oven dried. Both will be required (Caitcheon et al., 2012; 
suspended and surface sediment samples Collins et al., 2001; Collins et al., 2013a; 
were weighed transferred into planchets Collins et al., 2013b; Wilkinson et al., 
and set aside for analysis. 2 1 0Pb 2013). The level of significance of surface 
determination was by alpha spectrometry and bank material contributions to the 
following similar procedures of Ogundare fluvial sediment transport was statistical 
& Adekoya (2015) at the Ghana Atomic tested by performing the F-test using the R 
Energy Commission, Accra. Samples were programme. 
counted for 200 min to determine alpha 
activity concentration using the low Results and discussion 
background Gas-less Automatic Alpha Tables 1–4 illustrate the mean 
210
counting system (Canberra iMatic™) concentration of unsupported Pb for each 
calibrated with alpha (241Am) standards. of the surface samples as well as the mean 
The system uses a solid state silicon and median concentration of all the 
(Passivated implanted Planar Silicon, samples in all sub-basins. The mean 
PIPS) detector. Two measurements of concentration levels of bank sediments and 
central tendency, mean and median were stream sediments are also provided for 
extracted from results of the analysis. each sub-catchment. Except within the 
Ofin catchment, the median values were 
To quantify the relative contributions of 
210 generally lower than the means (Table 4: 
bank and surface sources, values of Pb Mean = 1.54; Median = 1.69 Bq/kg). Both 
concentration were analysed using a the mean and median values were higher 
simple mixing model (equation 1) based on than bank samples in the Oda Basin (Table 
210
mean Pb values of samples measured. 1–4). The lowest mean concentration of 
Equation (1) has been used effectively by  210unsupported Pb for bank materials was 
Mukunda et al. (2009), Nagle and Ritchie, 0.89 Bq/kg and was recorded in the Birim 
(2004) and Nagle et al. (2007); Basin and the highest value of 3.09 Bq/kg 
Cs =               …………….. (1), occurred in the Pra Sub-basin (Tables 1 and 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 77
TABLE 1
Mean and Median 210Pb concentration levels for stream sediment and potential source materials in Birim 
Sub-basin
Tracer Galamsey Road & Farm Settlement Gully Gutter Mean sur- Bank Stream 
(n = 7) Bridge (n = 8) (n = 8) (n = 3) (n= 2) face soil Sediment Sediment
 (n = 7) (n = 47) (n = 12)
Mean 1.45 1.79 1.43 1.65 1.06 0.74 1.34 0.89 1.10
Median of 
surface soils
Median 0.4 1.47 0.68 0.94 1.29 0.74 0.84 0.8 1.10
Units ((Bq/kg)
TABLE 2
Mean and Mean 210Pb concentration levels for stream sediment and potential source materials in the Pra 
Sub-basin
Tracer Galamsey Road & Farm Settlement Gully Gutter Mean sur- Bank Stream 
(n = 9) Bridge n = 7) (n = 9) ((n = 1) (n=10) face soil Sediment Sediment
(n = 21) (n = 72) (n = 15)
Mean 2.32 4.02 2.88 3.31 0.68 4.25 3.40 3.09 3.18
Median of 
surface soils
Median 2.42 1.40 2.74 3.73 0.68 3.83 2.58 2.28 3.18
Units ((Bq/kg)
TABLE 3
Mean and Median 210Pb concentration levels for stream sediment and potential source materials in Oda Sub-
basin
Tracer Galamsey Road & Settlement Gully Gutter Mean sur- Bank Stream 
(n = 5) Bridge (n = 4) (n = 2) (n = 2) face soil Sediment Sediment
(n= 4)  (n=29) (n=12)
Mean 0.44 0.55 1.01 2.34 2.96 1.77 2.12 2.24
Median of 
surface soils
Median 0.01 0.55 0.54 2.34 2.97 0.55 1.84 2.24
Units ((Bq/kg)
78 West African Journal of Applied Ecology, vol. 25(1), 2017
TABLE 4
Mean 210Pb concentration levels for stream sediment and potential source materials in Ofin Sub-basin
