Impact of projected climate variability on agricultural water demand of key crops in drought-prone central India region

Irrigation is a key component of agriculture to balance the food security of human beings. Nowadays, climate change has become a significant issue affecting the availability of future agricultural water and is responsible for the increased frequency of drought events. Major areas around the world are also dependent on rain-fed agriculture due to the changing climate. It is essential to accurately assess the impact of climate change on the future IWR and crop water requirement (CWR) of various crops in the drought-prone central India region. Some studies have also been performed to find the accurate behavior of climate change around the world. Various models represented the accuracy of local climate in earlier studies to see their effect on irrigation. Such as Wang et al. (2016), Rajendran et al. (2017), Gondim et al. (2018), Yahaya et al. (2018), Sahoo et al. (2018), Goodarzi et al. (2019), and Zhang et al. (2020) utilized the global climate models (GCMs) over the various study region around the world, where Bekele et al. (2019) and Durodola & Mourad (2020) assessed the future climate data from regional climate models (RCMs) or using both GCMs and RCMs to conclude the most effective model for projecting accurate IWR as well as CWR for various crops. Yahaya et al. (2018) determined the impacts of potential climate change on the CWR of Plantain in Ondo State, Nigeria, for periods 2050 and 2080s using the six downscaled GCMs. Most of the models concluded the increasing future CWR and IWR of Plantain based on the predicted climate for the periods 2050 and 2080s over the study region. Bekele et al. (2019) found the climate change effects on crop water demand (CWD) and surface water availability in the Birr watershed, Ethiopia. The author used the latest CORDEX model – Africa data output of Hadley Global Environment Model 2-Earth System (HadGEM2-ES) for representative concentration pathway (RCP) 2.6, 4.5 and 8.5 scenarios to evaluate future IWR for Maize crops. The results indicated the increasing rate of IWR for Maize in the study region.

Prior studies conducted for India provided useful information about the potential future agricultural water demand. Pramod et al. (2018) utilized the NorESM1-M model of the CMIP5 for the RCP 4.5 scenario over India to evaluate the IWR for wheat for two climatic periods (2021–2050 and 2051–2080). Results revealed that in the initial climate period, water requirement for wheat decreased over most of the districts in the country while increasing water needs for wheat-growing districts were found for later climate periods over the country. Behera et al. (2016) found the influence of climate change on future crop water requirements (CWR) of paddy, soybean and vegetables in the Sunei medium irrigation project, Odisha, India. Reference evapotranspiration (ETo) and CWR were evaluated with the help of CROPWAT 8.0. The study also predicted the future IWR for 2025, 2050 and 2080 using the GCM model HadCM3. Jaiswal et al. (2021) also evaluated the climate change impact on crop water requirements (CWR) of paddy crops over the Tandula command of Chhattisgarh, India. The authors assessed the average CWR of paddy for the historical period (1971–2014) and three future periods (2020–2035, 2046–2064 and 2081–2100). Compared to the present condition, the study concluded the maximum amount of additional water needed during 2046–2064 for paddy crops. Poonia et al. (2021) also utilized the four best GCMs, CNRM-CM5 (Centre National de Recherches Météorologiques- Coupled Model Phase 5), CCSM4 (Community Climate System Model, version 4), MPI-ESM-LR (Centre National de Recherches Météorologiques Coupled Global Climate Model, version 5) and ACCESS1-0 (Australian Community Climate and Earth-System Simulator version 1.0) to find the effect of climatic variation on irrigation water needs over the eastern Himalayas.

Earlier studies utilized the data from old climate models, which were not found effective for the central India region. Some investigations have been done with the improvements of GCM models to predict climate change and concluded the accuracy and applicability of various models for India (Sharmila et al. 2015; Panjwani et al. 2020). Certain studies have also revealed the performance of RCM models along with the GCMs for the Indian climate (Mishra et al. 2014, 2018; Choudhary et al. 2018). Some region-specific studies like Balvanshi & Tiwari (2019), and Sreelatha & Anand Raj (2022) have also been carried out over the regions of India. Vishwakarma et al. (2020) assessed the performance of the latest generation GCMs under the CMIP5 as well as RCMs under the CORDEX-SA over the drought-prone Bundelkhand region in central India. The study was performed for the individual as well as ensembles of both bias-corrected GCM and RCM models. The study found that the bias-corrected GCMs performed better than the CORDEX-South Asia RCMs over the central India region. It was also concluded that the bias-corrected GCM, ESM-2M outperformed the ensembles of best GCMs and RCMs. Hence, based on the bias-corrected GCM, ESM-2M data, it is possible to efficiently project the future CWR and IWR of various crops for the Bundelkhand region in central India, which would be further beneficial for the policymakers and the farmers. The current study focuses on assessing the CWR and IWR of major crops for future years in Bundelkhand. It utilizes bias-corrected data from the best general circulation model (GCM), ESM-2M to examine how water requirements may change over time and determine the additional water needed compared to the current scenario in the study region.

