Machine-learning-based regional-scale groundwater level prediction using GRACE

Prévision du niveau des eaux souterraines à l’échelle régionale basée sur l’apprentissage automatique à l’aide de GRACE

Predicción del nivel de las aguas subterráneas a escala regional usando GRACE

利用机器学习方法和GRACE数据预测区域地下水水位

Previsão do nível de água subterrânea em escala regional baseada em aprendizado de máquina usando GRACE

Abstract

The rapid decline of groundwater levels (GWL) due to pervasive groundwater abstraction in the densely populated (~1 billion) Indus-Ganges-Brahmaputra-Meghna (IGBM) transboundary river basins of South Asia, necessitates a robust framework of prediction and understanding. While few localized studies exist, three-dimensional regional-scale characterization of GWL prediction is yet to be implemented. Here, ‘support vector machine’, a machine-learning-based method, is applied to data from the Gravity Recovery and Climate Experiment (GRACE) and data on land-surface-model-based groundwater storage and meteorological variables, to predict the GWL anomaly (GWLA) in the IGBM. The study has three main objectives, (1) to understand the spatial (observation well locations) and subsurface (shallow vs. deep observation wells) variability in prediction results for in-situ GWLA data for a large number of observation wells (n = 4,791); (2) to determine its relationship with groundwater abstraction, and; (3) to outline the advantages and limitations of using GRACE data for predicting GWLAs. The findings, based on individual observation well results, suggest significant prediction efficiency (median statistics: r > 0.71, NSE > 0.70; p < 0.05) in most of the IGBM; however, the study identifies hotspots, mostly in the agriculture-intensive regions, having relatively poor model performance. Further analysis of the subsurface depth-wise prediction statistics reveals that the significant dominance of pumping in the deeper depths of the aquifer is linked to the relatively poor model performance for the deep observation wells (screen depth > 35 m) compared with the shallow observation wells (screen depth < 35 m), thus, highlighting the limitation of GRACE in representing spatial and depth-dependent local-scale pumping.

Résumé

La baisse rapide des niveaux des eaux souterraines (NES) due à l’extraction omniprésente des eaux souterraines dans les bassins fluviaux transfrontaliers densément peuplés (~1 milliard) de l’Indus, du Gange, du Brahmapoutre et du Meghna (IGBM) d’Asie du Sud, nécessite un cadre solide de prévision et de compréhension. Bien qu’il existe peu d’études localisées, une caractérisation tridimensionnelle à l’échelle régionale de la prévision du NES est encore à réaliser. Ici, ‘les machines à vecteurs de support’, une méthode basée sur l’apprentissage automatique, est appliquée aux données de GRACE (Gravity Recovery and Climate Experiment) et aux données sur le stockage des eaux souterraines basées sur un modèle de surface terrestre et sur les variables météorologiques afin de prévoir les anomalies du NES (ANES) dans les IGBM. L’étude a trois objectifs principaux, (1) comprendre la variabilité dans l’espace (localisation des puits d’observation) et en profondeur (puits d’observation peu profonds vs. puits d’observation profonds) des résultats prédictifs des données sur les ANES in-situ sur un grand nombre de puits d’observation (n = 4,791); (2) déterminer sa relation avec l’exploitation des eaux souterraines et (3) cerner les avantages et les limites du recours aux données de GRACE pour prédire les ANES. Les constatations, basées sur les résultats des puits d’observation individuels, suggèrent une efficacité de prédiction significative (médianes statistiques: r > 0.71, NSE > 0.70; < 0.05) dans la plupart des IGBM; cependant, l’étude identifie des points sensibles, surtout dans les régions d’agriculture intensive, qui présentent des performances relativement médiocres du modèle. Une analyse plus poussée des statistiques de prédiction en profondeur sous la surface révèle que la prédominance significative du pompage aux profondeurs les plus importantes de l’aquifère est liée à la performance relativement médiocre du modèle pour les puits d’observation profonds (profondeur de la crépine > 35 m) par rapport aux puits d’observation peu profonds (profondeur de la crépine < 35 m), mettant en évidence les limites de GRACE à représenter l’impact de pompages tant spatialement qu’en fonction de la profondeur à l’échelle locale.

