PhD candidate using satellites and machine learning to combat drought in Africa
Katie Fankhauser, a PhD candidate in environmental engineering, is the lead author that identifies impacts of high groundwater use in response to drought in the Horn of Africa through satellite data, remote-sensors and machine learning analysis tools.
We asked her about the research, her time at ÃÛÌÒ´«Ã½Æƽâ°æÏÂÔØ and where the work goes from here.
Question: How did you get to ÃÛÌÒ´«Ã½Æƽâ°æÏÂÔØ? What drew you to the university?
Answer: I previously worked in large-scale water, sanitation, and hygiene (WASH) and sustainable energy programs in East Africa with Associate Professor Thomas. My research towards a PhD in Environmental Engineering – and this project in particular – presented an opportunity to re-engage in this region of Africa with an approach toward improved water insecurity and livelihood resilience founded on data and evidence-based decision making.
Q: How would you describe this project and research?
A: Drought is one of the most persistent, expansive, and damaging of natural disasters. It is also increasing in prevalence and severity. And when coupled with weak policies and management, it can lead to regional water and food insecurity, disease, and conflict for billions of people globally.
Groundwater is an important resource of water for people, livestock, and agriculture during drought in the Horn of Africa. In this work, areas of high groundwater demand in drought-prone Kenya were identified, and the conditions forecasted prior to the dry season. For this project we used hydrologic and land surface conditions – derived from satellites and mechanistic models – combined with advanced statistics and machine learning to better describe a population’s reliance on groundwater during drought.
The maps we developed are very accurate and are now available for stakeholders including the Kenya National Drought Management Authority (NDMA) and the Famine Early Warning Systems Network (FEWS NET). These maps represent the first operational spatially-explicit sub-seasonal to seasonal estimates of groundwater use and demand in the literature.
Q: What are some of the applications of this work?
A: One of the primary reasons we want to research this dynamic is that knowledge of historical and forecasted groundwater use will improve decision-making and resource allocation for a range of early warning and early action applications.
The arid and semi-arid lands of Kenya have faced regular drought since at least 2016 with below average rainfall placing 18 million people at risk. The use of groundwater is an effective drought mitigation strategy and operational boreholes may reduce exposure and vulnerability to drought for affected populations. Yet, during the 2016-2017 drought, 55% of the pumps needed to access groundwater were non-functional in Kenya. The ability to direct limited resources to repair, maintain, or site the most critically needed boreholes based on projected use will enable responsible stewardship of water resources and community resilience to drought. Furthermore, observing and predicting trends in groundwater use could be an important indicator of a developing drought itself.
It is all part of the Drought Resilience Impact Platform (DRIP) – an integrated systems-based approach to reducing drought impacts and improving water quality and soil health that is being researched and explored on campus right now.
Q: Where will the research go from here?
A: At the beginning of July, near the start of the dry season in Kenya, we introduced the groundwater data products and service to stakeholders at several meetings and workshops. They have begun to integrate the information into their workflows and planning such as drought risk bulletins. Their feedback will inform iterative design of the models and data visualization.
With the dataset, we also plan to improve trend and anomaly detection of groundwater use and demand.
Q: What was is it like working with Associate Professor Evan Thomas and others in the Mortenson Center?
A: The Mortenson Center in Global Engineering is a diverse and welcoming group of people that have supported my research and personal growth while challenging me to think more critically about the impact of my work and the responsibility I have to promote anti-racism, decolonization of aid, localization and equity. This community has made my time at ÃÛÌÒ´«Ã½Æƽâ°æÏÂÔØ meaningful and continues to motivate me to be a better researcher, professional and friend.