IMPROVING THE SIMPLACE MODELLING FRAMEWORK FOR SUNFLOWER SIMULATION UNDER SALT STRESS
W. Z. Zeng1,2,3*, T. Ma1, G. Q. Lei1, Y. H. Fang4, Y. G. Zhang5,6, Y. Y. Zha1, J.W. Wu1 and J. S. Huang1
1State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan 430072, China
2Crop Science Group, Institute of Crop Science and Resource Conservation (INRES), University of Bonn, Katzenburgweg 5, D-53115 Bonn, Germany
3State Key Laboratory of Simulation and Regulation of the Water Cycle in River Basins, China Institute of Water Resources and Hydropower Research, Beijing 100038, P.R. China
4State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, China
5Department of Hydrology and Water Resources, The University of Arizona, Tucson, Arizona, 85721, USA
6Department of Soil, Water and Environmental Science, The University of Arizona, Tucson, Arizona, 85721, USA
*Corresponding author’s email: zengwenzhi1989@whu.edu.cn
ABSTRACT
Soil salinization is a major environmental challenge for crop growth. To monitoring salt stress in crops and benefit agricultural managements, SIMPLACE Modelling Framework was improved by modifying (i) water stress coefficient (TRANRF); (ii) radiation use efficiency (RUE); (iii) specific leaf area (SLA) and (iv) both RUE and SLA. The improved model was calibrated and validated using two years field trial data of sunflower collected at the Yichang experimental station in Hetao Irrigation District (HTD), Inner Mongolia, China. Sunflowers were planted in six salt affected plots and the salinity levels were denoted as saturated electrical conductivity (ECe). Crop characteristics included biomass and leaf area index (LAI) were measured several times during the growth period. We firstly obtained the TRANRF, RUE, and SLA for sunflower in potential conditions and then modified the TRANRF and scale factors of RUE and SLA (designated by prefix S to indicate SRUE and SSLA) to fit the measurements respectively. Results indicated that all these modifications could give relative accurate simulations for biomass and LAI but modifying SRUE and SSLA together gave us the most accurate predictions. Meanwhile, the simulation accuracy also varied with target variables, with more accurate biomass simulation by employing biomass as a target variable, and vice versa. In addition, we averaged the SRUE and SSLA results by using biomass and LAI as target variables respectively to balance the simulation accuracy and found that the averaged SRUE and SSLA results could obtain both acceptable biomass (RMSE=56.4-205.3 g·m-2) and LAI (RMSE=0.44-1.47) simulations. In addition, SRUE and SSLA had linear relationship with soil salt (ECe) and natural logarithm of ECe respectively. Meanwhile, using ECe of 0-10 cm depth could obtain higher accuracy for both biomass (averaged RMSE=268.6 g·m-2 vs 274.9 g·m-2) and LAI (averaged RMSE=1.89 vs 1.95) predictions of sunflower than using ECe of 0-100 cm depth.
Key words: Modelling, radiation use efficiency, salt stress, sunflower, specific leaf area |