Evaluating the climate change effects on temperature, precipitation and evapotranspiration in eastern Iran using CMPI5

Excessive use of fossil fuels, land use change and increasing world population and, consequently, the increasing expansion of industrial activities have led to gradual changes that we can refer to in the Earth's climate after the Industrial Revolution (Lu et al. 2019; Tian et al. 2020). The most obvious of these changes are the increase in the average temperature of the earth, and the increase in climatic phenomena such as floods, hurricanes, hail, tropical storms, heat waves, rising sea levels, melting of polar ice, drought, etc. Global warming and its effect on the water cycle is an important concern in water resources (Cheng et al. 2016; Yang et al. 2020; Quan et al. 2021), agriculture (Awan et al. 2020; Si et al. 2020; Sepahvand et al. 2021), and environmental science (Ebadi et al. 2020; Maina et al. 2020; Nnaemeka 2020; Nwankwo et al. 2020; Qayyum et al. 2020; Ma et al. 2021).

Climate change is the variation in the average long-term values of a climatic and statistical parameter, of which here the average is taken at certain intervals, for example in some decades (Zhang et al. 2019a, 2019b; Sun et al. 2021). Climate change is one of the global challenges that, with its long-term adverse effects, is slowing down the process of achieving sustainable development around the world. Climate change is a constraint on sustainable development. The fluctuation of water resources is largely a function of climate change, because the need for these resources increases with increasing evapotranspiration in warmer, drier and sunnier conditions (Huang et al. 2021). Climate change can lead to changes in the hydrological cycle of water and create special conditions in regional water resources (Zarenistanak et al. 2014; Lalehzari & Kerachian 2020a, 2020b). Climate change has different effects on temperature and rainfall distribution, and consequently on the spatial and temporal distribution of water resources, as well as the water needs of plants and water consumption in the agricultural sector (Jia et al. 2020; Rasouli et al. 2020).

As a result, recognizing the temporal and spatial fluctuations of meteorological parameters (such as temperature, precipitation, relative humidity, etc.) and their impacts on the agricultural sector and adopting appropriate strategies are essential in order to estimate and thus make the right decisions and plans from future climatic conditions, and research has been done (Zhang et al. 2011a; Karasakal et al. 2020a, 2020b; Li et al. 2021). The Intergovernmental Panel on Climate Change (IPCC) has stated that global warming is projected to cause changes in rainfall. Based on that, global hydrological cycle activity as well as atmospheric moisture content are expected to change significantly. Although predictions of changes in precipitation patterns vary from model to model, all general circulation models show less snowfall and shorter snowmelt intervals due to rising temperature (Changchun et al. 2007). Global warming leads to changes in the balance of radiation from the earth's surface, atmospheric circulation, change in spatial and temporal distribution and intensity of precipitation and accelerates the rotation of water vapor, as well as affecting hydrological properties such as evaporation, runoff and soil water. If appropriate measures are not taken to reduce the effects and adapt to climate change, this phenomenon can affect most water resources, agriculture and industry (Ross & Matthews 2009). At the end of the twentieth century, most of the studies that have been done and are being done have studied the phenomenon of climate change. The Intergovernmental Panel on Climate Change (IPCC) reported in 2001 that the climate was changing and global warming was taking place. General circulation models are used to analyze the effects of climate change. General circulation models fall into two general categories: AGCM models that consider atmospheric interaction and OGCM models that consider the interaction of oceans. Public circulation models usually include a combination of AGCM and OGCM categories. Then all atmospheric and ocean sub-models are paired together to form Ocean–Atmospheric General Circulation Models (AOGCM) and simulated in programs. Due to computational limitations, analyses related to public climate forecasts are performed by limited centers that are equipped with supercomputers dedicated to such calculations (Willby & Dawson 2013).

The changes in minimum and maximum daily temperatures in South America showed that there are no constant variations in indices based on annual maximum temperature, while there is a significant trend in daily minimum temperature indices (Vincent et al. 2005). Taye & Willems (2013) measured the effect of exponential microscopic methods on the image of the effects of climate change on the hydrological limit conditions of the Blue Nile Basin. They used two Lars-WG exponential microscale methods and the Advanced Factor of Change (QPM) method on ten GCM climate models under two emission scenarios A1B and B1. All models indicated temperature increases for the periods of 2050 and 2090. The results also showed a decrease in precipitation or an increase in evaporation.

Loo et al. (2015) examined the effect of climate change on seasonal monsoon rains in Asia and its effect on monsoon rainfall change in Southeast Asia. The results showed that 70% of the rainfall will be below normal and the temporal displacement of monsoon rains of about 15 days in Southeast Asia is a possible phenomenon in the future. Furthermore, changes in rainfall in the SEA area, including the irregular intensity and frequency of monsoon floods, will have serious consequences for human life, the economy and infrastructure and food security in this area in the future. Izadi et al. (2017) predicted monthly changes in the maximum temperature of Shahrekord region in future periods under different climate change scenarios and the results of the model forecast in future periods also indicate an increase in temperature for all months and scenarios of the study area over the base course. Accordingly, the maximum increase in monthly maximum temperature in the coming periods of 2030–2011 and 2046–2065 was observed to be 1.69 and 3.62 °C, respectively, under scenarios A2 and A1B.

