Research papers
Impact of renewed solar dimming on streamflow generation in monsoon dominated tropical river basins

https://doi.org/10.1016/j.jher.2022.02.002Get rights and content

Abstract

From 1950s to 1980s, various observational studies around the globe found a significant decrease in surface solar radiation (SSR), which reversed in late 1980s for most of the countries including India. SSR observations at 12 stations located across India revealed that a much stronger dimming has reappeared during the last decade (2006–2015) after a brightening during 1996–2005. In the present study, effects of renewed solar dimming on actual evapotranspiration and runoff were analyzed using a semi-distributed hydrological model, Soil and Water Assessment Tool (SWAT) in 24 river basins (ranging from 1260 to 40000 km2) located in peninsular India. For these river basins, calibration (2003–2009) and validation (2010–2014) were performed using the observed daily discharge data, obtained from water resources information system (WRIS) of India, with a 3 year warm up period (2000–2002). The sequential uncertainty domain parameter fitting algorithm (SUFI-2) of SWAT-CUP (calibration and uncertainty program) was used with modified Nash–Sutcliffe efficiency (MNS) as the objective function to calibrate 13 model parameters, which can potentially affect streamflow. In nearly all the river basins, the p- and r-factor of 95 percentage prediction uncertainty (PPU) were more than 0.7 and less than 1, respectively. At daily timescale, MNS values were more than 0.5 in most of the river basins, reaching up to 0.66 and 0.71 during calibration and validation periods, respectively. Calibrated model was used to analyze the water balance of these river basins and different sets of experiments (with observed SSR trends) were performed to find SSR impacts on it. The model was simulated with and without the observed declines in SSR trends. The average change in SSR (in terms of evaporation equivalent) was −267.93 ± 100.92 mm/day/year (−5.62 ± 2.12%) with maximum reaching up to −417.12 mm/day/year (−8.99%). Due to this SSR change, actual evaporation was reduced resulting in 18.97 ± 9.78 mm/day/year (4.13 ± 2.50%) change in percolation. The percolation changes were higher for river basins having areas covered by forests and cropland/woodland, and having loam and sandy-clay soils. The increase in runoff generated was 6.90 ± 3.42 mm/day/year (2.14 ± 1.58%) with a maximum of 15.25 mm/day/year (7.56%) whereas corresponding increase in streamflow was found to be 9.93 ± 5.27 mm/day/year(4.21 ± 2.38%) with a maximum of 26.71 mm/day/year (11.86 %). The study reveals that the recent observed SSR changes are significant enough to have resulted in increased streamflow in the monsoon dominated tropical river basins of India.

Introduction

Various observational and modeling studies have reported a gradual but significant decrease in surface solar radiation (SSR) from1950s to 1980s all around the globe, a phenomenon commonly known as “Global Dimming” (Wild et al., 2009). Partial recovery in SSR from late 1980s has been observed in many European countries, Japan and the USA, which is termed as “Brightening” (Wild et al., 2009). Unlike dimming, brightening is not global. The literature suggested that the dimming continued to exist in a few countries, including India (Wild et al., 2009, Padma2007 et al., 2007, Soni2012 et al., 2012, Padma et al., 2010). However, Soni2016 et al. (2016) reported a brightening phenomenon over India from 2001–2010. They analyzed the SSR data from 1971 to 2010 and found that, despite overall dimming from 1971 to 2010, the trend has weakened and brightening has begun since 2001, especially for clear sky days. Recently, using observed SSR data from India meteorological department (IMD), we found that despite a brightening during the period 1996–2005, a much stronger dimming exists over India during the past decade (2006–2015) (Soni et al., 2019).

The reported changes in SSR during the past decade are likely to have important implications for actual evapotranspiration (hereafter referred to as ‘AET’) and associated water cycle components (Roderick et al., 2002, Pieruschka et al., 2010, Seneviratne et al., 2010). The magnitude of terrestrial AET is not well understood but is estimated to be about 60–70% of the total precipitation (Oki et al., 2006, Trenberth et al., 2009). Using observation-driven Penman–Monteith-Leuning Model for the period 1981–2012, a recent study found AET to be 67% of mean annual precipitation globally (Zhang et al., 2016).

In spite of increasing temperatures due to greenhouse gas forcing, many observational studies around the globe have shown a declining trend in AET which could be attributed to changes in SSR (Mcvicar et al., 2012). Moreover, they also reported that AET changes are highly dependent on hydroclimatological behavior (water-limited or energy-limited) of the catchment. Kumar2016 et al. (2016) found that terrestrial hydrological sensitivity (change in runoff ratio per unit change in dryness index) is 3 times greater in regions where the hydrological cycle is energy limited rather than water limited. Using simulations from a global climate model, Douville et al. (2013) showed that modeled historical AET can be influenced by combination of aerosol anthropogenic emissions and greenhouse gases over the twentieth century. Using data from eddy-covariance flux tower network (FLUXNET) and a multi-model re-analysis (GSWP-2), Teuling et al. (2009) suggested that trends in AET are broadly consistent with those of shortwave radiation in central Europe and some parts of North America.

