Groundwater storage change detection from in situ and GRACE- based estimates in major river basins across India
Soumendra N. Bhanja1,2, Abhijit Mukherjee1,3, Matthew Rodell4
1Department of Geology and Geophysics, Indian Institute of Technology Kharagpur, West Bengal 721302, India
2Interdisciplinary Centre for Water Research, Indian Institute of Science, Bangalore, Karnataka 560054, India
3School of Environmental Science and Engineering, Indian Institute of Technology Kharagpur, West Bengal 721302, India
4Hydrological Sciences Laboratory, NASA Goddard Space Flight Center, Greenbelt, Maryland 20771, USA
Abstract
India has been the subject of many recent groundwater studies due to the rapid depletion of groundwater in large parts of the country. However, few if any of these studies have examined groundwater storage conditions in all of India’s river basins individually. Herein we assess groundwater storage changes in all 22 of India’s major river basins using in situ data from 3420 observation locations for the period 2003–2014. One-month and 12-month standardized
precipitation index measures (SPI-1 and SPI-12) indicate fluctuations in the long-term pattern. The Ganges and Brahmaputra basins experienced long-term decreasing trends in precipitation in both 1961–2014 and the study period, 2003–2014. Indeterminate or increasing precipitation trends occurred in other basins. Satellite-based and in situ groundwater storage time series exhibited similar patterns, with increases in most of the basins. However, diminishing groundwater storage (at rates of >0.4 km3/year) was revealed in the Ganges-Brahmaputra river basin based on in situ observations, which is particularly important due to its agricultural productivity.
Keywords
groundwater storage; Indian river basins; GRACE; standardized precipitation index (SPI)
1 Introduction
Groundwater storage (GWS) has been diminishing rapidly in some of the world´s most densely populated areas (Rodell et al. 2009; Bhanja et al. 2014; Feng et al. 2013; Voss et al.
2013; Bhanja et al. 2017b; Rodell et al. 2018; Scanlon et al. 2018). In terms of groundwater withdrawals, those in India exceed the combined withdrawals of China and the USA, the
NASA Public Access
Author manuscript
Hydrol Sci J. Author manuscript; available in PMC 2021 January 01.
Published in final edited form as:
Hydrol Sci J. 2020 ; 65(4): 650–659. doi:10.1080/02626667.2020.1716238.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
second and third countries on the list, respectively (Margat and van der Gun 2013). In India, agriculture accounts for approx. 89% of the total groundwater consumption (Mukherjee et al. 2015). Rapid groundwater depletion has been reported in parts of India (Rodell et al.
2009; Tiwari et al. 2009; Panda and Wahr 2016; Bhanja et al. 2017b; Asoka et al. 2017;
Rodell et al. 2018) and is linked to the Ganges river summer drying (Mukherjee et al., 2018).
However, a ray of hope still exists, as parts of India have experienced groundwater
replenishment due to implementation of sustainable water management strategies (Bhanja et al. 2017b).
From 2002 to 2017 the Gravity Recovery and Climate Experiment (GRACE) mission monitored changes in Earth’s gravity field with unprecedented accuracy (Tapley et al. 2004;
Famiglietti and Rodell 2013; Rodell et al. 2018). Temporal variations in Earth’s gravity field are attributed to changes in atmospheric, oceanic, and terrestrial water and solid earth mass.
Terrestrial water storage (TWS) changes can be estimated after first removing the influences of solid earth processes (Wahr 2007) and atmospheric and oceanic circulation (Flechtner 2007). GRACE provided global maps of TWS anomalies (departures from the long-term mean) with approximately 400 km horizontal resolution using a spherical harmonics based approach, with fields often provided on a 1° × 1° grid for convenience (Landerer and Swenson 2012). Recent products based on the mass concentration (mascon) approach have improved accuracy and horizontal resolutions that approach 300 km (Watkins et al. 2015;
Bhanja et al. 2016; Wiese et al. 2016).
Despite comprising of only 2.28% (297 × 106 ha) of global land area, India is home to 17.80% (1.24 billion as of 2011) of the global population (FAO, 2013). Hence the continued availability of freshwater in India is of serious concern. Soni and Syed (2015) estimated satellite-based GWS change in four Indian river basins. Sinha et al. (2017) presented a GRACE-based drought index in four different geographic regions in India. A total of 22 major river basins are present in India (Table 1; Fig. 1) (India-WRIS, 2012). Bhanja et al.
