The GRIND Lab develops responsible geospatial AI to understand and mitigate coupled nature-human risks — fostering a more resilient, equitable, and healthy society.
We combine geospatial thinking, artificial intelligence, and domain expertise to tackle pressing challenges at the intersection of nature, society, and technology.
Designing fair, explainable, and trustworthy geospatial AI models that account for spatial heterogeneity and social equity in predictions.
Leveraging multi-source satellite imagery (SAR, optical, LiDAR) and deep learning for land cover mapping, change detection, and disaster monitoring.
Developing predictive and post-event models for wildfire, flood, hurricane, and compound hazards, with a focus on community resilience and recovery.
Studying how human activities and natural processes interact across spatial and temporal scales to generate and amplify socio-environmental risks.
Advancing methods for spatial statistics, graph neural networks, and spatio-temporal modeling to extract knowledge from complex geospatial datasets.
Examining how environmental hazards and climate change disproportionately affect marginalized communities, and developing equitable GeoAI solutions.
Dr. Zhou's research develops responsible geospatial AI to understand and mitigate coupled nature-human risks. He designs advanced GeoAI models for disaster detection, forecasting, and impact explanation across space and time.
Before joining UTK, Dr. Zhou completed his Ph.D. at Texas A&M University and holds degrees from Tongji University and Wuhan University.
Stay up to date with the latest research findings, awards, and events from the GRIND Lab.
The GRIND Lab has been awarded an NSF grant to develop next-generation GeoAI models for wildfire risk assessment and community vulnerability mapping.
Read morePLACEHOLDER: Brief summary of the paper published in [Journal Name]. This paper introduces a novel framework for responsible GeoAI in disaster risk modeling.
Read moreLab members presented three papers at the American Association of Geographers Annual Meeting on GeoAI fairness, multi-hazard risk, and remote sensing.
Read morePLACEHOLDER: The GRIND Lab is actively seeking motivated PhD students with backgrounds in geography, computer science, data science, or related fields.
Apply nowOngoing and recently completed research projects addressing critical challenges in geospatial AI and nature-human risk.
Developing explainable GeoAI models to assess wildfire risk and identify socially vulnerable communities across the western U.S.
Deep learning models for automated flood damage assessment using SAR and optical imagery with bias mitigation across socioeconomic regions.
Foundational research examining sources of geographic and demographic bias in GeoAI models and developing fairness-aware frameworks.
We are actively recruiting PhD students, postdocs, and undergraduate researchers with a passion for geospatial AI, remote sensing, and disaster risk science.