عنوان مقاله [English]
Climate change is one of the major challenges which society will face during current century. Temperatures are projected to increase up to 2 ºC by 2100, which is expected to result in major changes in the atmosphere’s energy balance and the hydrological cycle. The term soil moisture generally refers to temporary storage of precipitation at a depth of 1 to 2 meters from the soil profile). Soil moisture in the root zone well known as a very important factors that severely effects on crop productivity. Hence, the study of soil moisture changes is crucial for the planning and managing water resources in the coming periods. Accordingly, this study was conducted in order to predict and stimulate soil moisture in upcoming period (2020-2039) according to baseline (1992-2011).
Materials and Methods
The study area in this research was Neyshabour plain located in Khorasan Razavi province. At first, SWAP model implemented using meteorological and agronomic data from study area, and soil moisture of field capacity is simulated for 30 cm soil depth. Then, to ensure the results of moisture simulation, the moisture content of the model is calibrated and validated using the measured soil moisture content of the 2008-2009. After ensuring the accuracy of the model results, soil moisture is simulated for the baseline (1992-2011). Then, to estimate moisture content in the upcoming period (2020-2039) six models (CanESM2, GFDL, MIROC, IPSL, CSIROMK3.6, and GISS-ES-R) and two emotion scenarios (RCP4.5 and 8.5) were used, as well as the ratio of the weekly and yearly values of the meteorological parameters of the upcoming period, which estimated using the baseline. In addition, to evaluate SWAP model accuracy, Root Mean-Squared Error (RMSE), Mean Absolute Error (MAE), and Coefficient of Determination (R2) were used. Data obtained by Neyshabur Synoptic Station (1992-2011) were utilized in order to determine daily weather patterns and future weather parameters. Temperature and precipitation parameters determined by the LARS-WG model.
Results and Discussion
In order to use the SWAP simulated moisture, initially the model calibrated and validated with measured soil moisture data from 2008-2009 growth season. So, after soil moisture simulation for 2008-2009, the measured data by TDR used for calibration and validation model. The coefficient of determination factor (R2) between simulated and measured results was 79.5%. The climate parameters used in this study include minimum and maximum temperatures and precipitation, which are known as the most important factors affecting the soil moisture. According to table 6, the minimum and maximum temperatures and precipitation for the upcoming period will increase compared to the baseline. On the other hand, According to table 9 and table 10, under two scenarios, the means, maximum and minimum soil moisture decrease compared to baseline. The weekly and yearly uncertainty in soil moisture under two scenarios showed in figures 4 and 5, respectively. Under the 4.5 scenario, the MIROC model due to the lower band thickness and the IPSL model, due to the higher bandwidth thickness, have the highest and lowest accuracy, respectively. Whereas, under the 8.5 scenario, the IPSL model has the highest accuracy and the CanESM2 and GISS-ES-R models have the lowest accuracy compared to other models. Results of weekly and yearly uncertainty showed less and more uncertainty for weekly and yearly soil moisture, respectively.
In general, results of this study revealed that minimum, maximum temperature and precipitation will increase in upcoming period compared to baseline. Soil moisture decrease in upcoming period compared to baseline under two scenarios (4.5 and 8.5). Moreover, temperature was higher and precipitation was lower under 8.5 scenario rather than 4.5 scenario. Therefore, 8.5 scenario showed worse condition compared to 4.5 scenario. Results of this study showed lower soil moisture in 1, 3, 6, 13, 18, and 19 years in upcoming period under two scenarios. Therefore, these years known as warning condition compared to baseline.