Modeling impacts of film mulching on rainfed crop yield in Northern China with DNDC

Abstract Water stress is a major factor threatening agricultural production across a wide range of rainfed croplands in China. Drought threats would become worse along with climate change, especially in Northern China where the projected climate change scenarios indicated decreases in precipitation in the arid or semi-arid agricultural areas. Pilot experiments have been launched to search alternative farming management practices for adaptation of the climate change in China. Plastic film mulching (FM) has recently been tested at a number of sites in China with encouraging results although no any regional assessment has been done yet. This paper reports how we met the gap by testing a process-based, biogeochemical model, Denitrification-Decomposition or DNDC, against observations and then utilizing the model to upscale the simulations to a large region in China. DNDC was first modified by including a new module, which tracked variations of soil climate under the film mulching conditions. Two new input parameters, i.e., FM-covering fraction and duration, worked in conjunction with daily weather data to define the daily soil temperature and moisture profiles. By varying the FM coverage or duration, we could simulate a variety of FM settings and their impacts on the soil climate. A 3-year dataset of soil climate as well as crop yield measured at a rainfed corn field in Shaanxi Province in Northwestern China were used to serve the model validation tests. The measured and modeled results were in agreement with each other and both indicated that the FM practice substantially improved the soil moisture as well as the crop yield. Sensitivity tests were conducted with the revised DNDC by varying each of four factors, i.e., precipitation, temperature, soil texture and fertilizer application rate, in its range commonly observed in Northern China while keeping other input factors constant. Results from the sensitivity tests indicated that the effectiveness of FM was mainly related to precipitation. Efficiency of FM increased with decrease of precipitation. The FM effectiveness was evaluated at regional scale by linking DNDC to the databases holding spatially differentiated climate, soil and management data for all the 1.17 million ha of rainfed corn fields in the entire province of Shaanxi, across which the annual average precipitation decreased from 940 mm in the south to 390 mm in the north. Results from the regional simulation indicated that (1) corn production increased by 1.79 million tons or 16% with FM applications in the domain of Shaanxi; (2) the FM-induced increases in corn yield mainly occurred in the northern counties of the province where precipitation was lower than 700 mm; and (3) the effectiveness of FM decreased with increase in precipitation from the northern to the southern areas in the domain. The study concluded that film mulching practice could play an important role in elevating rainfed crop yields in the arid or semi-arid regions in China.