Tracer Galamsey Road & Farm Settlement Gully Gutter Mean sur- Bank Stream 
(n = 2) Bridge (n = 2) (n= 2) (n= 2) (n= 3) face soil Sediment Sediment
(n = 8) (n=27) (n = 8)
Mean 2.25 2.22 1.13 2.84 0.85 0.52 1.54 1.01 1.16
Median of 
surface soils
Median 2.25 2.47 1.13 2.84 0.85 0.01 1.69 0.84 1.16
Units ((Bq/kg)
TABLE 5
Contribution of bank material and surface soil sources to suspended sediment load in the various sub-
catchments
River Basin Birim Pra Oda Ofin
Source Type Surface Bank Surface Bank Surface Bank Surface Bank 
soil sediment soil sediment soil sediment soil sediment
210Pb (%) 47 53 29 71 34 66 28 72
210
2). However the lowest mean and median Pb radionuclide fallouts. Samples from 
surface soil values were recorded in Birim gutters/ditches were also high in 
210Pb 
Basin (1.34 Bq/kg – Table 1) whiles higher concentration levels with a minimum 
levels were in the Pra Sub-basin (3.40 value of 0.52 Bq/kg in the Ofin Basin 
Bq/kg) (Table 2). (Table 4) and a maximum of 4.25 Bq/kg in 
Roads that are not tarred (Plate 1a) and the Pra Sub-basin (Table 2). This is so 
exposed urban landscapes (Plate 1b) were because, recent eroded sediments at the 
found to have the highest mean values of ends of ditches are residuals of sediments 
unsupported 210Pb in surface samples as of radionuclide fallout being transported. 
In all the basins, unsupported 210compared to other surface samples. The Pb mean 
rationale for this high concentration of levels in bank sediments were found to be 
unsupported 210Pb in these samples is due lower than mean unsupported
 210Pb 
to the well exposed and less disturbed state samples of most surface sites. Lower 
of these landscapes as compared to soils of values were recorded in sites of intense 
farms, galamsey, and gullies which are surface and bank erosion (Tables 6; 8 and 
frequently being mixed/tilled and or 9).Areas of active bank erosion do not
210
eroded. Generally lower levels of 210Pb in promote the accumulation of Pb fallouts 
surface samples were found in disturbed in the soils of river banks as they are 
sites. The exposed road and urban soils constantly being washed away into the 
promote the accumulation of unsupported river. Nagle et al., 2007 also attributed low 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 79
(1a)  Untarred road leading to Ofin River (1b) Urban erosion - Eroded plant root at Konongo 
in the Pra sub-basin
(1c) Evidence of cantilever bank failure along the (1d) Mine waste along the Birim River at Ekorso 
bank of the Oda River at Asaago – Kumasi Brimso
levels of an environmental radionuclide ongoing (Table 6 and Plate 2a). Soils are 
137
Cs in bank sediments to active bank dug from pits of 20 m or more by these 
erosion. Another  reason for lower Pb210 miners where 
210Pb fallouts do not reach, 
210
levels on stream banks is that on a vertical because maximum concentrations of Pb 
bank, much of the rainfall will run off and in soils are usually found at the surface 
not soaked into the bank face hence does decreasing exponentially with depth, and 
not enhance radionuclide accumulation. reaching undetectable levels at depths 
Similarly, recent illegal mine sites greater than 10 cm. The mixing of top soils 
(galamsey – Plate 2a) also recorded lower rich in 
210Pb with underground mined soils 
values of unsupported 210Pb (Tables 6–9). or the burial of top soils due to the mining 
For instance at Kibi, an abandoned illegal activities account for the low unsupported 
mine site that is becoming vegetated (Plate lead-210 values in recent mine sites. These 
2b) had almost 3 Bq/kg 210Pb concentration observations are in line with literature. 
