The rapid decline in Urmia lake levels, the largest inland wetland in Northwest of Iran and the third hyper-saline lake in the world, as the biggest environmental disaster has been of great concern. In recent decades, a frequent occurrence of dry periods has experienced in the North West. In early 2000 the depth of the lake has dropped about 6 meters, which is mainly due to changing patterns of water for agricultural uses, construction of numerous dams and dry periods (Deljo, 2013).
Abbaspour and Javid (2012) by hydrodynamic model which developed to simulate the condition of Urmia Lake and validated using the 35-yr data of annual precipitation, evaporation, run off, river discharges and ...view middle of the document...
Reference evaportranspiration, a key hydrological variable, is one of these variables which can be used to calculate water requirement, schedule irrigation and prepare input data for hydrological models. (Li et. al.2012).The principal weather parameters affecting evapotranspiration are radiation, air temperature, humidity and wind speed. The reference crop evapotranspiration represents the evapotranspiration from a standardized vegetated surface which water is abundantly available at it (FAO 56).
As the evaporation power of the atmosphere is expressed by the reference crop evapotranspiration (ETo), understanding the current trend, future changes of this variable can be use to project the future condition of Urmia Lake levels.
The objective of this study are (i) to analyze potential changes in climatic variables and in ETo on monthly ad annual time steps from 1971 to 2009 (ii) to project the temporal changes in ETo during 2011-2099 with statistical downscaling method.
-Data and Methodes
Climate data from Urmia weather station at latitude 37.5° N and longitude 45° E, 1316 m above mean sea level and GCM grid outputs (2011-2099) are used to generate and analyze the ETo in this region. The measured climate data for Tman (°C), Tmin(°C), Tdew (°C), wind speed (m/s) at 10 m , sun shine (h), air pressure (hPa), Precipitation (mm) were available on a daily time step from 1951 – 2005. The wind speed at 10 m transformed to wind speed at 2m height by the wind profile equation (Allen et. al., 1998).
SDSM (Statistical DownScaling Model) version 4.2 is employed to provide future scenarios of ETo. SDSM as a hybrid of regression based and stochastic weather generator downscaling method (Wilby et.al. 2002) has been used in many climate change studies.
There are 26 daily atmospheric predictor variables, describing atmospheric circulation, thickness and moisture content, derived from: (i) interpolated daily reanalysis dataset of NCEP for 1961-2001 at a spatial scale of 2.5° (longitude) × 2.5° (Latitude) (ii) outputs of scenarios A2 (medium – high emission) and B2 (medium-low emission) of HadCM3 (Hadly Centre Coupled Model, version 3) from 1961 to 2099 with a spatial resolution of 3.75° (longitude) × 2.5° (Latitude) are employed to provide future predictors.
By the SDSM model, the empirical relationship was established between a local-scale variable of Urmia weather station as a predictant (in this study ETo) and large-scale variables obtained from NCEP during 1961-2001. For developing and validating of regression equations, two sub periods including 1961-1990 and 1991-2001 were used respectively.
By trying different values of inflation and bias correction, as set-up parameters of the SDSM, the best combination of predictor variables were selected during calibrating model.
In order to get the time series of local variable for future climate, in the calibrated SDSM model, the corresponding HadCM3 predictors under A2 and B2 scenarios replaced the NCEP...