Changes in quantity, quality, and pattern of farmland in a rapidly developing region of China: a case study of the Ningbo region

摘要

Analyses of the quantity, quality, and pattern of farmlands are important to adequately understand the changing trajectories of farmlands, and the effectiveness of farmland protection policies in rapidly developing China. In this study, an integrated approach incorporating remote sensing images, random forest (RF) classification, geographic information system (GIS), digital elevation model (DEM), land quality score (SQ), and landscape pattern indices was applied to analyze the features of farmland use change in China, taking the Ningbo region as an example. RF classification was chosen to extract farmland use information from Landsat images; GIS and DEM were applied to quantify changes in farmland areas and their spatial variations; SQ and landscape pattern indices were used to detect changes in farmland quality and pattern. The results showed a trend of decrease followed by increase in farmland quantity, a continuously decreasing trend in farmland quality, and continuously increasing trends in both farmland fragmentation and dispersion in the Ningbo region. Besides, there was also a distinct trend of farmland advancing upwards into hills with slopes larger than 6 degrees and elevations higher than 20m. Farmland protection policies, represented by the overall land use plans of Ningbo, proved to be effective in protection of only farmland quantity, rather than quality and pattern. Suggestions were given for the optimization of farmland protection policies by considering not only quantity and quality but also the farmland pattern in China.