Skip to main content
ARS Home » Plains Area » El Reno, Oklahoma » Oklahoma and Central Plains Agricultural Research Center » Livestock, Forage and Pasture Management Research Unit » Research » Publications at this Location » Publication #409930
Research Project: Integrated Agroecosystem Research to Enhance Forage and Food Production in the Southern Great Plains

Location: Livestock, Forage and Pasture Management Research Unit

Title: Modeling vegetation phenology of tallgrass prairie using machine learning algorithms

Author
item Wagle, Pradeep
item GOPICHAND, DANALA - University Of Oklahoma
item JENTNER, WOFGANG - University Of Oklahoma
item Moffet, Corey
item Gunter, Stacey
item XIANGMING, XIAO - University Of Oklahoma

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 2/15/2024
Publication Date: 4/30/2024
Citation: Wagle, P., Gopichand, D., Jentner, W., Moffet, C., Gunter, S.A., Xiangming, X. 2024. Modeling vegetation phenology of tallgrass prairie using machine learning algorithms [abstract]. Nepalese Agriculture Professionals of Americas Biennial International Scientific Conference. 2024:5.

Interpretive Summary:

Technical Abstract: The vegetation phenology is an important aspect of tallgrass prairie ecology. It varies from year to year, depending on climatic conditions. Modeling vegetation phenology using satellite and climate data can be a useful tool for understanding and predicting how tallgrass prairies will respond to climate change and other disturbances. Machine learning algorithms are well-suited for modeling phenology by identifying patterns and relationships between climatic factors and vegetation phenology using historical data. This study evaluated the performance of several machine learning algorithms [decision tree, random forest, support vector regressor, K-nearest neighbors, and linear regression] in modeling patterns of the Moderate Resolution Imaging Spectroradiometer-derived enhanced vegetation index (EVI) in tallgrass prairie. Air and soil temperatures, heating and cooling degree days, and solar radiation were all strongly correlated with the EVI, demonstrating that climate is a major driver of vegetation phenology in tallgrass prairie. All tested algorithms, except the decision tree, performed well to accurately model EVI, with R2 of 0.77-0.83 and root mean squared error (RMSE) of 0.05-0.06. Random Forest performed the best, with R2 of 0.83 and RMSE of 0.05. Linear regression also performed well, with R2 of 0.78 and RMSE of 0.06. This study provides insights into the key climate factors and underlying processes that control the phenology of tallgrass prairie ecosystems. Our machine learning models can be a valuable tool for developing new strategies to manage tallgrass prairie ecosystems in the face of climate change.