Soil erosion potential and its control in the Sebeya catchment, Rwanda

Abstract

viii SUMMARY Soil erosion is one of the major causes of land degradation because it results into soil denudation with main consequences of low agriculture productivity and food insecurity. Located in the Western Province of Rwanda, the Sebeya catchment is characterized by high altitude varying between 1,462 m to 2,979 m a.s.l. (meters above sea level). The Sebeya catchment is also characterised by steep slopes and abundant rainfall (1,200 mm to 1,700 mm per year). The combination of the geological formation and soil data characterizes the Sebeya catchment as a fragile ecosystem susceptible to heavy erosion. The transported sediments lead to scour on bridge, increase pollutants in water bodies and reduce the performance of hydropower plants constructed along the Sebeya river. Soil erosion is an environmental issue which can be avoided by maintaining a protective cover on the soil to prevent the impact of rain drops or to reduce the runoff and the infiltration of rain water. A similar result can also be achieved by modifying the landscape to control runoff amounts and rates. The main objective of this study is the assessment of the spatial-temporal distribution of soil erosion and future-based remedial measures to control soil erosion on agricultural land of the Sebeya catchment to contribute to sustainable agriculture and environment protection. This research has specifically focused on the following specific objectives: (1) Assessment of soil erosion process and LULC changes in the Sebeya catchment; (2) Development of runoff coeffients and soil erodibility factors in the Sebeya catchment; (3) Assessment of Sebeya river sediment transport and its impacts in relation to rainfall events; (4) Proposal of future-based remedial measures on agricultural land against soil erosion; (5) Development of an erosion control model for farming systems and soil loss in the Sebeya catchment. During the study period, different site visits were frequently conducted to get primary data through informal and structural interviews on main cultivated crops and agriculture practices, main features of the Sebeya river, site topography, hydrographic network, soil characteristics and about the existing soil erosion control measures in the Sebeya catchment. Literature review was used to know about the status of soil erosion and its control measures throughout the world and particularly in the Sebeya catchment. DEM data have been collected from the Center of Geographical Information System (CGIS Rwanda) for delineation of the Sebeya catchment. ArcGIS software was used to delineate the catchment and its sub-catchments and for mapping all USLE factors. The

ix input data including Digital Elevation Model (DEM), soil data, rainfall data, land use / land cover data were collected from the Center of Geographical Information System of University of Rwanda (CGIS UR). To assess the impacts of the Sebeya river sediment load on Gihira water treatment plant, 3 hydropower plants (Keya, Gihira and Gisenyi) and Lake Kivu, water samples were handcollected from five different sampling points along the Sebeya river during rainy and dry seasons. Laboratory analysis for turbidity, total suspended solids, sieve analysis for bed load materials and NPK content of water and bed load samples were conducted. Agro-climatological data were required for planning crops cultivation and field experiments in soil erosion plots. The proposed methodology was much concerned with the performance assessment and suitability of various best management practices of soil erosion control for the Sebeya catchment. In an effort to achieve the objectives of this PhD research, the 1st activity has gathered worldwide literature views on various causes, effects and remedial measures of soil erosion. From 75 farmers interviewed during this PhD research, it was clear that the main natural causes of soil erosion in the Sebeya catchment are heavy rainfall (69.33%) and the slope steepness (28%) (Section 4.A.1). This study also assessed the various preventive measures against the soil surface crusting and the development of runoff coefficients in order to minimize the soil loss in the Sebeya catchment agricultural fields. The results from the field experiments showed that the mulched plot had high moisture content with low runoff and soil loss compared to 2 other plots (Section 4.A.2). This research revealed also that the average turbidity and suspended solids concentrations in the the Sebeya river were found to be high during rainy season compared to dry season. At all 5 sampling sites and during rainy season compared to dry season, the average mass flowrates of bed load materials were 7.84 and 2.96 kg/hr respectively. Sieve analysis results showed that all bed load sediments were mainly composed of high proportions of fine sand from which more than 70% adversely cause damages on hydropower turbines. In addition, the effect of high soil erosion rates on Gihira water treatment plant was indicated by high turbidity of raw water and excessive costs of reagents. At the Sebeya river outlet, nutrients load and high turbidity reduce significantly the aesthetic quality of Lake Kivu, having a harmful impact on recreation and tourism, aquatic life and affect the livelihoods of people living in the vicinity (Section 4.A.3). Farmers in the Sebeya catchment have not yet implemented sufficient soil erosion control measures in such a way that the estimated average soil loss from the whole catchment area was 130.724 t/ha/yr in 2012 and 100 t/ha/yr in 2018, which is extremely high. The same farmers interview has revealed that among

x the 22 existing soil erosion control measures, about 4.57% of farmers confirmed their existence while 95.43% expressed the need of their implementation in the Sebeya catchment (Section 4.A.4). Simulating a combination of 0, 1, 2 and 3 soil erosion control measures on each of the 259,673 parcels, the average of the simulated annual soil loss from the Sebeya catchment was 849.94; 143.27; 88.64 and 28.59 t/ha/yr respectively. Soil Loss and Crop Yield (SOLCY) model has been developed to predict soil loss and crop yields on each of the 13 main cultivated crops in the Sebeya catchment. A combination of 3 soil erosion control measures such as (bench terrace + mulching + drainage channels) has been found to be the most effective in reducing soil erosion on each parcel with a slope range of (16-60) % (Section 4.A.5). Assessing the sediment load fluctuations in the Sebeya river is useful for water managers and planners to adjust operations accordingly at the water treatment and hydropower plants. Finally, soil erosion control measures including agro-forestry, terraces, mulching, tree planting, contour bunds, vegetative measures for slopes and buffer zones, check dams, riverbanks stabilization are proposed and recommended to be implemented in the Sebeya catchment. Public and private land conservation agencies should be more involved in soil management centred on farmer’s awareness and capacity building. The majority of farmers suggested trainings and mobilization of a specialized technical team to assist them in implementing soil conservation measures and make sure that the fertilizers application is properly done in the whole catchment. Farmers and agriculture technicians can use SOLCY model to focus on the parcel soil loss and predicted crop yields for sustainable land management decision makings. Worldwide researchers should develop similar models on other catchments based on SOLCY model design concept. Various factors influencing farmers’ perceptions of soil erosion causes, effects, and willingness to adopt soil erosion control measures were analyzed using descriptive statistics and SPSS (Version 20), including t-tests, chi-square tests, and a binary logistic regression model. Chisquare test results indicate that gender, farmer age, land ownership, farmland size, social media access, and credit access were strongly associated (p < 0.05) with the adoption of SEC measures, while marital status and education were not. A binary logistic regression model showed that among farmers’ socioeconomic characteristics, farming experience (B = 0.749; p = 0.020) and access to socio media (B = 2.107; p = 0.027) were positively correlated, while age (B = –0.642; p = 0.035) and gender (B = –2.034; p = 0.032) were negatively correlated (p < 0.05) with the adoption of SEC measures (Section 4.B.1).

xi Simulating the existing and the site-based recommended soil erosion control measures, soil loss was reduced significantly from 73 t/ha/yr to 29 t/ha/yr. To highlight the implication of the site-based recommended SEC measures in improving agricultural productivity, this study suggests field investigations in soil erosion plots and prediction of crop yields from an established linear correlation model between soil loss and crop yields in the Sebeya catchment. For effective action in reducing high soil erosion rates to tolerable rates in the Sebeya catchment, the present research recommends implementing the site-based recommended soil erosion control measures with mulching and drainage channels on the same farmland (Section 4.B.2).