SWaDE: Satellite-based water demand estimation
THE QUESTION
Can AI make solar mini grids a more reliable energy source for powering irrigation in Ethiopian farms?
LOCATION: Ethiopia
SECTOR: Agriculture
TECH: Satellite data, ground data
TIMELINE: September 2024 - Present
PIONEER: Nina Hissen, Sehr Syed
PARTNERS: SWaDE
The Challenge
Agriculture is Ethiopia’s leading economic activity, accounting for a substantial portion of national GDP. Yet, the country is vulnerable to droughts induced by climate change. Droughts have led to severe food shortages in some parts of the country, as 90% of Ethiopia’s agricultural production is rainfed. After four years of adverse weather conditions, building resilience in the country’s agriculture sector is highly important.
Irrigation is critical to fostering steady agricultural production. However, irrigation requires farmers to have access to affordable energy, something that is limited in the rural areas of the country. High and increasing fuel costs for diesel pumps make these unaffordable for most farmers and trap people in poverty. There is a need to provide affordable energy access to irrigate rural areas, otherwise the cycle of food shortages and poverty is likely to continue.
The Idea
What if we could use solar mini-grids and pumps in order to provide a cheap and environmentally friendly source of energy? While that may sound like an easy solution, in practice there are several problems with doing so.
Issues with the deployment and management of these solar mini grids has led to inaccurate estimations of energy demand, creating problems for farmers. Investors don’t trust the solar mini-grids and the unreliable forecasting methods used to work out demand. Furthermore, conflicts have emerged between upstream/downstream farmers over water usage.
But there is hope for a solution using frontier tech - by applying AI-based software called Satellite Based Water Demand Estimation (SWaDE), we may be able to forecast water and energy use better than any human can. SWaDE uses cutting-edge AI algorithms to gather and interpret satellite imagery from a wide range of sources. It also uses agricultural data - for example soil moisture, surface temperature - to calculate expected energy usage.
By using SWaDE, we can demonstrate that solar mini-grids and pumps are a reliable tool in irrigation technology. This will please commercial investors and local authorities, and hopefully lead to more people using solar mini grids and pumps. The overall result is a better deal for farmers.
Our learnings and stories so far
This pilot hasn’t started to publish yet, but there are plenty of other blogs to read below. Check back soon!
Explore related pilots
Curious about how frontier technologies are impacting agriculture and farmers? Click below to read about our related work 👇🏽
In this edition of our Launchpad series, Grace outlines our adaptive learning approach to evidence generation, and why it is a core part of the FT Hub’s learning methodology.