In this study, an attempt has been made to determine the changes of temperature, precipitation, and evapotranspiration using GCM models of the Fifth Climate Change Report from 2020 to 2039. Moreover, the range of changes and certainty of different models of meteorological variables have been evaluated.

Neishabour city (case study) is located between 58°19′ to 59° 30′ longitude and 35° 40′ to 36° 39′ latitude on the eastern edge of the central desert of Iran (Figure 1). The climate of the region is semi-arid to arid and the average rainfall in the whole basin is 234 mm. In terms of temperature, Neishabour city has an average temperature of 13.9 °C, and 100 days of frost per year with an average minimum temperature of at least 7.1 °C and a highest minimum temperature of 22.5 °C.

For this study, precipitation, maximum temperature, minimum temperature and evapotranspiration data with BCSD suffix were obtained from the IPCC site using the GSM model with the published organization of this model in microscale form. And for Neishabour region, the latitude and longitude of Neishabour station is input into MATLAB software and the averaging of four networks around the station is determined. Each data network has an area of about 103 square kilometres in the longitudinal and latitudinal directions. In this study, because annual changes are considered, the data are not re-determined daily by exponential microscale methods. In its fifth AR5 evaluation report, IPCC used the new RCP scenarios to represent the four key emissions trajectories of RCP 2.6, RCP 4.5, RCP 6 and RCP 8.5. In this study, changes in the four parameters of precipitation, minimum temperature, maximum temperature and evapotranspiration during the future period 2020–2039 under the two scenarios of RCP 4.5 and RCP 8.5 for Neishabour region were evaluated.

Without adopting any mitigation policies and dealing with the consequences of the climate, the planet climate will proceed in the course of this scenario so that the continuation of this trend leads to the induction of radiation at the rate of 8.5 watts per square metre (W/m²) in 2100. This scenario was developed and designed by the MESSAGE modeling team and the International Institute for Systems Analysis (IIASA) in Austria, and it is characterized by an increasing trend of greenhouse gases. In this research, a number of CMIP5 models have been used to predict the temperature parameters in the future period 2020–2039 for the basic period 1992–2011, which can be seen in Table 1.

This scenario was designed by the MiniCAM modeling group and in it the emission of greenhouse gases before 2100 remains constant at 4.5 watts per square metre. In this scenario, the population growth rate is estimated to be lower than in the RCP 2.6 scenario. The characteristics of the models and scenarios used in this research are presented in Table 1.

At present, identifying temperature fluctuations can be considered as a decision-making parameter in agricultural and biological experiments (Li et al. 2017, 2019, 2020; Sun et al. 2019; Zhang et al. 2019a, 2019b, 2020; Chen et al. 2021; Sarvestani & Charehjou 2021; Liu et al. 2021). In Figure 2 the maximum temperature changes for the period from 2020 to 2039 for climate change models in the two scenarios RCP 4.5 and RCP 8.5 have been shown. Both scenarios for almost all models estimated an increase in maximum temperature, and the RCP 8.5 scenario proportionally predicted a higher degree of maximum temperature than the scenario RCP 4.5. The maximum temperature changes for all models were between 22 °C and 25 °C. The MIROC-ESM and CANESM models in both the RCP 4.5 and RCP 8.5 scenarios estimated the highest maximum temperature increase, and the GFDLCM2 and GISS E2 models for the RCP 4.5 scenario and MPI and EC-EARTH models for the RCP 8.5 scenario estimated the lowest maximum temperature. Furthermore, the GISS E2 and GFDLCM2 models estimated the lowest increase in both scenarios (Chaouche et al. 2010).

Minimum temperature changes in the near future (2020–2039) in both the scenarios RCP 4.5 and RCP 8.5 are shown in Figure 3. All models estimate a slight increase or decrease in the minimum temperature for both the RCP 4.5 and RCP 8.5 scenarios, and the RCP 8.5 scenario estimated a higher minimum temperature rise than the scenario RCP 4.5 (Chu et al. 2010). The MIROC-ESM and CANESM models in both the RCP 4.5 and RCP 8.5 scenarios predicted the highest minimum temperature increase, and the GFDLCM2 and GISS E2 models for the RCP 4.5 scenario and NORESM and EC-EARTH models for the RCP 8.5 scenario predicted the lowest minimum temperature increase (Marengo & Camargo 2008).

Figure 4 shows the precipitation changes for climate change models in the two scenarios, RCP 4.5 and RCP 8.5. Most models in the future period for both scenarios RCP 4.5 and RCP 8.5 estimated a decrease in precipitation in 2023–2025 and an increase in precipitation in 2025–2028. Fang & Pomeroy (2008) also predicted an increase in temperature. Moreover, the MIROC-ESM and GFDLCM2 models in the RCP 4.5 scenario and IPSL and GFDLCM2 models in the RCP 8.5 scenario predicted the lowest rainfall in the long-term average, and the HADGEM and GISS E2 models for both the scenarios RCP 4.5 and RCP 8.5 forecast the highest rainfall in the long-term average.

