Quantifying theagreement and accuracy characteristics of four satellite-based LULC productsfor cropland classification in China

Various land use and land cover (LULC) productshave been produced over the past decade with the development of remote sensingtechnology. Despite the differences inLULC classification schemes, there is a lack of research on assessing theaccuracy of their application to croplands in a unified framework. Thus, this study evaluated the spatial andarea accuracies of cropland classification for four commonly used global LULCproducts (i.e., MCD12Q1 V6, GlobCover2009, FROM-GLC and GlobeLand30) based onthe harmonised FAO criterion, and quantified the relationships between fourfactors (i.e., slope, elevation, field size and crop system) and croplandclassification agreement. The validationresults indicated that MCD12Q1 and GlobeLand30 performed well in cropland classificationregarding spatial consistency, with overall accuracies of 94.90 and 93.52%,respectively. The FROM-GLC showed theworst performance, with an overall accuracy of 83.17%. Overlaying the cropland generated by the fourglobal LULC products, we found the proportions of complete agreement anddisagreement were 15.51 and 44.72% for the cropland classification,respectively. High consistency wasmainly observed in the Northeast China Plain, the Huang-Huai-Hai Plain and thenorthern part of the Middle-lower Yangtze Plain, China. In contrast, low consistency was detectedprimarily on the eastern edge of the northern and semiarid region, theYunnan-Guizhou Plateau and southern China. Field size was the most important factor for mapping cropland. For area accuracy, compared with ChinaStatistical Yearbook data at the provincial scale, the accuracies ofdifferent products in descending order were: GlobeLand30, FROM-GLC, MCD12Q1,and GlobCover2009. The croplandclassification schemes mainly caused large area deviations among the fourproducts, and they also resulted in the different ranks of spatial accuracy andarea accuracy among the four products. Our results can provide valuable suggestions for selecting croplandproducts at the national or provincial scale and help cropland mapping and reconstruction, which is essential for foodsecurity and crop management, so they can also contribute to achieving theSustainable Development Goals issued by the United Nations.