Resumen

La rápida profundización de los niveles de las aguas subterráneas (GWL) debido a la extracción generalizada de aguas subterráneas en las cuencas fluviales transfronterizas densamente pobladas (~1.000 millones) Indus-Ganges-Brahmaputra-Meghna (IGBM) del Asia meridional, requiere un marco sólido de predicción y comprensión. Aunque existen pocos estudios locales, todavía no se ha realizado una caracterización tridimensional a escala regional de la predicción del GWL. En este caso, la “herramienta de vectores de apoyo”, un método basado en el análisis de datos mediante computadoras, se aplica a los datos del Gravity Recovery and Climate Experiment (GRACE) y a los datos sobre el almacenamiento de aguas subterráneas basado en modelos de terreno y en variables meteorológicas, para predecir la anomalía del GWL (GWLA) en el IGBM. El estudio tiene tres objetivos principales: (1) comprender la variabilidad espacial (ubicación de los pozos de observación) y del subsuelo (pozos de observación someros frente a profundos) en los resultados de la predicción de los datos in situ de GWLA para un gran número de pozos de observación (n = 4,791); (2) determinar su relación con la extracción de aguas subterráneas, y; (3) esbozar las ventajas y limitaciones de la utilización de los datos GRACE para predecir la GWLA. Los resultados, basados en los datos de los pozos de observación individuales, sugieren una eficiencia de predicción significativa (mediana: r > 0.71, NSE > 0.70; p < 0.05) en la mayoría de los GWLA; sin embargo, el estudio identifica puntos calientes, principalmente en las regiones de agricultura intensiva, que tienen un rendimiento del modelo relativamente pobre. Un análisis más detallado de las estadísticas de predicción de la profundidad del agua subterránea revela que el significativo predominio del bombeo en las profundidades del acuífero está vinculado al rendimiento relativamente deficiente del modelo para los pozos de observación profundos (profundidad de los filtros > 35 m) en comparación con los pozos de observación poco profundos (profundidad de los filtros < 35 m), lo que pone de relieve la limitación de GRACE para representar el bombeo a escala local dependiente del espacio y la profundidad.

摘要

在人口密集(约10亿)的印度河-恒河-布拉马普特拉河-梅克纳河(IGBM)跨界流域,由于地下水的大量抽取,地下水位(GWL)迅速下降,因此需要一个强有力的预测和理解地下水位动态的框架。尽管当地已有一些研究,但是三维区域地下水位动态特征预测研究仍有待开展。本文将支持向量机模型应用于GRACE数据和基于陆面模型的地下水储量和气象数据,对IGBM的地下水位变化(GWLA)进行预测。这项研究有三个主要目的,(1)了解大量观测井(n = 4,791)实地地下水位变化数据预测结果的空间(观测井位置)和地下(浅层和深层观测井)的变异性;(2)确定地下水位变化与地下水开采之间的关系;(3)概述用GRACE数据预测地下水位变化的优点与局限性。基于独立观测井的结果表明,在IGBM的大多数地区预测效率显著(中值统计:r > 0.71,NSE>0.70;< 0.05),然而在农业密集地区的大多数地区,模型性能较差。对地下不同深度地下水位预测统计的深入分析表明,与浅部观测井(滤管深度<35m)相比,大量深层含水层开采与深部观测井(滤管深度>35m)的模型性能较差,因此,凸显了GRACE在表征空间和深度主导的局部尺度抽水方面的局限性。

