Ecological Forecasting Through Machine Learning: Predicting the Spread of Oncosiphon piluliferum and Cenchrus ciliaris in the Arid Landscapes of the Sonoran Desert

2024/09/12 by Goolsby, Doan Charles, Jones, Scott Andrew, Mitchell, Rachel +1
#ArcGIS Pro #Biodiversity #Botany #Buffelgrass #Cenchrus ciliaris #Climate Variables #Ecological Forecasting #Ecology and Evolutionary Biology #GIS #Habitat Suitability #Invasive Species #Life Sciences #Pima County #Plant Sciences #R Programming #Random Forest #Species Distribution Modeling (SDM) #Weed Science

paper · doi:10.17605/osf.io/rekyn

Abstract

This project leverages machine learning, specifically Random Forest models, to develop a robust tool for land managers, aimed at the early detection and response to the spread of Cenchrus ciliaris (buffelgrass) and Oncosiphon piluliferum (Stinknet) in the arid landscapes of the Sonoran Desert. By integrating key environmental variables, the model identifies areas most at risk of invasion, offering an effective, data-driven solution for proactive land management. The focus of the study is to provide land managers with a practical and reliable tool to guide early intervention strategies, minimizing the ecological impact of these invasive species.

Related

• Development of a Forecasting and Warning System on the Ecological Life-Cycle of Sunn Pest
• Top-down Green-ups: Satellite Sensing and Deep Models to Predict Buffelgrass Phenology
• Machine learning for biosecurity: A probabilistic framework for invasive species management
• Identification of suitable areas for the growth of Teucrium polium species using machine learning models (a case study of Khalil Abad and Kashmar Counties, Iran)
• Regional impact risk classification (RIRC): A predictive framework for plant invasion impacts