The intensification of the hydrological cycle due to global warming is expected to result in the wet regions getting wetter and dry regions getting drier. However, significant discrepancies from the expected pattern of change have been observed in some recent studies (Kumar et al., 2015, Roderick et al., 2014, Byrne et al., 2015). Using Global Precipitation Climatology Project (GPCP) gridded dataset of monthly precipitation, model data from the Coupled Model Intercomparison Project Phase 5 (CMIP5) archive and RCP8.5 scenario, Polson et al. (2017) found that mean precipitation changes follow this pattern for fixed wet and dry regions, however when shifting locations of the wet and dry regions are considered, this signal is reduced. Similarly, using GCMs suitable for climate projections over Egypt, Hamed et al. (2021) showed a possible decrease in annual and winter precipitation in the range of 0 to −50% in the high rainfall regions in the north and a large increase in the low rainfall (abour 5 mm) region in the south. Similar results were simulated by Tabari et al., 2018, Nashwan et al., 2022 over Middle east and Egypt, respectively.

Using land use land cover data from MODIS, Li et al. (2017) showed that average annual AET reduced at a rate of −0.6 mm/yr from 2001 to 2013 in China. Hasan et al. (2018) performed sensitivity study by integrating the concept of the aridity index and climate elasticity over the Nile river basin. They showed that arid zones are more sensitive to precipitation as compared to tropical regions. Using three conceptual rainfall-runoff models (GR4J, AWBM, and IHACRES_CMD), Guo et al. (2017) found that runoff sensitivity is more than sevenfold difference in per unit change in annual average PET between different models.

Water resource is one of the most important factors for economic development of any country (Goswami et al., 2017). It can be significantly impacted by changes in climatic variables during the 20th century, and even larger changes are expected in the coming decades (Milly et al., 2005). Studies have used different models to estimate changes in streamflow due to changes in AET or SSR, but the results vary depending upon the nature of the catchment. Simulations from a land surface model suggested an increased annual river flow (up to 25%) for European river basins due to global dimming (Gedney et al., 2014). In a similar study, using community land model (CLM), Oliveira et al. (2011) found an increase of 5% and a decrease of 7–10% in runoff during dimming (1960–1990) and brightening (1990–2004) periods, respectively, around the globe. Using the SWAT model over Gyeonggi Province, Korea, Lee and Chung (2007) found that for 10% decrease in SSR, runoff increases by about 12%. In a statistical study, Yang et al. (2011) found that for every 1% decrease in SSR, runoff increases by 0.3–1.9% over river basins of China. However, using water energy balance equations, Xia et al. (2014) found smaller (4–7%) changes in runoff due to 10% change in SSR.

As of now, no study has evaluated the affect of SSR changes on the streamflow of the monsoon dominated tropical monsoon river basins of India. The main focus of the present study is to analyze the effects of recent solar dimming on AET and runoff at basin scale using a hydrological model, namely, SWAT (Soil and Water Assessment Tool) (Arnold et al., 1993) model. The SWAT model was selected for the study because it has been extensively used for Indian river basins with satisfactory performance (Wilk et al., 2002, Tripathi et al., 2003, Dhar et al., 2009, Lakshmanan et al., 2011, Garg et al., 2012). The present study has the following objectives -.

  • 1.

    To analyze the water balance of 24 river basins in peninsular India using the SWAT model;

  • 2.

    To assess the effects of recent SSR trends on the water balance of these river basins.

Section snippets

Study area and data used

Twenty-four river basins (Fig. 1 and Table 1) that satisfy the following conditions were selected for impact assessment- (1) basin area should be more than 1000 km2, limited by the input data resolution; (2) data should be available for the study period (2000–2014), limited by the input data availability; and (3) no major dams or reservoirs that regulate flows should be present in the river basin, limited by lack of flow regulation observations; Observed streamflow data were available only for

Methodology

The NSRDB data represent spatial variability in SSR at a fine resolution (10 km), but do not exhibit recent solar dimming present in IMD observed data. Since data from only 12 IMD stations are available, NSRDB and IMD SSR data were combined (as shown in Fig. 3(a)) for assessing the effects of renewed solar dimming on water balance of 24 river basins in peninsular India. For each NSRDB grid, annual trend from 2006–2014 was estimated and the time-series was de-trended to generate Control (CTRL)

Calibration of SWAT model parameters

We found that r_CN2 is the most sensitive parameter. The other sensitive parameters are v_ESCO, r_SOL_AWC, v_GW_REVAP, v_GW_DELAY, v_GWQMN, r_OV_N, and v_ALPHA_BF. Objective function was found to be relatively insensitive to the five parameters, namely, v_EPCO, r_HRU_SLP, v_REVAPMN, r_SLSUBBSN and v_SURLAG (Refer to Fig. S3 in Supplementary material).