Resumo

O rápido declínio dos níveis de água subterrânea (NAS) devido à captação generalizada de água subterrânea nas bacias hidrográficas transfronteiriças densamente povoadas (~1 bilhão) Indus-Ganges-Brahmaputra-Meghna (IGBM) do Sul da Ásia, necessita de uma estrutura robusta de previsão e compreensão. Embora existam poucos estudos localizados, a caracterização em escala regional tridimensional da previsão de NAS ainda está para ser implementada. Aqui, ‘máquina de vetor de suporte’, um método baseado em aprendizado de máquina, é aplicado a dados do Experimento de Recuperação de Gravidade e Clima (GRACE) e dados sobre armazenamento de água subterrânea baseado em modelo de superfície terrestre e variáveis ​​meteorológicas, para prever anomalias no NAS (ANAS) na IGBM. O estudo tem três objetivos principais, (1) compreender a variabilidade espacial (localizações de poços de observação) e de subsuperfície (poços de observação rasos vs. profundos) nos resultados de previsão para dados ANAS in-situ para um grande número de poços de observação (n = 4,791); (2) determinar sua relação com a captação de águas subterrâneas, e; (3) descrever as vantagens e limitações do uso de dados GRACE para a previsão de ANAS. Os achados, com base em resultados de poços de observação individual, sugerem eficiência de predição significativa (estatística média: r > 0.71, EPN > 0.70; p < 0.05) na maior parte do IGBM; no entanto, o estudo identifica pontos quentes, principalmente nas regiões de agricultura intensiva, com desempenho de modelo relativamente baixo. Uma análise mais aprofundada das estatísticas de previsão em profundidade de subsuperfície revela que a dominância significativa de bombeamento nas profundidades mais profundas do aquífero está ligada ao desempenho do modelo relativamente pobre para os poços de observação profundos (profundidade do filtro > 35 m) em comparação com os rasos poços de observação (profundidade do filtro < 35 m), destacando assim a limitação do GRACE em representar o bombeamento em escala local dependente da profundidade e espacial.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

References

  1. Aeschbach-Hertig W, Gleeson T (2012) Regional strategies for the accelerating global problem of groundwater depletion. Nat Geosci 5:853–861. https://doi.org/10.1038/ngeo1617

    Article  Google Scholar 

  2. Alley WM, Reilly TE, Franke OL (1999) Sustainability of ground-water resources. US Geol Surv Circ 1186. http://pubs.usgs.gov/circ/circ1186/pdf/circ1186.pdf. Accessed 10 Sept 2020

  3. Azur MJ, Stuart EA, Frangakis C, Leaf PJ (2011) Multiple imputation by chained equations: what is it and how does it work? Int J Methods Psychiatr Res 20:40–49. https://doi.org/10.1002/mpr.329

    Article  Google Scholar 

  4. Bhanja SN, Mukherjee A (2019) In situ and satellite-based estimates of usable groundwater storage across India: implications for drinking water supply and food security. Adv Water Resour 126:15–23. https://doi.org/10.1016/j.advwatres.2019.02.001

    Article  Google Scholar 

  5. Bhanja SN, Mukherjee A, Saha D, Velicogna I, Famiglietti JS (2016) Validation of GRACE based groundwater storage anomaly using in-situ groundwater level measurements in India. J Hydrol 543:729–738. https://doi.org/10.1016/j.jhydrol.2016.10.042

    Article  Google Scholar 

  6. Bhanja SN, Mukherjee A, Rodell M, Wada Y, Chattopadhyay S, Velicogna I, Pangaluru K, Famiglietti JS (2017a) Groundwater rejuvenation in parts of India influenced by water-policy change implementation. Sci Rep 7:7453. https://doi.org/10.1038/s41598-017-07058-2

    Article  Google Scholar 

  7. Bhanja SN, Rodell M, Li B, Saha D, Mukherjee A (2017b) Spatio-temporal variability of groundwater storage in India. J Hydrol 544:428–437. https://doi.org/10.1016/j.jhydrol.2016.11.052

