Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Moist heat stress extremes in India enhanced by irrigation

Abstract

Intensive irrigation in India has been demonstrated to decrease surface temperature, but the influence of irrigation on humidity and extreme moist heat stress is not well understood. Here we analysed a combination of in situ and satellite-based datasets and conducted meteorological model simulations to show that irrigation modulates extreme moist heat. We found that intensive irrigation in the region cools the land surface by 1 °C and the air by 0.5 °C. However, the decreased sensible heat flux due to irrigation reduces the planetary boundary layer height, which increases low-level moist enthalpy. Thus, irrigation increases the specific and relative humidity, which raises the moist heat stress metrics. Intense irrigation over the region results in increased moist heat stress in India, Pakistan, and parts of Afghanistan—affecting about 37–46 million people in South Asia—despite a cooler land surface. We suggest that heat stress projections in India and other regions dominated by semi-arid and monsoon climates that do not include the role of irrigation overestimate the benefits of irrigation on dry heat stress and underestimate the risks.

Main

India has been a focal point for the massive impacts of severe heat stress1,2,3, which are projected to worsen in the future4,5,6,7 due to the combined effects of rising temperature and population3. However, heat stress is mediated by moisture, and it is well-established that moist heat stress metrics are relevant for health, safety and productivity risk characterizations8,9. Many different moist heat stress metrics are in use, and they weight moisture in different ways and have different histories and uses10,11. Consequently, to develop an understanding of and predict heat stress and its human impacts requires a better grasp of the controls on lower atmospheric moisture and specifically on the interactions between atmospheric dynamic and thermodynamics and the direct anthropogenic effects from land-use change (irrigation) decisions12.

Understanding the possible impacts of irrigation in India is critical because a massive irrigation expansion took place there after the ‘green revolution’ and the net irrigated area increased from 31 to 60 million hectares between 1970 and 200713. The irrigated area is likely to expand in the future14. Widespread and intensive irrigation in India15 has been shown to alter near-surface temperature16,17 and humidity18,19, in keeping with the responses to irrigation seen in global studies20,21. However, the role of irrigation and climate warming on dry and moist heat stress in India and the associated physical mechanism remains relatively less explored1,2,3. In this study, we investigated the role of irrigation in modulating both dry and moist heat stress11 using in situ observations, reanalysis and model simulations that utilized the high-resolution Weather Research and Forecasting (WRF) model driven by reanalysis (European Centre for Medium-Range Weather Forecast Reanalysis v5 (ERA5)) with and without irrigation scenarios in India. We show that irrigation can reduce dry heat stress, which is an uncontroversial result, but it can also increase moist heat stress, which has far-reaching implications.

Irrigation and surface temperature

The Indo-Gangetic Plain is one of the most intensively irrigated regions in the world with more than 60–70% of the area irrigated with surface and groundwater22 (Fig. 1a,c). Both gross irrigated area (total area under crops irrigated once or more than once in a year) and net irrigated area (actual irrigated land) increased considerably after 1951 in India (Fig. 1c). Satellite data clearly show that irrigation decreases land surface temperature (TLS) in the region during the observational period of 1982–2015. The extensively irrigated areas experienced a cooling (0.5–1.5 °C) in TLS (Fig. 1b,d) and irrigation-induced cooling is more dominant in the premonsoon (February–May) and postmonsoon (October–December) seasons (Supplementary Fig. 1). Regions with less than 10% of the area under irrigation did not experience cooling (Fig. 1d and Supplementary Table 1). Moreover, non-irrigated regions (forest and water) did not experience cooling of more than 0.2 °C (Fig. 1d and Supplementary Table 1). Investigation of the trends in soil moisture, TLS, and vegetation metrics further supports the role of irrigation in land surface cooling in India (Supplementary Figs. 2 and 3 and Supplementary Table 2).

Fig. 1: Irrigation driven cooling in the Indo-Gangetic Plain.
figure1

a, Irrigated area (%) in India (data from the Food and Agriculture Organization). b, Changes in mean annual TLS (°C) from the National Oceanic and Atmospheric Administration’s AVHRR (Advanced Very High Resolution Radiometer) for the period 1982–2015. c, Increase in gross and net irrigated area during 1951–2015 in India. d, Changes in TLS for irrigated and non-irrigated regions in the Indo-Gangetic Plain. The area in the region outlined in red in a shows the highly irrigated region of the Indo-Gangetic Plain. The changes in b and d were estimated using a non-parametric Mann–Kendall test and Sen’s slope method.

Source data

Changes in dry and moist heat

Most of the heat-related mortality occurs during the summer season2,3,23. We analysed the changes in mean and extreme dry and moist heat stress metrics during the summer (April–May) in India over the period 1979–2018 using ERA5 reanalysis. The mean daily maximum air temperature at 2 m (a dry heat metric, T2) during the summer reaches above 40 °C in the central and western parts of India (Supplementary Fig. 4 and Supplementary Table 3), which are substantially hotter than that experienced in the Indo-Gangetic Plain region. In response to the observed warming over 1979–2018, the mean daily maximum temperature in the summer has increased by about 1 °C, predominantly in central, western and northern parts of India3 (Supplementary Fig. 4d). However, a moderate cooling in the summer daily maximum temperature was observed over the Gangetic Plain (Supplementary Fig. 4d). The increase (0.66 °C, P > 0.05; P < 0.05 means statistically significant at the 5% level) in the all-India averaged mean daily maximum temperature is higher than that of the Indo-Gangetic Plain region (0.40 °C, P > 0.05) (Supplementary Fig. 4g).

