Rainwater harvesting has proven to be a valuable strategy for mitigating the effects of global climate change. However, most of the techniques reported in the academic literature are based on capturing surface runoff generated by rainfall, and therefore, ponds are located close to surface currents, implying investments in infrastructure to conduct water to the place where it is utilized. The aims of this research are the following: i) develop a methodology for the detection of ponds for rainwater harvesting through hydrological and geospatial analyses based on terrain curvature, and ii) identify relevant variables to establish the potential storage volume in ponds. For this purpose, the study was performed in the Tolima region of Colombia with an area of 23,663 km², where 32% corresponds to arid and semi-arid areas. The detection of ponds from the terrain curvature was carried out using geographic information systems and the estimation of daily storage in the rainwater harvesting pond (RWHP) system. Likewise, the principal component analysis identified key variables in the process of detecting and selecting ponds. The results show that the proposed methodology allows the detection of potential ponds for rainwater harvesting near cultivation and water demanding zones for various uses, which could introduce relevant variables in decision-making processes and, decrease water supply distribution costs in semi-arid areas.

雨水收集已被证明是减轻全球气候变化影响的宝贵战略。但是，学术文献中报道的大多数技术都是基于捕获降雨产生的地表径流，因此，池塘的位置靠近地表水流，这意味着在基础设施方面进行了投资，以将水引导到利用水的地方。这项研究的目的如下：i）开发一种方法，该方法通过基于地形曲率的水文和地理空间分析来检测收集雨水的池塘，并且ii）识别相关变量以建立池塘中的潜在储水量。为此，该研究在哥伦比亚的托利马地区进行，面积为23,663 km ²，其中32％对应于干旱和半干旱地区。使用地理信息系统从池塘曲率中检测池塘，并估算雨水收集池（RWHP）系统中的每日存储量。同样，主成分分析确定了池塘检测和选择过程中的关键变量。结果表明，所提出的方法可以检测潜在的池塘，以收集各种用途的耕作和需水区域附近的雨水，这可以在决策过程中引入相关变量，并减少半干旱地区的供水成本。