Estimating soil erosion in sub-Saharan Africa based on landscape similarity mapping and using the revised universal soil loss equation (RUSLE)
Estimating soil erosion in sub-Saharan Africa based on landscape similarity mapping and using the revised universal soil loss equation (RUSLE)
复制标题
DOI：
10.1007/s10705-015-9674-9
复制
发表时间：
2015-01
影响因子：
3.1
通讯作者：
L. Tamene;Q. B. Le
中科院分区：
文献类型：
--
作者：
L. Tamene;Q. B. Le
土壤侵蚀是撒哈拉以南非洲地区土地退化的主要形式之一，严重影响着农业生产力。由于缺乏适当分辨率的可靠数据，以及所用方法的差异，大陆和流域一级的土壤侵蚀估计数存在差异。本研究试图对现有的基于两阶段方法的区域土壤侵蚀估算做出贡献。首先，我们使用大规模数据作为侵蚀驱动因素的代理，将SSA划分为环境单元，即所谓的相似环境约束包络线（SecE）。SECE的目的是提供空间框架，扩大模拟侵蚀的结果。其次，使用空间分布修正的通用土壤流失（RUSLE）模型对白色沃尔特河和尼罗河两个选定流域的土壤侵蚀进行了估计。整个撒南地区的SECE划分提供了空间上有区别的集群，这些集群受类似环境条件和土壤侵蚀风险水平的影响。基于RUSLE的估计表明，白色沃尔特盆地的土壤侵蚀范围为0至120吨公顷/年（总平均值为35吨公顷/年），尼罗河盆地为0至650吨公顷/年（总平均值为75吨公顷/年）。土壤流失估计数与在这两个流域进行的其他研究总体一致。我们的方法提供了指导，经验估计土壤侵蚀为一个给定的SECE可以外推到类似的SECE的可接受的信心和更精细的SECE的子单元应被定义为进一步崩溃的空间变异性的驱动程序的侵蚀。
Soil erosion is one of the major forms of land degradation in sub-Saharan Africa (SSA) with serious impact on agricultural productivity. Due to the absence of reliable data at appropriate resolution and differences in the methods used, there are discrepancies in soil erosion estimates at both continental and basin levels. This study attempts to contribute to the existing regional soil erosion estimates based on a two-stage approach. First, we partitioned SSA into environmental units, so-called similar environmental constraint envelops (SECEs), using broad scale data as proxies of erosion drivers. The SECEs are intended to provide spatial frame for scaling out modeled erosion results. Second, soil erosion estimate is made at two selected basins of the White Volta and the Nile using spatially distributed revised universal soil loss (RUSLE) model. The delineation of SECEs across SSA provided spatially differentiated clusters governed by the existence of similar environmental conditions and soil erosion risk levels. The RUSLE-based estimates show that soil erosion ranges between 0 to 120 t ha-1 yr-1 (overall mean of 35 t ha-1 yr-1) in the White Volta basin, and 0–650 t ha-1 yr-1 (overall mean of 75 t ha-1 yr-1) in the Nile basin. The soil loss estimates show an overall agreement with other studies conducted in the two basins. Our approach provides guidance on where empirically estimated soil erosion for a given SECE can be extrapolated to similar SECE’s with acceptable confidence and where finer SECE’s sub-units should be defined to further collapse the spatial variability of drivers of erosion.