Spatio-temporal Variation Of Main Crops Planting Area In The North China Plain Using Remote Sensing Data

Agricultural land uses affect land surface energy and water balance. The North China Plain(NCP) is one of the most important agricultural regions in China and is experiencing a severe water shortage due to decades of excessive extraction of groundwater for crop irrigation. The precise determination of different crops in NCP does not only provides the basic information needed for evaluating agricultural water consumption in space but also improves agricultural planning for sustainable water resources management.During the process of monitoring multi-year continuous variation of different crop planting area using remote sensing, varied training samples or classification rules cannot reflect change in spatiotemporal pattern of crop planting area very well.In this study, we established a method for agricultural land use classification using MODIS NDVI time-series data. Raw 16-day composite NDVI data were first processed using Savitzky-Golay filtering and then sampled the time-series pattern of NDVI for different croplands based on ground truth data. The typical phenological characteristics of the main crops were determined based on field-monitored phenological data in Agro-meteorological Stations in the study area. Then multiple cropping index(MCI) was induced based on the NDVI time-series to distinguish land areas under wheat-corn double cropping system.Finally, we extracted cropping area of winter wheat, corn, cotton, fruit in the North China Plain for the period of 2000-2013 by proposing a numerical model using the CART algorithm.The result was validated with county agricultural statistics data and sample data, further the motivation and changing of main crop spatial pattern in NCP in 2000-2013 were analyzed. Draw the following main conclusions:(1) The result was tested with county agricultural statistices data, which showed a statistically significant correlation in winter wheat, corn and cotton for the 13-year period with correlation coefficient of 0.78, 0.71 and 0.83 at 95% confidence level, respectively. The result was validated with field sample data, which showed that Kappa coefficient ranged from 0.72 to 0.96 and the total classification accuracy ranged from 79% to 97% in 2000-2013. In particular, the total classification accuracy less than 85% only appeared 2 years.(2) Spatial distribution of crops showed, winter wheat/summer corn was mainly distributed in the Yellow River irrigation areas of Shangdong Province and Henan Province and the piedmont of Mount Taihang. Fruit tree was distributed in major fruit producing areas and around Beijing. Cotton was concentrated in the central of NCP. During the fourteen years, cropping area of winter wheat and corn showed a downward trend in NCP, which from 5.53 million hectare to 5.30 million hectare and from 6.42 million hectare to 5.69 million hectare, respectively. The most significant decrease trend appeard in the North Hebei Plain. Winter wheat abandonment was appeared in Tianjin-Langfang-Baoding. The 14-year average cotton planting area about 0.84 million hectare. Especially in the South Hebei Plain, cotton planting area was more than half of the total area(53%). It showed significant fluctuations and increasing trend over years. The average fruit tree planting area about 0.63 million hectare. Nearly 14 years, the fruit tree planting areas of all four partitions of NCP showed increasing trends, and the main factors probably were economic benefits and government-oriented.(3) From the point of view of the planting transformation between economic crop and food crop, the exchanging area of fruit tree and food crop was 0.95 million herctare(accounting 11.15% of arable land in NPC), and mainly located in the area which has fruit cultivation history, such as Tangshan, Cangzhou, Binzhou and around Beijing. The exchanging area of cotton and food crop was 0.64 million herctare(accounting 7.58% of arable land in NPC), and concentrated in The South Heibei Plain and Wuqiao-Dongguang-Nanpi.(4) Agricultural land use frequently changed(about yearly) due to crop rotation, price fluctuation, water limitation, etc. This made it is difficult to evaluate agricultural land use change only through comparison of agricultural land use maps for any set of years. Thus in this study, the land use change for different crops was evaluated based on multi-year cropping probability comparison, which composed of actual counts of number of years of crop cultivation in the study period. the high cultivation probability reflects the dominant crop distribution. It can provide a potential method for improving the applicability of remote sensing in multiyear crop planting area monitoring at large scale.