Reveal the severe spatial and temporal patterns of abandoned cropland in China over the past 30 years

Abstract

The abandonment phenomenon is familiar in China. It threatens food security and seriously affects the sustainable development of society, the economy, and the natural environment. However, monitoring and mapping abandoned cropland on a large scale remains a challenge because of its complex land-use change process. According to the World Food and Agriculture Organization, cropland not used for agricultural production for more than 5 years is defined as abandoned cropland. This study uses the Landsat high-precision long-time series product to detect cropland nationwide from 1990 to 2019 by using the sliding window method on Google Earth Engine to profile the spatial distribution, intensity, trend, frequency, and recultivation of abandoned cropland. Between 1992 and 2015, the results illustrate that the total area of abandoned cropland in China was 559,170.26 km², 18.59 % of the cropland area. Excluding the recultivated cropland area, China's abandoned cropland area amounted to 392,156.24 km², 13.03 % of the total cropland. Cropland abandonment is mainly distributed in the second terrain gradient in midwestern and southwestern regions such as Inner Mongolia and Gansu. It rarely happens in western and eastern coastal areas. A high abandonment rate area usually has high elevation and slope. The light index is negatively correlated with the abandonment rate in suburban areas. This study is the first to map the spatiotemporal distribution of abandoned cropland in China with high precision at 30 m resolution. It provides an important basis for policies regarding the recultivation of abandoned cropland.

Keywords: Abandonment; Arable land; Google Earth Engine; Recultivation; Sliding window.