Analysis of cultivated land change, pressure index and its prediction in Shaanxi province

Abstract: Cropland is a very important land resource for maintaining peoples living and grain safety. Many researchers study cropland change and predict future trends at the different scales, and the results are not ideal because the short time series statistic data cannot show cropland change characteristics and the regression model method is not perfect for simulating future cropland trends. Shaanxi is an agriculture province in western China and is representative of all agriculture production. Statistical yearbooks are very important for analysis of long time series in China. We obtained 60 years' worth of cropland statistics from the statistical yearbooks of Shaanxi province between 1949 and 2010.The area change characteristics and pressure conditions of the cropland from 1949 to 2010 were analyzed for service agriculture production and food security based on these data. The cropland pressure index was calculated and GM (1, 1) model was used to forecast the trend of cropland and cropland acreage per capita for the future ten years. In order to ensure that the results were credible, non-parametric analysis was used to analyze the interaction between different parameters. The results showed that cropland consisted of arid land, irrigated land, and marshy fields. Arid land was the main type of cropland in Shaanxi province, and accounted for 94.63% of cropland acreage in 1949 and 63.40% in 2010. There was 438.47×104 hm2 cropland in 1949 and 286.05×104 hm2 in 2010, which decreased 152.41×104 hm2 and reduced 2.46×104 hm2 per year from 1949 to 2010. The acreage of irrigated land and marshy fields increased, with the increased acreage amount of irrigated land being more than that of marshy fields. The acreage of irrigated land increased 77.51×104 hm2 between 1949 and 2010. Arid land had an effect on cropland acreage change. Cropland structure became more reasonable and experienced the increased stage and reduced stage repeatedly between 1949 and 2010. Cropland acreage per capita was 0.38 hm2 in 1949 and 0.10 hm2 in 2010 and decreased 0.28 hm2, the change of which during 1949 to 1980 was more obvious than that after 1980. According to non-parameter analysis, cropland acreage per capita was affected by population that added 2417.17×104 peoples during the past 60 years. Based on the GM (1,1) models of cropland acreage per capita and cropland acreage, the amount of cropland and cropland acreage per capita will be 286.10×104 hm2 and 0.090 hm2 in 2013, 281.68×104 hm2 and 0.087 hm2 in 2015, and 270.92×104 hm2 and 0.079 hm2 in 2020, respectively. Under different projected conditions, excessive pressure of cropland happened in different times. Smin (420) condition was in the earliest time (2014) and Smin (400) in 2017. GM（1, 1）was suitable to simulate the future cropland change. Some suggestions were provided to prohibit food security issues and maintain the amount of cropland at the higher level in Shaanxi province.