Food and Agriculture Organization (FAO) Irrigation and Drainage Paper No. 24 gives general lengths and the total growing period for the four distinct growth stages of several crops for varying climates and locations. Single (time-averaged) crop coefficients (K_c), mean maximum plant heights for non-stressed and rooting depth and depletion fraction for major crops grown over the Bundelkhand region are listed in Table 1. The information is intended for use with the FAO Penman–Monteith ETo on well-managed crops in sub-humid regions. The atmospheric evaporation capacity is utilized for determining the percentage p (depletion fraction), which signifies the critical soil moisture level where initial signs of drought manifest, influencing both crop evapotranspiration and yield. These values are expressed as a fraction of the total available water (TAW).

The current study compiles crop data covering four distinct growth stages and the total growing period, sourced from the FAO Irrigation and Drainage Paper No. 56, as depicted in Table 2.

The upper part of Bundelkhand, which comes under the UP state, is more affected by deficient rainfall (<1,000 mm) than its lower parts of MP state. The Black Clay soil is found majorly in most of the MP in Bundelkhand (especially in Sagar, Damoh and Chhatarpur districts). In contrast, this soil is located majorly in the Mahoba district of UP, Bundelkhand. Red sandy loam and sandy loam soil are dominantly sprayed over the remaining districts of Bundelkhand, especially in most of the parts of UP.

The daily gridded rainfall data have been gathered from the Indian Meteorological Department (IMD) for the 82 rain gauge stations under the 13 districts of the study region from 1982 to 2010 (https://www.imdpune.gov.in/Clim_Pred_LRF_New/Grided_Data_Download.html). Simultaneously, daily gridded data of maximum and minimum temperatures have been collected from NASA (https://power.larc.nasa.gov/data-access-viewer/) from 1982 to 2010. Each category of data is used for this study on a common grid 0.5° × 0.5° resolution. Subsequently, these data are bilinearly interpolated to point data, which is specific to the locations of 82 rain gauge stations in Bundelkhand. The climate data operator (CDO), a Linux tool, has been employed for bilinear interpolation of the gridded data. Table 3 shows the description of 82 rain gauge stations in the Bundelkhand region.

The future rainfall and temperature data have been utilized from bias-corrected GCM, ESM-2M. This bias-corrected CMIP5 GCM was found suitable out of five CORDEX-RCMs and their driving latest generation CMIP5 GCMs for the Bundelkhand region in the recent study of Vishwakarma et al. (2020, https://doi.org/10.2166/wcc.2020.177). The description of the model is shown in Table 4.

IWR and CWR have been evaluated using the CROPWAT tool developed by FAO, Italy (Smith 1992). The standard methods involved to assess the IWR, CWR and other factors are detailed as follows,

The current research employed a fixed K_c value to estimate ET, as well as IWR and CWR. However, it's important to note that climate change can significantly alter the crop growth cycle. Maintaining a constant K_c throughout the future neglects the potential impacts of climate change on crop development stages. Future studies may benefit from considering these factors.

Similarly, the steepness of the slope of the projected minimum temperature for RCP 8.5 (y = 0.0469x + 18.549) has been found to be more than the RCP 4.5 scenario (y = 0.0127x + 20.304), as shown in Figure 4. Hence, it can be concluded that the maximum temperature will rise in the coming years while considering both scenarios. The rate of temperature rise is higher for RCP 8.5 compared to RCP 4.5.

Two major grown rabi crops, wheat and gram, have been chosen to evaluate their district-wise IWR and CWR over the Bundelkhand region. Rabi crops are typically grown from October to March, whereas vegetables are cultivated from February to June across the study region. This temporal pattern is observed and utilized within the study. Monthly averaged historical data of rainfall and potential evapotranspiration (PET) from 1982 to 2010 have been used for the study. PET has been evaluated using the FAO-recommended Penman–Monteith equation. IWR and CWR have been assessed using the CROPWAT 8.0 tool for the expected date of sowing of both the crops and for major soil groups founded in the districts of Bundelkhand. IWR and CWR have been evaluated using the soil and rainfall data and crop data from FAO Irrigation and Drainage Paper No. 56 (Table 1). The results of historical observations for rabi crops are shown in Table 5.

