Communications in Soil Science and Plant Analysis
ISSN: 0010-3624 (Print) 1532-2416 (Online) Journal homepage: https://www.tandfonline.com/loi/lcss20
Simple Formulation of the Soil Water Effect on
Residue Decomposition
S. G. K. Adiku , N. K. Amon , J. W. Jones , T. A. Adjadeh , F. K. Kumaga , G. N.
Dowuona & E. K. Nartey
To cite this article: S. G. K. Adiku , N. K. Amon , J. W. Jones , T. A. Adjadeh , F. K. Kumaga ,
G. N. Dowuona & E. K. Nartey (2010) Simple Formulation of the Soil Water Effect on Residue
Decomposition, Communications in Soil Science and Plant Analysis, 41:3, 267-276, DOI:
10.1080/00103620903460781
To link to this article:  https://doi.org/10.1080/00103620903460781
Published online: 05 Feb 2010.
Submit your article to this journal 
Article views: 71
Citing articles: 2 View citing articles 
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=lcss20
Communications in Soil Science and Plant Analysis, 41:267–276, 2010
Copyright # Taylor & Francis Group, LLC
ISSN: 0010-3624 print / 1532-2416 online
DOI: 10.1080/00103620903460781
Simple Formulation of the Soil Water Effect on
Residue Decomposition
S. G. K. ADIKU,1,2 N. K. AMON,2 J. W. JONES,1
T. A. ADJADEH,2 F. K. KUMAGA,3 G. N. DOWUONA,2
AND E. K. NARTEY2
1Department of Agricultural and Biological Engineering, University of
Florida, Gainesville, Florida, USA
2Department of Soil Science, University of Ghana, Legon, Accra, Ghana
3Department of Crop Science, University of Ghana, Legon, Accra, Ghana
Soil water content, h, is a major factor affecting residue decomposition, but simple
formulation of this factor is often lacking. We observed that h significantly (P ,
0.001) affected the residue decomposition constant, kd. When h varied from
0.09 g g21 to 0.23 g g21, kd ranged from 0.009 to 0.013 d
21 and from 0.009 to
0.022 d21 for residues with carbon to nitrogen ratio (C/N) . 30 and C/N , 25,
respectively. A h factor was formulated in terms of the field capacity hFC and the air-
dry hd in the form fw 5 (h 2 hd) / (hFC 2 hd), and this was used to modify the
potential kd as h varied. Coupling fw with a first-order residue decomposition
equation resulted in the prediction of the decomposition of four residue types in the
greenhouse (R2 5 0.94; relative root mean square error, RRMSE, 5 0.06) and in
the field (R2 5 0.93; RRMSE 5 0.11).
Keywords Prediction, residue decomposition, soil water
Introduction
Residues constitute one of the major ways by which organic matter or carbon (C) is
introduced into the soil. In tropical agricultural systems, where application of
fertilizer to crops is often negligible, the soil organic matter (or organic C) is the main
reservoir of nutrients for plant growth (Bandaranayake et al. 2003). Hence,
managing of residues to increase soil organic carbon (SOC) is economically relevant
in tropical agriculture. Though residue addition and its subsequent conversion to soil
C attracted much research attention in the past, there is now a renewed interest in the
subject in response to global warming and climate change, which is attributable to
the increasing carbon dioxide (CO2) load of the atmosphere. It is thought that
increasing soil C by increasing residue additions to the soil might mitigate climate
change (Lal 2004). If residues must be managed to enhance the buildup of soil C,
then the factors that affect the conversion process must be assessed. The bulk of
Received 11 April 2008; accepted 12 August 2009.
Address correspondence to S. G. K. Adiku, c/o Dr. J. W. Jones, Department of
Agricultural and Biological Engineering, University of Florida, Gainesville, FL 32611-0570,
USA. E-mail: s.adiku@ufl.edu; s_adiku@hotmail.com
267
268 S. G. K. Adiku et al.
evidence indicates that residue decomposition is determined by two factors: (i) the
residue quality (Palm and Sanchez 1991; Seneviratne 2004) and (ii) environmental
factors, such as soil moisture and temperature (Stroo et al. 1989; Stott et al. 1990).
In this study, we focused on the role of soil water, given that in the tropics the
variability in rainfall, and hence soil water, far exceeds the variability in temperature;
therefore, residue decomposition patterns would largely be determined by soil water.
To predict residue decomposition dynamics, we needed to formulate the soil water
effect. In the past, indirect approaches were taken to formulate the soil water effect,
apparently because of the paucity of soil water measurements in residue
decomposition studies. For example, several researchers (e.g., Gregory et al. 1985;
Douglas and Rickman 1992; Steiner et al. 1999) used the concept of decomposition
days (DCD), a composite variable of temperature and precipitation, to describe the
time variation of residue mass loss under field conditions. In this approach, a
precipitation coefficient was used as proxy for a soil water factor. The precipitation
coefficient varied when the precipitation was less than or exceeded set thresholds.
Bristow et al. (1986) formulated the water factor in terms of soil and residue water
content; however, data on residue water content is often not available.
Our aim was to develop a simple formulation of the water factor that would
require only the minimum measured or estimated soil water information. We
assumed that the decomposing residue is in direct contact with the soil and would
therefore equilibrate with the current soil moisture. Thus, the knowledge of soil
water content alone would be sufficient to describe the effect of water on residue
decomposition. Further, given that the residue decomposition is rapid at field
capacity, slow when soil moisture declines to less than 40% field capacity, and stops
when the soil is air-dry (Vigil and Sparks 2002), we hypothesized a linear decline of
the decomposition rate between the field capacity and the air-dry soil water content.
The aim of this study was to test these hypotheses and derive a simple soil water
factor that could be used for predicting residue decomposition under variable soil
water conditions.
Materials and Methods
Data Sources
Residue decomposition was studied under both greenhouse and field conditions in
Ghana from 2004 to 2005. From June to December 2004, greenhouse studies were
conducted at the Departments of Soil Science and Crop Science, University of
Ghana. Four residue types, (i) elephant grass (Pennisetum purpureum), R1; (ii)
pigeon pea (Cajanus cajan), R2; (iii) cowpea (Vigna unguiculata), R3; and (iv)
mucuna (Mucuna puriens), R4, were obtained from a long-term maize–fallow
rotation study aimed at characterizing the effect of fallow residue management on
maize productivity and soil carbon accretion (Table 1). The residues, which were
harvested at their mature stages, were chopped into 2- to 3-cm lengths and oven
dried. Then, 8-g portions were placed in litter bags (10 cm 6 6 cm) made from 1-mm
nylon mesh for subsequent incubation studies. Subsamples of the plant materials
were used to determine C and total nitrogen (N) using standard analytical
procedures.
About 400 kg of a sandy loam topsoil (0–20 cm), sampled from the site of the
maize–fallow trial, was air dried for 72 h, ground, and passed through a 2-mm mesh
Soil Water Effect on Residue Decomposition 269
Table 1. Description and composition of plant residues
Type Description C (%) N (%) C/N ratio
R1 Elephant grass 47.0 1.58 29.8
R2 Pigeon pea 40.5 2.03 19.9
R3 Cowpea 42.1 1.79 23.5
R4 Mucuna 33.5 1.79 18.7
before used in the subsequent greenhouse experiment. The field capacity (FC) of the
soil was determined by saturating samples in pots, allowing drainage for 2 days, and
then measuring gravimetric soil water content (Table 2). The air-dry water content
was also determined gravimetrically. Next, 800-g portions of the sieved soils were
packed to a bulk density of 1.10 g cm23 into 1.2-L plastic pots, leaving about 2.0 cm
at the brim. The litter bags and their contents were buried in the potted soils to a
depth of about 2.0 cm, and thereafter, three water treatments were imposed as
follows: W1 (100% FC or h 5 0.23 g g21), W2 (70% FC or h 5 0.16 g g21), and W3
(40% FC or h 5 0.09 g g21). The water treatments were maintained throughout the
study period by regular weighing and replenishing any loss of water detected. There
were 96 pots in all, arranged in a completely randomized design. The greenhouse
temperature was between 28 and 32 uC, with an average of 30 uC.
Triplicate litter bags for each water treatment and residue type were retrieved at
10, 20, 30, 50, 80, 120, 150, and 180 days after the commencement of the experiment.