(2016) estimated validation statistics and showed good match of GRACE-based GWS with in situ measurements in 12 of those river basins. However, to our knowledge, no study has reported groundwater storage condition in all of the major river basins in India. Therefore, the aim of this study is to estimate long-term (2003–2014), basin-wide change in
groundwater storage rates using in situ and satellite-based measurements. We compute long- term trends in GWS and discuss them within the context of precipitation variations. We also investigate long-term (1961–2014) changes in precipitation patterns in the region.
2 Data and methods
The detailed methodology is provided below. For ease of understanding the methodology, a flowchart is presented in the Supplementary Material (Fig. S1).
2.1 Precipitation data
Gridded (0.25° × 0.25°) daily precipitation data (IMD4) are used from the archives of the India Meteorological Department (IMD) for the period between 1961 and 2014 (Pai et al.
2014). This dataset uses the maximum number of in situ daily raingauge measurement in India, with information from 6995 locations (Pai et al. 2014). For point to grid-scale
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
conversion, an inverse distance weighted interpolation (IDW) scheme is used (Pai et al.
2014).
The 1-month standardized precipitation index (SPI-1) provides a normalized distribution of monthly precipitation patterns (NDMC 2018a). Fluctuations in SPI-1 mostly influence near- surface conditions such as soil moisture and crop stress (NDMC 2018a). SPI values of more than 2 indicate excess precipitation conditions and those less than 2 reflect comparatively lower precipitation than other years. Wetness conditions including floods and droughts depend on the magnitude of precipitation incident at a region. Twelve-month SPI (SPI-12) variations are associated with fluctuations in stream flow, reservoir levels, and groundwater levels (NDMC 2018b).
2.2 Ground-based groundwater level measurements
We used seasonal groundwater level (GWL) measurement data from the period 2003–2014 from 19 278 observation wells across India, as archived by the Central Ground Water Board of India (CGWB). More than 87% of the wells are screened at the uppermost shallow, unconfined aquifer (CGWB 2014). In order to maintain the temporal continuity, we selected the wells with at least three measurements per year for further analyses. Outliers in the data were discarded following Tukey’s approach (Tukey 1977). These criteria resulted in the selection of 3420 wells. Specific yield (Sy) data were obtained following Bhanja et al.
(2016) for the individual wells. Groundwater level anomalies for the individual wells were computed after removing the location specific long-term mean GWL (GWLM) value from the individual GWL measurements. Groundwater storage anomalies (GWSA) were computed by multiplying groundwater level anomalies with the specific yield at each location:
GWSA = (GWL − GWLM) × Sy (1)
2.3 GRACE-based measurement
Data of RL05 spherical harmonic-based terrestrial water storage (TWS) were retrieved from the National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (JPL) Tellus archive (Landerer and Swenson 2012). Gridded (1° × 1°) data files for 133 months were retrieved for the period 2003–2014. TWS data were used together in
combination of three independent TWS solutions, from the Center for Space Research at the University of Texas at Austin, the NASA JPL and the German Space Agency
(Geoforschungszentrum, GFZ). The detailed steps to obtain TWS from gravity
measurements are available online1. The Satellite Laser Ranging approach is used to replace degree 2 and order 0 coefficients in the RL05 spherical harmonics data (Cheng and Tapley 2004). Swenson et al. (2008) described a process to derive the degree 1 coefficients. The effect of subsurface elastic deformation related to post glacial rebound is accounted
following A et al. (2013). A destriping filter is applied to remove correlated errors (Swenson et al., 2006). The data are processed with a Gaussian filter of 300 km width. In order to
1http://grace.jpl.nasa.gov/data/get-data/monthly-mass-grids-land/ [Accessed 26 April 2016]
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
account for signal damping caused by this processing, scale factors are multiplied with the TWS data.
