AI framework estimates urban tree biomass using LiDAR and optical data

Researchers have developed a new framework for estimating above-ground biomass (AGB) of individual trees in urban environments using airborne LiDAR and optical imagery. This method, applied to an 810 km² area in Ontario, Canada, utilizes a dual-stream cross-attention network to delineate tree crowns and assign functional types. The framework achieved an R² of 0.609 for AGB prediction on a large test set, identifying crown delineation as a key source of uncertainty. The system requires no manual annotation and produces a public database of urban tree biomass, with estimates showing a net carbon gain over five years. AI

IMPACT This research demonstrates a novel application of AI for environmental monitoring and carbon stock assessment in urban areas.

RANK_REASON Academic paper detailing a new methodology for biomass estimation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• Canada
• Jose David Bermudez Castro
• Lambert et al.
• lidar
• Niagara Escarpment
• Ontario

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →