Water harvesting has been widely applied in different social-ecological contexts, proving to be a valuable approach to sustainable intensification of agriculture. Global estimates of the potential of water harvesting are generally based on purely biophysical assessments and mostly neglect the socioeconomic dimension of agriculture. This neglect becomes a critical factor for the feasibility and effectiveness of policy and funding efforts to mainstream this practice. This study uses archetype analysis to systematically identify social-ecological regions worldwide based on >160 successful cases of local water harvesting implementation. We delineate six archetypal regions which capture the specific social-ecological conditions of the case studies. The archetypes cover 19% of current global croplands with hotspots in large portions of East Africa and Southeast Asia. We estimate that the adoption of water harvesting in these cropland areas can increase crop production up to 60–100% in Uganda, Burundi, Tanzania and India. The results of this study can complement conventional biophysical analysis on the potential of these practices and guide policy development at global and regional scales. The methodological approach can be also replicated at finer scales to guide the improvement of rainfed agricultural.

集水已广泛应用于不同的社会生态环境中，被证明是实现农业可持续集约化的一种有价值的方法。对集水潜力的全球估计通常是基于纯粹的生物物理评估，并且大多忽略了农业的社会经济方面。这种忽视成为政策和供资努力使这种做法主流化的关键因素。这项研究使用原型分析，基于160多个成功实施本地集水的案例，系统地确定了全世界的社会生态区域。我们划定了六个原型区域，这些区域捕获了案例研究的特定社会生态条件。原型覆盖了目前全球19％的耕地，其中热点分布在东非和东南亚的大部分地区。我们估计，在乌干达，布隆迪，坦桑尼亚和印度，这些农田地区采用集水技术可使农作物的产量提高60-100％。这项研究的结果可以补充对这些做法潜力的常规生物物理分析，并指导全球和区域规模的政策制定。该方法学方法也可以在较小的规模上推广，以指导改进雨养农业。这项研究的结果可以补充对这些做法潜力的常规生物物理分析，并指导全球和区域规模的政策制定。该方法学方法也可以在较小的规模上推广，以指导改进雨养农业。这项研究的结果可以补充对这些做法潜力的常规生物物理分析，并指导全球和区域规模的政策制定。该方法学方法也可以在较小的规模上推广，以指导改进雨养农业。