The RL05 mascon solutions were also used to derive TWS change (Watkins et al. 2015;
Wiese et al. 2016). Similar techniques (as for the spherical harmonics – SH – products) are applied on the data for generating TWS information2. The mascon (MS) approach is different from the SH approach in terms of post-processing filter application. For example, in JPL’s mascon approach, the entire globe is characterized as ~3° spherical mass
concentration blocks with nearly equal area (Watkins et al. 2015). Use of a priori information facilitates correlated noise removal, which limits the use of post-processing filters (Watkins et al. 2015). Mascon products are not too much dependent on application of scale factors comparing the SH approach (Watkins et al. 2015). We applied scale factors with the TWS solutions.
Satellite-based groundwater storage anomaly signals can be disaggregated from the TWS anomalies (TWSA) after removing soil moisture (SMA) and surface water (SWA) anomalies:
GWSA = TWSA − SMA − SWA (2)
Snow has been rarely observed in the northern-most part of the study region; hence, we ignore snow in this analysis. Continuous, ground-based measurements of soil moisture and surface water equivalents are very scarce in the region. We used Global Land Data
Assimilation System (GLDAS) simulation outputs from NASA archives for estimating SMA and SWA (Rodell et al. 2004). An ensemble of three different models – the Community Land Model (CLM), Variable Infiltration Capacity (VIC), and Noah – is used to overcome the uncertainty (can appear as a function of different model physics) associated with any single model. Bhanja et al. (2016) showed better performance of the combination of models in contrast to any single model output comparing ground-based measurement in the study region. The satellite-based GWSA estimates in smaller basins are likely to have large errors because they are too small for GRACE to resolve and are only shown for comparison with the in situ observations.
3 Results and discussions
3.1 Precipitation patterns
The river basins are subjected to differential precipitation rates and patterns (Fig. 2, Supplementary material Fig. S2). Seasonal precipitation rates show distinct temporal variation with highest precipitation occurring in monsoon times, June-September, in most parts of the country; southern parts of India do receive substantial amount of precipitation in October-November (Mukherjee et al. 2015). Precipitation rates are not equal in all of the basins, making them climatologically distinct (Fig. 2, Fig. S2). Substantial differences are observed in pre- and post-monsoon precipitation across different basins. The basins are
2http://grace.jpl.nasa.gov/data/get-data/jpl_global_mascons/ [Accessed 26 April 2016]
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
hydrogeologically different (Fig. 3). The basins with higher dominance of alluvial soils have comparatively higher specific yield values and the basins located in fractured formations have lower specific yield values (Fig. 3).
A negative trend in long-term (1961–2014) SPI-1 was observed in basins 2a, 2b, 8, 12, 13, 18, 2c and 10 (Fig. 4, Fig. S3). A recent decrease in SPI-1 was observed in basins 2b and 2c (Fig. 4, Fig. S3). SPI-12 was found to be decreasing in 2003–2014 in basins 2a, 2b, 2c (Fig.
5, Fig. S4). Basins 2b and 2c have experienced meteorological drought in recent years (Fig.
5, Fig. S4). One intense drought event is observed in the period 2000–2005 in basins 5, 18 and 19 (Fig. 5, Fig. S4). The SPI-12 patterns show periodic nature with simultaneously high and low precipitation occurrences in Ganges basin (basin 2a) (Fig. 5, Fig. S4). However, in the Indus basin (basin 1), SPI-12 shows distinct high and lows extending for a long period of time (10–15 years; Fig. 5, Fig. S4). The basins (4, 14 and 19) are experiencing above normal precipitation in the study period (Fig. 5, Fig. S4).
