The Effects of Desiccation and Climatic Change on the Hydrology of the Aral Sea

Abstract

Anthropogenic desiccation of the Aral Sea between 1960 and the mid-1990s resulted in a substantial modification of the land surface that changed air temperature in the surrounding region. During the desiccation interval, the net annual rate of precipitation minus evaporation (P ? E) over the Aral Sea?s surface became more negative by ?15%, with the greatest changes occurring during the summer months. In addition, Aral Sea surface temperatures (SST) increased by up to 5°C in the spring and summer and decreased by up to 4°C in the fall and winter. A series of coupled regional climate?lake model experiments were completed to evaluate if the observed hydrologic changes are caused by desiccation or instead reflect larger-scale climatic variability or change, or some combination of both. If the P ? E changes are the result of desiccation, then a positive feedback exists that has amplified the anthropogenic perturbation to the hydrologic system. The effects of desiccation are examined by varying the simulated area, depth, and salinity of the Aral Sea in different model experiments. The simulated changes in SST resulting from desiccation are similar to the observed changes?both simulated and observed SSTs have increased during the spring and summer and have decreased during the fall and winter. The simulated changes in the annual cycle of P ? E resulting from desiccation are also similar to observed changes, but the simulated net annual decrease in P ? E is only ?30% of the observed decrease. Warming has been observed across central Asia during the desiccation interval. The hydrologic response to this large-scale climatic variability or change was assessed by perturbing the meteorological boundary conditions (1.5°C cooling with constant relative humidity) but leaving the Aral Sea characteristics unchanged. The simulated effects of warming do not closely match the observed changes on the monthly timescale?SST changes are positive and the P ? E changes are negative in all months. However, the annual change in P ? E is similar to the observed value. The simulated hydrologic response to the combined effects of desiccation and warming matches the observed SST and P ? E changes more closely than the response to each forcing alone. This result indicates that a combination of both desiccation and climatic change or variability has produced the observed hydrological changes?the primary effect of desiccation is to alter the annual cycle of SST and P ? E whereas warming has modified the hydrologic budget on the annual timescale.