Mapping Forest Disturbance and Resilience in a Changing World
We develop AI-driven remote sensing approaches to monitor disturbance and post-disturbance forest recovery. By integrating ground plots, UAV surveys, and multi-source satellite observations, our team builds transferable models that scale from local field sites to large scale ecosystems.
Our research seeks to understand how forest ecosystems respond to intensifying disturbance regimes under climate change — and to provide the spatial intelligence needed to anticipate where resilience may falter.
News
New Research Published in Remote Sensing of Environment
The group proposed a novel deep learning framework combining time-series satellite imagery and foundation models for multi-temporal forest disturbance classification.
Lab Hosts Seminar on Geospatial AI Innovation
The group invited researchers and students for an academic seminar focusing on WebGIS applications, deep learning segmentation, and satellite data pipelines.
New Grant Awarded and Welcoming 2026 Cohort
Our lab secured a new research project on eco-restoration monitoring and welcomed new graduate students joining the group.
Forest Disturbance and Recovery Dynamics Monitoring
Integrating high-resolution multi-source satellite imagery and time-series analysis to quantify forest logging, fire disturbances, and subsequent vegetation recovery patterns.
Deep Learning for Multi-Temporal Geospatial Analysis
Developing custom neural network architectures and adapting foundation models to process multi-temporal Earth observation data for automated semantic segmentation.
Vegetation Phenology and Climate Change Impacts
Leveraging long-term satellite observations to track seasonal vegetation phenology shifts and evaluate forest ecosystem resilience under global climate change.
Interactive WebGIS and Spatial Data Analytics
Building performant WebGIS applications and cloud-based data pipelines to deliver interactive, real-time visualization and spatial analytics of remote sensing products.
ZH 
EN