Advancing Image Analysis, AI, and Ground Truth Data Enhances Wildfire Risk Management and Biomass Prediction

Wednesday, September 18, 2024

4:30 PM – 5:00 PM MST

Location: Embassy Suites, River Birch C

Speaker(s)

Puskar Khanal, n/a (he/him/his)

Asst. Professor
Clemson University
Clemson, South Carolina, United States
Sanjeev Sharma, n/a

Graduate Research Assistant
Clemson University
Clemson, South Carolina, United States

Co-Author(s)

JO

Justin O Beslity
LR

Lindsey Rustad
PT

Peter T. Manos
LJ

Lacy J. Shelby
AB

Andrew B Reinmann
CK

Churamani Khanal

This study evaluates remote sensing and GIS tools for natural resource management, highlighting their integration with in-situ validation. It assesses machine learning models for forest fire prediction in South Carolina, finding Decision Tree models most accurate. Additionally, it validates a Random Forest model for biomass estimation, emphasizing its effectiveness.

Learning Objectives:

Upon completion, participants will learn about different machine learning techniques in forestry application.
Upon completion, participants will learn about spatial distribution of forest fire and the associated factors.