At each time of retrieval, the bags were gently shaken to remove all attached soil
particles, after which the contents were emptied into paper envelopes, dried at 60 uC
for 72 h, and weighed.
Assuming that the rate of residue mass loss was proportional to the current mass,
then
dYt
~{kd Y ð1Þ
dt
where Yt is the residue mass (g) at time t (d) and kd is the decomposition rate constant.
The analytical solution to equation (1) is
Y ~Y e{kd tt o ð2Þ
with Yo being the initial mass (g). In this study, the kd was expressed either per day (d
21)
or degree day [(uC d)21]. Curves were fitted to the mass loss with time, and Excel was used
to determine the kd for the different residue types and soil water treatments.
The field studies were carried out at the University of Ghana Farm, Legon, from
July to September 2005. Daily weather data, including rainfall and temperature,
Table 2. Description and composition of soil
Sand Silt Clay hFC hd Bulk density
Soil (%) (%) (%) (g g21) (g g21) (g cm23)
Greenhouse 65.0 15.0 20.0 0.23 0.02 1.10
Field 64.0 25.0 5.0 0.19 0.02 1.35
270 S. G. K. Adiku et al.
were obtained from the Ghana Meteorological Services Department, which was
situated about 4 km from the farm. The soil at the farm was loamy sand with
moisture constants given in Table 2. The field capacity and air-dry soil water
contents were determined using the same procedure as outlined for the greenhouse
study. There was no irrigation during the field study.
For the field study, only residue types R1, R2, and R4 were used. Three hundred
fifty grams of plant residues were chopped to about 30 cm long and placed in litter
bags made of 1-mm nylon mesh with dimensions of 72 cm 6 42 cm. The bags were
buried at a depth of 3.0 cm below the soil surface. The litter bags were retrieved at 10,
20, 30, 50, and 80 d after the beginning of the study. At each retrieval time, the
remaining residue was oven dried at 60 uC for 72 h and weighed.
Formulation of the Soil Water Factor
To formulate the effect of soil water on residue decomposition, we expressed the rate
constant kd in terms of a potential rate constant under nonlimiting soil water
conditions, kdpot, and a soil water factor, fw:
kd~kdpot fw ð3Þ
Further, assuming that kd declines linearly with soil water content between the field
capacity hFC, and air-dry, hd, then fw m
ay be expressed as
h{hd
fw~ ð4Þ
hFC{hd
In the greenhouse, where h was maintained at specified levels throughout, the
calculation of the fw was straightforward. With kdpot taken as the rate constant at
field capacity for each residue type, the coupling of Eqs. (2) to (4) enabled the
prediction of residue mass loss with time for the different water treatments.
However, h in the field varied with time and had to be calculated at each time step
(in this case daily). We applied a simple bucket-type water balance equation to
estimate the daily change in the soil water storage, DS, to the depth of 100.0 cm, using
+DS~Pr{ET{D{R ð5Þ
where P is rainfall rate, ET is evaporation rate, D is drainage rate, and R is runoff rate.
All the terms in Eq. (5) have units of mm d21. With the exception of Pr, no
measurements of the other terms in Eq. (5) were made. We proceeded to estimate
D + R 5 0.20 6 Pr, considering that the field was flat with low runoff potential, acting
as a large soil bucket that would minimize drainage. The potential evaporation ETo
was calculated from the weather data using Cropwat (Smith 1992). The ET equaled
ETo when soil water was at field capacity. However, when soil water fell below field
capacity, ET was obtained as the product of ETo and fw. The daily average change in
water content of the soil was calculated as Dh 5 ¡DS/z with z 5 100.0 cm.
With the h known, kd and fw were calculated using Eqs. (3) and (4). Equation (1)
was then solved numerically using the Euler method, with a time step of 1 d in a
Fortran program (IBM, Armonk, N.Y.). Input variables were kdpot for different
residue types, field soil moisture constants, and the weather data. Without any soil
water measurements at the beginning or during the field study, the simulations
Soil Water Effect on Residue Decomposition 271
Figure 1. Rainfall (bars) and mean temperature (line) during the field experiment.
started by assuming that the soil was initially at field capacity, given the ample
rainfall in the month of June prior to the study (Figure 1).
