Assessment of the effect of climate change on the yields and water footprint of crops in arid area

Given Iran's arid and semi-arid climate, water is a crucial resource and a major limiting factor for agricultural production. It is crucial to understand the water requirements for agricultural systems to ensure their success. This knowledge is essential for optimizing crop growth and yield. The amount of water needed and the water footprint of crop yield vary depending on the specific region and its climate. Furthermore, individual water usage patterns, agricultural practices, and water use efficiency in the region all contribute to this variation. As a result, it is essential to develop a framework that allows for an accurate estimation of water demands for each crop yield. For this purpose, the water footprint is used as a comprehensive index that shows the actual amount of water used by each crop in the situation and climate of each region. Hoekstra & Hung (2002) first introduced this index, and in recent years, it has attracted the attention of experts from different parts of the world. The water footprint of each crop includes three colored components: blue water, green water, and gray water. Bluewater refers to the water of rivers, lakes, groundwater sources, etc., which is used for irrigation. Green water is rainwater that is temporarily stored in saturated soil, on the soil surface, or coverage herbal. Gray water and its traces also indicate the volume of polluted water. Calculating the water footprint is used as a suitable method to check the water use of crops and measure the efficiency of agricultural water and crop production (Morillo et al. 2015). In recent years, this index has been widely assessed by researchers and experts around the world. The footprint of various crops has been calculated, including tea (Jefferies et al. 2012), rice (Chapagain & Hoekstra 2011), wheat (Hoekstra & Chapagain 2007; Aligholinya et al. 2019), cotton (Chico et al. 2013), grapes (Ene et al. 2013), corn (Nana et al. 2014), and potato (Herath et al. 2014; Rodriguez et al. 2015). In their 2006 study, Chapagain et al. (2006) investigated the water use footprint of cotton production worldwide. They categorized water usage into three types: green water (evaporation of infiltrated rainwater to cotton growth), blue water (surface water or groundwater used for irrigation), and water contamination during the cotton growth period. The study aims to assess the impact of cotton production on water resources and identify areas for improvement. Their study found that cotton requires 256 Gm³ (a billion cubic meters) of water annually, with 42% coming from blue water, 39% from green water, and 19% from gray water. In their research, Ercin & Hoekstra (2014) examined various factors such as population growth, economic growth, production-business patterns, and technology development on a global scale to predict the water footprint scenario by 2050. Their study results showed that if the pattern of water use changes, the water footprint will also reduce despite the increase in population. Masud et al. (2018) simulated the yields and water consumption of the crop by using the soil and water evaluation tool along with the geographic information system (SWAT-GIS) model. Their findings showed that due to increased performance and reduced water consumption, the water footprint is also reduced. Fulton et al. (2019) conducted a study on the water footprint and economic value of almond production in 12 regions of California. Their findings revealed that the average water footprint for almonds was 10.24 m³/kg, while the economic value was 0.42 $/m³. A comparison of these results with 43 other crops in California revealed that almonds have the highest economic value among nuts and berries. In a study conducted by Bazrafshan et al. (2019), the virtual water and water footprint of saffron production in Iran were examined. The study revealed that the average water footprint of saffron production in Iran is 4,659 m³/kg, with green, blue, white, and gray water footprints accounting for 12, 42, 40, and 6%, respectively. Also, the total water footprint of saffron production was 1,541 million m³ per year, the share of exported virtual water is 1,354.6 million m³ per year, and the average economic water footprint of saffron production is 1.3 $/m³. Rahimipour Anaraki et al. (2020) in a study assessed the virtual water and water footprint of crops in Qalaganj. The results showed that the most cultivated area was related to cereals and vegetables, which covered more than half of the 48,000 hectares of cultivated area and more than 60% of the total water footprint, and their yields were low. Arunrat et al. (2022) addressed the effect of climate change on the water footprint and yields of major crops in Thailand. This study was conducted over three time periods: short term, medium term, and long term, and examined two different scenarios. The results revealed that both scenarios showed an increase in minimum and maximum temperatures and rainfall across all three time periods. In addition, the researchers predicted that rice crop yield will increase in all three periods under the SSP2.4.5 scenario and will also increase in the middle period under the SSP5.8.5 scenario. On the other hand, drought, climate change, and incorrect management policies have severely affected existing water sources and caused a negative balance in some parts of the world. This is in direct contradiction to the principles of sustainable water usage. The impact of climate change is felt across various sectors, including agriculture and water sources. As such, it is crucial to accurately predict these effects to minimize vulnerability and effectively adapt to them. Several studies have examined the changes in water footprints under the conditions of climate change. Pour Salehi et al. (2015) in a study calculated the virtual water value of eight crops in Birjand Plain and investigated the optimal cultivation pattern using the LINGO linear programming model. Their study results showed that in the study area, compared to other crops, saffron cultivation with an area equal to 2010 hectares under the first scenario and 12,188.65 hectares under the second scenario had a higher virtual water value and the cultivated area. Elbeltagi et al. (2020) in a study investigated the effect of climate change under RCP2.6, RCP4.5, and RCP8.5 scenarios on water footprint of wheat and corn in the Nile Delta in Egypt using a deep neural network model. The results showed that the coefficient of determination (R²) of evapotranspiration for two crops varies from 0.92 to 0.97, and the prediction values of wheat and corn yields were in the range of −3.21 to 3.47% and −4.93 to 5.88%, respectively. Also, by reducing precipitation in the future, under three scenarios, the green water footprint is reduced. Kouzegaran et al. (2020) in a study predicted the effect of climate change on saffron yields. The results of predicting extreme climatic events in the future showed that the increase in minimum and maximum temperature and the decreasing trend in the precipitation parameter are among the factors reducing the yields of saffron. As a result, yield reduction during future periods under the RCP8.5 scenario was higher than the same periods under the RCP4.5 scenario. Mali et al. (2021) in a study assessed the effect of climate change on the water footprint of cereals in India. To assess the effect of climate change on water footprint, RCP4.5 and RCP6 scenarios were used by the hybrid-delta model. The results showed that the WF of cereals will change in the range of −2.3 to 6.3% under two scenarios. Govere et al. (2022) in that study assessed the effects of climate change under RCP4.5 and RCP8.5 scenarios on wheat yields, crop water use, and wheat water footprint in Zimbabwe using the AquaCrop model. The results showed that water footprint reduced under the scenarios in the future and wheat yields increased in the future, while water use reduced due to climate change. Pilevneli et al. (2023) investigated the effect of climate change on agricultural production and income in 25 river basins in Turkey. The results of this research showed that as a result of the increase in the need for irrigation water in the future, the risk of water shortage will decrease from an excess of 14.6 km³ per year under the RCP8.5 scenario to 3.57 km³. Also, the results indicate that the performance and income value of crop yield in this basin are associated with a 100% decrease. The review of the aforementioned research indicates the existence of increasing tension on freshwater sources in different parts of the world along with the increase in demand for agricultural crop yield and climate change. Therefore, the importance of a new attitude and the use of comprehensive criteria such as water footprint to determine the actual amount of water used by crops is necessary for planning and optimal management of water use in the agriculture sector. The main purpose of this research is to investigate the water footprint of barberry and jujube crop yield and to investigate the amount of water consumed by these two main crop yields of the agricultural sector in Birjand Plain. Also, in this research, the water footprint and amount of water consumed by these crop yields in the future will be calculated under three scenarios of the sixth climate change report SSP1-2.6, SSP2-4.5, and SSP5-8.5 during the statistical period of 2022–2038.