    Article  Google Scholar 

  8. Bhanja SN, Mukherjee A, Rodell M (2018) Groundwater storage variations in India. In: Mukherjee A (ed) Groundwater of South Asia. Springer, Singapore, pp 49–59

    Google Scholar 

  9. Bhanja SN, Mukherjee A, Rangarajan R, Scanlon BR, Malakar P, Verma S (2019a) Long-term groundwater recharge rates across India by in situ measurements. Hydrol Earth Syst Sci 23:711–722. https://doi.org/10.5194/hess-23-711-2019

    Article  Google Scholar 

  10. Bhanja SN, Malakar P, Mukherjee A, Rodell M, Mitra P, Sarkar S (2019b) Using satellite-based vegetation cover as indicator of groundwater storage in natural vegetation areas. Geophys Res Lett 46:8082–8092. https://doi.org/10.1029/2019GL083015

    Article  Google Scholar 

  11. Bhanja SN, Mukherjee A, Rodell M (2020) Groundwater storage change detection from in situ and GRACE-based estimates in major river basins across India. Hydrol Sci J 65:650–659. https://doi.org/10.1080/02626667.2020.1716238

    Article  Google Scholar 

  12. Bonsor HC, MacDonald AM, Ahmed KM, et al (2017) Hydrogeological typologies of the Indo-Gangetic basin alluvial aquifer, South Asia. Hydrogeol J 25:1377–1406. https://doi.org/10.1007/s10040-017-1550-z

  13. Castellazzi P, Martel R, Galloway DL, Longuevergne L, Rivera A (2016) Assessing groundwater depletion and dynamics using GRACE and InSAR: potential and limitations. Groundwater 54:768–780. https://doi.org/10.1111/gwat.12453

    Article  Google Scholar 

  14. Faridabad BM Jha (ed) (2007) Ground water scenario, India. CGWB, Ministry of Water Resources, Government of India, New Delhi

  15. Central Ground Water Board (CGWB) (2014) Ground water year book: India 2013–14. Ministry of Water Resources, Government of India, New Delhi, 76 pp

  16. Central Ground Water Board (CGWB) (2017) Dynamic groundwater resources. Ministry of Water Resources, Government of India, New Delhi, 306 pp

  17. Chen J, Li J, Zhang Z, Ni S (2014) Long-term groundwater variations in Northwest India from satellite gravity measurements. Glob Planet Chang 116:130–138. https://doi.org/10.1016/j.gloplacha.2014.02.007

    Article  Google Scholar 

  18. Chen H, Zhang W, Nie N, Guo Y (2019) Long-term groundwater storage variations estimated in the Songhua River Basin by using GRACE products, land surface models, and in-situ observations. Sci Total Environ 649:372–387. https://doi.org/10.1016/j.scitotenv.2018.08.352

    Article  Google Scholar 

  19. Climatic Research Unit (CRU) (2018) University of East Anglia. https://crudata.uea.ac.uk/cru/data/hrg/cru_ts_4.01/. Accessed 30 July 2018

  20. Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20:273–297. https://doi.org/10.1023/A:1022627411411

    Article  Google Scholar 

  21. Coulibaly P, Anctil F, Aravena R, Bobée B (2001) Artificial neural network modeling of water table depth fluctuations. Water Resour Res 37:885–896. https://doi.org/10.1029/2000WR900368

    Article  Google Scholar 

  22. Cuthbert MO, Gleeson T, Moosdorf N, Befus KM, Schneider A, Hartmann J, Lehner B (2019) Global patterns and dynamics of climate–groundwater interactions. Nat Clim Chang 9:137–141. https://doi.org/10.1038/s41558-018-0386-4

