Simulation of sediment transport to Sawah rice fields by applying the water erosion prediction project (WEPP) model to a watershed in Ghana

Vol. 4(3), pp. 46-53, June 2013 DOI 10.5897/JSSEM12.002 ISSN 2141-2391 ©2013 Academic Journals http://www.academicjournals.org/JSSEM

Journal of Soil Science and Environmental Management

Full Length Research Paper

Simulation of sediment transport to Sawah rice fields by applying the water erosion prediction project (WEPP) model to a watershed in Ghana E. T. Atakora1*, N. Kyei-Baffour1, E. Ofori1 and B. O. Antwi2 1

Department of Agricultural Engineering, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. Soil and Water Management Division, CSIR- Soil Research Institute of Ghana, Academy Post Office, Kumasi, Ghana.

2

Accepted 16, October 2012

Inland valleys are being used under the Sawah technology for rice production to reduce rice imports to Ghana. Sawah technology is assumed to benefit from geological fertilization. However, there is no quantitative information on runoff and sediment flows in the agricultural watershed of Ghana. This study was carried out at Biemso in the southern part of the country. The aim was to estimate runoff and sediment transport using the water erosion prediction project (WEPP) model (version 2006.500), from hillslope to the valley bottom where rice is cultivated using the Sawah technology. A digital elevation model (DEM) was created from ground survey and used to select the various plots (hillslopes) and to select slope input parameters. Four plots (hillslopes) were selected for the model simulation. Data on local daily values of rainfall and on minimum and maximum temperatures were used to set a CLIGEN model station file to determine climate input parameters for the model. Rainfall characteristics (erosivity and distribution) were analysed. Soil erodibility was also determined. Soil and crop management input parameters required by the model were identified and or estimated from field measurements and secondary sources. The model was run for two management scenarios: Fallow and continuous maize systems. The results of the simulation showed that 2.9 to 3.9 and 6.8 to 10.2 t/ha/year of sediments were eroded from upper catchment to valley bottom under fallow system and maize, respectively. The range of values for runoff produced under fallow was 17.4 to 40 mm whereas that under maize system is 158.7 to 233.62 mm. The study has shown that land use system in the study area has a great influence on geological fertilization. In addition, the valley bottom where rice is produced under the Sawah system is enriched with organic matter from upslope. Key words: Geological fertilization, sediment, sawah, inland valleys, runoff, water erosion prediction project (WEPP) (2006.500) model. INTRODUCTION Soil erosion is a dynamic process whereby sediments are detached, transported, and accumulated at a place different from where they are originated, mostly in lowlying areas or in water bodies downstream. Sediment transport in a catchment can have both negative and positive impacts. *Corresponding author. E-mail: [email protected].

The negative impacts include the removal of nutrient-rich topsoils in upland areas and subsequent reduction of agricultural productivity in those areas, reduction of irrigation conveyance capacities in canals and reservoir storage volumes, reduction in irrigation water quality increasing water turbidity, changes in river channels and

Atakora et al.