Birjand plain was assessed during the basic period and climate change using the CropWat 8.0 model to calculate water requirement and evapotranspiration of selected crops in this study (barberry and jujube). Also, to investigate the yields of crops in the future under the influence of climate change, the nonlinear input–output (NIO) time series model was used. Meteorological data including precipitation, minimum temperature, maximum temperature, sunny hours, and relative humidity were prepared from the Birjand synoptic station and sorted for the basic statistical period (2005–2021). The yield data were prepared from the site of the Agricultural Jihad Organization of South Khorasan Province.

The main objectives of CMIP6 models are to assess organized models, investigate the responses of the earth's structure to various forces regarding the origin and quantification of climate changes, and assess the uncertainty of scenarios. The reason for the improvement of CMIP6 models compared to CMIP5 models is the improvement of the number of vertical layers, which have a more accurate perspective in the stratosphere, and in these models, the number of scenarios for future periods has been considerably increased (Gupta et al. 2020; Su et al. 2021). In this study, first, the minimum temperature, maximum temperature, and precipitation data for Birjand Station in the future daily time series were downloaded by the MIROC-ES2L model under three scenarios SSP1-2.6, SSP2-4.5 and SSP5-8.5 from the site https://cds.climate.copernicus.eu/cdsapp#!/dataset/projections-cmip6?tab=form, and the specifications are presented in Table 1. Given that to assess the predicted global climates in hydrological systems, different methods and exponential downscale models are used, and in this study, the climate model data for hydrologic modeling (CMhyd) were used for the parameters of minimum and maximum temperature and precipitation. This model has eight different methods of downscaling for temperature and precipitation data. Some of the methods available in this software are not suitable for the exponential downscaling of precipitation or temperature, so the downscaling method should be selected according to the objective. This model includes three data inputs, which include observation data, data from the past period of the climate model (historical), and scenario data (future) of the climate model. In this model, the data can be used in two formats, NetCDF and text or ASCII, which downscale the data at the station level. In this study, the exponential downscaling of data was done in ASCII format daily, and the linear scaling method was also used for exponential downscaling. The historical data are from 1990 to 2014, and the simulated output of meteorological parameters for the future under three scenarios SSP1-2.6, SSP2-4.5, and SSP5-8.5 is considered during 2022–2050.