    Article  Google Scholar 

  23. Famiglietti JS (2014) The global groundwater crisis. Nat Clim Chang 4:945–948. https://doi.org/10.1038/nclimate2425

    Article  Google Scholar 

  24. Famiglietti JS, Lo M, Ho SL, Bethune J, Anderson KJ, Syed TH, Swenson SC, De Linage CR, Rodell M (2011) Satellites measure recent rates of groundwater depletion in California’s Central Valley. Geophys Res Lett 38:L03403. https://doi.org/10.1029/2010GL046442

    Article  Google Scholar 

  25. Famiglietti JS, Cazenave A, Eicker A, Reager JT, Rodell M, Velicogna I (2015) Satellites provide the big picture. Science 349:684–685. https://doi.org/10.1126/science.aac9238

    Article  Google Scholar 

  26. Fendorf S, Michael HA, Van Geen A (2010) Spatial and temporal variations of groundwater arsenic in south and Southeast Asia. Science 328:1123–1127. https://doi.org/10.1126/science.1172974

    Article  Google Scholar 

  27. Garduño H, Romani S, Sengupta B, Tuinhof A, Davis (2011). India groundwater governance case study. Water papers, World Bank, Washington, DC. https://openknowledge.worldbank.org/handle/10986/17242. Accessed February 2021

  28. Gibrilla A, Anornu G, Adomako D (2018) Trend analysis and ARIMA modelling of recent groundwater levels in the White Volta River basin of Ghana. Groundw Sustain Dev 6:150–163. https://doi.org/10.1016/j.gsd.2017.12.006

    Article  Google Scholar 

  29. Girotto M, GJM DL, Reichle RH, Rodell M, Draper C, Bhanja SN, Mukherjee A (2017) Benefits and pitfalls of GRACE data assimilation: a case study of terrestrial water storage depletion in India. Geophys Res Lett 44:4107–4115. https://doi.org/10.1002/2017GL072994

    Article  Google Scholar 

  30. Global Land Data Assimilation System (GLDAS) (2021) https://ldas.gsfc.nasa.gov/gldas. Accessed 26 April 2016

  31. Harris I, Osborn TJ, Jones P, Lister D (2020) Version 4 of the CRU TS monthly high-resolution gridded multivariate climate dataset. Sci Data 7:109. https://doi.org/10.1038/s41597-020-0453-3

    Article  Google Scholar 

  32. Huang Z, Pan Y, Gong H, Yeh PJF, Li X, Zhou D, Zhao W (2015) Subregional-scale groundwater depletion detected by GRACE for both shallow and deep aquifers in North China Plain. Geophys Res Lett 42:1791–1799. https://doi.org/10.1002/2014GL062498

    Article  Google Scholar 

  33. Indian Meteorological Department (IMD) (2018) https://mausam.imd.gov.in/imd_latest/contents/rainfallinformation.php. Accessed 25 July 2018

  34. ISO (2021) ISO codes. https://www.iso.org/obp/ui. Accessed February 2021

  35. Kumar M, Mishra DC, Singh B (2013) Lithosphere, crust and basement ridges across ganga and Indus basins and seismicity along the Himalayan front, India and Western Fold Belt, Pakistan. J Asian Earth Sci 75:126–140. https://doi.org/10.1016/j.jseaes.2013.07.004

  36. Lapworth DJ, MacDonald AM, Krishan G, Rao MS, Gooddy DC, Darling WG (2015) Groundwater recharge and age-depth profiles of intensively exploited groundwater resources in Northwest India. Geophys Res Lett 42:7554–7562. https://doi.org/10.1002/2015GL065798

    Article  Google Scholar 

  37. MacDonald AM, Bonsor HC, Taylor R, Shamsudduha M, Burgess WG, AhmedKM MA, Zahid A, Lapworth D, Gopal K, Rao MS, Moench M, Bricker SH, Yadav SK, Satyal Y, Smith L, Dixit A, Bell R, van Steenbergen F, Basharat M, GoharMS TJ, Calow RC, Maurice L (2015) Groundwater resources in the Indo-Gangetic Basin: resilience to climate change and abstraction. BGS open report, OR/15/047, British Geological Survey, Keyworth, UK, 58 pp