3.2 Satellite-based groundwater storage anomalies
The satellite-based GWS anomalies (GWSAsat) reveal spatial and temporal variability in the study area (Supplementary material Figs S5 and S6). Both of the satellite-based estimates (SH and MS) show similar spatial patterns. A north–south distinction in the patterns of GWS anomaly was observed, particularly in the initial and final years (Figs 6 and 7, Figs S5 and S6). In the initial years (2003–2005), GWSAsat displayed strong positive values in northern India and negative values were observed in western and parts of southern India (Figs S5 and S6). However, in the final years (2012–2014), the GWSAsat anomaly shows strong negative values in northern India (Figs S5 and S6). This suggests rapid groundwater depletion in northern India similar to the findings of Rodell et al. (2009) (17.7 km3/year depletion in a 438 000-km2 region of northern India between 2002 and 2008) and Tiwari et al. (2009) (54 km3/year depletion in a 2 700 000-km2 region of northern and eastern India between 2002 and 2008). Groundwater replenishment was observed in western India and parts of southern India (Figs 6 and 7), similar to the observation of Bhanja et al. (2017b). The GWSAsat shows rapid decreasing trends in basins 2a and 2b at rates of −13.81 (−9.22) and −5.56 (−3.84) km3/year, respectively, from both of the satellite-based estimates – GRACE-MS (GRACE- SH) – in the period 2003–2014 (Figs 6-8). South Indian basins (3 and 4) are experiencing GWSAsat renewal at rates of 3.67 (2.91) and 1.83 (2.01) km3/year, respectively, from GRACE-MS (GRACE-SH) estimates (Figs 6-8). The results are also consistent with those reported by Rodell et al. (2018) and the groundwater level fluctuation observations of the Indian government authorities (CGWB 2014). The groundwater depletion in northern India is reported to be linked with irrigation water withdrawals, on the other hand, groundwater replenishment is found to be associated with the implementation of sustainable water management strategies (Bhanja et al. 2017b; Rodell et al. 2018).
Although we have shown GRACE based GWS change in all of the major river basins including smaller basins (Table 1; Fig. 1), the reader is advised to be cautious to use GRACE data in smaller basins. The native resolution of GRACE does not allow users to apply data for smaller basins directly. Therefore, GWSAsat computed for basins smaller than
approximately 150 000 km2 is likely to have large errors and values are only shown for
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
comparison with the in situ observations. Recent strategies including the simulation of land surface models along with the GRACE data will allow users to use it for finer resolutions (Landerer and Swenson 2012; Watkins et al. 2015). Other strategies could be useful for making GRACE-SH data available at higher resolution (Dutt Vishwakarma et al. 2016).
Scarcity of in situ measurement of the hydrological components has made the separation of GWSA from TWSA challenging (Scanlon et al. 2015).
3.3 Observed groundwater storage anomalies
In situ observations (GWSAobs) show depleting groundwater storage trends in most of the wells within the Indus-Ganges-Brahmaputra (IGB) basins in the period 2003–2014 (Fig. 8).
In contrast, parts of western and southern India show replenishment in most of the wells in 2003–2014 (Fig. 8). GWSAobs shows seasonal variability in all of the basins. In general, GWSAobs is lowest during the pre-monsoon season and highest during monsoon and post- monsoon seasons (Fig. 9). GWSAobs shows an increasing trend in recent years (2012–2014) in the Ganges (basin 2a) and Brahmaputra (basin 2b) basins, but trends over the entire study period are decreasing. The patterns of GWSAobs are similar to those reported by Panda and Wahr (2016), Bhanja et al. (2017b), Asoka et al. (2017), and Bhanja et al. (2018). GWSAsat and GWSAobs exhibit similar patterns of groundwater depletion and replenishment in most of the study basins (Table 2). One distinct difference in the two types of estimates is seen in basin 1. The observation wells are not evenly distributed in the basin, particularly in the northern part, which might explain the discrepancy. GWSA shows replenishment trends in most of the basins except for 2a, 2b, 5, 6 and 7 (Table 2). Despite similar magnitudes of GWS between GWSAsat and GWSAobs, the magnitudes of GWSAobs trends are found to be distinctly lower than those of GWSAsat. The net change in storage is found to be smaller in GWSAobs than in GWSAsat at the five largest basins. This might be due to the uneven distribution of observation wells. For example, many of the wells are located in the southern parts in basins 1 and 2a, where the trend is neutral or positive than in the north, where the trend is strongly negative according to GRACE (Figures 1, 6 and 7). Another reason might be associated with the shallower observation wells that under sample the deeper aquifers where larger trends exist. GWSA has rising trends in basins 4, 14 and 19 in satellite and in situ based observations for the period 2003–2014. The SPI indexes (both SPI-1 and SPI-12) indicate decreases in the early years and increases in final years of the study period in these basins (Figs 4 and 5). Increasing precipitation could be a reason for the rising GWS in basins 4, 14 and 19 (consistent with Rodell et al. 2018). GWSA has rising trends from both satellite and in situ estimates in basins 3 and 20, which have benefitted from sustainable water resource management policies (Bhanja et al. 2017b). Basins 2a and 2b have lost groundwater at rates of −1.02 and −0.40 km3/year, respectively, in the period 2003–2014. The IGB basin (basins 1, 2a and 2b) is mostly made up of unconsolidated sediments (Mukherjee et al. 2015) and includes areas of intense irrigation. It is also one of the most densely populated regions the world (Mukherjee et al. 2015). Groundwater demand has been increasing over the years due to the prevalence of water intensive crops (FAO 2013). Irrigated area to cropped area percentage is very high in the Indian states located at the IGB basin, i.e., Punjab (98%), Haryana (85%), Uttar Pradesh (76%), Bihar (61%), and West Bengal (56%) (MoA 2012).
