Bagher Ghermezcheshmeh; Aliakbar Rasuli; Majid Rezaei-Banafsheh; Alireza Massah; Alimohammad Khorshiddoost
Abstract
In the statistical downscaling methods which is based on the relationship between AOGCMs data and ground based climatic variables (such as rain and temperature), the future period of those variables are simulated. Since in the simulation, all effective parameters cannot be modeled, estimated values suffers ...
Read More
In the statistical downscaling methods which is based on the relationship between AOGCMs data and ground based climatic variables (such as rain and temperature), the future period of those variables are simulated. Since in the simulation, all effective parameters cannot be modeled, estimated values suffers from be uncertainty. The outputs of downscaling models are used as inputs to agriculture and water resources models; therefore, identifying the models inputs’ error or uncertainty is essential to realize the reliability of obtained results. In this research, an attempt is made to investigate the uncertainty of Artificial Neural Network (ANN) as a downscaling model in a case study in the northwest of Iran. For this purpose, precipitation, minimum and maximum temperature variables were used in the designed ANN model, and the NCEP data was employed for its calibration and validation. The HadCM3 was the selected AOGCM in this study. Observed daily time series were gathered at all stations in the study period and on the basis of bootstrap method the 99% confidence interval was calculated for all the variables. In the next step, the simulated (downscaled) mean and variance of the variables by the ANN model, compared to the calculated confidence interval. To compare the results, the criterion of the number of station-month was used. The results showed that the average maximum temperature at 14 station-months were within the confidence interval. The results of monthly analysis showed that the accuracy of ANN model in summer was low and its uncertainty is more than the other seasons. In the simulation of minimum temperature based on this criterion, 18 station-months were within the confidence interval. The accuracy of ANN to estimate the minimum temperature in summer was low with high uncertainty in almost all the stations. Moreover, in June and August in any of the stations estimated values were not within the confidence interval. Due to the high variability of rainfall in relation to temperature, confidence range was very high, and in some stations was more than 50% of average monthly precipitation. Because of the high confidence rang of precipitation, in 53 Stations-month cases were within the confidence interval.
Bagher Ghermezcheshmeh; Aliakbar Rasuli; Majid Rezaei Banafsheh; Alireza Massah Bovani; Alimohammad Khorshiddust
Abstract
Increasing Green House Gases (GHG) may change the climate in different areas. Investigation of parameters are difficult due to induced changes in climate parameters, such as precipitation and temperature. For predicting global climate change, different climate scenarios are defined, using AOGCM models. ...
Read More
Increasing Green House Gases (GHG) may change the climate in different areas. Investigation of parameters are difficult due to induced changes in climate parameters, such as precipitation and temperature. For predicting global climate change, different climate scenarios are defined, using AOGCM models. AOGCMs are able to simulate global atmospheric circulation patterns. However, the spatial resolutions of such models are coarse; for example HadCM3 has spatial resolutions of 3.75 and 2.5 in longitude and latitude, respectively. Therefore, to study climate change in a given area, the outputs of the used AOGCMs must be downscaled properly. For this reason, statistical and dynamical methods have been developed. Statistical methods establish a relationship between AOGCM outputs and climate parameters such as precipitation and temperature. For example, many statistical methods use multiple regressions to predict future climate parameters. However, the accuracy of downscaling procedure varies depending on the geographical position of the studied station in relative to the nearby AOGCM grids. In this research, the accuracy of SDSM was tested in different synoptic stations of northwest Iran. This area has a complex topography and climate due to intrusion of different rain bearing weather systems to the region. First of all, daily climate data (precipitation, maximum and minimum temperature) were collected and their time series created. HadCM3 data for the girds over the studied area was obtained and SDSM model was applied for each climate parameters of all synoptic stations in the region. Then, the difference between the SDSM outputs and observed parameters were evaluated for all the stations and the performance of the downscaled outputs were evaluated by comparing the mean and variance of the model outputs and those of the NCEP/NCAR for the present climate. The morpho-climatic parameters were derived for each station and their relations with the magnitude of the model error were evaluated. Results showed that the error in precipitation has significant relation with the distance to the grid center, whereas the error in maximum temperature is related to the difference between the elevation of a given station and the mean elevation of the HadCM3 grids. For example, in Urmia station, the error is the highest of 104 mm while in Saqez the error is the lowest of 9.4 mm. Also, the maximum temperature accuracy in stations with elevation near to mean elevation of the grid is higher. Pars Abad station with 32 m elevation and with high elevation difference with the grid mean elevation, showed 1.14 ºC of error and Tabriz station with less elevation difference to grid mean elevation, showed 0.0.08 ºC of error.
