While LCLUC information is a key component for better understanding land-atmosphere interactions, most LCLU maps lack time-series information, which limits their value for climatological studies. Yaqian He, a colleague in my department (Tim Warner), and Eungul Lee produced a continuous series of annual LULC maps of China from 1982 to 2013 using random forest classification derived from the NOAA AVHRR NDVI data [He et al., 2017 Remote Sensing of Environment]. Based on a comparison with Google Earth images, the overall accuracy of a simplified eight-class version of our 2012 map of land use and land cover is 73.8%, which is not significantly different from the accuracy of the NASA MODIS map of the same year. These annual maps of land cover will be an important dataset for climate studies, and the methodologies used in the classification and validation of LCLU can be applied to other geographical regions where availability of continuous time series of LCLU maps is limited.

By using our newly developed time-series of annual LCLU maps, Yaqian He, colleagues in my department (Tim Warner and Brenden McNeil), and Lee have aggregated the nominal LCLU maps to a coarser scale to produce fractional cover of croplands, forests, and grasslands to identify fractional changes in the three major LCLU types in China [He et al., 2018 Remote Sensing].