Skip to main content
ARS Home » Southeast Area » Fayetteville, Arkansas » Poultry Production and Product Safety Research » Research » Research Project #441675

Research Project: Land Suitability Analysis for Olive and Maize Crop in Agroforestry System by Using Machine Learning Algorithms

Location: Poultry Production and Product Safety Research

Project Number: 6022-63000-006-001-N
Project Type: Non-Funded Cooperative Agreement

Start Date: Feb 28, 2022
End Date: Feb 28, 2024

Objective:
Landscape transformation from prevailing annual crop systems towards increased diversity, perenniality, and integrating livestock and trees has potential to improve the resiliency of agroecosystems. This collaboration will foster climate resilience, ecosystem services and health, and profitability by identifying and quantifying benefits of diverse perennial circular systems across major regions of the world affected by climate change. The objective of this collaboration is to use advanced big data approaches to evaluate land suitability at multiple scales under a changing climate.

Approach:
Materials: Available soil data from legacy soil surveys and past and current research collected over time continue to be assembled and curated to perform various statistical and spatial analysis. In addition, available remote and proximal sensing soil data that are complimentary to the existing data continue to be collected. Procedure: Machine learning models will be implemented to conduct analysis to determine the controlling factors of soils development/formation and properties and crop response in diverse crop-tree systems. Multiple predictive models will be used to to improve efficiency and resilience, particularly against the backdrop of increased agricultural production demands for a growing population under greater climatic stochasticity. Expected Output: Detailed maps of soil functional units and interpretations will be generated. The procedures, methodological developments and lessons learned will be shared via papers published in peer review journals. All output will become available to users and public at large.