The three-day maximum temperature during the summer in central and western India exceeds 45 °C, which creates extreme dry heat conditions that can result in high mortality2,23 (Fig. 2 and Supplementary Table 3). However, extreme dry heat is less intense over the Indo-Gangetic Plain in comparison with that in central and western India (Fig. 2 and Supplementary Table 3). Over the past 40 years of observational records, the three-day maximum temperature has increased in central, western, and northern parts of India along with cooling over the Indo-Gangetic Plain (Fig. 2d). The all-India averaged three-day summer maximum temperature has increased by 0.65 °C (P > 0.05), whereas the Indo-Gangetic Plain experienced a much smaller rise of 0.26 °C (P > 0.05) during 1979–2018 (Fig. 2g). This localized cooling in extreme dry heat centred over the region is attributable to intensive irrigation24,25 and consistent with the satellite observations (Fig. 1b,d).

Fig. 2: Changes in three-day maximum heat indicators in India during the summer (April–May) for the 1979–2018 period.
figure2

ac, Mean three-day maximum T2 (°C) (a), HI (°C) (b) and TW (°C) (c) for the summer during 1979–2018. df, Changes in the three-day maximum T2 (°C) (d), HI (°C) (e) and TW (°C) (f) for the summer between 1979 and 2018. gi, All-India (brown) and Indo-Gangetic Plain (green) averaged three-day maximum T2 (°C) (g), HI (°C) (h) and TW (°C) (i) during the summer. Changes in df and gi were estimated using a non-parametric Mann–Kendall test and Sen’s slope method. P < 0.05 shows a statistically significant change at the 5% significance level.

Source data

Summertime heat index (HI) and wet-bulb temperature (TW), the metrics for moist heat stress, are consistently higher in the coastal and Indo-Gangetic Plain regions (Supplementary Fig. 4b,c and Supplementary Table 3). Over the period 1979–2018, the mean daily moist heat increased significantly (P < 0.05) (Supplementary Fig. 4e,f). Moreover, the increase in the mean daily moist heat in the Indo-Gangetic Plain was slightly higher and statistically significant (P < 0.05) compared with the rest part of India (Supplementary Fig. 4h,i). The estimated three-day maximum HI and the corresponding TW taken as measures of extreme moist heat during the summer exceed the danger levels, that is, above 45 °C for HI and 29 °C for TW with the latter being noticeable in the eastern coastal regions and parts of the Indo-Gangetic Plain (Fig. 2b,c).

We observed a significant increasing trend (P < 0.05) in the extreme moist heat measures across India over the period 1979–2018 (Fig. 2 and Supplementary Table 3). The extreme moist heat increased at a faster rate over the Indo-Gangetic Plain in comparison with that of the all-India average (Fig. 2h,i). Although in situ observations of TW are limited over India, changes in the three-day maximum T2 and TW from the ERA5 dataset are in agreement with the in situ observations over the Indo-Gangetic Plain (Supplementary Fig. 5d,e and Supplementary Table 4). A decline in dry heat but an increase in moist heat over the Indo-Gangetic Plains was observed during 1979–2018 (Supplementary Fig. 4 and Fig. 2).

We analysed the relationship between the area affected by extreme dry and moist heat and human mortality (Supplementary Fig. 6). The geographical area under a three-day maximum TW (which exceeded 27 °C) and three-day maximum HI (which exceeded 45 °C) in India during the summer is more strongly (correlation = 0.44 and 0.45, respectively, P < 0.1, statistical significance tested at the 10% level) related to human mortality than that based on the three-day maximum T2 (correlation = 0.30, P > 0.1), the latter being a measure of dry heat (Supplementary Fig. 6)9. Thus, moist heat stress can be a more useful indicator to understand the linkage between heat stress and human mortality in India2. The Indo-Gangetic Plain and coastal regions frequently experience a three-day maximum TW > 27 °C and a three-day maximum HI > 45 °C (Supplementary Fig. 7), which highlights its predilection for a high moist heat stress. The area affected by the three-day maximum TW and HI crossing the respective danger levels significantly (P < 0.001) increased during 1979–2018 (Supplementary Fig. 7), which signifies the rising moist heat stress under a warming climate4,12. However, the relationship between the area affected by dry and moist heat and human mortality is based on limited observations (Supplementary Fig. 6).

The decline in dry heat and increase in moist heat over the Indo-Gangetic Plain can be attributed to the combined impact of the large-scale influence of climate warming in the region and the localized effect of irrigation12,15,19,20. The increase in mean and extreme moist heat in the summer across India indicates the potential role of anthropogenic warming4,26,27. To further diagnose this aspect, we estimated the changes in specific and relative humidity over 1979–2018 (Supplementary Fig. 8). The mean specific humidity during the summer increased significantly (P < 0.05) across India and more predominantly so over the Indo-Gangetic Plain and central India (Supplementary Fig. 8a). Moreover, the three-day maximum specific humidity also increased over the majority of India (Supplementary Fig. 8b). Such changes are roughly in agreement with the prediction from observations theory28,29. It is more surprising that the mean and extreme relative humidity increased during the summer over the majority of India, but more predominantly so over the Indo-Gangetic Plain (Supplementary Fig. 8c,d). This large increase in relative humidity is important to interpret the moist heat stress trends and was therefore investigated in greater depth using modelling.