    Google Scholar 

  38. MacDonald AM, Bonsor HC, Ahmed KM, Burgess WG, Basharat M, Calow RC, Dixit A, SSD F, Gopal K, Lapworth DJ, Lark RM, Moench M, Mukherjee A, Rao MS, Shamsudduha M, Smith L, Taylor RG, Tucker J, van Steenbergen F, Yadav SK (2016) Groundwater depletion and quality in the Indo-Gangetic Basin from in situ observations. Nat Geosci 9:762–766. https://doi.org/10.1038/ngeo2791

    Article  Google Scholar 

  39. Malakar P, Mukherjee A, Sarkar S (2018) Potential application of advanced computational techniques in prediction of groundwater resource of India. In: Mukherjee A (ed) Groundwater of South Asia. Springer, Singapore, pp 643–655

    Google Scholar 

  40. Malakar P, Sarkar S, Mukherjee A, Bhanja SN, Sun AY (2020a) Use of machine learning and deep learning methods in groundwater. In: Mukherjee A (ed) Global groundwater. Elsevier, Amsterdam, pp 545–557

    Google Scholar 

  41. Malakar P, Mukherjee A, Bhanja SN, Saha D, Ray RK, Sarkar S, Zahid A (2020b) Importance of spatial and depth-dependent drivers in groundwater level modeling through machine learning. Hydrol Earth Syst Sci. https://doi.org/10.5194/hess-2020-208. Accessed February 2021

  42. Malakar P, Mukherjee A, Bhanja SN, Ganguly AR, Ray RK, Zahid A, Sarkar S, Saha D, Chattopadhyay S (2021) Three decades of depth-dependent groundwater response to climate variability and human regime in the transboundary Indus-Ganges-Brahmaputra-Meghna mega river basin aquifers. Advances in Water Resources 149:103856

  43. Minor Irrigation Census (2014) Report of the 4th minor irrigation census (2006–2007). Ministry of Water Resources, Government of India, New Delhi

  44. Minor Irrigation Census (2017) Report of the 5th minor irrigation census (2013–2014). Ministry of Water Resources, Government of India, New Delhi

  45. MoA (Ministry of Agriculture) (2012) State of Indian agriculture 2011–12. Ministry of Agriculture, Government of India, New Delhi

    Google Scholar 

  46. Mukherjee A, Ramachandran P (2018) Prediction of GWL with the help of GRACE TWS for unevenly spaced time series data in India: analysis of comparative performances of SVR, ANN and LRM. J Hydrol 558:647–658. https://doi.org/10.1016/j.jhydrol.2018.02.005

    Article  Google Scholar 

  47. Mukherjee A, Fryar AE, Howell PD (2007) Regional hydrostratigraphy and groundwater flow modeling in the arsenic-affected areas of the western Bengal basin, West Bengal, India. Hydrogeol J 15:1397–1418. https://doi.org/10.1007/s10040-007-0208-7

    Article  Google Scholar 

  48. Mukherjee A, Saha D, Harvey CF, Taylor RG, Ahmed KM, Bhanja SN (2015) Groundwater systems of the Indian sub-continent. J Hydrol Reg Stud 4:1–14. https://doi.org/10.1016/j.ejrh.2015.03.005

    Article  Google Scholar 

  49. Mukherjee A, Bhanja SN, Wada Y (2018) Groundwater depletion causing reduction of baseflow triggering Ganges River summer drying. Sci Rep 8:12049. https://doi.org/10.1038/s41598-018-30246-7

    Article  Google Scholar 

  50. NASA (2016) Monthly mass grids: global mascons (JPL RL06_v02). http://grace.jpl.nasa.gov/data/get-data/jpl_global_mascons/. Accessed 26 April 2016