Results
Variation of kd with Soil Water Content
The kd values for the different residue types were comparable to those obtained in
other studies (Table 3). For the grass residue R1, kd values ranged from 0.009 to
0.013 d21 when the water treatment increased from W3 to W1. For example,
Mubarak et al. (2002) obtained the kd value of 0.099 week
21, equivalent to 0.014 d21
for the field decomposition of maize straw, which was close to our observation at
W1. Others obtained k values between 0.0002 to 0.0006 uC d21d for corn–fallow
Table 3. Decomposition rate constants k 21d (d ) under varying water treatments
Residue type
Water R1 R2 R3 R4
W1 0.013a 0.021a 0.020a 0.022a
(0.0007)a (0.0011) (0.0012) (0.0012)
W2 0.012a 0.016a 0.014b 0.014b
(0.0006) (0.0008) (0.0007) (0.0007)
W3 0.009a 0.008b 0.008c 0.009c
(0.0005) (0.0004) (0.0004) (0.0005)
LSD (P , 0.05) 0.011 0.009 0.005 0.003
aNumbers in parentheses indicate kd in (uC d)
21.
Note. For columns, numbers with the same letters are not significantly different at P , 0.05.
272 S. G. K. Adiku et al.
rotations (Ma et al. 1999). Even though our values are somewhat greater [0.0005 to
0.0007 (uC d)21], the order of magnitude is similar. In the case of legume residue
decomposition, Mubarak et al. (2002) obtained a k 21d value of 0.158 week (or
0.023 d21), whereas others obtained values between 0.027 and 0.036 d21 for soybean
under wet soil conditions (Thonninsen et al. 2000). Again, these data are in
agreement with our data for the legume residues (R2, R3, and R4).
Soil water content had a significant (P , 0.001) effect on the decomposition rate
constants (Table 3). When soil water was at field capacity (W1), mucuna residue (R4)
had the highest kd of 0.022 d
21, whereas the grass residue (R1) had the lowest kd of
0.013 d21. Decreasing soil water content resulted in the decline of the kd for all
residue types, but this decline was more rapid for the residues with lower C/N ratios
(R2, R3, and R4) than for the grass residue (R1). Although the analysis of variance
(not shown) indicated that the effect of residue type on kd was not significant, we
must note that the range of C/N ratios in the residues studied was rather narrow. In
general, the variation of the kd with soil water could be described by a linear function
(Figure 2):  
kd h
~0:82 {0:18, R2~0:85 ð6Þ
kdpot hmax
where hmax is the h at W1. The scatter in the data may be attributed to the differences
in residue types.
Figure 2. Variation of the decomposition rate constant with soil water content.
Soil Water Effect on Residue Decomposition 273
Discussion
Prediction of Residue Decomposition in Varying Soil Water Conditions
The residue mass declined exponentially with time in all treatments under
greenhouse conditions (Figure 3). For water treatment W2, all treatments lost more
than 80% of the initial mass by the end of the greenhouse studies. The residue types
R2, R3, and R4 in particular lost more than 90% of the initial mass. For water
treatment W3, residue decomposition was reduced but did not cease completely, with
greater amounts of residues remaining undecomposed after 180 d than was the case
in water treatment W2. About 25% of the R1 residue and slightly more than 25% of
the other residue types remained after 180 d. These patterns of residue decomposi-
tion with time and soil water were well predicted (R2 5 0.94; RRMSE 5 0.06)
despite the slight overestimation for R1 at W3.
A comparison of the observed and predicted time course of residue
decomposition (Figure 4) showed that the simple formulation of the soil water
factor was also adequate for describing field-scale residue decomposition (R2 5 0.93;
RRMSE 5 0.11). The residues lost between 65 to 80% of their initial mass by 80 d,
which was well captured by the model. Although we lacked data to test the model
performance in the dry periods after September, the simulated trends indicate that
the model would predict a further but very slow decline in residue mass under such
conditions.
It is worth noting that the simple assumptions made about the terms in Eq. (5)
led to good prediction of the residue mass loss with time. Undoubtedly, a more
detailed soil water model, which would provide h as a function of soil depth, would
enable the use of the topsoil h to control the kd, rather than the average h of the
whole profile used in this study. Such a detailed soil water model, however, would
not invalidate our concept for formulating the soil water factor.