Modelling of irrigation influence on moist heat stress

Having empirically demonstrated a plausible role for irrigation in increasing moist heat stress from observations, we employed modelling to provide causal attribution. We conducted simulations at a 0.25° spatial resolution using WRF coupled with Noah land surface scheme for irrigation ‘on’ and ‘off’ scenarios using ERA5 data as the boundary conditions. Specific humidity and maximum temperature, the two main factors in moist heat, are well simulated by the WRF at the all-India level and over the Indo-Gangetic Plain (Supplementary Fig. 9). Also, WRF simulations show a consistent performance for the planetary boundary layer (PBL) height against the observed and reanalysis datasets (ERA5 and MERRA (Modern-Era Retrospective Analysis for Research and Applications)) for all-India and the Indo-Gangetic Plain region (Supplementary Fig. 10). In addition, the WRF-simulated PBL height shows a satisfactory comparison against radiosonde observations during the summer (Supplementary Fig. 11).

We calculated the differences in the summer energy budget, temperature, specific humidity, relative humidity, sea level pressure (SLP), and PBL height between irrigation-on and irrigation-off simulations (Fig. 3 and Supplementary Table 5). The changes in the vertically integrated moisture flux (\({\bf{UQ}}\) and \({\bf{VQ}}\), where U and V represent the wind direction components of moisture flux Q and the bold font indicates a vector quantity) were also analysed. Irrigation causes an increase in the latent heat flux and a nearly compensating decline in the sensible heat flux during the summer in India (Fig. 3a–c). The influence of irrigation is consistent across India (Fig. 1a); however, it is more dominant in the Indo-Gangetic Plain (Fig. 3 and Supplementary Table 5). Increased latent heat and evaporation lead to increased evaporative cooling24, which, in turn, reduces TLS and T2 (Fig. 3 and Supplementary Table 5). Cooling in TLS due to irrigation is higher than that for T2, which is commensurate with the reduced sensible heat flux (Fig. 3d,e). Irrigation results in an increased SLP30, as well as high specific and relative humidities and a reduced PBL height (Fig. 3 and Supplementary Table 5).

Fig. 3: The role of irrigation on summer heat fluxes, temperature, humidity, SLP and PBL height.
figure3

Difference between the irrigation-on and irrigation-off (irrigation – no-irrigation) scenarios during summer for 2000–2018 for latent heat (W m–2) (a), sensible heat flux (W m–2) (b), the sum of latent and sensible heat fluxes (W m–2) (c), TLS (°C) (d), T2 (°C) (e), specific humidity (kg kg–1) (f), SLP (hPa) (g) and PBL height (m) (h). The vectors UQ and VQ in g show the difference (irrigation – no irrigation) in integrated UQ and VQ (from the surface to 850 hPa) for the irrigation-on and irrigation-off scenarios. The role of irrigation was estimated with WRF simulations for the 2000–2018 period using ERA5 as the boundary conditions.

Source data

Surface cooling due to enhanced evapotranspiration might increase relative humidity, directly through cooling of the air, but without being associated with increased specific humidity. If relative humidity was approximately constant, as one might expect in PBLs, the resulting specific humidity may decrease. Thus, to explain the large increase in specific humidity, specifically, requires either an increase in the absorbed solar radiation (the energy input) or a decrease in the depth of the layer over which the evaporative flux is distributed. We found no evidence in our analysis for a considerable increase in net radiation (Supplementary Fig. 12). Instead, an explanation was found in the massive reduction in the height of the PBL (Fig. 3h and Supplementary Figs. 1114) during the summer in the irrigated case, which is consistent with a long-term decline in the PBL height in the summer during the 1979–2018 period over the Indo-Gangetic Plain in ERA5 reanalysis data (Supplementary Fig. 14). The decrease in PBL height due to irrigation is shown by our sensitivity experiments (Supplementary Fig. 13), which suggest that the presence of irrigation reduces the PBL height. It is well-established in other studies that a shallower boundary layer31 leads to an enhancement of low-level moist enthalpy. This mechanism works when the atmospheric boundary layer is not convectively coupled with the free troposphere. If deep convection occurs, the high moist enthalpy air is ventilated into the free troposphere. Irrigation reduces the sensible heat flux and increases the latent heat flux, as shown by our simulations (Fig. 3a–c). The reduction in sensible heat flux leads to surface air cooling, and so air descends. As a result, the PBL collapses32 (Fig. 3h). The subsidence leads to an increase in sea level pressure and the development of anticyclonic circulation (Fig. 3g). The reduced boundary-layer thickness and hence total heat capacity allows the moist enthalpy to increase faster during the day, which results in an increased TW. Our results, consistent with previous studies24,25,33, show that irrigation alters the land surface energy budget, temperature and humidity, as mediated by this PBL deflation mechanism32,34 (Fig. 3 and Supplementary Table 5), which can influence extreme moist heat stress in India.

