Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #424574
Research Project: Improving Resiliency of Semi-Arid Agroecosystems and Watersheds to Change and Disturbance through Data-Driven Research, AI, and Integrated Models

Location: Water Management and Systems Research

Title: Topographic position index predicts within-field yield variation in a dryland cereal production system

Author
item MACDONALD, JACOB - Colorado State University
item Barnard, David
item Mankin, Kyle
item Miner, Grace
item Erskine, Robert
item Poss, David
item MEHAN, SUSHANT - South Dakota State University
item Mahood, Adam
item Mikha, Maysoon

Submitted to: Agronomy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/28/2025
Publication Date: 5/27/2025
Citation: Macdonald, J., Barnard, D.M., Mankin, K.R., Miner, G.S., Erskine, R.H., Poss, D.J., Mehan, S., Mahood, A.L., Mikha, M.M. 2025. Topographic position index predicts within-field yield variation in a dryland cereal production system. Agronomy. 15(6). Article e1304. https://doi.org/10.3390/agronomy15061304.
DOI: https://doi.org/10.3390/agronomy15061304

Interpretive Summary: Many farming systems show a lot of variation in crop yields, even within the same field. Better crop management requires better understanding of what causes this variation. We used machine learning to see how topography, soil properties, and weather affected yield over several years from 18 fields in northeastern Colorado. We found that yield wheat and millet yields were higher in parts of the fields with smaller Topographic Position Index values (in other words, relatively lower elevations). Yield was also higher in areas with less sand in the soil, more carbon in the soil, and higher nitrogen fertilizer applied. It is noteworthy that Topographic Position Index was so good at predicting yield, because most similar research had not used it before. Even though we had a large and detailed dataset, our models were only able to explain about 25% of the variation in crop yields. This shows that more research is needed to fully understand the factors that influence crop yields.

Technical Abstract: Many agricultural systems exhibit a large degree of within-field yield variability. We require a better understanding of the drivers of yield variability in order to optimally manage croplands. Here, we utilize a multi-year dataset, spanning 18 fields, from a dryland system in northeastern Colorado, including high resolution topographic data, densely sampled soil properties, and on-site weather data to investigate drivers of spatial variability of hard red winter wheat (Triticum aestivum L., variety Langin), corn (Zea mays L.), and proso millet (Panicum milaceum L.) yields at a sub-field scale. We modeled yield for each crop separately using random forest regression, and evaluated model performance using spatially-blocked cross-validation. Topographic position index (TPI) and percent sand had a strong negative effect on yield, while nitrogen application rate (N) and total soil carbon had strong positive effects on yield in both the wheat and millet models. Remarkably, TPI had almost as large of an effect size as N, and outperformed other more commonly used topographic predictors of yield such as topographic wetness index (TWI), elevation, and slope. Despite the size and quality of our dataset, cross-validation results revealed that our models can only account for approximately one quarter of the total yield variance, highlighting the need for continued research.