subsequent increase in flood vulnerability of the floodplain farmlands and residential areas. Sediment transport, however, is not always a negative environmental process. Whenever soil erosion occurs, there may be downstream benefits such as deposition of rich sediments for promotion of agricultural activities. One system that benefits from such rich sediment eroded from uplands is the Sawah technology. The Sawah technology involves bunding, puddling and leveling of rice fields, provision of irrigation and drainage facilities. The benefit of growing rice under the Sawah system is assumed to be the flow of nutrients and sediment from the upper catchment (Wakatsuki et al., 2005). This in principle is supposed to increase rice yields. However, while making a case for the sawah ecotechnology as a means of coping with the widespread land degradation problem and the attendant decline in soil productivity in West Africa, Obalum et al. (2012) recently decried the lack of data on the extent of geological fertilization of the lowlands in the region as one of the areas needing research. Soil erosion has been of much concern especially to countries like Ghana whose economies depend on agriculture. Investigations carried out by the Soil Research Institute of Ghana in 1989 revealed that 29.5% of the country is subject to slight to moderate interrill and rill erosion, 43.3% to severe interrill and rill and gully erosion, and 23% to very severe interrill and rill and gully erosion (Quansah et al., 1989). Several studies conducted in the country indicates that erosion by water has contributed immensely to the loss of rich soil nutrients from upland soils to low-lying areas (Asiamah et al., 2000; Quansah, 2001; Quansah et al., 2002; Akyea, 2009; Amegashie, 2009). In Ashanti Region for example, reports by Quansah et al. (2000) and Adama (2003) indicate that eroded sediments which are transported to low-lying areas contains higher concentrations of soil organic matter and plant nutrients than the soil from which it was eroded. Estimating soil erosion rates and sediment transport to low-lying areas has been challenging since they are affected by multiple factors. As a result, many erosion and sediment yield models have been developed to estimate runoff and sediment flows in a catchment. However, many of such models fail to consider the depositional portions of a landscape profile. Again, the practical uses of such models are limited due to uncertainties in their input requirement (Bogardi et al., 1985). The Water Erosion Prediction Project (WEPP) model (Flanagan et al., 2001) is one well-validated erosion prediction tool that is considered as an important analytical tool for global studies (Favis-Mortlock et al., 1996; Bhuyan et al., 2002). The WEPP, a continuous physically-based simulation model, can estimate erosion and soil loss in “Sawah” site since it has the capability of estimating the spatial and temporal distribution of soil

47

loss for an entire hillslope or for each point on a hillslope. Again, it has been demonstrated to produce reasonable results in Uganda (Biteete-Tukahirwa, 1995), Kenya (Ampofo et al., 2002; Saenyi, 2002) and Tanzania (Chitrakar, 2002). Therefore, the WEPP model was considered appropriate for this study. The main objective of this study was to use the WEPP model to estimate runoff and soil losses from hillslopes to the valley bottom where rice is grown under the “Sawah” technology. The WEPP model WEPP is a process-oriented model which is based on modern hydrological and erosion science to calculate runoff and erosion on a daily basis. It is based on fundamentals of infiltration, surface runoff, plant growth, residue decomposition, hydraulics, tillage, management, soil consolidation and erosion mechanics (Flanagan et al., 2001). The WEPP model accommodates many processes related to soil erosion, including rill and interrill erosion, infiltration, percolation, sediment transport and deposition, surface runoff, residue and canopy effects, tillage effects, and evapotranspiration. Therefore, it can provide reasonably accurate and quick estimates for soil erosion (Renschler and Lee, 2005). The most notable advantages of the model include the capability of estimating spatial and temporal distributions of net soil loss and extrapolation to a broad range of conditions, which may not be practical or economical to test in the field (Flanagan and Nearing, 1995; Flanagan et al., 2001). The hillslope component of the WEPP erosion model requires a minimum of four input data files to run. These are: climate, slope, soil and crop/management files. The climate file includes daily values for precipitation, temperatures, solar radiation, and wind information. A stand-alone programme called CLIGEN is used to generate either continuous simulation climate files or single storm climate files. The slope file requires information on slope orientation, slope length, and slope steepness at points down the profile. Many types of nonuniformities on a hillslope can be simulated through the use of strips or Overland Flow Elements (OFE’s). Each OFE on a hillslope is a region of homogeneous soils, cropping, and management. The soil file contains information on the number of soil layers, depth of soil layer from the soil surface and soil properties, such as interrill and rill erodibilities, critical shear, percentages of sand, clay, and organic matter, cation exchange capacity, hydraulic conductivity and gravel concentration. The management file contains information needed to define initial conditions, tillage practices, plant growth parameters, residue management, and crop management.

