Gully erosion is a major environmental problem in Gombe town, a large area of land is becoming unsuitable for human settlement, hence the need for a gully erosion susceptibility map of the study area. To generate a gully inventory map, a detailed field exercise was carried out, during this investigation one hundred gullies were identified and studied extensively within the study area of about 550 km2. In addition to the mapped gullies, Google EarthPro with high-resolution imagery was used to locate the spatial extents of fifty (50) more gullies. Ten gully erosion predisposing factors were carefully selected considering the information obtained from literature, and multiple field survey of the study area, the factors include elevation, slope angle, curvature, aspect, topographic wetness index (TWI), soil texture, geology, drainage buffer, road buffer and landuse. In this study, a GIS-based Frequency Ratio (FR) and Analytical Hierarchy Process (AHP) models were employed to predict areas prone to gully erosion in Gombe town and environs. The result obtained from FR shows that drainage, soil texture, and slope have the highest correlation with gully occurrence, while the AHP model revealed that drainage buffer, soil texture, geology have a high correlation with the formation of a gully. Gully erosion susceptibility maps (GESM) were produced and reclassified into very high, high, moderate, and low zones. The overall accuracies of both models were tested utilizing area under the curve (AUC) values and gully density distribution.FR and AHP model have AUC values of 0.73 and 0.72 respectively, the outcome indicates that both models have high prediction accuracy. The gully erosion density distribution values revealed that gullies are concentrated in the very high susceptibility class and it decreases towards the low class, therefore the GESM produced using these models in this study area is reliable and can be used for land management and future planning.

中文翻译：

---

基于GIS的尼日利亚贡贝镇及周边地区沟壑侵蚀敏感性模型，采用双变量统计方法和层次分析法

沟壑侵蚀是贡贝镇的一个主要环境问题，大面积的土地变得不适合人类居住，因此需要研究区域的沟壑敏感性地图。为了生成沟壑清单图，进行了详细的野外演习，在此调查期间，在约550 km2的研究区域内识别并研究了一百个沟壑。除了映射的沟壑之外，还使用具有高分辨率图像的Google EarthPro来定位另外五十（50）个沟壑的空间范围。结合文献资料和研究区多场调查，精心选择了十个沟壑侵蚀诱因，包括海拔，坡度，曲率，纵横比，地形湿度指数（TWI），土壤质地，地质，排水缓冲区等。 ，道路缓冲和土地利用。在这项研究中，基于GIS的频率比率（FR）和层次分析模型（AHP）模型被用来预测贡贝镇和周围地区容易发生沟壑侵蚀的区域。FR的结果表明，排水，土壤质地和坡度与沟壑发生的相关性最高，而AHP模型则表明排水缓冲，土壤质地，地质与沟壑的形成高度相关。绘制了沟壑侵蚀敏感性图（GESM），并将其重新划分为非常高，高，中和低区域。利用曲线下面积（AUC）值和沟壑密度分布测试了两个模型的整体精度.FR和AHP模型的AUC值分别为0.73和0.72，结果表明这两个模型都具有较高的预测精度。