Urganch-农作物种植结构-2019

1）数据内容：种植结构指的是一个地区或国家在农作物种类种植比例上的问题，一般以粮食作物为主，其他经济类作物为辅，本数据描述10M分辨率的灌区种植结构的空间分布。2）数据来源及加工方法：哨兵数据，随机森林法。3）数据质量描述：Kappa系数80%。4）数据应用成果及前景：各种水文生态模拟分析的基础数据，精细计算农业蒸散、农业需水、下渗、灌溉需求，农业结构达到田块级别，为了能够推动农业种植健康发展，调整和优化各因素，并明确各要素在农业种植结构中的作用显得尤为重要。5）种植结构是在GEE平台上利用随机森林算法并结合采集的样本点数据计算得出的，为了区分方便，在计算过程中我们将每一种相似作物类型用一个阿拉伯数字表示，计算完的.tif结果再用分区统计的方式链接到提取的耕地上，在这个过程中，我们把表示作物类型的字段保留下来，即max字段，每个阿拉伯数字对应的作物类见说明文档。

1) Data Content: Planting structure refers to the planting proportion of crop species in a region or country, generally taking grain crops as the mainstay and other economic crops as supplements. This dataset describes the spatial distribution of irrigation area planting structures at a 10-meter resolution.
2) Data Source and Processing Method: Sentinel data, Random Forest algorithm.
3) Data Quality Description: Kappa coefficient reaches 80%.
4) Data Application Achievements and Prospects: This dataset acts as the fundamental data for various hydro-ecological simulation analyses, allowing accurate calculation of agricultural evapotranspiration, agricultural water demand, infiltration and irrigation water demand. With the planting structure refined to the field scale, adjusting and optimizing relevant factors and clarifying the role of each element in the agricultural planting structure are particularly critical to promoting the healthy development of agricultural planting.
5) The planting structure was calculated using the Random Forest algorithm combined with collected sample point data on the Google Earth Engine (GEE) platform. For the convenience of differentiation, each similar crop type was represented by an Arabic numeral during the calculation. The generated .tif result was then linked to the extracted farmland via zonal statistics. In this process, the field representing crop types, namely the "max" field, was retained. The crop types corresponding to each Arabic numeral are provided in the instruction document.