48

J. Soil Sci. Environ. Manage.

The WEPP hillslope profile erosion model uses a steady-state sediment continuity equation to calculate erosion on a rill-interrill basis as: ∂G ∂x

= D

i

+ D

(1) r

where x represents distance downslope (m); G is -1 -1 sediment load (kgs m ); Di is interrill sediment delivery to -1 -2 the rill (kgs m ), and Dr is rill erosion rate (kgs-1m-2). The rill erosion rate is calculated using the equation:

Ghana Meteorological Agency. The mean monthly precipitation, mean monthly maximum and minimum temperatures for the period were calculated and used as monthly climatic parameters. For the simulation year, 2004, daily values of precipitation and maximum and minimum temperatures were used. Also, rainfall erosivity and rain distribution were estimated using the Modified Fournier Index (MFI) method (Arnoldus, 1980) and Precipitation Concentration Index (PCI) method (Oliver, 1980) respectively as follows: MFI =

PCI

 G D r = D c +  1 − Tc 

  

Dc =Kr +(τ −τc)

(3)

where Kr is a rill erodibility parameter (sm-1), τ is flow shear stress acting on soil particles (Pa), τc is the rill detachment threshold parameter (critical shear stress of the soil) (Pa). Finally, the Interrill erosion rate is calculated using the equation: R  D i = K adj I eσ ir SDR r  s   w 

= 100

(2)

where Dc is detachment capacity by rill flow (kgs-1m-2), and Tc is sediment transport capacity in rill (kgs-1m-1). The Dc is calculated as:

(4)

where Kadj is the adjusted interrill erodibility (kgs m-4); Ie is the effective rainfall intensity (ms-1); σ ir is the interrill runoff rate (ms-1); SDRr is the sediment delivery ratio in (%); Rs is the rill spacing (m), and w is the width of the rills (m).

∑

p2 P

∑ P

p

(5) 2

(6)

2

where p is monthly precipitation and P is annual precipitation. Climate dataset required by the model was built by setting up a CLIGEN par file. To build a CLIGEN input parameter file for the study area, the existing international precipitation and temperature data from Accra, and in GDS file format was downloaded from: http://hydrolab.arsusda.gov/nicks/nicks.htm. This was further used with local data collected from Ghana Meteorological Agency to search for a similar US station and also to set up the CLIGEN par file from the local data. The values were then used to generate the needed WEPP climate file using CLIGEN 4.3 to set up a CLIGEN par file for the study area. Crop/management Based on secondary data analysis and field visits the major crop which is maize, was identified at the hillslope. Two management scenarios were considered for model simulation: fallow land and continuous maize cropping. Data were collected on plant growth and harvest parameters, temperature and radiation parameters, canopy, leaf area index (LAI) and root parameters, residue parameters, tillage parameters and initial plant conditions. Initial plant conditions and plant growth parameters were determined using field measurement and personal communication with technicians from the Soil Research Institute. However, harvest parameters, temperature and radiation parameters, canopy, leaf area index (LAI) and root parameters, residue parameters and tillage parameters were obtained from the WEPP User’s Manual (Flanagan et al., 2001).

MATERIALS AND METHODS Description of study site

Slope

This study was conducted at a ‘Sawah’ technology experimental site located at Biemso in the Ashanti Region of Ghana. This study area is located in the Ahafo Ano South District of Ashanti Region (Figure 1). Its geographic location is within N 06°, 52’ 53.2” and W 001°, 50’ 47.3”. The mean annual precipitation in the region is 1301 mm (averaged for the period from 1974 to 2004). The rainfall pattern in the area is bimodal (with two peaks). The geology of this study site consists of rock of the Lower Birimian formation comprising phyllites, greywackes, schists and gneiss, with the soils classified under the Akumadan–Bekwai/Oda Complex association in the local soil surveys (Adu, 1992). Agriculture is the main land use in the area. The valley is mostly used for rice cultivation under the system called “Sawah”. The most important crop commonly grown in the adjoining uplands (hillslopes) is maize (Zea mays). Maize cultivation covers about 50% of the land in this study area.

