Effects of Catchment Soil Erosion on Rice Production: A Rusle-Gis Based Case Study of Nyiramageni Marshland, Gisagara District, Rwanda

Abstract

Aims: This study aimed to quantify catchment soil erosion, assess sediment transport pathways, and determine their effects on rice production in Nyiramageni Marshland, Gisagara District, Rwanda. Study Design: A mixed-methods research design integrating geospatial modelling and statistical analysis was adopted. Place and Duration of Study: The study was conducted in Nyiramageni Marshland and its contributing catchments in Gisagara District, Southern Province, Rwanda. Spatial and rice production data covering the period 2020–2025 were analysed. Methodology: Soil erosion was estimated using the Revised Universal Soil Loss Equation (RUSLE) integrated with a Geographic Information System (GIS). The five RUSLE factors (rainfall erosivity, soil erodibility, slope length and steepness, cover management, and conservation practices) were derived from rainfall records, soil data, Sentinel-2 satellite imagery, SRTM Digital Elevation Model data, and field observations. Sediment transport pathways were evaluated using the Sediment Transport Index (STI). Rice production data from COOPRORIZ Nyiramageni Cooperative were analysed using descriptive statistics, one-way analysis of variance (ANOVA), and Tukey HSD post-hoc tests. Results: The estimated annual soil loss ranged from 0 to 1,462.54 t ha⁻¹ yr⁻¹, with a mean value of 326.35 t ha⁻¹ yr⁻¹, indicating severe erosion conditions in the catchment. The highest erosion rates were associated with steep slopes, inadequate vegetation cover, and limited conservation practices. STI analysis revealed strong sediment connectivity between upland erosion hotspots and marshland deposition zones. Rice productivity differed significantly among marshland zones, particularly during Season B (F (2,15) = 15.17, p = 0.000249, η² = 0.669), with the Middle Zone recording significantly lower production than the Upper Zone. Conclusion: The study demonstrates that upstream catchment soil erosion significantly influences spatial variability in rice productivity through sediment transport and deposition processes. Topographic amplification (LS-factor) and the near-total absence of structural conservation practices (P-factor) were identified as the principal controllable drivers of erosion, while rainfall erosivity (R) and soil erodibility (K) act as comparatively fixed background conditions. Integrated watershed management interventions, including bench terracing, cover-management improvement, and targeted erosion control on slopes draining into the Middle Zone’s second-order channels, are recommended to sustain rice production in Nyiramageni Marshland. The findings are based on a six-year secondary production record and a zone-stratified statistical design; the limitations of this approach are discussed, and results should therefore be interpreted as indicative of a robust catchment-scale pattern rather than a fully generalisable predictive model.