Next, we evaluated the role of irrigation on extreme dry and moist heat in India using the WRF simulations with irrigation-on and irrigation-off scenarios (Fig. 4, Supplementary Fig. 15 and Supplementary Table 6). To represent extreme dry and moist heat conditions, we estimated the 95th percentiles of daily T2, HI and TW during the summer for the 2000–2018 period. The general patterns of extreme dry and moist heat in the irrigation-off scenarios in the WRF simulations (Fig. 4a–c) are consistent with those from ERA5 (Fig. 2b,c) and show a high moist heat over the Indo-Gangetic Plain. Irrigation reduces extreme dry heat based on the difference between the irrigation and no-irrigation scenarios in the WRF simulations (Fig. 4d). However, the influence of irrigation on moist heat indicators is not as straightforward as for that for maximum T2. The changes in extreme moist heat due to irrigation depend on the amount of cooling and moistening12. Our experiment with WRF shows that irrigation reduces extreme HI during the summer in India and the reduction is more prominent over the Indo-Gangetic Plain (Fig. 4e). As HI is positively correlated with specific humidity and temperature11, the amount of cooling due to irrigation dominates the amount of moistening. We found an increase in extreme TW due to irrigation based on the WRF simulations (Fig. 4f and Supplementary Fig. 15). Moreover, irrigation leads to an increase in extreme and mean TW in central India (Fig. 4f, Supplementary Fig. 15 and Supplementary Table 6). The changes in extreme TW due to irrigation are controlled by the extent of cooling and increase in humidity due to irrigation.

Fig. 4: Influence of irrigation on dry and moist heat stress in India.
figure4

ac, 95th percentile of T2 (°C) (a), HI (°C) (b) and TW (°C) (c) for the summer (April–May) during 2000–2018 under the no-irrigation scenario. df, Difference (irrigation – no irrigation) in the 95th percentile of T2 (°C) (d), HI (°C) (e) and TW (°C) (f). Data for af are based on the WRF simulations with irrigation-on and irrigation-off scenarios for the 2000–2018 period.

Source data

Implications of increase in extreme moist heat

Extreme dry and moist heat in the summer increased in the majority of India during 1979–2018. Although extreme moist heat increased in the Indo-Gangetic Plain, extreme dry heat declined. This contrast in extreme dry and moist heat, as we hypothesized, is at least partly attributable to intensive irrigation over the Indo-Gangetic Plain. The changes in extreme dry and moist heat over the Indo-Gangetic Plain and the other parts of the country are the manifestation of large-scale changes driven by anthropogenic warming4,26,27,35 and the localized influence of irrigation24,25,33,36. The rise in mean and extreme temperature in the majority of India during the summer can be attributed to anthropogenic warming3,37. Observed summer cooling over the Indo-Gangetic Plain can be attributed to irrigation20,24,25. The extent of cooling due to irrigation is associated with the fractional area irrigated, soil moisture dynamics and the response of cloud to irrigation19. Furthermore, near-surface humidity is increased due to anthropogenic warming and irrigation25,29. Anthropogenic warming can be a major factor in rising maximum temperature and surface specific humidity in the summer4, especially in the regions that are not heavily irrigated. Changes in the depth of the planetary boundary caused by irrigation-induced moistening may play a fundamental role in governing moist heat stress extremes. Both the warming climate and irrigation contribute to the increased extreme moist heat in parts of India.

Anthropogenic influences, both the global imprint of climatic warming and the localized impact of irrigation, have large and complex interactions in determining extreme moist heat stress in India. This has implications for agriculture, labour efficiency, mortality and public health. The consideration of the role of irrigation is vital as the severity of heat stress has increased in the past2 and is likely to increase in the future1,38. Our findings have direct implications for the reported benefits of green roofs and green cities to mitigate heat and urban heat island effects20,39. The effects of urban irrigation on near-surface temperature and energy budget components are limited40. However, excess moisture from plants on a green roof can provide evaporative cooling41 and mitigate dry heat. Green roofs and green urban areas (due to increased moisture) can increase the surface humidity and worsen the extreme heat stress conditions, especially under a warmer climate4,12. Therefore, the combination of an increased irrigated area in both urban and agricultural regions, which causes a significant (P < 0.05) increase in surface humidity, with an increased surface temperature can be detrimental for extreme heat stress. The increased heat stress due to irrigation is not limited to the geographical boundaries of India. Our results show that intensive irrigation over the Indo-Gangetic Plain increases the extreme heat stress in Pakistan, as well as western parts of Afghanistan, which indicates that the transboundary implications of irrigation-induced heat stress (Supplementary Fig. 16). Using the gridded population estimates, we found that the increased heat stress due to irrigation over the Gangetic Plain can affect about 37–46 million people in South Asia. Notwithstanding the negative impacts of irrigation on moist heat stress and associated mortality, irrigation over the Gangetic Plain remains a major factor to ensure food security in the region. Overall, our findings have implications for other regions dominated by a semi-arid or monsoon climate in which variations in humidity dominate heat stress as measured by TW (ref. 11).

