DIWASA Rainfed and Irrigated Cropland Map for Africa

A framework integrating the Budyko model has been developed to distinguish between rainfed and irrigated cropland areas across Africa. This expands on remote sensing land cover products available for agricultural water studies in Africa and thereby helps address the need for deeper insights into cropland patterns. Validation against an independent dataset revealed an overall accuracy of 73% with high precision and specificity scores. These results validate the framework’s effectiveness in identifying irrigated areas while minimizing errors in misclassifying rainfed areas as irrigated.

本研究开发了一套框架，该框架整合了 Budyko 模型，旨在区分非洲地区的雨养和灌溉农田。此举丰富了非洲农业水资源研究可用的遥感土地覆盖产品，从而有助于满足深入探究农田分布模式的迫切需求。通过与独立数据集的验证，该框架在识别灌溉区域方面展现出73%的整体准确率，并获得了高精密度和高特异性评分。这些结果验证了该框架在识别灌溉区域方面的有效性，同时最大限度地减少了将雨养区域误分类为灌溉区域的错误。