  51. Pai DS, Sridhar L, Rajeevan M et al (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 65:1–18

    Google Scholar 

  52. Panda DK, Wahr J (2016) Spatiotemporal evolution of water storage changes in India from the updated GRACE-derived gravity records. Water Resour Res 52:135–149. https://doi.org/10.1002/2015WR017797

    Article  Google Scholar 

  53. Raghavendra S, Deka PC (2014) Support vector machine applications in the field of hydrology: a review. Appl Soft Comput 19:372–386. https://doi.org/10.1016/j.asoc.2014.02.002

    Article  Google Scholar 

  54. Reichstein M, Camps-Valls G, Stevens B, Jung M, Denzler J, Carvalhais N, Prabhat A (2019) Deep learning and process understanding for data-driven earth system science. Nature 566:195–204. https://doi.org/10.1038/s41586-019-0912-1

    Article  Google Scholar 

  55. Resche-Rigon M, White IR (2018) Multiple imputation by chained equations for systematically and sporadically missing multilevel data. Stat Methods Med Res 27:1634–1649. https://doi.org/10.1177/0962280216666564

    Article  Google Scholar 

  56. Rodell M, Houser PR, Jambor U, Gottschalck J, Mitchell K, Meng CJ, Arsenault K, Cosgrove B, Radakovich J, Bosilovich M, Entin JK, Walker JP, Lohmann D, Toll D (2004) The global land data assimilation system. Bull Am Meteorol Soc 85:381–394. https://doi.org/10.1175/BAMS-85-3-381

    Article  Google Scholar 

  57. Rodell M, Velicogna I, Famiglietti JS (2009) Satellite-based estimates of groundwater depletion in India. Nature 460:999–1002. https://doi.org/10.1038/nature08238

    Article  Google Scholar 

  58. Rodell M, Famiglietti JS, Wiese DN, Reager JT, Beaudoing HK, Landerer FW, Lo MH (2018) Emerging trends in global freshwater availability. Nature 557:651–659. https://doi.org/10.1038/s41586-018-0123-1

    Article  Google Scholar 

  59. Sathre, R, Antharam, MS, Catena M, Buluswar S (2018) Technology breakthroughs for global water security: a deep dive into South Asia. Institute for Transformative Technologies (ITT). www.transformativetechnologies.org. Accessed February 2021

  60. Scanlon BR, Longuevergne L, Long D (2012) Ground referencing GRACE satellite estimates of groundwater storage changes in the California Central Valley, USA. Water Resour Res 48:1–9. https://doi.org/10.1029/2011WR011312

    Article  Google Scholar 

  61. Scanlon BR, Zhang Z, Save H, Sun AY, Schmied HM, LPH VB, Wiese DN, Wada Y, Long D, Reedy RC, Longuevergne L, Döll P, Bierkens MFP (2018) Global models underestimate large decadal declining and rising water storage trends relative to GRACE satellite data. Proc Natl Acad Sci U S A 115:E1080–E1089. https://doi.org/10.1073/pnas.1704665115

    Article  Google Scholar 

  62. Schewe J, Heinke J, Gerten D, Haddeland I, Arnell NW, Clark DB, Dankers R, Eisner S, Fekete BM, Colón-González FJ, Gosling SN, Kim H, Liu X, Masaki Y, Portmann FT, Satoh Y, Stacke T, Tang Q, Wada Y, Wisser D, Albrecht T, Frieler K, Piontek F, Warszawski L, Kabat P (2014) Multimodel assessment of water scarcity under climate change. Proc Natl Acad Sci U S A 111:3245–3250. https://doi.org/10.1073/pnas.1222460110

    Article  Google Scholar 

  63. Shah T (2009) Climate change and groundwater: India’s opportunities for mitigation and adaptation. Environ Res Lett 4:035005. https://doi.org/10.1088/1748-9326/4/3/035005