Methods

We identified irrigated areas over India based on the Global Map of Irrigated Areas42 developed by the Food and Agriculture Organization (http://www.fao.org/nr/water/aquastat/irrigationmap/index10.stm). The spatial resolution of the irrigated map is approximately 10 km, which was aggregated to 0.25°. Weekly (7-day composite) TLS data at a 4 km spatial resolution were obtained from the NOAA’s Satellite Applications and Research programme for the period 1982–2015 (https://www.star.nesdis.noaa.gov/smcd/emb/vci/VH/vh_ftp.php). Land cover data for India was obtained from the National Remote Sensing Center, which is available at a 56 m spatial resolution for the year 2007–2008. The land cover data were classified into three dominant land cover classes: agriculture, forest and water. The land cover map was resampled using a majority class to a 4 km spatial resolution to make it consistent with the spatial resolution of the TLS data. Finally, we classified grid cells according to irrigated area fractions (0–1 with ten categories) for each dominant land cover class. The mean change in TLS based on fractional irrigated area was estimated by taking the mean of changes in all the grid cells that fall within a particular category.

We used a non-parametric Mann–Kendall43 method with Sen’s slope44 to estimate trends in TLS datasets during 1982–2015. Changes in the dry and moist heat indicators were estimated for the 1979–2018 period. Changes were estimated based on the trend slope multiplied by the total period.

We obtained daily maximum temperature at a 0.5° spatial resolution from the India Meteorology Department (IMD) for the period 1951–2015. The daily maximum temperature data from IMD is based on 395 stations45. Apart from the daily IMD dataset, we obtained the hourly air temperature at 2 m from the ERA5 reanalysis46 (https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5) for the period 1979–2018. The ERA5 dataset is available at around a 31 km spatial resolution, which was regridded to 0.25°. Furthermore, the hourly dew point temperature and surface pressure fields were obtained from the ERA5 datasets at the same spatial and temporal resolutions. These auxiliary datasets were then used to estimate the hourly near-surface (specific and relative) humidity values for the period 1979–2018. We used mean and 3-day summer (April–May) maximum T2 to characterize dry heat, whereas the 3-day summer maximum of TW was used to represent moist heat. In addition, we used mean and 3-day maximum HI to evaluate the moist heat conditions during the summer. Our measures of extreme dry and moist heat are based on 3-day summer maximum values of the indicators (maximum T2, TW and HI).

We calculated the daily maxima of these indicators based on their respective hourly values (T2, HI and TW). In this way, we retained the covariability of hourly variables (for example, specific humidity and air temperature), and also it allowed us to characterize the specific time of day when a selected heat indicator exhibited its maximum value. Subsequently, the 3-day maximum values were estimated with a moving-average approach based on the respective daily values for each year over the period 1979–2018. The wet bulb temperature and HI47 were estimated following the method described by Buzan et al.11, which uses hourly fields of air temperature, relative humidity, and near-surface pressure fields. The method uses an efficient and iterative approach to compute the TW and HI estimates with a first guess being estimated based on the Davies-Jones48 method and the latter iterated to estimate the final TW values11. The HI represents a feel-like temperature based on the level of discomfort, which was developed by the United States National Weather Service for a heat stress early warning system. HI considers the effects of both temperature and humidity and is based on a polynomial fit to the comfort model developed by Steadman49. Four different categorical scales to determine heat stress based on HI ranges were 27–32 °C caution, 33–39 °C extreme caution, 40–51 °C danger, and above 52 °C extreme danger. The HI was used in several recent studies to analyse heat stress effects50.

We used simulations from the WRF model (WRF 4.0)51, which was developed by the National Center for Atmospheric Research (NCAR) in collaboration with other agencies. The WRF model was used to evaluate the impacts of irrigation on surface fluxes and land–atmosphere interactions. We conducted WRF simulations from 2000 to 2018 to understand the influence of irrigation on dry and moist heat stress in the entire Indian region. To do so, using ERA5 as a boundary condition, we conducted two simulations with irrigation-on and irrigation-off scenarios. The NOAH land surface model52 coupled with a rapid radiative transfer model53, Mellor–Yamada–Janjic54 planetary boundary schemes and Kain–Fritsch55 convective parameterization schemes, which produce a more realistic pattern and extent of precipitation56, were used in our simulations. To conduct the WRF simulations the NOAH land surface scheme datasets are required at the surface and model level. We used the ERA5 surface 2 m temperature, 10 m U and V wind components, mean sea level pressure, sea surface temperature, skin temperature, snow density, snow depth, soil temperature at different levels, surface pressure, volumetric soil water, and model level variables (specific humidity, temperature, U and V components of the wind, and geopotential) to conduct the control (no-irrigation) simulations at a 0.25° spatial resolution. The fractional area under irrigation from the Food and Agriculture Organization57 irrigation map was obtained to implement the irrigation scheme in the Noah land surface model coupled with WRF. Irrigation was applied as precipitation when the root-zone soil moisture fell below the field capacity56. More details on the implementation of the irrigation scheme can be obtained from Qian et al.56. We used 6 hour fields of ERA5 data as input for the simulations conducted between 2000 and 2018 under the two scenarios (control and irrigation) with model output at hourly time steps. The WRF simulations were further used to understand the role of irrigation on dry and moist heat stress in India.