Numerous studies have been conducted in which accurate estimation of evapotranspiration has played an influential role in their results (Deng et al. 2021a, 2021b). The rate of evapotranspiration under the two scenarios of RCP 4.5 and RCP 8.5 for the future period (2020–2039) is shown in Figure 5. According to the figure, all models estimated an increase in evapotranspiration, and the evapotranspiration rate for both the RCP 4.5 and RCP 8.5 scenarios was estimated in the same way. The MIROC-ESM and CANESM models in both the RCP 4.5 and RCP 8.5 scenarios estimated the highest evapotranspiration rates, and the CESM and GISS E2 models for the RCP 4.5 scenario and EC EARTH and CESM models for the RCP 8.5 scenario estimated the lowest rate of evapotranspiration. In 2034, the decrease in rainfall and, as a result, the relative humidity of the air increases daily evapotranspiration.

Figure 6 shows the annual changes in the studied parameters for the two scenarios. The minimum temperature, maximum temperature and evapotranspiration for both scenarios has a constant trend and has a slight increase and decrease (Zhang et al. 2011b). For the predicted future period (2020–2039) both scenarios were slightly different and the RCP 8.5 scenario estimated the minimum temperature, maximum temperature, and evapotranspiration as slightly higher than the RCP 4.5 scenario.

The IPSL and GISS E2 models in the RCP 8.5 scenario and GFDLCM2 and CANESM models in the RCP 4.5 scenario estimated the lowest and highest rainfall, respectively (Figure 7). The CANESM model estimated the maximum rate of minimum temperature for both the RCP 4.5 and RCP 8.5 scenarios (Sahoo et al. 2011). As shown in Figure 7, the RCP 8.5 scenario estimated the maximum temperature rate as more than the RCP 4.5 scenario. The GISS E2 model for the RCP 4.5 scenario and the MPI model for the RCP 8.5 scenario estimated the lowest rate of maximum temperature. Previous studies also estimated temperature increase (Liu et al. 2011; Khayatnezhad & Nasehi 2021; Ren & Khayatnezhad 2021). Furthermore, the EC EARTH and CANESM models for the RCP 8.5 scenario and GISS E2 and CANESM models for the RCP 4.5 scenario estimated the lowest and highest evapotranspiration rates, respectively.

To compare between GCM models and evaluate the accuracy of the estimated values, for each of the variables, a box diagram can be used to examine the certainty of the models. Figure 8 shows a box diagram for predicted values of 11 GCM climate change models and the two scenarios, RCP 4.5 and RCP 8.5. As shown in Figure 8(a), the GFDLCM2, MPI and IPSL models have more certainty for precipitation. Furthermore, the lowest and highest rainfall were estimated by the MIROC-ESM and CANESM models, respectively. Pre-season estimation of precipitation was one of the main factors in sustainable irrigation planning (Lalehzari et al. 2016, 2020). The GISS E2 and GFDLCM2 models have more certainty for minimum temperature and maximum temperature than the other models (Figure 8(b) and 8(c)). Moreover, the GISS E2 model has the lowest and the CANESM model has the highest minimum and maximum temperature (Tatsumi et al. 2015). As shown in Figure 8(d), the CANESM and GISS E2 models predicted the highest and lowest evapotranspiration rates for the period 2020–2039.

In this study, the trend of precipitation changes, maximum temperature, minimum temperature and evapotranspiration of climate change models during the 20-year future period were examined for two scenarios, RCP 4.5 and RCP 8.5. According to the results, the minimum temperature, maximum temperature and evapotranspiration have an almost steady trend and have a slight decrease and increase, but precipitation has many fluctuations. The results also showed that the RCP 8.5 scenario estimated the four parameters for the next period to be more than the RCP 4.5 scenario. Moreover, the three parameters of minimum temperature, maximum temperature, and evapotranspiration for both scenarios have an almost constant trend during the study period and have a slight increase and decrease, but this is completely different in the case of precipitation, which increased and decreased intensively. The RCP 8.5 scenario estimated four parameters as more than the RCP 4.5 scenario. Then the changes of the four parameters of precipitation, minimum temperature, maximum temperature, and evapotranspiration were investigated for the climate change model and emission scenarios and the results showed the two scenarios RCP 4.5 and RCP 8.5 estimated almost the same results for the four parameters. And the RCP 8.5 scenario estimated four parameters as more than the RCP 4.5 scenario. By comparing the box diagram of climate change models for the four parameters, the results showed that the GFDLCM2, MPI, and IPSL models have more certainty about precipitation and the GISS E2 and GFDLCM2 models have more certainty about minimum temperature, maximum temperature, and transpiration evaporation. The lowest values of precipitation were predicted by the MIROC-ESM model and those of minimum temperature, maximum temperature, and evapotranspiration by the GISS E2 model.

All relevant data are included in the paper or its Supplementary Information.