As a result, a groundwater decline of 4 m or more has been observed during the last decade in Rajasthan, Punjab, Haryana, Delhi and West Bengal (CGWB 2014).
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
3.4 Relationship between precipitation indices and groundwater storage anomaly The trends of precipitation in the form of SPI-1 and SPI-12 values do not explain the GWSAobs trends in these two basins. For example, basins 2a and 2b experienced recent (2012–2014) droughts (SPI values), but the GWSAobs trends do not conform to the precipitation patterns (Fig.s 4, 5 and 9). We further studied the relationship between precipitation indices and the GWSAobs and the results are presented in Table 3. Good statistically significant (p < 0.05) correlation was observed in only six basins at zero lag (Table 3). Cross-correlation analysis indicates best relationship at zero lag only for SPI-1 (Table 3). Furthermore, statistically significant correlation (p < 0.05) was obtained for SPI-12 in both zero lag and 1-season lag at 13 of 22 basins (Table 3). In general, SPI-12 has better association with GWSAobs than SPI-1; this is because it is normalized over the 12- month period. Short-term (monthly) precipitation fluctuation has a smaller influence on groundwater storage than the long-term (12-monthly) fluctuation (NDMC 2018b).
4 Conclusions
Groundwater storage change has been studied in the 22 major river basins of India for the period 2003–2014. The basins are hydrogeologically and climatically heterogeneous. The standardized precipitation index (SPI) for 1-month data indicates a negative long-term (1961–2014) trend in basins 2a, 2b, 8, 12, 13, 18, 2c and 10, while positive long-term trends are observed in other basins. The 12-month SPI shows a decrease in precipitation in basins 2a, 2b, and 2c for 2003–2014; either mixed results (ups and downs) or increasing trends are obtained in other basins. The results indicate that the densely populated basins 2a (Ganges) and 2b (Brahmaputra) have lost groundwater in the period 2003–2014 at rates of −1.02 and
−0.40 km3/year based on in situ observations, respectively. The southern Indian basins display rising trends in both the satellite and in situ data. The results presented herein could be used to understand and predict long-term groundwater storage conditions across India and their relationships with patterns of precipitation, and hence could be used to formulate sustainable groundwater management strategies across the water stressed regions of India.
Supplementary Material
Refer to Web version on PubMed Central for supplementary material.
Acknowledgements
This manuscript uses freely-available data of the Central Ground Water Board (CGWB), Government of India. The opinion expressed in the paper is of authors’ own and not of the affiliated departments. We acknowledge CGWB India for providing water level data. GRACE land data were processed by Sean Swenson, supported by the NASA MEaSUREs Program, and are available at http://grace.jpl.nasa.gov. The GLDAS data used in this study were acquired as part of the mission of NASA's Earth Science Division and archived and distributed by the Goddard Earth Sciences (GES) Data and Information Services Center (DISC). Precipitation data are obtained from the India Meteorological Department (IMD).
References
Geruo A, Wahr J, and Zhong S, 2013 Computations of the viscoelastic response of a 3-D compressible Earth to surface loading: an application to Glacial Isostatic Adjustment in Antarctica and Canada.
Geophysical Journal International, 192(2), 557–572.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Asoka A, Gleeson T, Wada Y, and Mishra V, 2017 Relative contribution of monsoon precipitation and pumping to changes in groundwater storage in India. Nature Geoscience, 10(2), 109–117.
Bhanja SN, Mukherjee A, Rodell M, Velicogna I, Pangaluru K and Famiglietti JS, 2014 December.