The WRF simulations in the control (no-irrigation) scenario were compared at the all-India level and on the regional scale (Indo-Gangetic Plain). We compared specific humidity, air temperature at 2 m, and PBL height from the WRF simulations against the ERA5 and MERRA58. We used the Integrated Global Radiosonde Archive station observations for the period 2000–201859 to estimate the PBL height. The radiosonde observations contain several atmospheric parameters at different pressure levels, such as geopotential height, temperature gradient, potential temperature, and humidity60. We employed the maximum vertical gradient of potential temperature to calculate the PBL height60, which indicates a transition of regions from a less stable to a stable convective zone61.

Data availability

All datasets are available in the manuscript or the Supplementary Information, and on the Zenodo database https://zenodo.org/record/3999048#.X0Sdni1h2so. Temperature observations from the India Meteorological Department are available from http://imdpune.gov.in/ndc_new/Request.html. The ERA5 reanalysis dataset is available from https://go.nature.com/2FB53G8. Source data are provided with this paper.

Code availability

All the codes used in this study will be provided by the corresponding author upon reasonable request.

References

  1. 1.

    Im, E. S., Pal, J. S. & Eltahir, E. A. B. Deadly heat waves projected in the densely populated agricultural regions of South Asia. Sci. Adv. 3, e1603322 (2017).

    Google Scholar 

  2. 2.

    Mazdiyasni, O. et al. Increasing probability of mortality during Indian heat waves. Sci. Adv. 3, 1–6 (2017).

    Google Scholar 

  3. 3.

    Mishra, V., Mukherjee, S., Kumar, R. & Stone, D. A. Heat wave exposure in India in current, 1.5 °C, and 2.0 °C worlds. Environ. Res. Lett. https://doi.org/10.1088/1748-9326/aa9388 (2017).

  4. 4.

    Coffel, E. D., Horton, R. M. & de Sherbinin, A. Temperature and humidity based projections of a rapid rise in global heat stress exposure during the 21st century. Environ. Res. Lett. 13, 014001 (2017).

    Google Scholar 

  5. 5.

    King, A. D. et al. Emergence of heat extremes attributable to anthropogenic influences. Geophys. Res. Lett. 43, 3438–3443 (2016).

    Google Scholar 

  6. 6.

    Knutson, T. R. & Ploshay, J. J. Detection of anthropogenic influence on a summertime heat stress index. Clim. Change 138, 25–39 (2016).

    Google Scholar 

  7. 7.

    Matthews, T. K. R., Wilby, R. L. & Murphy, C. Communicating the deadly consequences of global warming for human heat stress. Proc. Natl Acad. Sci. USA 114, 3861–3866 (2017).

    Google Scholar 

  8. 8.

    Kjellstrom, T. et al. Heat, human performance, and occupational health: a key issue for the assessment of global climate change impacts. Annu. Rev. Public Health 37, 97–112 (2016).

    Google Scholar 

  9. 9.

    Sherwood, S. C. How important is humidity in heat stress? J. Geophys. Res. Atmos. 123, 11808–11810 (2018).

    Google Scholar 

  10. 10.

    Horton, R. M., Mankin, J. S., Lesk, C., Coffel, E. & Raymond, C. A review of recent advances in research on extreme heat events. Curr. Clim. Change Rep. 2, 242–259 (2016).

    Google Scholar 

  11. 11.

    Buzan, J. R., Oleson, K. & Huber, M. Implementation and comparison of a suite of heat stress metrics within the Community Land Model version 4.5. Geosci. Model Dev. 8, 151–170 (2015).

    Google Scholar 

  12. 12.

    Kang, S. & Eltahir, E. A. B. North China Plain threatened by deadly heatwaves due to climate change and irrigation. Nat. Commun. 9, 2894 (2018).

    Google Scholar 

  13. 13.

    Shankar, P. V., Kulkarni, H. & Krishnan, S. India’s groundwater challenge and the way forward. Econ. Political Wkly 46, 37–45 (2011).

    Google Scholar 

  14. 14.

    Amarasinghe, U. A., Shah, T. & Anand, B. K. India’s water supply and demand from 2025-2050: business-as-usual scenario and issues. In Proc. Workshop on Analyses of Hydrological, Social and Ecological Issues of the National River Linking Project (eds Amarasinghe, U. A. & Sharma, B. R.) 23–61 (IWMI, 2007).

  15. 15.

    Ambika, A. K., Wardlow, B. & Mishra, V. Remotely sensed high resolution irrigated area mapping in India for 2000 to 2015. Sci. Data 3, 160118 (2016).

    Google Scholar 

  16. 16.

    Cook, B. I., Puma, M. J. & Krakauer, N. Y. Irrigation induced surface cooling in the context of modern and increased greenhouse gas forcing. Clim. Dyn. 37, 1587–1600 (2011).

    Google Scholar 

  17. 17.

    Thiery, W. et al. Present-day irrigation mitigates heat extremes. J. Geophys. Res. Atmos. 122, 1403–1422 (2017).

    Google Scholar 

  18. 18.

    Boucher, O., Myhre, G. & Myhre, A. Direct human influence of irrigation on atmospheric water vapour and climate. Clim. Dyn. 22, 597–603 (2004).

    Google Scholar 

  19. 19.

    Lobell, D. et al. Regional differences in the influence of irrigation on climate. J. Clim. 22, 2248–2255 (2009).

    Google Scholar 

  20. 20.

    Kumar, R. et al. Dominant control of agriculture and irrigation on urban heat island in India. Sci. Rep. 7, 14054 (2017).

    Google Scholar 

  21. 21.

    Mueller, N. D. et al. Cooling of US Midwest summer temperature extremes from cropland intensification. Nat. Clim. Change 6, 317–322 (2015).

    Google Scholar 

  22. 22.

    Asoka, A., Gleeson, T., Wada, Y. & Mishra, V. Relative contribution of monsoon precipitation and pumping to changes in groundwater storage in India. Nat. Geosci. 10, 109–117 (2017).

    Google Scholar 

  23. 23.

    Azhar, G. S. et al. Heat-related mortality in India: excess all-cause mortality associated with the 2010 Ahmedabad heat wave. PLoS ONE 9, e91831 (2014).

    Google Scholar 

  24. 24.

    Marcella, M. P. & Eltahir, E. A. B. Introducing an irrigation scheme to a regional climate model: a case study over West Africa. J. Clim. 27, 5708–5723 (2014).

    Google Scholar 

  25. 25.

    Puma, M. J. & Cook, B. I. Effects of irrigation on global climate during the 20th century. J. Geophys. Res. Atmos. 115, D16120 (2010).

    Google Scholar 

  26. 26.

    Willett, K. M. & Sherwood, S. Exceedance of heat index thresholds for 15 regions under a warming climate using the wet-bulb globe temperature. Int. J. Climatol. https://doi.org/10.1002/joc.2257 (2012).

  27. 27.

    Sherwood, S. C. & Huber, M. An adaptability limit to climate change due to heat stress. Proc. Natl Acad. Sci. USA 107, 9552–9555 (2010).

    Google Scholar 

  28. 28.

    Byrne, M. P. & O’Gorman, P. A. Trends in continental temperature and humidity directly linked to ocean warming. Proc. Natl Acad. Sci. USA 115, 4863–4868 (2018).

    Google Scholar 

  29. 29.

    Willett, K. M., Gillett, N. P., Jones, P. D. & Thorne, P. W. Attribution of observed surface humidity changes to human influence. Nature 449, 710–712 (2007).

    Google Scholar 

  30. 30.

    Bollasina, M. & Nigam, S. The summertime ‘heat’ low over Pakistan/northwestern India: evolution and origin. Clim. Dyn. 37, 957–970 (2011).

    Google Scholar 

  31. 31.

    Gentine, P., Holtslag, A. A. M., D’Andrea, F. & Ek, M. Surface and atmospheric controls on the onset of moist convection over land. J. Hydrometeorol. 14, 1443–1462 (2013).

    Google Scholar 

  32. 32.

    Kang, S. & Eltahir, E. A. B. Impact of irrigation on regional climate over eastern China. Geophys. Res. Lett. 46, 5499–5505 (2019).

    Google Scholar 

  33. 33.

    Kueppers, L. M., Snyder, M. A. & Sloan, L. C. Irrigation cooling effect: regional climate forcing by land-use change. Geophys. Res. Lett. 34, L03703 (2007).

    Google Scholar 

  34. 34.

    Alter, R. E., Im, E. S. & Eltahir, E. A. B. Rainfall consistently enhanced around the Gezira Scheme in East Africa due to irrigation. Nat. Geosci. 8, 763–767 (2015).

    Google Scholar 

  35. 35.

    Im, E. S. & Kang, S. & Eltahir, E. A. B. Projections of rising heat stress over the western Maritime Continent from dynamically downscaled climate simulations. Glob. Planet. Change https://doi.org/10.1016/j.gloplacha.2018.02.01 (2018).

  36. 36.

    Sacks, W. J., Cook, B. I., Buenning, N., Levis, S. & Helkowski, J. H. Effects of global irrigation on the near-surface climate. Clim. Dyn. 33, 159–175 (2009).

    Google Scholar 

  37. 37.

    Dileepkumar, R., Achutarao, K. & Arulalan, T. Human influence on sub-regional surface air temperature change over India. Sci. Rep. 8, 8967 (2018).

    Google Scholar 

  38. 38.

    Seneviratne, S. I. et al. Land radiative management as contributor to regional-scale climate adaptation and mitigation. Nat. Geosci. 11, 88–96 (2018).

    Google Scholar 

  39. 39.

    Sharma, A. et al. Green and cool roofs to mitigate urban heat island effects in the Chicago metropolitan area: evaluation with a regional climate model. Environ. Res. Lett. 11, 064004 (2016).

    Google Scholar 

  40. 40.

    Georgescu, M., Moustaoui, M., Mahalov, A. & Dudhia, J. An alternative explanation of the semiarid urban area ‘oasis effect’. J. Geophys. Res. Atmos. https://doi.org/10.1029/2011JD016720 (2011).

  41. 41.

    Zipper, S. C., Schatz, J., Kucharik, C. J. & Loheide, S. P. Urban heat island-induced increases in evapotranspirative demand. Geophys. Res. Lett. https://doi.org/10.1002/2016GL072190 (2017).

  42. 42.

    Siebert, S., Henrich, V., Frenken, K. & Burke, J. Update of the Digital Global Map of Irrigation Areas to Version 5 (FAO, 2013); https://doi.org/10.13140/2.1.2660.6728

  43. 43.

    Mann, H. B. Nonparametric tests against trend. Econometrica 13, 245–259 (1945).

    Google Scholar 

  44. 44.

    Sen, P. K. Estimates of the regression coefficient based on Kendall’s Tau. J. Am. Stat. Assoc. 63, 1379–1389 (1968).

    Google Scholar 

  45. 45.

    Srivastava, A. K., Rajeevan, M. & Kshirsagar, S. R. Development of a high resolution daily gridded temperature data set (1969–2005) for the Indian region. Atmos. Sci. Lett. 10, 249–254.

  46. 46.

    Dee, D. P. et al. The ERA-Interim reanalysis: configuration and performance of the data assimilation system. Q. J. R. Meteorol. Soc. 137, 553–597 (2011).

    Google Scholar 

  47. 47.

    Haldane, J. S. The influence of high air temperatures No. I. J. Hyg. (Lond.) 5, 494–513 (1905).

    Google Scholar 

  48. 48.

    Davies-Jones, R. An efficient and accurate method for computing the wet-bulb temperature along pseudoadiabats. Mon. Weather Rev. 136, 2764–2785 (2008).

    Google Scholar 

  49. 49.

    Steadman, R. G. The assessment of sultriness. Part I. A temperature–humidity index based on human physiology and clothing science. J. Appl. Meteorol. 18, 861–873 (1979).

    Google Scholar 

  50. 50.

    Brooke Anderson, G., Bell, M. L. & Peng, R. D. Methods to calculate the heat index as an exposure metric in environmental health research. Environ. Health Perspect. 121, 1111–1119 (2013).

    Google Scholar 

  51. 51.

    Skamarock, C. et al. A Description of the Advanced Research WRF Model Version 4 (NCAR, 2019); https://doi.org/10.5065/1DFH-6P97

  52. 52.

    Mitchell, K. et al. Noah Land Surface Model (LSM) User’s Guide (NCAR, 2005).

  53. 53.

    Iacono, M. J. Radiative forcing by long-lived greenhouse gases: calculations with the AER radiative transfer models. J. Geophys. Res. Atmos. https://doi.org/10.1029/2008JD009944 (2008).

  54. 54.

    Janzic, Z. I. The step-mountain eta coordinate model: further developments of the convection, viscous sublayer, and turbulence closure schemes. Mon. Weather Rev. 122, 927–945 (1994).

    Google Scholar 

  55. 55.

    Kain, J. S. & Kain, J. The Kain–Fritsch convective parameterization: an update. J. Appl. Meteorol. 43, 170–181 (2004).

    Google Scholar 

  56. 56.

    Qian, Y., Huang, M., Yang, B. & Berg, L. K. A modeling study of irrigation effects on surface fluxes and land–air–cloud interactions in the Southern Great Plains. J. Hydrometeorol. 14, 700–721 (2013).

    Google Scholar 

  57. 57.

    Siebert, S. et al. Development and validation of the global map of irrigation areas. Hydrol. Earth Syst. Sci. 9, 535–547 (2005).

    Google Scholar 

  58. 58.

    Rienecker, M. M. et al. MERRA: NASA’s modern-era retrospective analysis for research and applications. J. Clim. 24, 3624–3648 (2011).

    Google Scholar 

  59. 59.

    Durre, I. & Yin, X. Enhanced radiosonde data for studies of vertical structure. Bull. Am. Meteorol. Soc. 89, 1257–1262 (2008).

    Google Scholar 

  60. 60.

    Seidel, D. J., Ao, C. O. & Li, K. Estimating climatological planetary boundary layer heights from radiosonde observations: comparison of methods and uncertainty analysis. J. Geophys. Res. Atmos. 115, D16113 (2010).

    Google Scholar 

  61. 61.

    Basha, G. & Ratnam, M. V. Identification of atmospheric boundary layer height over a tropical station using high-resolution radiosonde refractivity profiles: comparison with GPS radio occupation measurements. J. Geophys. Res. Atmos. 114, D161010 (2009).

    Google Scholar 

Download references

Acknowledgements

We acknowledge the data availability from the IMD ERA5, and MERRA reanalysis. The work was funded by the National Water Mission, Ministry of Environment, Forest, and Climate Change (MoEFCC) and the Ministry of Earth Sciences.

Author information

Affiliations

Authors

Contributions

V.M. conceived and designed the study and discussed it with M.H. A.K.A. conducted the WRF simulations. V.M., A.K.A., A.A., R.K. and S.A. performed the analysis. V.M. wrote the first draft and all the authors contributed to the discussion.

Corresponding author

Correspondence to Vimal Mishra.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Primary handling editor: Tamara Goldin.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Tables 1–6 and Figs. 1–16.

Source data

Source Data Fig. 1

Data for all the panels of Fig. 1.

Source Data Fig. 2

Data for all the panels of Fig. 2.

Source Data Fig. 3

Data for all the panels of Fig. 3.

Source Data Fig. 4

Data for all the panels of Fig. 4.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Mishra, V., Ambika, A.K., Asoka, A. et al. Moist heat stress extremes in India enhanced by irrigation. Nat. Geosci. 13, 722–728 (2020). https://doi.org/10.1038/s41561-020-00650-8

Download citation

Further reading

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing