Landscape deformation monitoring for landslide susceptibility assessment in Mount Elgon, Eastern Uganda

Abstract

Bududa district, located on the southwestern slopes of Mt. Elgon, continues to experience landslides with significant losses of lives and damages to property. The recorded landslides of Mt. Elgon are those associated with large-scale material movement, yet typically, landslides start as small, unnoticeable ground deformations with unknown velocity. Therefore, this study: i) mapped the deformation rate of Mt. Elgon in Bududa district; ii) determined the factors that predispose Mt. Elgon to deformation; and iii) determined the future landslide susceptibility by 2039. The study used Sentinel 1B data and the Persistent Scatterer Interferometry Synthetic Aperture Radar (PSInSAR) technique to map the deformation rate from 2018 to 2021. The relative frequency ratio model (RFRM) was used to study the characteristics of predisposing factors. An ordinary least squares (OLS) regression was used to fit the effects of different factors on the deformation rate. The OLS was mapped to reveal susceptibility based on factors and deformation rate. The future susceptibility was determined based on the OLS equation by replacing the current with future variables. The results revealed that deformations in Bududa district are occurring at rates of 0–54 mmyr-1, with a major peak deformation rate of 100–200 mm at the end of the monitoring period. The deformations are in the form of subsidence and uplift, with average velocities of -2.85±0.03 mmyr-1 and 2.45±0.88 mmyr-1, respectively. High deformation rates were most pronounced on the middle slopes of six sub-counties: Bushika, Nakatsi, Bulucheke, Bushiyi, Bumayoka, and Bukalasi. Soil, rainfall, land use land cover (LULC), and slope angle had high predictive rates for deformation. Lithology, distance from the road, profile curvature, and altitude had medium predictive rates, while slope aspect and distance from streams had low predictive rates. Factors such as soil, rainfall, LULC, altitude, slope, distance from streams, lithology, distance from the road, profile curvature, and altitude significantly (p<0.05) influenced the rate of deformation. Current to future (2039) susceptibility results revealed a gradual outward expansion and increased intensity of likely affected areas. Under RCP 4.5, areas predicted to have high susceptibility by 2039 are expected to rise by 70%, compared to 35% under RCP 8.5 in the same period. The most affected areas will include Bushika, Bulucheke, Bumayoka, Bubiita, Bushiyi, and Bubiita. This study has shown that Bududa is experiencing a differentially increasing deformation rate, which is therefore likely to increase the rate, frequency, and intensity of landslides in the district. The study recommends that the Government of Uganda, district local governments, and disaster risk management agencies should conduct sensitization programs in high-risk areas, educating communities about impending landslide signs and mitigation measures. Additionally, establishing routine monitoring systems utilizing advanced remote sensing technologies is vital for real-time ground displacement understanding and effective landslide risk assessment. Sustainable land management practices should be reinforced through collaborative efforts to foster resilient and disaster-resilient landscapes in affected regions. Further studies should explore material motion evolution and landslide initiation time, considering the time and space heterogeneity of conditioning factors.