SPR-837

Description:

The primary objective of this project is to develop an automated, multi-scale decision-support framework that transforms satellite and vehicle-acquired remote sensing data into actionable roadside vegetation management workflows. Specifically, the project will achieve the following technical milestones: (1) Develop a Multi-Scale Sensing Architecture: Combine wide-area daily satellite imagery for macro-scale screening with meso-scale mobile sensor rigs mounted to daily Arizona Department of Transportation (ADOT) fleet vehicles (trucks, sweepers, mowers) for high-resolution edge diagnostics. (2) Engineer Multi-Modal Phenological Classification Models: Utilize Visible & Near-Infrared (VNIR) spectral bands (400–900 nm) to track rapid, post-monsoon green-up windows unique to invasive grasses and invasive forbs across all phenological growth stages, alongside Short-Wave Infrared (SWIR) imaging (900–1700 nm) and Long-Wave Infrared (LWIR) thermal sensing to calculate accurate Live Fuel Moisture Content (LFMC). (3) Construct a Predictive Prioritization Matrix: Build an automated roadside inventory system that fuses structural vegetation canopy data (via oblique terrestrial LiDAR/RGB), fuel moisture maps, wildfire risk indices, and traffic data to automate maintenance work orders and preserve clear zones, seamlessly integrated with an Interactive Fire Risk Dashboard.

Visit SPR-837 for further information.