The dataset of main grain land changes in China over 1985–2020

Continuous, Accurate, and detailed information on main grain land (MGL) areas is crucial for provisioning food security and making policies affecting sustainable agricultural production. Here, we developed the change map of MGL with resolution 30m in China for the period 1985–2020 using the Landsat image-based random forest algorithm on the GEE platform. Finally, the planting intensity, gain time and loss time of MGL was calculated. Results indicate that our mapping results are highly consistent with the annual planting area of various grain crops according to national statistics. The proportion of different MGL types in 2020 is presented as follows: single maize (38.55%) > single rice (23.23%) > single wheat (15.42%) > wheat & maize (12.60%) > double rice (7.65%) > wheat & rice (2.48%). Due to over half of MGL expansion occurring after 2000, the average planting intensity of MGL in China is 19.74 years in the past 36 years. The consistent, high-resolution data of MGL change can support progress toward sustainable agricultural management and food production.针对主要粮田（Main Grain Land, MGL）区域的连续、精准且详尽的信息，对于保障粮食安全、制定可持续农业生产相关政策具有至关重要的意义。本研究基于谷歌地球引擎（Google Earth Engine, GEE）平台，采用基于Landsat影像的随机森林算法，构建了1985-2020年中国范围内分辨率为30米的主要粮田变化图谱。最终，本研究还计算了主要粮田的种植强度、增益时段与流失时段。结果表明，本研究生成的图谱结果与国家统计的各类粮食作物年度种植面积具有高度一致性。2020年各类主要粮田类型占比依次为：单季玉米（38.55%）>单季水稻（23.23%）>单季小麦（15.42%）>麦玉轮作（12.60%）>双季稻（7.65%）>稻麦轮作（2.48%）。由于超过半数的主要粮田扩张发生在2000年之后，过去36年间中国主要粮田的平均种植强度为19.74年。这套连续一致的高分辨率主要粮田变化数据，可为可持续农业管理与粮食生产的发展提供有力支撑。