Developing a precision irrigation framework to facilitate smallholder dry-season farming in developing countries: A case study in northern Ghana

Abstract

Changing climate has resulted in increasingly unreliable weather patterns with prolonged dry-seasons in some parts of Sub-Saharan Africa. Food production in these areas is under threat because the people depend mostly on rain-fed farming. Enabling dry-season farming, in light of the prolonged dry-seasons, is central to sustainable food production and poverty alleviation in these areas. Efficient water management is key to successful dry-season farming. Ideally, efficient irrigation water management should involve real-time monitoring of soil moisture (SM) to guide irrigation scheduling. However, farmers in these areas are mostly poor smallholder farmers without the financial capacity to instrument their farms for real-time SM monitoring. We present a precision irrigation framework (PIF) as a low-cost alternative to site-specific SM monitoring to guide irrigation scheduling. PIF applies machine leaning to integrate multi-scale ground-truth data and satellite imagery to create irrigation water management zones for an entire region. We demonstrate the strategy in the Pwalugu area in northern Ghana.