AUTHOR=Sui Haochen , Chen Dawei , Yan Jiaguo , Li Bin , Li Wei , Cui Baoshan TITLE=Soil Salinity Estimation Over Coastal Wetlands Based on Random Forest Algorithm and Hydrological Connectivity Metric JOURNAL=Frontiers in Marine Science VOLUME=Volume 9 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2022.895172 DOI=10.3389/fmars.2022.895172 ISSN=2296-7745 ABSTRACT=Owing to climate warming and human activities (irrigation, reservoirs), sea level rise and runoff reduction have been threatening the coastal ecosystem by increasing the soil salinity. However, short-term sparse in-situ observations limited the study on the response of coastal soil salinity to external forcings and thus its effect on coastal ecosystem. In this study, based on hydrological connectivity metric and random forest algorithm (RF), we developed a coastal soil salinity inversion model with in-situ observations and satellite-based datasets. Using Landsat images and ancillary variables as input, we produced the 30 m monthly grid datasets of surface soil salinity over Yellow River Delta. Based on the cross-validation result with in-situ observations, the proposed RF model performed higher accuracy and stability with the determination coefficient of 0.89, the root means square error of 1.48 g·kg-1 and the mean absolute error of 1.05 g·kg-1. The proposed RF model could gain the accuracy improvements of about 11–43% over previous models at different conditions. The spatial distribution and seasonal variabilities of soil salinity was sensitive to change signals of runoff, tide and local precipitation. In conclusion, by combining spatiotemporal collaborative information with the hydrological connectivity metric, the proposed RF model in this research could accurately estimate surface soil salinity, especially in natural reserved regions. The modelling results of surface soil salinity could be significant for exploring the effect of seawater intrusion and runoff reduction to the evolution of coastal salt marsh ecosystems.