Researchers Awarded $3M NSF Grant for Next-Generation Forest Mapping and Monitoring

Story by:

• Kimberly Mann Bruch- [email protected]

Share This:

Article Content

Researchers at the University of California San Diego, University of Florida and Arizona State University have been awarded $3.28 million from the U.S. National Science Foundation (NSF) to build OpenForest4D - a web-based cyberinfrastructure platform for next-generation 4D forest mapping and monitoring. The project's goal is to advance the mapping and monitoring of global forest ecosystems by fusing the most up-to-date, multi-source remote sensing data and novel artificial intelligence (AI) models to generate research-grade estimates of forest structure and above-ground biomass across a range of timescales.

The multi-disciplinary OpenForest4D team is led by principal investigators (PIs) Viswanath Nandigam at the San Diego Supercomputer Center (SDSC), part of the School of Computing, Information and Data Sciences at the University of California San Diego; Carlos Silva at the School of Forest, Fisheries and Geomatics Sciences at the University of Florida; and Chelsea Scott at the School of Earth and Space Exploration, Arizona State University.

Accurate forest mapping and monitoring across large regions and over time is critical for understanding and ultimately mitigating the impacts of climate and natural hazards, managing natural resources and protecting vital ecosystems. Forest structure encompasses the physical arrangement of trees and vegetation within a forest, including various layers such as the canopy and understory, the spatial distribution of trees and species composition. Understanding forest structure is crucial for studying forest management, conservation and ecosystem dynamics. Above-ground biomass refers to the total mass of all living vegetation and dead organic matter above the soil surface, including stems, stumps, branches, bark, seeds and foliage. This metric is vital for assessing carbon stocks, managing ecosystems and understanding the role of forests in the global climate system.

"There are currently no frameworks to calculate forest structure and above-ground biomass estimates using the latest available remote sensing data nor an accessible platform that enables efficient data processing using validated and fit-for-purpose statistical models to generate these products on demand," said Nandigam, PI and director of Advanced Cyberinfrastructure Development Lab at SDSC. "We plan to develop a cyberinfrastructure framework to bridge the expertise gap between domain and data sciences, enabling researchers to leverage essential data products, statistical modeling and machine learning technologies for use in their research."

"Breakthrough advances in the application of AI on fused multi-sourced lidar, optical and Synthetic Aperture Radar imagery datasets will enable academics to conduct transformative research in forest sciences and ecology," said Silva. "Applying AI to enhance the spatial and temporal resolutions of global forest structure and aboveground biomass products will stimulate new research and applications in both government and commercial sectors."

Scott added, "OpenForest4D is likely to be a valuable STEM educational resource across broad discipline ranges and educational levels. We plan to develop educational resources, published curriculum materials and live webinars centered around the OpenForest4D gateway and domain science that will contribute to developing a diverse, globally competitive STEM workforce."

Previously these sorts of calculations were limited to research professionals with advanced expertise. By providing these cyberinfrastructure services via an easily accessible and user-friendly science gateway, users of all expertise levels can now access advanced forestry analysis tools regardless of their technical skill level, democratizing scientific computing and enabling on-demand generation of forestry-related products. Forestry products like above ground biomass have well known applications beyond academia in the commercial sector, as well as natural hazards mitigation. This is particularly crucial as the frequency and severity of wildfires and deforestation continue to rise globally, with significant impacts on human health.

The Advanced Cyberinfrastructure Development Lab is part of the Cyberinfrastructure and Convergence Research (CICORE) Division at SDSC which provides expertise in data science, AI and advanced computing to partnerships and projects focused on achieving scientific and societal impact. This award by the Office of Advanced Cyberinfrastructure is jointly supported by the Division of Biological Infrastructure within the Directorate for Biological Sciences and NSF's National Discovery Cloud for Climate initiative (NSF award nos. 2409885, 2409886 & 2409887).

Learn more about research and education at UC San Diego in: Climate Change, Artificial Intelligence

Two distinct data and processing workflows in OpenForest4D - 1. Airborne lidar data sources to produce grid-level (e.g., canopy height) and tree-level (e.g., crown height/diameter) forestry products and 2. Spaceborne data products to produce grid-level (e.g., canopy cover, above ground biomass) forestry products. Credit: Viswanath Nandigam