Open AccessRelationship between soybean yield from high and low yielding field sites and selected soil characteristics
Author(s) -
Bandara Ananda Y.,
Weerasooriya Dilooshi K.,
MurilloWilliams Adriana,
White Charles M.,
Collins Alyssa A.,
Bell Terrence H.,
Esker Paul D.
Publication year - 2020
Publication title -
agrosystems, geosciences and environment
Language(s) - English
Resource type - Journals
ISSN - 2639-6696
DOI - 10.1002/agg2.20126
Subject(s) - rhizoctonia solani , biology , soybean cyst nematode , rhizoctonia , agronomy , growing season , principal component analysis , soil texture , yield (engineering) , soil water , ecology , cultivar , mathematics , materials science , metallurgy , statistics
The sporadically occurring high and low yielding field sites is a commonly observed manifestation of the within‐farm spatial heterogeneity of soybean [ Glycine max (L.) Merr.] yield in Pennsylvania. Understanding the causes behind this phenomenon is important for implementing site‐specific management practices, leading to greater attainable yields at farm scale. To explore the problem, bulk soil samples were systematically collected in 2018 from farmer‐designated high and low yielding sites at V1 (early season) and R8 (late season) soybean growth stages from 14 farms in Pennsylvania. Soil organic matter; cation exchange capacity; pH; P, K, Mg, Ca, S, Zn, and Cu; soil texture; site slope; number of putatively pathogenic fungal colonies ( Fusarium , Pythium , and Phytophthora species and Rhizoctonia solani ); and plant pathogenic nematodes were assayed. No measured variables differed significantly between two site types at V1 or R8. Principal component analysis did not distinctly cluster high and low sites based on the variables included in the analysis. Hierarchical clustering on principal components showed situations where sites from different farms are more similar than sites from the same farm. Correlation analysis performed for variables collected with high sites showed that yield is significantly and positively correlated with V1‐spiral nematode count, R8‐ R. solani colony count, and soil K content at V1 and R8. For low sites, yield was significantly and positively correlated with V1‐spiral nematode count and soil P/Zn content at R8. Taken together, our findings indicate that the causes behind within‐farm spatial heterogeneity of soybean yields are complex and beyond the variables investigated in this study.