Development of a soil-plant phosphorus simulation model for calcareous and weathered tropical soils
Views
0% 0
Downloads
0 0%
Timeless limited access
Loading...
View/Open
Corresponding Author
Date
2003-03-08
Date Issued
2003-06-01
ISI Journal
Impact factor: 4.212 (Year: 2003)
Citation
Samira H. Daroub, Argyrios Gerakis, Joe T. Ritchie, Dennis K. Friesen, John Ryan. (1/6/2003). Development of a soil-plant phosphorus simulation model for calcareous and weathered tropical soils. Agricultural Systems, 76 (3), pp. 1157-1181.
Abstract
Phosphorus (P) is a limiting nutrient affecting crop yields in many regions of the world, in particular in areas with highly weathered acid soils and calcareous alkaline soils. Given the many factors associated with P behavior in a range of soil environments, there is a need to go beyond site-specific situations. The use of crop simulation models is a valuable tool to evaluate the efficacy of applying P fertilizers under different soil, management and climatic conditions. A computer model was developed to simulate P in the soil–plant system adapted to soils with high P limitations. The soil P module is operated with two comprehensive crop simulation models (CERES and CROPGRO) within the DSSAT software. The P module comprises inorganic and organic P pools estimated from measured P fractionation data and works on a daily time step. The rate constants for P movement between the pools follow first order kinetics. The P module was calibrated and tested using three data sets from Colombia, Syria, and Tanzania. The limited testing showed that the P module simulated accurately grain yield and P uptake by wheat grown under semi-arid conditions. The wheat crop responded little to fertilization although measured Olsen P was as low as 2.6 mg kg−1. The P module overestimated P uptake for both soybean and bean crops grown in acidic soils, but predicted with a good degree of accuracy labile P in the soil and P uptake for maize grown under the same acidic conditions. Testing with more data sets is needed to improve model predictions.