| Popis: |
Water resources are becoming increasingly scarce, and improving water productivity (WP) is crucial to the development of dryland farming on the Chinese Loess Plateau. Fertilization and cropping systems are important measures to improve WP, and tracking their long-term effects on water balance components is helpful for water use and risk management. In this study, the calibrated and validated Hydrus-1D model was used to simulate the dynamic changes of water balance components under four different fertilization treatments (without fertilization, CK; 120 kg ha−1 N and 26.2 kg ha−1 P, NP; 75 t ha−1 manure, M; 120 kg ha−1 N, 26.2 kg ha−1 P, and 75 t ha−1 manure, NPM) and three cropping systems (winter wheat monoculture, WWM; spring maize monoculture, SMM; wheat-wheat-maize rotation, WMR) from 1985 to 2020, with differences in crop yields and WP analyzed. Compared to CK, NPM, NP, and M treatments increased the average yield by 201.9%, 161.7%, and 130.6%; increased the inter-annual yield variations by 191.2%, 149.3%, and 144%; increased transpiration by 80.4%, 58.7%, and 37.5%; reduced evaporation by 18.8%, 5.6%, and 2.3%; decreased deep percolation in the 3 m profile by 44.7%, 60.7%, and 41.1%; and increased WP by 174.5%, 130.8%, and 110.2%, respectively. Values under the three cropping systems occurred in the order SMM>WWM>WMR for normalized crop yield, SMM>WMR>WWM for transpiration, WWM>WMR>SMM for evaporation, and SMM>WWM>WMR for annual average WP. Our results suggest fertilization significantly increases crop yield and WP, with NPM fertilization treatment having the highest WP, but results in large inter-annual yield variations in dryland farming. Spring maize monoculture had the highest yield and WP in this study, while the rotation system balanced the water use differences. These results provide useful information for formulating water management practices for regional dryland farming. |
Full Text from ScienceDirect