摘要：During the past three decades, sea water level (SWL) in the Caspian Sea has declined by about 2 m and sea area has decreased by about 15 000 km2. This has affected coastal communities, the environment and economically important gulfs of the sea (e.g. Dead Kultuk). To assess the effects of coastline change and evaluate zones vulnerable to desiccation, we simulated SWL using total inflow from feeder rivers and precipitation and evaporation over the sea. We determined potential vulnerable areas of the sea over the past 80 years by comparing the minimum and maximum annual water body maps (for 1977 and 1995). We then determined the linear regression between SWL rise and covered potential vulnerable area (CVA), using annual Normalised Difference Water Index (NDWI) maps and SWL data from 1977 to 2018. Combining SWL-CVA regression and SWL simulation model enabled us to determine desiccated areas in different regions of the Caspian Sea due to changes in precipitation, evaporation and total inflow. The results showed that 25 000 km2 of the sea is potentially vulnerable to SWL fluctuations in terms of desiccation, with 70% of this vulnerable area located in Kazakhstan. Potential vulnerable area per kilometre coastline was found to be 6 km2 in Kazakhstan, 4 km2 in Russia and whole of Caspian Sea, 1.5 km2 in Iran, 1 km2 in Azerbaijan and 0.5 km2 in Turkmenistan. The results also indicated that SWL in the Caspian Sea is sensitive to evaporation and that e.g. a 37.5 mm decrease in mean annual net precipitation would lead to a 1875 km2 decrease in the sea area, while a 1 km3 decrease in mean annual inflow would lead to a 1400 km2 decrease in the sea area. Thus the developed framework enabled the spatial consequences of changes in water balance parameters on sea area to be quantified. It can be used to assess future changes in SWL and sea area due to anthropogenic activities and climate change.
关键词：Normalised Difference Water Index (NDWI); Sea water level; Sensitivity analysis; Simulation; Water balance
