Analysis of land use change impacts on flash flood occurrences in the Sosiani River basin Kenya

ABSTRACT

This paper investigates the impacts of land use changes on river peak discharges during floods and its contribution to flood occurrence in the Sosiani River basin in the Rift Valley province of Kenya. The land use maps used in the study were derived from the Landsat imagery employing remote sensing and GIS techniques. Two-dimensional Rainfall–Runoff–Inundation (RRI) model was utilized in simulating the flash flood events. The results indicated an increase in river peak discharge due to extensive deforestation in the past decades and increase in farmlands which covers up to 75% of the total watershed area. According to the observed discharge data and generated land use maps, expansion of farmlands from 15.3% (in the early 1970s) to 75.2% (2013) and urban areas from 0.4% (in the early 1970s) to 10% (2013) have triggered the observed river peak discharge to increase from 167 m³/s (1970) to 233 m³/s (2013). Two land use change scenarios were tested for urbanization and reforestation using the calibrated RRI model and modified land use maps accordingly. The obtained results indicated a high sensitivity for the variation in land use to the river peak discharges. The results of the study will be beneficial for future developments in the basin and its flood management activities.

KEYWORDS:

• Remote sensing
• land use change
• urbanization
• urban floods
• hydrological modelling

Acknowledgements

The authors acknowledge the academic and material support given by the International Centre for Water Hazards and Risk Management (ICHARM) at the Public Works Research Institute (PWRI) of Japan, National Graduate Institute for Policy Studies (GRIPS) of Japan and Japan International Cooperation Agency (JICA) who provided the financial support for the first author to study for her postgraduate degree in Japan. Moreover, authors are grateful to the reviewers and journal editors for their constructive comments and helpful suggestions, which resulted in this improved manuscript. Betty Namulunda Barasa conducted the simulations, analysed the results, and prepared land-use maps using remote sensing and GIS. Edangodage Duminda Pradeep Perera supervised the research, provided guidance and wrote the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.