Evaluating spatio-temporal soil erosion dynamics in the Winam Gulf catchment, Kenya for enhanced decision making in the land-lake interface

Olivier S Humphrey; Odipo Osano; Christopher M Aura; Andrew L Marriott; Sophia M Dowell; William H Blake; Michael J Watts

doi:10.1016/j.scitotenv.2021.151975

Evaluating spatio-temporal soil erosion dynamics in the Winam Gulf catchment, Kenya for enhanced decision making in the land-lake interface

Sci Total Environ. 2022 Apr 1:815:151975. doi: 10.1016/j.scitotenv.2021.151975. Epub 2021 Nov 26.

Authors

Olivier S Humphrey¹, Odipo Osano², Christopher M Aura³, Andrew L Marriott⁴, Sophia M Dowell⁵, William H Blake⁶, Michael J Watts⁴

Affiliations

¹ Inorganic Geochemistry, British Geological Survey, Nottingham NG12 5GG, UK. Electronic address: olih@bgs.ac.uk.
² School of Environmental Sciences, University of Eldoret, Eldoret, Kenya.
³ Kenya Marine and Fisheries Research Institute (KMFRI), P.O. Box 1881, Kisumu, Kenya.
⁴ Inorganic Geochemistry, British Geological Survey, Nottingham NG12 5GG, UK.
⁵ Inorganic Geochemistry, British Geological Survey, Nottingham NG12 5GG, UK; School of Geography, Earth and Environmental Sciences, Plymouth University, UK.
⁶ School of Geography, Earth and Environmental Sciences, Plymouth University, UK.

PMID: 34843789
DOI: 10.1016/j.scitotenv.2021.151975

Abstract

Soil erosion accelerated by poor agricultural practices, land degradation, deprived infrastructure development and other anthropogenic activities has important implications for nutrient cycling, land and lake productivity, loss of livelihoods and ecosystem services, as well as socioeconomic disruption. Enhanced knowledge of dynamic factors influencing soil erosion is critical for policymakers engaged in land use decision-making. This study presents the first spatio-temporal assessment of soil erosion risk modelling in the Winam Gulf, Kenya using the Revised Universal Soil Loss Equation (RUSLE) within a geospatial framework at a monthly resolution between January 2017 and June 2020. Dynamic rainfall erosivity and land cover management factors were derived from existing datasets to determine their effect on average monthly soil loss by water erosion. By assessing soil erosion rates with enhanced temporal resolution, it is possible to provide greater knowledge regarding months that are particularly susceptible to soil erosion and can better inform future strategies for targeted mitigation measures. Whilst the pseudo monthly average soil loss was calculated (0.80 t ha^-1 month^-1), the application of this value would lead to misrepresentation of monthly soil loss throughout the year. Our results indicate that the highest erosion rates occur between February and April (average 0.95 t ha^-1 month^-1). In contrast, between May and August, there is a significantly reduced risk (average 0.72 t ha^-1 month^-1) due to the low rainfall erosivity and increased vegetation cover as a result of the long rainy season. The mean annual gross soil loss by water erosion in the Winam Gulf catchment amounts to 10.71 Mt year^-1, with a mean soil loss rate of 9.63 t ha^-1 year^-1. These findings highlight the need to consider dynamic factors within the RUSLE model and can prove vital for identifying areas of high erosion risk for future targeted investigation and conservation action.

Keywords: GIS; Kenya; RUSLE; Remote sensing; Soil erosion; Winam Gulf.

MeSH terms

Conservation of Natural Resources
Decision Making
Ecosystem
Environmental Monitoring
Geographic Information Systems
Kenya
Lakes*
Soil
Soil Erosion*

Substances

Soil