Analysis of Temporal Change of Actual Evapotranspiration and Its Relationship with Temperature and Precipitation in East Azarbayejan Province Using MODIS

Introduction
The increasing expansion of industries and the use of fossil fuels have led to an increase in greenhouse gas emissions and, consequently, global warming and the occurrence of climate change phenomena. Climate change is a change in the climate behavior of an area relative to the behavior that is expected during a long-term period of observed or recorded information in that area. The ET element is considered as one of the most important components of water balance in nature; it is strongly influenced by important climate components such as temperature and under the influence of climate change, it can exhibit various reactions. Therefore, in calculating the water balance of each basin, ET calculation is important. Because ET along with the surface flow and water penetration in the soil are considered as components of the water balance. Due to the temporal variations of climate variables and consequently ET, the use of remote sensing methods can consider these changes as more favorable. Using Landsat satellite imagery and energy balance model, using the NOAA, AVHRR and SEBS algorithms, Baba Jafari et al (2015) estimated monthly ETA values for agricultural use in the Akhole region in Tabriz. Comparison of the results of the algorithm with observation values ​​indicates the accuracy of the model with a coefficient of 0.8 and the mean square root error of 9.64 millimeter per month (Baba Jafari et al., 2015: 1).
Materials and methods
In this study, for the analysis of the ETa of the East Azarbaijan province, we applied MOD16A2 remote sensing data in a time interval of 8 days in the period 2014-2000. Data on maximum, minimum and precipitation temperatures of 11 stations of the province were also used for the purpose. Since the relationship between temperature and ETa is linear, the correlation between the average of the maximum and minimum temperatures of 8 days in the stations of the province was calculated with the mean values ​​of 8 daily ETa for representative cells. As the purpose of the current research is to estimate ETa for explaining the water balance of the province, the correlation between precipitations of the stations in the province with the ETa of the representative cells was also evaluated. Given the fact that the data of the MODIS are 8 daily scales, with programming in MATLAB software, the data of the maximum, minimum and precipitation temperatures of the stations was converted into eight days interval. Investigations showed that MODIS provides only ETa data of the representative cells of Ahar, Jolfa, Sarab, Sahand, Marand and Meyaneh stations, and for other stations due to the placement of the cell in the class of ground cover construction, ETa data has not been recorded.
Results and discussion   
The findings of this study indicate that the average maximum and minimum mean and total mean of ETa data on 124546 pixels in the province for the period from 2000 to 2014 were 2.3, 0.5 and 0.8 millimeter per day. The time series ETa of the cells within the boundaries of the province also showed the highest changes in 2003 and 2010, at 9.1 millimeter a day, and the lowest in 2001, at 0.1 millimeter per day. In this regard, the time series of the ETa of the stations in the study years showed that the highest and lowest values ​​of ETa changes were in the Sarab and Sahand stations, and the value was 3.3 (millimeters per day) in 2012 and 0.01(millimeters per day) in 2001. Due to the fact that the temperature relationship with ETa is linear, the correlation between the average maximum and minimum temperatures of 8 days in the stations of the province with average ETa values ​​of 8 days for representative cells of the station was calculated. Investigations showed that there was a negative relationship between the mentioned parameters. Basically, there must be a positive and direct relationship between temperature and ETa. The only factor that can disrupt this relationship is the lack of adequate water in the area. Regression analysis on min and max temperature data matrix with ETa of 8 days in the stations at 95% confidence level showed that with 1 degree increase in temperature in the province, the ETa decreases by 0.02 mm per day. The ETa variation slope in relation to precipitation stations showed that at 95% of confidence level, with an increase of 1 millimeter of precipitation, the ETa would increase by 0.9 millimeter per day.
Conclusion
In this study, was investigated the correlation between the average maximum and minimum temperature of ground stations with the mean values of 8 daily ETa of the representative cells of the stations. The results of the investigations showed that there is a negative relationship between the mentioned parameters. The existence of a negative relationship between the above mentioned parameters reflects the fact that when the warm season approaches, the water resources are decreasing as a result of decreasing water, and also ETa values ​​decrease with increasing temperature. Because there is not enough water to evaporate in the area, the correlation between mean precipitation and ETa also showed a positive relationship. In other words, the higher the rainfall, the higher is the water available to the plant and the soil. Given that the average annual rainfall in the province is 267.27 mm and the annual loss of ETa is 32.4 mm. This amount of evaporation is significant compared with 267.27 millimeters of annual rainfall. The climate condition of the province leads to desert and dry land.