We considered minimum, maximum, and average temperatures and precipitation for input. We used crop yield data for output (target) data. The NIO time series model shows the future yields of barberry and jujube under three scenarios: SSP1-2.6, SSP2-4.5, and SSP5-8.5. The NIO model and MATLAB performed the calculations related to this section.

Since Cmhyd does not simulate the humidity values for future periods and the relative humidity is one of the inputs required by CropWat 8.0 to calculate evapotranspiration, the Minitab model was used to estimate the relative humidity in the future period. Then, using the linear regression relationship obtained between the minimum temperature, maximum temperature and average temperature, and relative humidity, the relative humidity in the future under the influence of three scenarios SSP1-2.6, SSP2-4.5, and SSP5-8.5 was calculated and considered an input to the CropWat 8.0 model for the future period.

The results include two parts. The first part includes the results related to the prediction of meteorological parameters by the CMhyd model and then compared with the data of the base period, and in the second part, the results of the water footprint during the two future and base periods are presented and assessed.

For precipitation, maximum temperature, and minimum temperature in the future under the three scenarios SSP1-2.6, SSP2-4.5, and, SSP5-8.5 compared to the base period, the change in these parameters has been calculated based on the total and annual average during the future period compared to the base period, the results of which are given in Table 3. As shown, the average rain is rising in the future (2022–2050) compared to the base (1990–2014) under SSP1-2.6 and SSP2-4.5. But, it decreased under the SSP5-8.5. The highest increase in precipitation under the SSP2-4.5 scenario is 144.85 mm per year. It can also be seen that the amount of precipitation in the future under three scenarios varies from −13.08 to 5.13 mm per year. The maximum temperature and the minimum temperature are increasing in the future compared to the base period under all three scenarios, and the highest increase in the maximum temperature compared to the base period under the SSP5-8.5 scenario is 26.22 °C on average during 2022–2050. Also, it is observed that the maximum temperature changes vary from 1.07 to 1.25 °C. The minimum temperature values in the future compared to the base period under the SSP5-8.5 scenario had the highest increase by 10.16 °C, and its changes ranged from 1.15 to 1.54 °C under all three scenarios compared to the base period.

During the recorded statistical period (1990–2021), wind speed data have been available, but climate change models cannot predict this parameter for future periods. Therefore, it is necessary to consider the same values of wind speed during the base and future periods when calculating changes in evapotranspiration. This ensures that any increase or reduction in evapotranspiration is not attributed to changes in wind speed. Furthermore, similar considerations were made for sunny hours in the future period in comparison to the base period (Montaseri et al., 2016).

The study calculated the future values of relative humidity under three scenarios using the Minitab model. This was done by considering the linear regression relationship between meteorological parameters and relative humidity during the base period. Furthermore, the study determined the green water footprint, blue water footprint, and total water footprint of barberry and jujube by incorporating the values of water requirement and evapotranspiration potential of crops in the future under the three scenarios. The volume of water use per unit weight of the crop (m³ per ton) was then calculated using Equations (4)–(7). The results of these calculations are provided in Table 4. The results of Table 4 show that the values of water requirement, blue water footprint, green water footprint, total water footprint, and finally the relative values of both crops have increased during the future period (2022–2038) compared to the base period (2005–2021) due to climate change under the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios. The water requirements for barberry and jujube have increased from 468.90 to 950.81 mm per year and from 483.58 to 992.40 mm per year, respectively. Also, the water footprint results in Table 4 show that out of the total footprint of 3,784.1 m³/ton in the base period of the barberry crop, 96.36% of it includes a blue water footprint and 3.63% green water footprint. In the future period, under the SSP1-2.6 scenario, out of the total footprint of barberry 7,392.9 m³/ton, 98.98% is blue water footprint and 1.1% is green water footprint. The total water footprint of the barberry crop yield under the SSP2-4.5 scenario was calculated as 8,033 m³/ton, of which 98.67% includes the blue water footprint and 1.32% green water footprint of this crop yield. In the SSP5-8.5 scenario, the total footprint is 7,691.7 m³/ton, of which 98.68 and 1.31% are blue water footprint and green water footprint, respectively. The amount of water footprint in the Jujube crop yield is 2,244 m³/ton in the base period, of which 96.46% is blue water footprint and 3.53% is green water footprint. In the SSP1-2.6 scenario, the total footprint of jujube is calculated to be 4,525.8 m³/ton, and its blue water footprint and green water footprint are 99 and 1.06%, respectively. The total footprint of jujube in the SSP2-4.5 scenario is 4,372.1 m³/ton, of which 98.71% includes blue water footprint and 1.28% includes green water footprint. Finally, it can be seen that the total water footprint of the jujube crop yield under the SSP5-8.5 scenario is calculated to be 4,465.6 m³/ton, of which 98.71% is the blue water footprint and 1.27% is the green water footprint. The results of water footprint calculation in two crop yield barberry and jujube under the influence of climate change and three scenarios of Shared Socioeconomic Pathways (SSPs) and increasing temperature and increasing water demand indicate that the amount of total water footprint has increased. Most types of water footprints include blue water footprints. The table results indicate a high dependence of two crop yields, barberry and jujube, on surface and groundwater resources, as seen in the green water footprint and total footprint. Therefore, according to the study results of water requirement and water footprint in the future, it can be concluded that climate change in the future in Birjand Plain will lead to the instability of water and its higher use, which can be attributed to the excessive use of surface water and groundwater sources due to the increase in water use (water footprint) that is caused by changes in water requirement of crops in the study area. The results of this section are similar to the study results of Montaseri et al. (2016). Therefore, the reduction in surface water and groundwater sources affected by climate change leads to the intensification of water stress in the Birjand Plain. For this reason, one of the most important solutions is to reduce and balance water use according to the potential of water sources in the region. In this context, the factors that reduce water footprint in the study area include low irrigation methods, reducing cultivated area, changing the agricultural calendar, and changing the crop cultivation pattern.

The main factors influencing water demand in the agricultural sector are climate changes, which have a significant impact on crop yield and global food security. It is crucial to evaluate the effects of climate change on the green and blue water footprints of crop yield to make informed decisions about climate adaptation strategies and mitigate its consequences. In this study, we aimed to investigate the impact of climate change on the water footprint in Birjand Plain. We extracted meteorological data, including minimum and maximum temperatures and precipitation, from the sixth climate change report using the MIROC-ES2L model. These data were then downscaled at the station level using the Cmhyd exponential downscale model and the linear scaling method under three scenarios: SSP1-2.6, SSP2-4.5, and SSP5-8.5 for the future period (2022–2050). We then compared the simulated meteorological parameters with the base period and found that the model accurately predicted the values of these parameters for the future period.

The results of the study investigating minimum temperature, maximum temperature, and precipitation indicate that precipitation is predicted to increase in the SSP1-2.6 and SSP2-4.5 scenarios, while it is expected to decrease in the SSP5-8.5 scenario. The changes in precipitation range from −13.08 to 5.13 mm per year in the future under all three scenarios. In addition, the study found that both minimum and maximum temperatures are projected to increase under all three scenarios. Specifically, the maximum temperature is expected to rise by 1.07–1.25 °C, and the minimum temperature is predicted to increase by 1.15–1.54 °C compared to the base period.

The evapotranspiration and water requirement values for the years 2022–2038, which will be affected by climate change, were calculated using the CropWat 8.0 model. The green and blue water footprint, total footprint, and yields of barberry and jujube, which are the most widely cultivated crops in the Birjand plain, were then calculated using an input–output nonlinear time series model. The results of the yield prediction for the future showed that barberry yields are expected to increase in the SSP1-2.6 scenario but decrease in the other two scenarios. On the other hand, jujube yields are predicted to increase in the SSP2-4.5 and SSP5-8.5 scenarios but decrease in the SSP1-2.6 scenario.

The results of calculating the water consumption for two crop yields, barberry and jujube, indicate that the ratio of blue water to green water in barberry crops has increased from 45.34 in the base period to 121.38 in the future period. The largest increase in this ratio was observed under the SSP1-2.6 scenario. Similarly, the ratio of blue water to green water in jujube crops has also increased from 46.69 in the base period to 125.89 in the future period. The highest ratio of blue water to green water for jujube crops was found under the SSP1-2.6 scenario.

Based on the investigation of water requirements and water footprint in the future, it can be concluded that climate change in the Birjand Plain will result in water instability and increased consumption. This is due to the excessive use of surface and groundwater resources, driven by changes in water demand for agricultural crop yield in the area. To address this issue, it is necessary to propose and implement methods such as reducing the area under cultivation, implementing less irrigation, changing cultivation patterns, and adjusting the agricultural calendar. In addition, improving irrigation systems and replacing traditional methods with modern ones can increase irrigation efficiency and reduce the water footprint of crops in the Birjand Plain. As climate change is inevitable, it is crucial to focus on efficient irrigation and optimal water management to decrease the water footprint of crops.

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.