Regional groundwater storage changes in the Indian subcontinent: The role of anthropogenic activities. In AGU Fall Meeting Abstracts.
Bhanja SN, Mukherjee A, Saha D, Velicogna I, and Famiglietti JS, 2016 Validation of GRACE based groundwater storage anomaly using in-situ groundwater level measurements in India. Journal of Hydrology, 10.1016/j.jhydrol.2016.10.042.
Bhanja SN, Mukherjee A, Rodell M, Wada Y, Chattopadhyay S, Velicogna I, Pangaluru K and Famiglietti JS, 2017 Groundwater rejuvenation in parts of India influenced by water-policy change implementation. Scientific reports, 7(1), 7453. [PubMed: 28785088]
Bhanja SN, Mukherjee A and Rodell M, 2018 Groundwater Storage Variations in India In Groundwater of South Asia (pp. 49–59). Springer, Singapore.
CGWB (Central Ground Water Board), 2014 Ground Water Year Book – India 2013-14. Central Ground Water Board, Government of India Ministry of Water Resources 76pp.
Cheng M, and Tapley BD, 2004Variations in the Earth’s oblateness during the past 28 years. J.
Geophys. Res, 109, B09402, doi:10.1029/2004JB003028.
Famiglietti JS, and Rodell M, 2013 Water in the Balance, Science, 340 (6138), 1300–1301, doi:
10.1126/science.1236460. [PubMed: 23766323]
Feng W, Zhong M, Lemoine JM, Biancale R, Hsu HT and Xia J, 2013 Evaluation of groundwater depletion in North China using the Gravity Recovery and Climate Experiment (GRACE) data and ground-based measurements. Water Resources Research, 49(4), 2110–2118.
Flechtner F, 2007 AOD1B Product Description Document for Product Releases 01 to 04 (Rev. 3.1, April 13, 2007). GRACE project document, 327–750.
Food and Agriculture Organization of the United Nations (FAO), 2013 FAO Statistical Yearbook 2013:
World Food and Agriculture, 289 pp.
India-WRIS, 2012 River Basin Atlas of India. RRSC-West, NRSC, ISRO, Jodhpur.
Landerer FW, and Swenson SC, 2012 Accuracy of scaled GRACE terrestrial water storage estimates.
Water Resources Research, 48, W04531, doi:10.1029/2011WR011453.
Margat J, and Van der Gun J, 2013 Groundwater around the world: a geographic synopsis. CRC Press.
MoA (Ministry of Agriculture), 2012 State of Indian agriculture 2011–12. Ministry of Agriculture, Government of India.
Mukherjee A, Saha D, Harvey CF, Taylor RG, Ahmed KM, and Bhanja SN, 2015 Groundwater systems of the Indian Sub-Continent. J. Hydrol. Regional Studies, 4, 1–14.
NDMC (National Drought Mitigation Center), 2018a Interpretation of 1-Month Standardized Precipitation Index Map. National Drought Mitigation Center, University of Nebraska, http://
drought.unl.edu/MonitoringTools/ClimateDivisionSPI/Interpretation/1month.aspx [Accessed 25 March 2018].
NDMC (National Drought Mitigation Center), 2018b Interpretation of 12-Month Standardized Precipitation Index Map. National Drought Mitigation Center, University of Nebraska, http://
drought.unl.edu/MonitoringTools/ClimateDivisionSPI/Interpretation/12month.aspx [Accessed 25 March 2018].
Pai DS, Sridhar L, Rajeevan M, Sreejith OP, Satbhai NS and Mukhopadhyay B, 2014 Development of a new high spatial resolution (0.25× 0.25) long period (1901–2010) daily gridded rainfall data set over India and its comparison with existing data sets over the region. Mausam, 65(1), pp.1–18.
Panda DK, and Wahr J 2016 Spatiotemporal evolution of water storage changes in India from the updated GRACE-derived gravity records. Water Resour. Res, 51, doi:10.1002/2015WR017797.
Rodell M, Houser PR, Jambor UEA, Gottschalck J, Mitchell K, Meng CJ, Arsenault K, Cosgrove B, Radakovich J, Bosilovich M and Entin JK, 2004 The global land data assimilation system. Bulletin of the American Meteorological Society, 85(3), 381–394.
