A multi-criteria remote sensing-based data-driven framework for monitoring lake drying and salinization and mapping its environmental impacts

Lakes are natural water resources that are affected by different factors. The increase in soil salinity is one of the major issues created due to the drying up of a lake. In this study, a two-step methodology was used to assess drought vulnerability and salinity variation of the Urmia Lake basin. In this regard, firstly, a multi-criteria intelligence method based on empirical wavelet transform-long short-term memory, which integrated 15 geo-environmental variables extracted from the in-situ observations and satellite datasets, was used for developing drought vulnerability maps of the basin. In the next step, the salinization progress of the basin and its impacts on the environment were investigated using satellite datasets. Results showed that the Southern and Eastern sections of the lake were more prone to severe droughts. It was found that temperature and precipitation variations did not lead to significant shrinkage of the lake, but human activities along with climate changes caused the basin to dry up. The results showed that the biomass production in the basin is affected by salinity, and there is a negative correlation between the salinity index and the normalized difference vegetation index. Also, a positive correlation was found between land subsidence and the density of drilled wells in the basin. The rate of land subsidence varied between − 1.8 and − 8.4 mm/year. The quality of groundwater was investigated for the existing wells in the basin. Results showed that the excessive use of groundwater resources has affected the quality of water.