Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Crop Production Systems Research » Research » Publications at this Location » Publication #385061

Research Project: Development of Productive, Profitable, and Sustainable Crop Production Systems for the Mid-South

Location: Crop Production Systems Research

Title: Teasing apart silvopasture system components using machine learning for optimization

item Kharel, Tulsi
item Ashworth, Amanda
item Owens, Phillip
item PHILIPP, DIRK - University Of Arkansas
item THOMAS, ANDREW - University Of Missouri
item Sauer, Thomas

Submitted to: Soil Systems
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/22/2021
Publication Date: 7/30/2021
Citation: Kharel, T.P., Ashworth, A.J., Owens, P.R., Philipp, D., Thomas, A.L., Sauer, T.J. 2021. Teasing apart silvopasture system components using machine learning for optimization. Soil Systems.

Interpretive Summary: Silvopastoral systems combine agroforestry and pasture/livestock to maximize ecosystem services and mitigate risk by diversifying markets. It is a complex system, with factors such as soil, topography, tree, and forage species interacting to influence production. Classical statistical methods are not designed to evaluate these complex interactions. Scientist from USDA-ARS, University of Arkansas, and University of Missouri compared a novel machine learning approach to classical statistical methods on a 3 year (2017-2019) grazing dataset from a silvopasture study site established at the University of Arkansas, Fayetteville for identifying the most important factors that affect overall system productivity. Results showed that the machine learning method performed better than classical approach for interpreting agricultural big data for ultimately identifying why cattle selectively graze one area of a landscape over another. Overall, soil wetness (depth) affected soil metal and nutrient distribution, which in turn affected tree and grass growth, and ultimately where cattle preferred to graze. The automation of methods outlined in this paper (machine learning approach for variable selection and interpretation) can help scientist select important variables and interpret their relationship for improving system productivity. Overall, machine learning can help explain soil-crop relationships and overall optimization of these complex, multifunctional systems.

Technical Abstract: Silvopasture systems combine tree and livestock production to minimize market risk and enhance ecological services. Our objective was to explore and develop a method for identifying driving factors linked to productivity in a silvopastoral system using machine learning. A multi-variable approach was used to detect factors that affect system-level output (i.e., plant production (tree and forage), soil factors, and animal response based on grazing preference). Variables from a three-year (2017–2019) grazing study, including forage, tree, soil, and terrain attribute parameters, were analyzed. Hierarchical variable clustering and random forest model selected 10 important variables for each of four major clusters. A stepwise multiple linear regression and regression tree approach was used to predict cattle grazing hours per animal unit (h ha-1 AU-1) using 40 variables (10 per cluster) selected from 130 total variables. Overall, the variable ranking method selected more weighted variables for systems-level analysis. The regression tree performed better than stepwise linear regression for interpreting factor-level effects on animal grazing preference. Cattle were more likely to graze forage on soils with Cd levels <0.04 mg kg-1 (126% greater grazing hours per AU), soil Cr <0.098 mg kg-1 (108%), and a SAGA wetness index of <2.7 (57%). Cattle also preferred grazing (88%) native grasses compared to orchardgrass (Dactylis glomerata L.). The result shows water flow within the landscape position (wetness index), and associated metals distribution may be used as an indicator of animal grazing preference. Overall, soil nutrient distribution patterns drove grazing response, although animal grazing preference was also influenced by aboveground (forage and tree), soil, and landscape attributes. Machine learning approaches helped explain pasture use and overall drivers of grazing preference in a multifunctional system.