Conclusions
We investigated the effect of soil water content on the rate of residue
decomposition and concluded that soil water significantly (P , 0.001) affected
Figure 3. Observed (symbols) and predicted (lines) patterns of residue mass loss with time for
the different residue types at water treatment W2 (top row) and W3 (bottom row).
274 S. G. K. Adiku et al.
Figure 4. Observed (symbols) and predicted (lines) of residue mass loss under field conditions.
Soil Water Effect on Residue Decomposition 275
the kd. Residue decomposition was enhanced by increasing soil water and vice
versa. Our analysis shows that the kd 2 h relation could be described by a linear
equation and that a simple soil water factor could be formulated to predict
residue decomposition in the greenhouse and in the field under variable soil
water conditions.
Acknowledgments
This study was made possible through support provided by the Office of Natural
Resource Management and Office of Agriculture in the Economic Growth,
Agriculture, and Trade Bureau of the U.S. Agency for International Development,
under the terms of Grant No. LAG-G-00-97-00002-00, and through the collaboration
with the Department of Agricultural and Biological Engineering, University of
Florida, USA. The authors are grateful for this support.
References
Bandaranayake, W., Y. L. Qian, W. J. Parton, D. S. Ojima, and R. F. Follet. 2003. Estimation
of soil organic carbon in turf grass systems using CENTURY MODEL. Agronomy
Journal 95:558–563.
Bristow, R. L., G. S. Campbell, R. I. Papendick, and L. F. Elliott. 1986. Simulation of heat
and moisture transfer through surface residue–soil system. Agricultural and Forest
Meteorology 36:193–214.
Douglas Jr., C. L., and R. W. Rickman. 1992. Estimating crop residue decomposition from air
temperature, initial nitrogen content, and residue placement. Soil Science Society
America Journal 56:272–278.
Gregory, J. M., T. R. McCarty, F. Ghidey, and E. E. Alberts. 1985. Derivation and evaluation
of a residue decay equation. Transactions of American Society of Agricultural Engineers
28:98–101.
Lal, R. 2004. Soil carbon sequestration to mitigate climate change. Geoderma 123:1–22.
Ma, L., G. A. Peterson, L. R. Ahuja, L. Sherrod, M. J. Shaffer, and K. W. Rajas. 1999.
Decomposition of surface crop residues in long-term studies in dryland agroecosystems.
Agronomy Journal 91:401–409.
Mubarak, A. R., A. B. Rosenani, A. R. Anuar, and S. Zauyah. 2002. Decomposition and
nutrient release of maize stover under tropical field conditions of Malaysia.
Communications in Soil Science and Plant Analysis 33:609–622.
Palm, C. A., and P. A. Sanchez. 1991. Nitrogen release from the leaves of some tropical
legumes as affected by the lignin and the polyphenolic contents. Soil Biology and
Biochemistry 23:83–88.
Seneviratne, G. 2000. Litter quality and nitrogen release in tropical agriculture: A synthesis.
Biology and Fertility of Soils 31:60–64.
Smith, M. 1992. Cropwat: A computer program for irrigation planning and management.
Rome, Italy: FAO.
Steiner, J. L., H. H. Schomburg, P. W. Unger, and J. Cresap. 1999. Crop residue
decomposition in no-tillage small grain fields. Soil Science Society of America Journal
63:1817–1824.
Stott, D. E., H. F. Stroo, L. F. Elliot, R. I. Papendick, and P. W. Unger. 1990. Wheat residue
loss from fields under no-till management. Soil Science Society of America Journal 54:
92–98.
Stroo, H. F., K. L. Bristow, L. F. Elliott, R. J. Papendick, and G. S. Campbell.
1989. Predicting rates of wheat decomposition. Soil Science Society of America Journal
53:91–99.
276 S. G. K. Adiku et al.
Thonninsen, C., D. J. Midmore, J. K. Lahda, D. C. Olk, and U. Schmidhalter. 2000. Legume
decomposition and nitrogen release when applied as green manure in tropical vegetable
production systems. Agronomy Journal 92:253–260.
Vigil, M. F., and D. L. Sparks. 2002. Factors affecting the rate of crop residue decomposition
under field conditions (USDA-ARS Conservation Tillage Fact Sheet No. 3-95).
Washington, D.C.: U.S. Government Printing Office.