Rodell M, Velicogna I, and Famiglietti JS, 2009 Satellite-based estimates of groundwater depletion in India.Nature, 460, 999–1002. [PubMed: 19675570]
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Rodell M, Famiglietti JS, Wiese DN, Reager JT, Beaudoing HK, Landerer FW and Lo MH, 2018 Emerging trends in global freshwater availability. Nature.
Scanlon BR, Zhang Z, Save H, Sun AY, Schmied HM, van Beek LP, Wiese DN, Wada Y, Long D, Reedy RC and Longuevergne L, 2018 Global models underestimate large decadal declining and rising water storage trends relative to GRACE satellite data. Proceedings of the National Academy of Sciences, p.201704665.
Sinha D, Syed TH, Famiglietti JS, Reager JT and Thomas RC, 2017 Characterizing drought in India using GRACE observations of terrestrial water storage deficit. Journal of Hydrometeorology, 18(2), pp.381–396.
Swenson SC , Chambers DP, and Wahr J, 2008 Estimating geocenter variations from a combination of GRACE and ocean model output. J. Geophys. Res.-Solid Earth, Vol 113, Issue: B8, B08410, doi:10.1029/2007JB005338.
Tapley BD, Bettadpur S, Ries JC, Thompson PF, and Watkins MM, 2004 GRACE measurements of mass variability in the Earth system. Science, 305, 503–505. [PubMed: 15273390]
Tiwari VM, Wahr J, and Swenson S, 2009 Dwindling groundwater resources in northern India, from satellite gravity observations. Geophys. Res. Lett. 36, L18401.
Tukey JW, 1977 Exploratory Data Analysis, Addison-Wesley.
Voss KA, Famiglietti JS, Lo M, Linage C. de, Rodell M, and Swenson SC, 2013 Groundwater depletion in the Middle East from GRACE with implications for transboundary water management in the Tigris-Euphrates-Western Iran region. Water Resour. Res, 49.
Wahr JM, 2007 Time-variable gravity from satellites. Treatise on Geophysics, 3, 213–237.
Watkins MM, Wiese DN, Yuan D-N, Boening C, and Landerer FW, 2015 Improved methods for observing Earth’s time variable mass distribution with GRACE using spherical cap mascons. J.
Geophys. Res. Solid Earth, 120, doi:10.1002/2014JB011547.
Wiese DN, Landerer FW and Watkins MM, 2016 Quantifying and reducing leakage errors in the JPL RL05M GRACE mascon solution, Water Resour. Res, 52, 7490–7502,
doi:10.1002/2016WR019344.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 1.
River basins numbers and groundwater observation wells in study region. Inset shows surrounding countries.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 2.
Basin-wise seasonal precipitation (mm) for the six largest basins. x-axis represents the seasons in 2003–2014 (four for each year). The basins are shown based on descending geographical area.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 3.
Boxplot of specific yield (Sy) in the studied basins. Mean Sy is represented as black dots.
The upper and lower limits of the boxes represent 1 standard deviation.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 4.
Basin-wise long-term (1961–2014) monthly SPI-1 values for the six largest basins. The basins are shown based on descending geographical area.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 5.
Basin-wise long-term (1961–2014) monthly SPI-12 values for the six largest basins. The basins are shown based on descending geographical area.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 6.
Annual trend in satellite-based GWS anomalies (cm) using GRACE-SH data.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 7.
Annual trend in satellite-based GWS anomalies (cm) using GRACE-MS data.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 8.
Linear regression trends for observed GWS (GWSobs) (cm) in the study region for the period 2003–2014.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Figure 9.
Basin-wise mean GWSAobs and GWSAsat (SH and MS) for the period 2003–2014.
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Table 1.
Basin name, ID, geographical area and number of observation wells used in this study. The basins are arranged based on descending geographical area.
