Irrigated agriculture is expanding rapidly across southern Africa. Because this expansion is occurring without much regulation or monitoring, it’s not exactly clear where and when expansion is happening, or the fate of expanded agriculture. In most expansionary cycles, there is a boom-and-bust dynamic, where many investments fail quickly and only a few persist. We’re interested in mapping these dynamics over the past 5-10 years across the world and monitoring them going forward.
Our project will set up the necessary infrastructure, generate appropriate label data, and disseminate that data for training a computer vision model that can be used to monitor and map agricultural expansion and irrigated water use.
The project pipeline is broken down into a series of steps: