Simulating crop yields and water productivity for three cotton-based cropping systems in the Texas High Plains

Abstract

Implementing appropriate cropping systems suited to specific soil types and climatic conditions is crucial for improving crop yield and conserving water in semi-arid environments. The Decision Support System for Agrotechnology Transfer (DSSAT) was applied to simulate crop yields of cotton, sorghum, and winter wheat across three cropping systems, including continuous cotton, cotton-sorghum, and cotton-wheat. Simulations were conducted for 48 fields with various soil types across six counties in the Texas High Plains, spanning growing seasons from 2000 to 2022. Cotton water productivity, derived from DSSAT-simulated cotton yield and crop evapotranspiration (ET), was compared among these cropping systems. The DSSAT demonstrated good performance (R² ​≥ ​0.79, nRMSE ​≤ ​15.74%, and d-index ​≥ ​0.95) in predicting yields of cotton, sorghum, and winter wheat. The CROPGRO-Cotton model showed slightly better accuracy in predicting cotton yield under the continuous cotton system than under the cotton-sorghum and cotton-wheat systems. Model performance was similar across different soil types, with slightly higher accuracy in fine-textured soils such as clay loam (R² ​≥ ​0.84, MAPE ​= ​12.35, and d-index ​= ​0.95) than in other soils (R² ​≤ ​0.82, MAPE ​≥ ​13.76, and d-index ​≤ ​0.94). Additionally, the model performance varied by season, showing high accuracy in years with adequate precipitation but generally underpredicting cotton yields in drought seasons. Among the three cropping systems, cotton yield and water productivity were the highest for the cotton-sorghum system (6.3 ​kg ​ha⁻¹ ​mm⁻¹), followed by the cotton-wheat and continuous cotton systems. Overall, the DSSAT models effectively captured the effects of management practices, soil types, and growing seasons in predicting crop yield and crop water productivity across three cotton-based cropping systems. The findings provide valuable information for decision support in adopting cropping systems across various soil types and environmental conditions, fostering sustainable agriculture and water conservation in semi-arid regions.

Keywords

Cotton / Cropping systems / DSSAT / Sorghum / Wheat / Yield and water productivity

Abbreviations

AcA Acuff loam (0-1% slope)

AfA Amarillo fine sandy loam (0-1% slope)

AIB Amarillo loam (1-3% slope)

AmB Amarillo loamy fine sand (0-3% slope)

Br Brownfield fine sand (0-3% slope)

CERES Crop Environment Resource Synthesis

CV (%) coefficient of variation in percentage

CWP crop water productivity

d-index agreement index

DSSAT Decision Support System for Agrotechnology Transfer

EcB Estacado clay loam (1-3% slope)

ET evapotranspiration

GRIDMET gridded meteorological dataset

LoA Lofton clay loam (0-1% slope)

MAPE mean absolute percentage error

NCCPI national commodity crop productivity index 3.0

nRMSE normalized root mean square error

OcA Olton clay loam (0-1% slope)

OtA Olton loam (0-1% slope)

PuA Pullman clay loam (0-1% slope)

PuB Pullman clay loam (1-3% slope)

R² coefficient of determination

Sh Springer loamy fine sand (0-1% slope)

SLPF soil fertility factor

SOC soil organic carbon

SpB Spur fine sandy loam (1-3% slope)

SSURGO soil geographic database

Tv Tivoli fine sand (0-1% slope)

Availability of data and materials

The management data for each field and each year of this study are available in the appendix as supplementary documents, weather data are available in GRIDMET, and soil data are available in the SSURGO database.

Authors’ contributions

B.G., O.A., G.R., and W.G.: Writing, reviewing, and editing; B.G., O.A., and W.G.: Data curation; B.G. and W.G.: Methodology, investigation, and conceptualization; G.R. and W.G.: Validation; B.G.: Writing of original draft, visualization, and data analysis; and W.G.: Supervision, software, resources, project administration, and funding acquistion.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Author Wenxuan Guo (Editorial Board member) was not involved in the journal's review nor decisions related to this manuscript.

Acknowledgements

We appreciate the financial support from Cotton Incorporated ​(​16-252TX), ​USDA NIFA and Cotton Board (2022-67013-36992), and USDA NIFA Hatch (9898).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.crope.2025.03.001.