Dryland winter wheat production and its relationship to fine-scale soil carbon heterogeneity - A case study in the US Central High Plains

Authors: RAMÍREZ, PAULINA - Oregon State University; CALDERÓN, FRANCISCO - Oregon State University; Vigil, Merle; Mankin, Kyle; Poss, David; FONTE, STEVEN - Colorado State University

Submitted to: Agronomy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/9/2023
Publication Date: 10/12/2023
Citation: Ramírez, P.B., Calderón, F.J., Vigil, M.F., Mankin, K.R., Poss, D.J., Fonte, S.J. 2023. Dryland winter wheat production and its relationship to fine-scale soil carbon heterogeneity - A case study in the US Central High Plains. Agronomy. 13(10). Article e2600. https://doi.org/10.3390/agronomy13102600.
DOI: https://doi.org/10.3390/agronomy13102600

Interpretive Summary: Precision management of dryland crop fields will require decision makers to understand the relationships between crop yield and site-specific soil and field factors. Using detailed soil, field, and winter wheat yield data on a 100-ft grid across 11 fields, we found wheat yield increased in response to soil levels of total carbon, clay, potassium, and phosphorus and decreased in response to soil levels of sand. However, soil explained less than 40% of the spatial variation in yield, meaning that other factors also affected yield that we didn't analyze. Even though the fields were relatively flat (< 3% slopes), yields decreased at higher elevations in two of the fields. These results highlight the continuing need for local and site-specific knowledge in developing dryland precision management strategies.

Technical Abstract: Understanding the spatial relationship between soil and yield is critical for the implementation of site-specific management strategies. This study evaluates the spatial variation of wheat yield in relation to soil attributes, including macro and micronutrients, soil C, and texture within 11 fields. Soil samples (0–15 and 0–30 cm depths) were obtained at 30 m grid-spacing, and wheat grain yield data were obtained from harvester-mounted monitors in 2018 and 2019. These yields were interpolated for each field using Empirical Bayesian kriging. Wheat yield was negatively affected by soil sand content in both years. Conversely, total C, clay, K, and Olsen P were positively correlated with wheat yields. A regularized random forest algorithm (RRF) showed that soil attributes contributed differently to the yield spatial variation during the two seasons studied. Soil was able to explain less than 40% of the spatial variation in yield, and the influence of soil properties on yield were markedly lower in the fields during the 2018 season relative to the 2019 season. The total C, P, and SO4-S were important determinants of yield in 2018, while Fe, Mn, and Olsen P were important in 2019. Even though the fields studied were relatively flat areas (< 3% slopes), two of the fields showed significant negative trends in grain yield. These results highlight the continuing need for local and site-specific knowledge in developing dryland precision management strategies.