Investigating uncertainty of climate change effect on entering runoff to Urmia Lake Iran

Abstract

Abstract. The largest lake in Iran, Urmia Lake, has been faced with a sharp decline in water surface in recent years. This decline is putting the survival of Urmia Lake at risk. Due to the fact that the water surface of lakes is affected directly by the entering runoff, herein we study the effect of climate change on the runoff entering Urmia Lake. Ten climate models among AOGCM-AR4 models in the future time period 2013–2040 will be used, under the emission scenarios A2 and B1. The downscaling method used in this research is the change factor-LARS method, while for simulating the runoff, the artificial neural network was applied. First, both the 30-yr and monthly scenarios of climate change, temperature, and precipitation of the region were generated and weighted by the Beta function (β). Then, the cumulative density function (cdf) for each month was computed. Calculating the scenarios of climate change and precipitation at levels of 25, 50, and 75% of cdf functions, and introducing them into LARS-wg model, the time series of temperature and precipitation in the region in the future time period were computed considering the uncertainty of climate variability. Then, introducing the time series of temperature and precipitation at different risk levels into the artificial neural network, the future runoff was generated. The findings illustrate a decrease of streamflow into Urmia Lake in scenario A2 at the three risk levels 25, 50, and 75% by, respectively, −21, −13, and −0.3%, and an increase by, respectively, 4.7, 13.8, and 18.9% in scenario B1. Also, scenario A2 with its prediction of a warm and dry climate suggests more critical conditions for the future compared to scenario B1 and its cool, humid climate.