Biodiversity is paramount for maintaining ecosystem health, reducing greenhouse gas emissions and providing commercial, agricultural, and recreational benefits. Yet more than 20,000 species have become extinct or endangered with the impending Sixth Mass Extinction. Invasive species stands as the second most common driver of such extinctions.
Within this context, this dissertation project delves into the invasive Tamarix and its biocontrol agent, Diorhabda. Through integrating multi-source satellite remote sensing, drone imagery, and in-situ observations, this dissertation project aims to investigate the Tamarix biocontrol program's effects on the riparian plant biodiversity of the Colorado River in Arizona.
The project has three core objectives: 1) assessing multi-scale time series in characterizing species-level plant phenology; 2) detecting Tamarix defoliation and vegetation regrowth timing with the COntinuous monitoring of Land Disturbance (COLD) model; 3) evaluating biodiversity change using the convolutional autoencoder (CAE)-based time series clustering. With these objectives, this project will provide comprehensive insights into the biocontrol program's biodiversity impacts. The results are set to be pivotal for understanding post-biocontrol biodiversity changes and guiding future conservation and restoration endeavors.