The historical CWR and IWR (1982–2010) of vegetables have also been evaluated for all the districts of Bundelkhand. The vegetable class (cucurbits, okra, cowpea, bottle gourd and sponge gourd) has been chosen as hot weather crops to see the effect of climate change on them. The results of the IWR and CWR of vegetables are shown in Table 6.

Historical observations tabulated in Tables 5 and 6 have been utilized to see the effect of climate change on future water needs of major crops over the study region.

Future water needs for major crops have been evaluated for three time periods as 2011–2040 (initial century), 2041–2070 (mid-century) and 2071–2100 (end century) to compare the future data with observed historical data (1982–2010) for the same date of sowing. The bias-corrected future data of meteorological variables were chosen for two scenarios RCP 4.5 and RCP 8.5. Thus, the IWR and CWR have been evaluated for both scenarios.

Besides the wheat crop, the climate change effect is not shown much on the gram crop in the study region. The additional water need found less for most of the study region for both the scenarios compared to the historical condition can be seen for both the scenarios, RCP 4.5 and RCP 8.5.

Overall, the climate change effect on vegetables in the future has been observed significantly less and could be grown with suitable water management strategies in the hot weather season over the Bundelkhand region.

In summary, the present study indicates that in the future, more water may need to be allocated for major rabi crops like wheat and gram, as well as for hot weather vegetables, due to the impact of climate change in Bundelkhand. Prior research has already assessed future water needs for crops globally, considering both crop-specific requirements and irrigation demands and has highlighted the influence of climate variability on these needs. Many studies have projected an increase in water demand for various crops in the coming years using multiple climate models tailored to regional accuracy. However, there is a notable gap in research regarding the effect of climate change on agriculture in Bundelkhand, a region severely affected by droughts linked to climate change. Therefore, the findings of the present study hold significance for both farmers and policymakers in Bundelkhand.

The effect of climate change has been observed on the future water requirements of major rabi and hot weather crops grown in the Bundelkhand region by utilizing the bias-corrected best-suited CMIP5 GCM (ESM-2M) data and compared to the historical observations. The study concluded the increasing trend of future water requirements had been found for most of the major rabi and hot weather crops grown in the Bundelkhand region. Some major crops were also found suitable to grow in the future without the arrangement of additional irrigation water. Those were gram under the rabi crop and vegetable, also called climate-secure crops. Under the category of rabi crops, additional water demand has been found more of ±27 mm in the future for wheat crops especially at the end of the 21st century (2071–2100), while the gram crop was found secure against the surplus water need in future over the Bundelkhand region.

The study also indicated a decline in future water availability for agriculture, attributed to a reduction in projected rainfall, as depicted in Figure 2. Therefore, essential steps are needed to save them. It is also mandatory to propose proper water management and irrigation practices against the future water crisis for wheat crops. This study will also help researchers and policymakers utilize the projected data of IWR and CWR for proper water management studies.

The results of projected crop and irrigation water requirements of various major rabi and hot weather crops can be utilized to propose the best planting period/dates of crops so that optimum yield can be achieved mainly in case of drought. It would also be advantageous to plan an optimum cropping pattern in future drought years. This study opens the door to utilizing the projected data to find the severity of drought and drought years (monsoon, non-monsoon and annual drought). The future drought projection will also help to see its effect on agriculture in drought-prone regions. The study also helps to choose or opt for the best climate-secure crops in case of future water scarcity.

The FAO-56-based crop coefficient approach is deemed suitable for projecting future CWD. However, it's important to recognize that climate change is causing a reduction in the duration of the crop-growing cycle. Consequently, the reliance on fixed growing periods and constant K_c values within the approach fails to consider the impact of climate change on the crop cycle. Moreover, the influence of atmospheric carbon dioxide on crop evapotranspiration (ET) remains unaccounted for in the current methodology, offering room for further enhancement.

The authors thank the IMD and the National Aeronautics and Space Administration (NASA) for providing meteorological data to carry out this research work.

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

The authors declare there is no conflict.