Monitor Stress Epidemiology
Researchers are working to detect and track pest,
insect and pathogen incidence with machine learning on multi-temporal
data and to monitor drought symptoms with multi-sensor platforms.
- John Couture, Assistant Professor, Entomology, Purdue University
- Douglass F. Jacobs, Fred M. van Eck Professor of Forest Biology, Forestry & Natural Resources, Purdue University
- Brady Hardiman, Assistant Professor, Forestry & Natural Resources, Purdue University
- Matt Ginzel, Professor, Entomology, Purdue University
- Philip Townsend, Professor, Forestry & Natural Resources, Purdue University
- Melba Crawford, Professor, Agronomy, Civil Engineering, Electrical & Computer Engineering, Purdue University
The main objective of this proposal is to integrate multi-spatial and temporal scale RS products with forest management scenarios. Specifically, we will focus on three areas of forest management:
- Tracking insect pest and pathogen incidence, severity, and spread,
- Early detection of drought stress-related symptoms, and
- Optimizing RS acquisitions to determine the number of collections appropriate to make an informed management decision.
We will integrate different forms of RS data, collected over multiple time points and measurement scales, at two experimental forestry field stations with plantations that have known biotic and abiotic stressors present at various levels. We will combine the RS data with high-resolution chemical and physiological data, and incorporate these data into different machine learning frameworks to better understand the utility of RS in forest health management. This project will be one of the first to explore relationships among genotypes, phenotypes, and RS data within a forest management framework, advancing our understanding of the ability of remote approaches to monitor forest health.