Jiakai Zhang
New Mexico Institute of Mining and Technology, NM·Earth & Environmental Science
Droughts, wildfires, flash floods and extreme heat are hitting communities across the United States more often, and they can cause serious economic and social disruption. Preparing for them requires knowing which places are most exposed to hazards and which have the least capacity to cope, which in turn means combining kinds of information that are rarely analyzed together.
- Fields
Earth & Environmental Science
Computer Science
Public Health & Epidemiology
- Topics
environmental risk modeling
extreme weather
climate risk
- Grants
- 1 known grant, through Aug 2028
What they work on
- Integrate large public spatial, hazard and socioeconomic datasets into a single AI-driven model of regional environmental risk and vulnerability across the United States.
- Capture the complex, dynamic structure of environmental risk and community resilience to droughts, wildfires, flash floods and extreme heat using deep learning and ensemble models that combine multiple predictors.
- Apply explainable AI methods that expose why the model makes a given prediction, and release the resulting risk tools publicly so researchers, planners and communities can use them for adaptation planning.
More about the research
This project builds an artificial intelligence (AI) model of regional environmental risk and vulnerability in the U.S. from large, varied public datasets: spatial data tied to geographic locations, hazard data on extreme weather events, and socioeconomic indicators describing a community's people and economy. The modeling approach uses deep learning, neural networks that find patterns in very large datasets; ensemble models, which combine several predictors for more reliable estimates; and explainable AI methods, which reveal why a model reached a particular conclusion so that planners can judge whether to trust it. Together these are meant to capture the complex, shifting structure of environmental risk and community resilience rather than a single static score. The resulting tools will be made publicly available to researchers, planners and communities to support adaptation strategies, and the project includes mentoring and training that prepares students for careers in AI and policy-relevant research.
Why it matters
Communities, planners and policymakers facing extreme weather often lack a clear picture of their local risk. Publicly available, interpretable risk and vulnerability tools could inform decisions about where to invest in adaptation and resilience measures, helping protect residents, infrastructure and local economies from drought, wildfire, flash flooding and extreme heat.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Richard Bixler
University of Texas at Austin, TX·Social Science
Identify at-risk Texas communities using social vulnerability data and mapping tools, then form reciprocal partnerships with 2-3 place-based communities to co-produce a resilience research agenda.
Lani Tsinnajinnie
University of New Mexico, NM·Earth & Environmental Science
Work with Indigenous and rural communities in New Mexico and Oklahoma, through a co-production process, to identify information needs about the land-water-air system under extreme weather.
Di Shi
New Mexico State University, NM·Engineering
Model multi-scale climate dynamics, including flooding, drought and temperature extremes, and link them to sociodemographic energy maps of three rural communities.
The school
New Mexico Institute of Mining and Technology: 3 labs on PI Pages, $80M a year on research, 21 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.