as compared to an adjacent site with a According to Walling et al., (1993) in 
value of 1.2 Bq/kg where mining was undisturbed soils, 
210Pb are concentrated in 
80 West African Journal of Applied Ecology, vol. 25(1), 2017
TABLE 6
Results of bank sediments and surface soil analyses in the Birim Basin
Sample No. Bank Galamsey Road & Bridge Farm Settlement Gully Gutter
Sediment
1 0.79 2.99 3.29 3.28 4.43 1.89 0.01
2 0.01 1.20 0.89 4.21 0.84 0.01 1.47
3 2.59 4.95 3.43 0.17 2.48 1.29
4 0.01 0.01 0.45 0.67 2.95
5 0.93 0.20 1.47 0.69 1.03
6 2.28 0.40 1.61 0.67 0.01
7 0.81 0.40 1.40 0.21 0.85
8 0.56 1.53 0.64
9 0.47
10 1.24
11 0.16
12 0.85
Mean 0.89 1.45 1.79 1.43 1.65 1.06 0.74
Median 0.8 0.4 1.47 0.68 0.94 1.29 0.74
Units ((Bq/kg)
Table 7
Results of bank sediments and surface soil analyses in the Pra Basin
Sample No. Bank Galamsey Road & Bridge Farm Settlement Gully Gutter
Sediment
1 2.28 2.71 2.55 0.01 4.05 0.68 6.03
2 0.01 1.42 0.01 0.16 3.85 5.73
3 2.71 2.65 3.81 6.29 1.12 0.99
4 0.32 0.55 2.73 1.72 3.73 1.45
5 5.59 1.42 0.84 3.75 4.64 1.33
6 6.68 0.83 3.6 5.47 1.85 6.86
7 2.11 4.01 0.63 2.74 6.49 3.69
8 2.41 4.85 1.40 3.25 11.88
9 1.01 2.42 0.60 0.83 3.96
10 0.84 0.61 0.55
11 2.44 0.51
12 5.03 1.0
13 1.72 1.24
14 11.03 1.52
15 2.21 2.27
16 0.73
17 1.8
18 22.55
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 81
19 0.72
20 30.83
21 4.41
Mean 3.09 2.32 4.02 2.88 3.31 0.68 4.25
Median 2.28 2.42 1.40 2.74 3.73 0.68 3.83
Units ((Bq/kg)
TABLE 8
Results of bank sediments and surface soil analyses in the Oda Basin 
Sample No. Bank Galamsey Road & Bridge Settlement Gully Gutter
sediment
1 0.01 0.01 0.05 0.01 1.36 5.7
2 1.24 0.01 3.05 0.01 3.32 0.2
3 1.35 0.01 1.07 3.67
4 1.99 0.91 2.95 2.26
5 2.68 1.25
6 2.27
7 8.27
8 2.27
9 1.10
10 1.69
11 2.54
12 0.03
Mean 2.12 0.44 0.55 1.01 2.34 2.96
Median 1.84 0.01 0.55 0.54 2.34 2.97
Units ((Bq/kg)
Table 9
Results of bank sediments and surface soil analyses in the Ofin Basin Basin 
Sample No. Bank Galamsey Road & bridge Farm Settlement Gully Gutter
sediment
1 3.05 2.81 4.74 1.36 3.57 0.01 0.01
2 0.01 1.69 3.37 0.89 2.1 1.69 0.01
3 0.56 0.01 1.53
4 0.01 0.01
5 1.64 4.08
6 1.11 1.56
7 0.58 0.48
8 1.09 3.48
Mean 1.01 2.25 2.22 1.13 2.84 0.85 0.52
Median 0.84 2.25 2.47 1.13 2.84 0.85 0.01
Units ((Bq/kg)
82 West African Journal of Applied Ecology, vol. 25(1), 2017
(2a) Illegal mining site where mining is on-going (2b). Abandoned illegal mining 
site becoming vegetated 
the upper 10 cm of the soil profile, whereas 
in cultivated soils this radionuclide will be transport in the main Pra Basin and its sub-
mixed throughout the plough layer and catchments with a p-value of 0.0003 and an 
surface concentrations will be much lower. R2-value of 0.99. The p-value of surface 
Also subsoil horizons below 25 cm and soil is however 0.08 with a higher AIC 
exposed river banks (apart from the upper value of -4.9 as compared to -27.3 for bank 
10 cm) will contain zero or only very low sediments. 