The slope data of the catchment were collected through ground survey. Data collection was done using a dumpy level, a leveling staff and a handheld GPS. The data obtained from ground survey were converted to x, y and z coordinates using Microsoft Excel. The file was then converted into a comma separated value (csv) file and saved. The csv data file was then imported manually into an ArcMap with the aim of converting the data into an ArcGIS shape files. These were then used to create a digital elevation model (DEM). From the DEM, topographic attributes such as slope gradient, aspect, hill shade land use map, flow direction and flow accumulation were determined. Four plots (hillslopes) were extracted through overlay operations.

Climatic data Climatic data for the period 1974 to 2004 were collected from the

Soil Based on the DEM created from ground survey, a soil survey was carried out. Two types of soil samples – disturbed and undisturbed – were collected from each plot. Samples were taken from both upslope and foot slope positions of each plot.

Atakora et al.

49

Figure 1. Location of this study area and sampling sites at Biemso in Ghana.

In the laboratory, soil bulk density was determined by the metal core sampler method (Blake and Hartge, 1986). The particle size distribution was determined by the method proposed by Van Reeuwijk (2002). The pH of the soil samples was potentiometrically measured in the supernatant suspension of a 1: 2.5 soil: liquid (pHH2O) as described by Peech (1965). The Walkley-Black procedure as described by Nelson and Sommers (1996) was used to determine organic matter concentration in the soil. The cation exchange capacity (CEC) was obtained by a method described by Black (1965). Infiltration rates were measured with minidisk infiltrometer and hydraulic conductivity was determined using the method proposed by Zhang (1997) for dry soils. In addition to soil parameters which were determined in the laboratory, the model requires data on soil erodibility, critical shear stress and soil albedo. These parameters were estimated using formulae documented by Flanagan and Livingston (1995) as follows: For cropland soils containing 30% or more sand, the equations are:
 = 2728000 − 192100 × 

(7)


 = 0.00197 + 0.00030  + 0.03863 . 

(8)

!

= 2.67 + 0.065#$%& − 0.058 

(9)

where; Ki = interrill erodibility parameter (kg s m-4); Kr = rill erodibility parameter (sm-1); Ʈc = critical shear parameter (Nm-2); VFS = percent very fine sand (%) ≤ 40% (if > 40%, use 40%); ORGMAT = organic matter in the soil surface (%) > 0.35% (if < 0.35%, use 0.35%); CLAY = percent clay (%) ≤ 40% (if > 40%, use 40%). For cropland soils containing less than 30% sand, the equations are:
 = 6054000 − 55130 #$%&

(10)


 = 0.0069 + 0.134 *.+*,-. ,

(11) (12)

= 3.5

Where in Equations 10 and 11, CLAY ≥ 10% (if < 10%, use 10%)

Albedo =

0.6 l

0 .4 ×OM

(13)

WEPP 2006.500 was used in continuous-simulation mode to predict runoff and sediment yield for ten years. Each plot was considered as an overland flow element (OFE) and had its own set of input files.

RESULTS AND DISCUSSION Topographical nature of the catchment As shown in Figure 2, the valley occupies the northern part of the catchment. Among the four plots selected forthis study, Plot 3 had the highest slope followed by Plot 4. Also, Plot 1 which had the lowest slope had the largest area (Table 1). Rainfall erosivity and rain distribution Figure 3 illustrates the Modified Fournier Index (MFI) for the individual years. The MFI values ranged from 118.43 mm (in 1983) to 229.88 mm (in 2002) with a mean value of 169.08 mm (Figure 4). Only 2 years, 1983 and 1996 had MFI values slightly less than 120 mm. The remaining

50

J. Soil Sci. Environ. Manage.

Figure 2. Schematic description of Biemso catchment.

had values above 120 mm. According to the conceptual scale for assessing the MFI proposed by CORNIE (cited by Lujan and Gabriels, 2005), values above 120 show high aggressive rains. Thus, the results obtained indicated that rains in the area are high enough to cause severe soil loss and nutrient transport especially from the upper catchment to the valley bottom. Figure 4 illustrates the Precipitation Concentration Index (PCI) values for the individual years as well as the mean PCI. The mean value was 13.02. The values ranged from 10.59 (1990) to 16.00 (1992). All the values were greater than 10. According to the scale for assessing PCI proposed by Oliver (1980), values from 11 to 15 denote a moderate seasonal distribution. This could be attributed to the variations in rainfall values. In addition, since precipitation concentration is related to soil erosion, and water management, the results also imply that soil erosion in the area is moderately difficult to control (Lujan and Gabriels, 2005)

Figure 3. Modified Fournier Index (MFI).

