Optimization Of The Nitrogen Fertilizer Management For A Winter Wheat-Summer Maize Rotation System In The North China Plain-from Field To Regional Scale

Excessive nitrogen fertilization is a common problem in a winter wheat-summer maize rotation system in the North China Plain, which cause not only the decrease of nitrogen use efficiency but also the increasing of the risk of environmental pollution. Therefore, optimizing N fertilization in field scale and as well as in regional scale is urgent. A two years survey with 561 farmers, 13 field experiments with 6-10 N levels and four replicates, and 57 field experiments with three treatments without replicates in 2 growing seasons were conducted from 2001 to 2003 in Huimin County, Shandong Province. The real-time N monitoring and regulating technique (RTNMRT) based on soil nitrate-N testing and N balance had been developed to optimize N fertilization and improved N fertilizer use efficiency. Effects of N fertilization on crop grain yield and apparent N losses had been evaluated. Based on regional N balance model and GIS technique, the regional optimized N fertilization management system, which takes the target yield, soil N supply and as well as N losses into consideration, had been established preliminarily.The results of farmers survey showed that, the average amount of 673 kg·hm-2(N), 244 kg·hm-2 (P2O5) and 98 kg·hm-2 (K2O) were applied to the winter wheat-summer maize rotation system in Huimin County, Shandong Province. Under conventional crop management condition, wheat grain yield was affected significantly by soil organic matter and spike number per acreage. Maize grain yield was affected significantly by the growth days after sowing.Under current fertilizer application method and time condition, optimized N fertilization, which based on soil nitrate-N quick testing and N balance, dramatically reduced N losses meanwhile maintaining crop yield. In the optimized N treatments, the average of N application rates were 85 kg·hm-2 (w=8) for winter wheat and 144 kg·hm-2 (w=5) for summer maize in all experiments, and the residual soil nitrate-N content in the top 90 cm of the soil profile after harvest varied from 87 to 139 kg·hm-2, and apparent N losses varied from -14 to 64 kg·hm-2. Compared with the conventional N fertilization treatments, the optimized N fertilization treatments saved 77% N fertilizer for wheat (n=4) and 41% N fertilizer for maize (n=5), and reduced 72% apparent N losses without crop yield decrease.The length of border was found affecting the amount of irrigation water and N losses. Compared with the longer border (180m), 30mm irrigation water was saved, and 44mm water drainage and 20 kg·hm-2 N losses were reduced in once irrigation by shortening border to 90m. Under conventional irrigation and N fertilization management condition, the spatial variability in the water and soil N didn't affect wheat grain yield because of excessive water and N input.A quick testing method for analyzing soil nitrate-N content by Merck Reflectance meter was developed and was compared with a routine method in the laboratory. A significant correlation of the soil nitrate-N content was found between the quick testing methods and the laboratory testing method(the correlation of topsoil and subsoil was 0.918 and 0.926, respectively). Therefore a quick testing method for soil nitrate-N could be used to replace the laboratory testing method for N recommendation.The spatial variability of soil total N increased with increasing spatial distance. The spatial variability of organic matter and exchangeable K was best described by exponential model, and the maximum correlative distance was 23.5 km and 16.6 km. The spatial variability of Olsen-P and pH had no relation with spatial distance inside the sampling distance. Compared with 1980's, the soil total N, organic matter content and Olsen-P in 2003 significantly increased, which was related to the surplus of N and p input during past 20 years.Apparent N mineralization in summer maize season was higher than that in winter wheat season, and both of them were linearly related to soil organic matter. Apparent N losses during wheat growth season in sandy loam soil (45 kg·hm-2) were...