This research will provide an early warning system for drought, enabling farmers and policymakers to react and decrease the severity of the effects of drought.

What challenge(s) does the project address?

Droughts can cause catastrophic disruptions to human livelihood and ecosystem services, which has stimulated research into developing advanced warning systems for droughts.

How will this research address the challenge(s)?

Using tipping point theory in a deep learning model enables a purely data-driven approach to the problem of detecting early warning signs of drought, using the data made available by recent advances in remote sensing technology.

Why does this research matter?

Droughts impact ecosystems, agriculture, and human livelihood in a rapidly warming world, but predicting them in any concrete sense remain a challenge. If droughts can be anticipated, resources can be better allocated to adapt to or prevent their occurrence.

What are the (new) methods (techniques, technologies, etc.) that the project team will use during the research?

This research will develop a deep learning model to anticipate the abrupt onset of droughts by harnessing the mathematical theory of tipping points. Combined with large data sets and the integration of available climate data, the deep learning model will have the ability to learn and be validated using real and simulated drought scenarios.

What impact will the project have on agriculture?

This research will provide an early warning system for agricultural and environmental sciences and present an opportunity for environmental monitoring extending beyond droughts to apply to any natural phenomena that exhibit a tipping point.

Collaborators & Students

Dr. Rozita Dara, Dr. Chris Bauch, and Dr. Daniel Dylewsky