Default values of CN2 in the SWAT model overestimated observed discharges in most of the basins resulting in negative r_CN2 values up to −0.4

Analysis of water balance

HRU level water balance revealed that combined non-monsoon precipitation (during the winter, premonsoon and postmonsoon seasons) is only about 10–15% of the annual precipitation. Due to different catchment attributes, river basins behave differently in terms of water balance. Generally percentage AET component is higher in the annual water balance for river basins having less precipitation. When averaged over all the river basins, AET, runoff and percolation components are 51.65%, 20.42% and

Summary and conclusions

The present study focused on finding the impacts of renewed solar dimming (2006–2015) on water balance of monsoon dominated tropical river basins in peninsular India. 24 river basins (areas varying from 1,260 km2 to 40,000 km2) were selected and impacts of SSR changes are investigated using the SWAT model. Four different experiments were conducted by generating four sets of SSR data using NSRDB gridded and IMD station data. The model was calibrated using the best representation of actual SSR

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors thank IMD and TRMM for the precipitation data. We also thank NSRDB for providing various meteorological data at high resolution. We would also thank two anonymous reviewers and the editorial team of Journal of Hydro-environment Research for their valuable suggestion to significantly improve the manuscript.

References (68)

  • M. Tripathi et al.

    Identification and prioritisation of critical sub-watersheds for soil conservation management using the swat model

    Biosyst. Eng.

    (2003)
  • S. Visakh et al.

    Inter-comparison of water balance components of river basins draining into selected delta districts of eastern india

    Sci. Total Environ.

    (2019)
  • R.G. Allen et al.

    Crop evapotranspiration-guidelines for computing crop water requirements-fao irrigation and drainage paper 56

    Fao, Rome

    (1998)
  • J.G. Arnold et al.

    Hydrologic model for design and constructed wetlands

    Wetlands

    (2001)
  • J.G. Arnold et al.

    Swat2000: current capabilities and research opportunities in applied watershed modelling

    Hydrol. Process.: Int. J.

    (2005)
  • J.G. Arnold et al.

    Swat: Model use, calibration, and validation

    Trans. ASABE

    (2012)
  • J.G. Arnold et al.

    Large area hydrologic modeling and assessment part i: Model development1

    J. Am. Water Resour. Assoc.

    (1998)
  • S. Babar et al.

    Streamflow response to land use–land cover change over the nethravathi river basin, india

    J. Hydrol. Eng.

    (2015)
  • M.A. Bollasina et al.

    Anthropogenic aerosols and the weakening of the south asian summer monsoon

    Science

    (2011)
  • M.P. Byrne et al.

    The response of precipitation minus evapotranspiration to climate warming: Why the wet-get-wetter, dry-get-drier scaling does not hold over land

    J. Clim.

    (2015)
  • Commission, C.W., 2019, ‘Reassessment of water availability in india-main report’. url:...
  • S. Dhar et al.

    Hydrological modelling of the kangsabati river under changed climate scenario: case study in india

    Hydrol. Process.

    (2009)
  • Dong, B., 2012. Impacts of climate change on the surface water balance of the central united states,...
  • H. Douville et al.

    Anthropogenic influence on multidecadal changes in reconstructed global evapotranspiration

    Nature Climate Change

    (2013)
  • A. El-Sadek et al.

    Evaluating the impact of land use uncertainty on the simulated streamflow and sediment yield of the seyhan river basin using the swat model

    Turkish J. Agricul. Forestry

    (2014)
  • FAO, 1997. Digital soil map of the world and derived soil...
  • K.K. Garg et al.

    Spatial mapping of agricultural water productivity using the swat model in upper bhima catchment, India

    Irrigation Drainage

    (2012)
  • N. Gedney et al.

    Detection of solar dimming and brightening effects on northern hemisphere river flow

    Nat. Geosci.

    (2014)
  • K.B. Goswami et al.

    The role of water resources in socio-economic development

    Int. J. Res. Appl. Sci. Eng. Technol.

    (2017)
  • D. Guo et al.

    Impact of evapotranspiration process representation on runoff projections from conceptual rainfall-runoff models

    Water Resour. Res.

    (2017)
  • M.M. Hamed et al.

    A novel selection method of cmip6 gcms for robust climate projection

    Int. J. Climatol.

    (2021)
  • Kothawale, D., Revadekar, J., 2017. Interannual variations of indian summer monsoon. url:...
  • S. Kumar et al.

    Revisiting trends in wetness and dryness in the presence of internal climate variability and water limitations over land

    Geophys. Res. Lett.

    (2015)
  • S. Kumar et al.

    Terrestrial contribution to the heterogeneity in hydrological changes under global warming

    Water Resour. Res.

    (2016)
  • Cited by (0)

    View full text