levels of unsupported 210Pb. Also lower Due to the good vegetative cover of the 
levels of 137Cs in certain plots relative to landscape, soils are well protected and 
others were explained by soil mixing due coupled with low gradient of the catchment 
to cultivation (Nagle & Ritchie, 1999). which is less than 20 degrees, surface 
The simple mixing model results of erosion will be quiet low. Over 81% of the 
equation (1) using mean values show that a catchment is protected by vegetative cover 
higher percentage contribution of c o m p o s i n g  o f  f o r e s t  r e s e r v e s ,  
sediments into the fluvial sediment open/secondary growth, savannah and 
transport is coming from bank sediments coastal scrub/grassland with only about 
with a range of 53–72% within the sub- 22% being covered by bushes/cropland 
catchments (Table 5). The lowest and built-ups/barelands (Kusimi et al., 
estimated bank sediment contribution is in 2014; Kusimi et al., 2015). Collins et al., 
the Birim Basin (53%) whiles the highest is 1997 and Collins et al., 2001 also observed 
recorded in the Ofin Basin (72%). that, surface erosion is restricted by higher 
Conversely, the lowest surface soil density provided by natural vegetation and 
sediment source (28%) is in the Ofin Basin plant litter. 
whiles the highest surface sediment source On the other hand, channel erosion at the 
is in the Birim Basin (47%) (Table 5). The middle section and activities of alluvial 
model results are buttressed by a statistical small scale mining activities within the 
test which showed that stream bank rivers entrain much bank materials into the 
erosion is a significant source of sediment rivers and this accounts for higher bank 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 83
Plate 3: Illegal alluvial gold mining of the river bed and bank of the Ofin River. 
material tracer sources.  At the peak of the results in the remobilization of deposited 
rainy season river banks were flooded to sediments into the sediment transport 
bank full stage making them susceptible to system. These processes generate plumes 
mass failure and slumping (Plate 1c and of sediments in the rivers which discolours 
Plates 4ab) due to bank wetness and the the water. There are hundreds of these 
removal of vegetative cover. Bank toe miners strung along all the major 
undercutting during the falling limb stage tributaries and most streams within the 
of flow hydrograph results in the removal basin. 
of bank material and bank collapse into As part of the same study, Kusimi et al., 
river channels. Bank erosion was observed 2014 found sediment yield to be increasing 
to be very widespread in the middle course downstream in all sub-catchments. This is 
of the tributaries. Wasson et al. (2010) also inconsistent with the inverse relationship 
found that in the mid to lower Daly River model between sediment delivery ratio and 
channel bank erosion is an important basin area which shows that sediment yield 
sediment source due to increases in annual per unit area generally declines with 
rainfall and discharge. increasing catchment area. It is well known 
Secondly, alluvial gold mining is being that for most large river systems, most of 
undertaken along channel banks and bed the sediment which enters the system does 
(Plates 1d and 3) and this entrains not get transported to the outlet of the river 
sediments directly into the river or it owing to decreasing slope and channel 
84 West African Journal of Applied Ecology, vol. 25(1), 2017
(4a) Flooded channel of the Birim River in the (4b) Flooded channel of the Oda River at ´
middle section where illegal mining is Bepotenten near an illegal mining site
occurring
(4d) Colour of the Oda River at Bepotenten in the 
(4c) Clear water in the upper course at a non- dry season – same section as 4b which is near 
galamsey site at Ejisu on Oda River. an illegal mining site ´
gradients with increasing basin size and and of low virtual clarity or transparency 
this increases opportunities for either a (Plates 4a, b and d), an indication of high 
temporary/ permanent deposition to be turbidity and sediment load transport. 
associa ted  wi th  r iver  channels ,  
floodplains, valleys or the base of slopes Conclusion
(Caitcheon et al., 2012; Prosser et al., A sediment source study using a sediment 
2001). Kusimi et al. (2015) observed that, 2 1 0m i x i n g  m o d e l  i n v o l v i n g  P b  
in the upper sections of the rivers where environmental radionuclide fallout is 
illegal mining was not taking place, the employed to determine that much 
water was clear (Plate 4c) as compared to suspended fluvial sediments in transport in 
the middle and lower sections were illegal the Pra River are originating from bank 
mining was occurring. At the middle and materials. Surface sediments in transport 
lower sections, the water was very dirty were relatively low owing to less surface 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 85
erosional processes. Combinations of very subsidized rate. Field samples were 
illegal mining (galamsey) along the river processed for analyze without any charges 
and bank erosion are the plausible at the Ecological Laboratory of the 
explanatory factors accounting for the Department of Geography and Resource 
entrainment of bank sediments into the Development, University of Ghana. 
sediment transport budget. The method Samples were analyzed at Ghana Atomic 
used in this study did not allow the Energy Commission, Accra. Lastly, I will 
discrimination between the contributions like to express my gratitude to Mr. 
of the two subsurface sources to the fluvial Jonathan Kusimi, Mrs. Bertha Kusimi, Mr. 
sediment in transport in view of the small N. S. Opata (Ghana Atomic Energy 
number of samples collected for the study. Commission, Accra) and Mr. Gabriel 
The model results are however consistent Appiah (Water Research Instute, CSIR – 
with field observations and sediment load Accra) for their assistance in the field data 
analysis (Kusimi et al., 2014) which collection and analysis.
showed that much of the bank sediment 
injection is as a result of illegal mining References 
activities.   Hence, the activities of illegal Akrasi S. A. (2011). Sediment Discharges from 
Ghanaian Rivers into the Sea. West African 
mining in the Pra Basin and for that matter 
Journal of Applied Ecology 18: 1–13.
the whole country need to be controlled by Akrasi S. A. (2005). The assessment of suspended 
enforcing established policies and sediment inputs to Volta Lake. Lakes & 
legislations on small scale mining Reservoirs: Research & Management 10 (3): 
(Minerals and Mining Act, 2006 and 179–186.
National Mining Policy, 2010), Water Use Akrasi S. A. and Ansa-Asare O. D. (2008). 
Assessing Sediment and Nutrient Transport in 
(LI 1692) and land use (National Land use the Pra Basin of Ghana. W A J. Appl Ecol. 13: 
P o l i c y,  1 9 9 9 )  a s  t h e i r  p r o p e r  1–11.
implementa t ion  wi l l  ensure  the  Alam J. B., Uddin M., Ahmed J. U., Cacovean H., 
sustainable exploitation of the natural Rahman M. H., Banik B. K. and Yesmin N. 
resources of the Pra Basin without (2007). Study of Morphological Change of River 
Old Brahmaputra and Its Social Impacts by 
compromising its water quality.
Remote Sensing. Geogr Tech. 2: 1–11.
Amisigo B. A. and Akrasi S. A. (1998). A 
Acknowledgement Preliminary Assessment of the Suspended 
This paper is an extract of a PhD project Sediment Yields of Rivers in the Volta River 
which was partially funded by the Basin System of Ghana. Ghana Journal of 
University of Ghana by awarding me a Science, 38: 7–13.
Bentil N. L. (2011). Kyebi Water Plant Shut Down 
Faculty Development Grant which was as a Result of Evil Effects of Galamsey. Daily 
very helpful in paying for the cost of parts Graphic, 2 August 2011. 
of the field data collection and analysis. I Boateng I., Bray M. and Hooke J. (2012). 
also wish to acknowledge Water Research Estimating the fluvial sediment input to the 
Institute of the Council for Scientific coastal sediment budget: A case study of Ghana. 
Geomorphology, 138: 100–110.
Research, Accra for giving out their 
Brigham M. E., McCullough C. J. and Wilkinson 
equipment for field data collection on a P. (2001). Analysis of Suspended-Sediment 
86 West African Journal of Applied Ecology, vol. 25(1), 2017
Concentrations and Radioisotope Levels in the Ghana News Agency (2005). Owabi Dam to be de-
Wild Rice River Basin, Northwestern silted, Modern Ghana News: Published: 
Minnesota, 1973–98. Water-Resources Thursday, March 17, 2005. Retrieved from: 
Investigations Report 01–4192, US Geological http://www.modernghana.com/news/74031/1/o
Survey. wabi-dam-to-be-de-silted.html: 13th June, 2010.
Caitcheon G., Olley J., Pantus F., Hancock G. R. Gruszowski K. E., Foster I. D. L., Lees J. A. and 
and Leslie C. (2012). The dominant erosion Charlesworth S. M. (2003). Sediment sources 
processes supplying fine sediment to three and transport pathways in a rural catchment, 
major rivers in tropical Australia, the Daly (NT), Herefordshire, UK. Hydrol Process. 17: 
Mitchell (Qld) and Cloncurry (Qld) Rivers. 2665–2681.
Geomorphology. 151–152: 188–195. Johansson L. Y. (2008). Determination of Pb-210 
Collins A. L., Walling, D. E. and Leeks G. J. L. and Po-210 in aqueous environmental samples. 
(1997). Source type ascription for fluvial PhD Dissertation, Leibniz University of 
suspended sediment based on a quantitative Hanover, Germany. 
composite fingerprinting technique. Catena. 29: Kusimi J. M. (2008). Analysis of Sedimentation 
l–27. Rate in the Densu River Channel: the result of 
Collins A. L., Walling D. E., Sichingabula H. M. erosion and anthropogenic activities in the Densu 
and Leeks G. J. L. (2001). Suspended sediment basin. W A J. Appl Ecol. 14: 1–14.
source fingerprinting in a small tropical Kusimi J. M., Amisigo B. A. and Banoeng-Yakubo 
catchment and some management Implications. B. K. (2014). Sediment Yield of a Forest River 
Appl Geogr. 21: 387–412. Basin in Ghana. Catena. 123: 225–235. 
Collins A. L., Zhang Y. S., Duethmann D., Walling Kusimi J. M., Yiran G. A. B. and Attua E. M. (2015). 
D. E. and Black K. S. (2013a). Using a novel Soil Erosion and Sediment Yield Modelling in 
tracing-tracking framework to source fine- the Pra River Basin of Ghana using the Revised 
grained sediment loss to watercourses at sub- Universal Soil Loss Equation (RUSLE). Ghana J 
catchment scale. Hydrol Process. 27: 959–974. Geogr. 7 (2): 38–57. 
Collins A. L., Zhang Y., Hickinbotham R., Bailey Matisoff G. and Whiting P. J. (2011). Measuring 
G., Darlington S., Grenfell S. E., Evans R. and Soil Erosion Rates Using Natural (7B, 210Pb) and 
Blackwell M. (2013b). Contemporary fine- Anthropogenic (137Cs, 239,240Pu) Radionuclides. In 
grained bed sediment sources across the River Handbook o f  Environmental  I so tope  
Wensum Demonstration Test Catchment, UK. G e o c h e m i s t r y - A d v a n c e s  i n  I s o t o p e  
Hydrol Process. 27: 857–884. Geochemistry. (M. Baskaran, ed.), pp.487–519. 
Davis C. M., Fox J. F. (2009). Sediment DOI 10.1007/978-3-642-10637-8 25.
Fingerprinting: Review of the Method and Mensah K. (2012). Pra River basin under threat-
Future Improvements for Allocating Nonpoint Nana Sarpong. Business and Financial Times, 31 
Source Pollution. J. Environ Eng. 135 (7): O c t o b e r ,  2 0 1 2 .  R e t r i e v e d  f r o m :  
490–504. http://www.ghheadlines.com/agency/ business-
Dickson K. B. and Benneh G. (1995). A New and-financial-times/20121031/ 399941/pra-
Geography of Ghana. Longman, England. river-basin-under-threat: 20th June, 2013.
Edwards T. K. and Glysson G. D. (1999). Field Molinaroli E., Blom M. and Basu A. (1991). 
Methods for Measurement of Fluvial Sediments. Methods of Provenance Determination Tested 
In: U.S. Geological Survey Techniques of with Discriminant Function Analysis. J Sediment 
Water-Resources Investigations, Book 3, Petrol. 61 (6): 900–908.
Chapter C2, U.S. Geological Survey, Nagle G. N. and Ritchie J. C. (1999). The use of 
Information Services, Reston, Virginia. tracers to study sediment sources in three streams 
Elwell H. A. and Stocking M. A. (1975). in North-eastern Oregon.  Phys Geogra. 20 (4): 
Parameters for estimating annual runoff and soil 348–366.
loss from agricultural lands in Rhodesia. Water Nagle G. N. and Ritchie J. C. (2004). Wheat field 
Resour Res. 11 (4): 60–605. erosion rates and channel bottom sediment 
J. M. Kusimi: The contribution of bank and surface sediments to fluvial sediment transport 87
sources in an intensively cropped Northeastern Stocking M. A. and Elwell H. A. (1976). Rainfall 
Oregon Drainage Basin. Land Degrad Dev. 15: erosivity over Rhodesia. T I Brit Geogr NS 1: 
15–26. 231–245.
Nagle G. N., Fahey T. J., Ritchie J. C., Peter B. Wallbrink P. J., Murray A. S. and Olley J. M. 
and Woodbury P. B. (2007). Variations in (1998). Determining sources and transit times 
sediment sources and yields in the Finger Lakes of suspended sediment in the Murrumbidgee 
and Catskills regions of New York. Hydrol River, New South Wales, Australia, using 
fallout 137Cs and 210Process, 21: 828–838. Pb. Water Resour Res. 34: 
Ohio EPA (2001). Sediment Sampling Guide and 879–887.
M e t h o d o l o g i e s ,  2 n d  E d i t i o n ,  Wilkinson S., Hancock G., Bartley R., Hawdon 
http://www.epa.ohio.gov/portals/35/guidance/s A. and Keen R. (2013). Using sediment tracing 
edman2001.pdf: Accessed: 11th June, 2011. to assess processes and spatial patterns of 
Obi M. E. and Salako F. K. (1995). Rainfall erosion in grazed rangelands, Burdekin River 
parameters influencing erosivity in southeastern basin, Australia. Agr Ecosyst Environ. 180: 
Nigeria. Catena 24: 275–287. 90–102.
Ogundare F. O. and Adekoya O. I. (2015). Gross Walling D. E., Woodward J. C. and Nicholas A. 
alpha and beta radioactivity in surface soil and P. (1993). A Multi-Parameter Approach to 
drinkable water around a steel processing Fingerprinting Suspended-Sediment Sources, 
Facility. J Radiat Res Appl Sci. 8: 411–417. Tracers in Hydrology (Proceedings of the 
Peart M. R. and Walling D. E. (1988). Techniques Yokohama Symposium, July 1993). IAHS, 
for Establishing Suspended Sediment Sources in Publ. 215: 329–338.
Two Drainage Basins in Devon, UK: A Walling D. E. (2004). Using environmental 
Comparative Assessment, Sediment Budgets radionuclides to trace sediment mobilisation 
(Proceedings of the Porto Alegre Symposium, and delivery in river basins as an aid to 
December 1988). IAHS Publ. 174: 269–279. catchment management. Proceedings of the 
Peng W., Zhou J., He Z. and Yang C. (2008). Ninth International Symposium on River 
Integrated Use of Remote Sensing and GIS For Sedimentation. Yichang, China. 
Predicting Soil Erosion Process. The Water Resources Commission. (2011). Basins – 
International Archives of the Photogrammetry, Pra Basin. Retrieved from:  http://wrc-
Remote Sensing and Spatial Information gh.org/en/basins/17/pra-basin: 14th May, 
Sciences.  XXXVII. Part B4. Beijing. 2013.
Prosser I. P., Rutherfurd I. D., Olley J. M., Young Yu L. and Oldfield F. (1989). A Multivariate 
W. J., Wallbrink P. J. and Moran C. J. (2001). Mixing Model for Identifying Sediment Source 
Large-scale patterns of erosion and sediment from Magnetic Measurements. Quaternary 
transport in rivers networks, with examples from Res. 32:.168–181.
Australia. Freshwater Mar Res. 52: 81–99.