Skip to main content
ARS Home » Plains Area » Temple, Texas » Grassland Soil and Water Research Laboratory » Research » Publications at this Location » Publication #386064
Research Project: Resilient Management Systems and Decision Support Tools to Optimize Agricultural Production and Watershed Responses from Field to National Scale

Location: Grassland Soil and Water Research Laboratory

Title: Simulating productivity of dryland cotton using APSIM, climate scenario analysis, and remote sensing

Author
item LI, ZHOU - Shanxi Agriculture University
item Menefee, Dorothy
item YANG, XUAN - Shanxi Agriculture University
item CUI, SONG - Middle Tennessee State University
item RAJAN, NITHYA - Texas A&M University

Submitted to: Agricultural and Forest Meteorology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/29/2022
Publication Date: 8/31/2022
Citation: Li, Z., Menefee, D.S., Yang, X., Cui, S., Rajan, N. 2022. Simulating productivity of dryland cotton using APSIM, climate scenario analysis, and remote sensing. Agricultural and Forest Meteorology. 325. Article 109148. https://doi.org/10.1016/j.agrformet.2022.109148.
DOI: https://doi.org/10.1016/j.agrformet.2022.109148

Interpretive Summary: Cotton yield in a Texas farm was simulated using both APSIM crop modeling software and a remote sensing-based method with good agreement between models. Cotton yields were then modeled for several climate change scenarios with most showing a decline in productivity.

Technical Abstract: Assessing the potential of using process-based models, remote sensing, and climate change scenarios on biomass production, development, lint yield, and precipitation productivity of cotton (Gossypium hirsutum L.) under dryland condition is crucial for major production areas, such as the East-Central Texas. Based on two-year field data obtained from a large dryland cotton farm (12 ha), APSIM accurately predicted cotton biomass and lint yield during the calibration (NRMSE: biomass, 17.6%; lint, 10.8%) and validation (NRMSE: biomass, 18.8%; lint, 13.1%) processes. The deviation of simulated days after sowing was less than 6 days across squaring, flowering, and boll maturity stages. A partial least square model constructed based on satellite NDVI data and days after sowing accurately predicted cotton biomass values (R2 = 0.93, P < 0.05), and the results agreed well with APSIM predicted values (R2 = 0.96, P < 0.05). Decreased yield were detected in almost all RCP scenarios, with the greater reductions in end-century (29.9-82.4%) scenarios than the mid-century (16.2-46.7%). For precipitation productivity, 14-80.3% of reduction was found across all future scenarios. Great reduction in reliable yield (45.0-92.0%) and reliable precipitation productivity (37.6-84.7%) were projected by future scenarios. Slight differences were detected in model validation between two dominant soil types, but the behaviors under future projections were very different (P < 0.05), largely caused by differences in water holding capacity during critical growth stages. The results from this study suggest that a combination of managerial and breeding effort focusing on water use efficiency enhancement should be promoted in the future.