Improving the cotton simulation model, GOSSYM, for soil, photosynthesis, and transpiration processes

Sahila Beegum, Dennis Timlin, Kambham Raja Reddy, Vangimalla Reddy, Wenguang Sun, Zhuangji Wang, David Fleisher, Chittaranjan Ray

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


GOSSYM, a mechanistic, process-level cotton crop simulation model, has a two-dimensional (2D) gridded soil model called Rhizos that simulates the below-ground processes daily. Water movement is based on gradients of water content and not hydraulic heads. In GOSSYM, photosynthesis is calculated using a daily empirical light response function that requires calibration for response to elevated carbon dioxide (CO2). This report discusses improvements made to the GOSSYM model for soil, photosynthesis, and transpiration processes. GOSSYM’s predictions of below-ground processes using Rhizos are improved by replacing it with 2DSOIL, a mechanistic 2D finite element soil process model. The photosynthesis and transpiration model in GOSSYM is replaced with a Farquhar biochemical model and Ball-Berry leaf energy balance model. The newly developed model (modified GOSSYM) is evaluated using field-scale and experimental data from SPAR (soil–plant–atmosphere-research) chambers. Modified GOSSYM better predicted net photosynthesis (root mean square error (RMSE) 25.5 versus 45.2 g CO2 m−2 day−1; index of agreement (IA) 0.89 versus 0.76) and transpiration (RMSE 3.3 versus 13.7 L m−2 day−1; IA 0.92 versus 0.14) and improved the yield prediction by 6.0%. Modified GOSSYM improved the simulation of soil, photosynthesis, and transpiration processes, thereby improving the predictive ability of cotton crop growth and development.

Original languageEnglish (US)
Article number7314
JournalScientific reports
Issue number1
StatePublished - Dec 2023

ASJC Scopus subject areas

  • General


Dive into the research topics of 'Improving the cotton simulation model, GOSSYM, for soil, photosynthesis, and transpiration processes'. Together they form a unique fingerprint.

Cite this