Basin name Basin
ID
Area (km2)
Wells used
Ganges 2a 808334.4 1029
Indus (Indian part) 1 453931.9 248
Godavari 3 302063.9 393
Krishna 4 254743.3 363
Brahmaputra 2b 186421.6 109
West flowing rivers of Kutch and Saurashtra including Luni basin
20 184441.1 118
Mahanadi 8 139659.2 138
West flowing rivers south of Tapi basin 14 111643.9 214
Narmada 12 92670.5 91
Cauvery 5 85624.4 178
Tapi 13 63922.9 79
East flowing rivers between Pennar and Cauvery basins
18 63646.2 77
Pennar 9 54243.4 64
Brahmani and Baitarni 7 51893.7 80
East flowing rivers between Mahanadi and Godavari basins
15 46243.1 65
Barak and other basins 2c 45622.4 -
East flowing rivers south of Cauvery basin 19 38646.1 49
Mahi 10 38336.8 19
Sabarmati Basin 11 30678.6 46
Subarnarekha Basin 6 25792.2 17
East flowing rivers between Krishna and Pennar Basin
17 23335.8 23
East flowing rivers between Godavari and Krishna Basin
16 10345.2 20
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Table 2.
Basin-wide estimates of linear regression trends for GWSAobs and GWSAsat (SH and MS, see Section 2.3).
Basin ID GWSAobs
(km3/year)
GWSAsat(km3/year)
SH MS
2a −1.02 −9.22 −13.81
1 1.06 2.11 −1.52
3 0.71 2.91 3.67
4 0.67 2.01 1.83
2b −0.40 −3.84 −5.56
20 1.46 0.32 0.96
8 0.28 0.31 1.18
14 0.05
12 0.13 0.75 0.52
5 −0.01 −0.42 0.21
13 0.20 0.75 0.61
18 0.08 0.30 0.63
9 0.09 0.45 0.56
7 −0.12 −0.09 −0.64
15 0.10
2c −0.69 −0.34
19 0.04 0.05 0.11
10 0.18 0.29 0.02
11 0.23 0.27 0.47
6 −0.09 −0.10 −0.33
NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt NASA A uthor Man uscr ipt
Table 3.
Cross-correlation analysis between precipitation indices SPI-1 and SPI-12 and the observed groundwater storage. Bold indicates statistically significant correlation.
Basin ID
SPI-1 SPI-12
Lag 0 season
Lag 1 season
Lag 2 seasons
Lag 3 seasons
Lag 0 season
Lag 1 season
Lag 2 seasons
Lag 3 seasons
2a −0.09 −0.06 0.09 0.08 0.37* 0.30* 0.20 0.04
1 0.10 −0.13 0.07 −0.01 0.09 0.05 0.03 −0.03
3 0.01 −0.10 −0.17 0.03 0.37* 0.28* 0.16 −0.03
4 0.26* −0.12 0.05 0.17 0.40* 0.29* 0.18 0.05
2b 0.28* −0.02 −0.14 0.16 0.18 0.22 0.20 0.16
20 0.38* −0.14 −0.19 0.25* 0.55** 0.41* 0.26* 0.08
8 0.03 −0.20 0.04 0.06 0.19 0.14 0.02 0.03
14 0.10 −0.33* 0.13 0.40* 0.19 0.12 0.13 0.09
12 0.05 0.15 0.04 −0.01 0.17 0.16 0.11 0.03
5 0.34* 0.10 0.16 0.15 0.52** 0.53** 0.50** 0.40*
13 0.07 −0.08 −0.04 −0.05 0.28 0.22 0.11 −0.04
18 0.15 0.10 0.23 0.02 0.56** 0.53** 0.49** 0.33*
9 0.17 0.16 0.29* 0.15 0.62** 0.50** 0.30* 0.05
7 −0.09 −0.03 0.03 0.07 0.08 0.07 0.03 0.06
15 0.20 −0.14 0.01 0.19 0.45** 0.40* 0.24 0.09
2c 0.10 −0.35* −0.06 0.29* 0.20 0.21 0.25* 0.22
19 0.18 0.04 0.15 0.12 0.47** 0.45** 0.41* 0.33*
10 0.34* −0.02 −0.01 0.00 0.47** 0.43* 0.25* 0.14
11 0.43* −0.18 −0.15 0.12 0.46** 0.39* 0.24 0.13
6 −0.10 −0.15 0.11 0.14 0.21 0.16 0.16 0.07
17 0.21 0.11 0.20 0.07 0.64** 0.54** 0.39* 0.13
16 0.06 0.18 0.13 0.02 0.33* 0.27* 0.17 0.03
*and
**denote statistically significant results at 0.05 and 0.01 significance level, respectively.
