Abstract:
With 141 000 km2 area and accounting for approximately 20% of China’s grain production, the North China Plain is one of the most important agricultural regions in China. Agricultural land uses affect land surface energy and water balance. With changes in natural condition and economics, the spatial distribution patterns of crops change accordingly, affecting the quantity and quality of regional food production. It is important to precisely determine the distributions of land areas under different crops. This lays the basis for not only adjustment and optimization of agricultural structure, but also the reduction of agricultural disaster and protection of groundwater resources. As an advanced technique, remote sensing is widely used in crop research in terms of changes in spatial patterns at different spatial and temporal scales. Remote sensing can also be used to accurately monitor crop planting in real time at low cost. To understand the spatiotemporal variations in crop planting in the North China Plain, this paper established a method for classifying agricultural land use using MODIS NDVI and TM/ETM data. Raw 16-day composite NDVI data were firstly processed using HANTS filtering and then combined with multiple cropping index (MCI) to extract planting areas of winter wheat/summer maize, single-cropping maize, forest and fruit trees, and cotton. The planting areas of vegetables and rice were next extracted by supervised classification using the TM/ETM image data. Finally, the spatiotemporal variations in the main crops planting areas in the North China Plain from 2000 to 2013 were analyzed. The results were as follows: 1) The planting areas of the main corps of the study area extracted from MODIS NDVI and TM/ETM data were stable and highly precise. 2) Winter wheat/summer maize mainly distributed in the piedmont plain of Taihang Mountain and in the Yellow River irrigation regions of Shandong and Henan Provinces. While single-cropping maize widely distributed in the north of Hebei Plain, rice was concentrated in Tianjin, Tangshan and along the coast of the Yellow River. Also vegetables mainly distributed in the suburbs whereas forest and fruit trees scattered over major fruit producing areas and around Beijing and Tianjin. Cotton concentrated in the central region of the North China Plain. 3) The planting areas of food crops (wheat, maize and rice) in the North China Plain decreased obviously. However, economic crops (forest and fruit trees, and vegetables) increased significantly. The percentage changes in area of forest and fruit trees, vegetables and rice were respectively 56.45%, 35.76% and 23.16%. Furthermore, vegetables and rice had an obvious shift in planting area. 4) Landscape pattern indexes of area weighted mean patch fractal dimension and Shannon’s evenness index showed that large-scale degree of winter wheat/summer maize planting enhanced in the south of Hebei Plain, while the planting area of winter wheat/summer maize increased in North Henan. Because different planting areas of several kinds of economic crops in the north of North Henan increased, regional crop diversity index increased. The results provided a critical reference for adjusting agricultural planting structure and reasonable utilization of resources.