18

PCI

16

Runoff generated by upper catchment

14

12

2004

2001

1998

1995

1992

1989

1986

1983

1980

1977

1974

10

Year

Figure 4. Precipitation concentration for each year Index (PCI) values for each year.

Table 2 shows the average amount of selected soilproperties in the upper catchment. The amount of average annual runoff produced on each plot is presented in Figure 5. The results from this study have shown that soil properties indeed affect the amount of runoff. Low organic matter content, low CEC and high clay content low hydraulic conductivity and low infiltration rates produced the greatest runoff as shown by plot 4 in Figure 5. On the other hand, high organic matter content, high clay content, high hydraulic conductivity and high infiltration rates produced lowest runoff. According to FAO (2005), soil organic matter and CEC improves soil

Atakora et al.

51

Table 1. Description of created plots at study site.

Plot Plot 1 Plot 2 Plot 3 Plot 4 Valley

2

Area (m ) 2488 2340 2216 2123 20644

Area occupied (%) 5.7 5.4 5.1 4.9 47.5

Average slope (%) 20.5 21.7 34.7 30 6.3

Table 2. Average values of selected soil properties in the upper catchment.

Land use Plot 1

Parameter Mean Std. Dev

SOM (%) 2.83 0.73

CEC 12.01 1.04

Clay (%) 20.01 0.47

Ks (mm/h) 31.43 9.5

If (mm/h) 182 89.16

Plot 2

Mean Std. Dev

2.97 1.1

13.78 1.4

16.04 1.23

43.49 4.25

419.78 101.67

Plot 3

Mean Std. Dev

4.1 1.35

14.24 0.6

20.08 0.61

60.06 15.65

601.78 191.71

Plot 4

Mean Std. Dev

2.48 1.2

5.67 1.06

22.03 1.11

4.11 1.22

41.78 14.49

SOM = Soil organic matter, CEC = cation exchange capacity, Ks = hydraulic conductivity, If = infiltration rate.

moisture, moisture holding capacity, infiltration capacity, diversity and activity of soil organisms. Also the results showed that runoff generated is highly influenced by land use. Runoff produced under maize cultivation was greater than runoff produced under fallow system. Fallow lands are rich in soil organic matter and therefore form stable aggregate thus reducing runoff. This study has shown that to reduce runoff from the upper catchment there is need to consider fallow period in the cropping calendar.

In general, the model results showed a fair estimate as the results conform to the results of sediment flow using cesium- 137 within a footslope at Kotokosu catchment in Ejura (Antwi, 2006). The radio nuclide predicted 4 to 10 tons of sediment per hectare using the Diffusion and Migration model. Conclusion

Sediment yield assessment

Based on the results obtained from the study and the discussions made, the following conclusions can be drawn:

Figure 6 presents average annual sediment yield from the hillslope profile under the two land use system (maize and fallow). Sediment yield ranged from a minimum of 2.9 t/ha/year to a maximum of 10.2 t/ha/year. This indicates that for every year between 2.85 to 10.2 tons of sediment per hectare leaving the hillslope are deposited at the ‘sawah’ fields. The results indicate that sediment yield generation is greatly affected by land use (Figure 6). Average sediment yield leaving the hillslope profile under maize cultivation is more than twice that leaving the profile under fallow system. This shows that landuse in this study site affects the movement of sediment from upper catchment to the lower catchment.

(1) The rains in this study area are quite aggressive to cause detachment and transport of soil particles. Mean Modified Fournier Index (MFI) and mean Precipitation Concentration Index (PCI) were 169.08 and 13.02, respectively. (2) Land use has great effect on sediment yield and runoff. Sediments leaving hillslope profile under maize hillslope under fallow system. Runoff produced under maize was more than seven times that produced under fallow. (3) Sediment and runoff are both affected by soil properties such as hydraulic conductivity, soil organic matter, cation exchange capacity and clay content.

