The effect of spatial heterogeneity in near-infrared (NIR) surface reflectance and the number of ground samples as well as their locations were analyzed • The threshold of spatial heterogeneity to meet a uncertainty of 0.01 for NIR surface reflectance was given for different sampling cases • Optimal sampling plays a more important role in reducing the uncertainty of ground "truth" than increasing the number of sample plots
Abstract. In situ measurements from sparsely distributed networks worldwide are a valuable source of the reference data for validating or correcting the bias of satellite products. However, the significant differences in spatial scale between in situ and satellite measurements make them incomparable except for that the underlying surface of in situ sites is absolutely homogeneous. Instead, in site measurements need to be upscaled to be matched with satellite pixel. Based on the upscaling model we proposed as well as the consideration that in-situ observation generally lacks spatial representativeness due to the widely distributed spatial heterogeneity, we have developed a coarse pixel-scale ground "truth" dataset based on ground measurements of 368 in situ sites from the sparsely distributed observation networks. Furthermore, the effectiveness of the dataset was carefully evaluated over the sites with different degrees of spatial representativeness. The results demonstrate that using this dataset in validation outperforms the direct comparison between satellite and in situ site measurements. The accuracy of the reference data employed for validation or bias correction can be enhanced by 3.5 % overall with this dataset. But the performance of the dataset show dependence on the degree of spatial heterogeneity. Specifically, the improvement of accuracy was the most significant over the regions with strong spatial heterogeneity, with the accuracy of reference data enhanced by 7.3 %. To the best of our knowledge, this dataset is unique in providing coarse pixel scale ground truth with the widest spatial distribution and longest time series. Its ability to capture both spatial and temporal variations of surface albedo at coarse spatial scales makes it an invaluable resource for validating and correcting global surface albedo products.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.