Abstract
In recent years, modeling gully erosion susceptibility has become an increasingly popular approach for assessing the impact of different land degradation factors. However, different forms of human influence have so far not been identified in order to form an independent model. We investigate the spatial relation between gully erosion and distance to settlements and footpaths, as typical areas of human interaction, with the natural environment in rural African areas. Gullies are common features in the Ethiopian Highlands, where they often hinder agricultural productivity. Within a catchment in the north Ethiopian Highlands, 16 environmental and human-related variables are mapped and categorized. The resulting susceptibility to gully erosion is predicted by applying the Random Forest (RF) machine learning algorithm. Human-related and environmental factors are used to generate independent susceptibility models and form an additional inclusive model. The resulting models are compared and evaluated by applying a change detection technique. All models predict the locations of most gullies, while 28% of gully locations are exclusively predicted using human-related factors.
Highlights
We focus on the role of land use and settlement activities in gully erosion
The most relevant human-related and environmental factors influencing the spatial distribution of gullies and gully heads are derived in the following
It is evident from this work that gullies tend to occur in the immediate vicinity of footpaths, but not necessarily in the direct vicinity of residential areas
Summary
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Gullies are linear depressions of constant grade incised deeper than 0.3 m, resulting from the removal of soil and weathered bedrock by concentrated runoff [1,2]. It poses a problem for the conservation of arable land and for longterm food production [3,4,5]. Poesen et al [6] conclude that 10%–94% of overall soil loss volume due to water erosion is caused by gullying. In order to predict and subsequently prevent gully erosion, several remote-sensing-based models are available that use machine learning algorithms and statistical methods, i.e., Random Forest (RF), Support Vector
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