52

J. Soil Sci. Environ. Manage.

Average annual sediment yield

Figure 5. Estimated average annual runoff.

12 10 8 6 4 2 0 Plots

Figure 6. Estimated average sediment yield.

ACKNOWLEDGEMENTS We are very grateful to Prof. T. Wakatsuiki (Kinki University, Japan) for providing funds for the work. We also thank the Soil Research Institute for providing some field and laboratory assistance.

REFERENCES Adama AI (2003). The Effect of Different Tillage Practices on Soil Erosion, Water Conservation and Yield of Maize in the SemiDeciduous Forest Zone of Ghana. M.Sc. Thesis. Department of Crop and Soil Sciences, KNUST, Kumasi, Ghana. Adu SV (1992). Soils of Kumasi region, Ashanti region, Ghana. Memoir No. 8. Ghana Soil Research Institute. P. 90. Akyea AO (2009). Effect of Different Land Preparation Methods on Erosion, Growth and Yield of Cassava. Unpublished M.Sc. Thesis. Department of Crop and Soil Sciences, KNUST, Kumasi, Ghana. Amegashie BK (2009). Assessment of catchment erosion, sedimentation and nutrient export into small reservoirs from their catchments in the Upper East region of Ghana. M.Sc. Thesis. Department of Crop and Soil Sciences, KNUST, Kumasi, Ghana. Ampofo EA, Muni RK, Bonsu M (2002). Estimation of soil losses within plots as affected by different agricultural management. Hydrol. Sci. J. 47(6):957-967. Antwi BO (2006). Using Cesium 210 and particle size fraction to model spatial soil erosion rates in the forest Savanna Transition ecology of Volta Basin of Ghana. Unpublished PhD Thesis. KNUST, Kumasi,

Ghana. P. 209 Arnoldus HMJ (1980). An approximation of the rainfall factor in the Universal Soil Loss Equation. In: De Boodt M. and D. Gabriels (Eds.) Assessment of Erosion. Wiley, Chichester, West Sussex, UK. pp. 127-132. Asiamah RD, Quansah C, Dedzoe CD (2000). Soil Degradation: Management and Rehabilitation in Ghana - An Overview Report, In: Proceedings of the FAO/ISCW Expert Consultation on Management of Degraded Soils in Southern and East Africa (MADS - SEA). Pretoria, South Africa. pp. 89-100 Biteete-Tukahirwa JM (1995). Measurement, Prediction and Social Ecology of Accelerated soil erosion in Kibale District of Southwest Uganda. PhD Thesis. Makerere University, Uganda. P. 210. Bhuyan SJ, Kalita PK, Janssen KA, Barnes PL (2002). Soil loss predictions with three erosion simulation models. Environ. Modell. Software 17:137-146 Black CA (1965). Methods of Soil Analysis. Part 2. Chemical and microbialogical properties. First edition. American Society of Agronomy and Soil Science Society of America. Madison, Wisconsin, USA. Agronomy 9:371-373 Blake GR, Hartge KH (1986). Bulk density. In: Klute, A. (ed.). Methods of soil analysis. Part 1. Physical and mineralogical methods. Second edition. American Society of Agronomy and Soil Science Society of America. Madison, Wisconsin USA. pp. 363-375 Bogardi I, Fogel M, Duckstein L, Bardossy A (1985). In: E. Plate and N. Buras (ed.). Estimating of sediment loading in Agricultural reservoirs with Design and Management of Water Resources Systems. pp. 8798 Chitrakar S (2004). Testing of the WEPP soil erosion model in the west Usambara mountains of Tanzania. MSc. Thesis. Erosion and Soil and Water Conservation Group. Wageningen Universiteit. The

Atakora et al.

Netherlands. FAO (2005). Importance of Soil Organic Matter. FAO Soil Bulletin. P. 94. Favis Mortlock DT, Quinton JN, Dickson WT (1996). The GCTE Validation of Soil Erosion Model for Global Studies. J. Soil. Water Conserv. 51:397-403. Flanagan DC, Laflen JM, Lane LG,. Nearing MA (2001). Water Erosion Prediction Project (WEPP) Computer Program (version 2006.5) for Windows. USDA- Water Erosion Prediction Project: Hillsope Profile and Watershed Model Documentation. NSERL Report No. 10. National Soil Erosion Research Laboratory, IN, USA. P. 340. Flanagan DC, Nearing MA (Eds). 1995. USDA-Water Erosion Prediction Project: Technical documentation. NSERL Report No. 10. National Soil Erosion Research Laboratory, IN, USA. P. 147. Flanagan DC, Livingston SJ (1995). USDA-Water Erosion Prediction Project: User summary. NSERL Report No. 11. National Soil Erosion Research Laboratory, IN, USA. Lujan DL, Gabriels D (2005). Assessing the rain erosivity and rain distribution in different agro-climatological zones in Venezuela. Sociedade and Natureza, Uberlândia, Special Issue: S&N-2009-632. Nelson DW, Sommers LE (1996). Total Organic Carbon and Organic Matter. In: Methods. In: Methods of Soil Analysis. Part 3. SSSA Book Series No. 5. SSSA Madison, WI. pp. 214-235. Obalum SE, Buri MM, Nwite, Hermansah JC, Watanabe CA, Igwe Y Wakastuiki T (2012). Soil Degradation-Induced Decline in Productivity of Sub-Saharan African Soils: The Prospects of Looking Downwards the Lowlands with the Sawah Ecotechnol. Appl. Environ. Soil Sci 2012, Article ID 673926, P. 10 doi:10.1155/2012/673926 Oliver JE (1980). Monthly Precipitation Distribution: A Comparative Index. Professional Geographer. 32(3):300-309. Peech C (1965). pH. In: Black, C.A. (Ed.). Methods of soil analysis. Part 2: Chemical and microbiological properties. First edition. American Society of Agronomy and Soil Science Society of America, Madison, Wisconsin USA. pp. 371-373. Quansah C (2001). Integrated soil management for sustainable agriculture and food security. Country case study: Ghana. FAO. Accra. pp.33-75. Quansah C, Baffoe-Bonnie E, Asiamah RD (1989). Soil erosion in Ghana and strategies for conservation. A paper presented at the African Ministerial Conference on the environment (RCU/SOFERNET) and soil conservation strategy for Africa. Accra, Ghana. 28-30 August, 1989.

53

Quansah C, Safo EY, Ampontuah EO, Amankwaah AS (2000). Soil Fertility erosion and the associated cost of NPK removal under different soil and residue management in Ghana. Ghana J. Agric. Sci. 33(1):33-42. Quansah C, Agodzo SK, Gyawu, M, Bonsu J, Dittoh S (2002). National Action programme to Combat Desertification in Ghana. Environmental Protection Agency, Accra, Ghana Renschler CS, Lee T (2005). Spatially distributed assessment of shortand long-term impacts of multiple best management practices in agricultural watersheds. J. Soil Water Conservat. 60(6):446-456. Saenyi WW (2002). Sediment Management in Masinga Reseroir, Kenya. PhD Thesis. University of Agricultural Sciences (BOKU), Vienna, Austria. P. 194 Van Reeuwijk LP (2002). Procedure for Soil Analysis. Technical Paper 9. Sixth edition. ISRIC-FAO. Wakatsuki T, Buri MM, Fashola. OO (2005). Ecological Engineering for Sustainable Rice Production and the Restorattion of Degraded Watershed in West Africa. In: World Rice Research Conference Publication. The International Rice Research Institute (in Press). Zhang R (1997). Determination of soil sorptivity and hydraulic conductivity from the disc infiltrometer. Determination of soil sorptivity and hydraulic conductivity from the disc infiltrometer. Soil Sci. Soc. Am. J. 61:1024-1030.