    Article  Google Scholar 

  64. Shamsudduha M, Taylor RG, Ahmed KM, Zahid A (2011) The impact of intensive groundwater abstraction on recharge to a shallow regional aquifer system: evidence from Bangladesh. Hydrogeol J 19:901–916. https://doi.org/10.1007/s10040-011-0723-4

    Article  Google Scholar 

  65. Siebert S, Burke J, Faures JM, Frenken K, Hoogeveen J, Döll P, Portmann FT (2010) Groundwater use for irrigation: a global inventory. Hydrol Earth Syst Sci 14:1863–1880. https://doi.org/10.5194/hess-14-1863-2010

    Article  Google Scholar 

  66. Singh AK, Tripathi JN, Kotlia BS, Singh KK, Kumar A (2019) Monitoring groundwater fluctuations over India during Indian Summer Monsoon (ISM) and Northeast monsoon using GRACE satellite: impact on agriculture. Quat Int 507:342–351. https://doi.org/10.1016/j.quaint.2018.10.036

    Article  Google Scholar 

  67. Slavíková L, Malý V, Rost M, Petružela L, Vojáček O (2013) Impacts of climate variables on residential water consumption in the Czech Republic. Water Resour Manag 27:365–379. https://doi.org/10.1007/s11269-012-0191-1

    Article  Google Scholar 

  68. Srivastava P, Pal DK, Aruche KM, Wani SP, Sahrawat KL (2015) Soils of the Indo-Gangetic Plains: a pedogenic response to landscape stability, climatic variability and anthropogenic activity during the Holocene. Earth Sci Rev 140:54–71. https://doi.org/10.1016/j.earscirev.2014.10.010

    Article  Google Scholar 

  69. Stocker T, editor. Climate change 2013: the physical science basis: Working Group I contribution to the Fifth assessment report of the Intergovernmental Panel on Climate Change. Cambridge university press; 2014 Mar 24

  70. Sun AY (2013) Predicting groundwater level changes using GRACE data. Water Resour Res 49:5900–5912. https://doi.org/10.1002/wrcr.20421

    Article  Google Scholar 

  71. Sun AY, Scanlon BR, Zhang Z, Walling D, Bhanja SN, Mukherjee A, Zhong Z (2019) Combining physically based modeling and deep learning for fusing GRACE satellite data: can we learn from mismatch? Water Resour Res 55:1179–1195. https://doi.org/10.1029/2018WR023333

    Article  Google Scholar 

  72. Swenson S, Wahr J (2002) Methods for inferring regional surface-mass anomalies from Gravity Recovery and Climate Experiment (GRACE) measurements of time-variable gravity. J Geophys Res Solid Earth 107:ETG 3-1-ETG 3-13. https://doi.org/10.1029/2001jb000576

    Article  Google Scholar 

  73. Swenson S, Wahr J, Milly PCD (2003) Estimated accuracies of regional water storage variations inferred from the Gravity Recovery and Climate Experiment (GRACE). Water Resour Res 39:1223. https://doi.org/10.1029/2002WR001808

    Article  Google Scholar 

  74. Tapley BD, Bettadpur S, Watkins M, Reigber C (2004) The gravity recovery and climate experiment: mission overview and early results. Geophys Res Lett 31:L09607. https://doi.org/10.1029/2004GL019779

    Article  Google Scholar 

  75. Taylor RG, Scanlon B, Döll P, Rodell M, Van Beek R, Wada Y, Longuevergne L, Leblanc M, Famiglietti JS, Edmunds M, Konikow L, Green TR, Chen J, Taniguchi M, Bierkens MFP, Macdonald A, Fan Y, Maxwell RM, Yechieli Y, Gurdak JJ, Allen DM, Shamsudduha M, Hiscock K, Yeh PJF, Holman I, Treidel H (2013) Ground water and climate change. Nat Clim Chang 3:322–329. https://doi.org/10.1038/nclimate1744

    Article  Google Scholar 

  76. Tiwari VM, Wahr J, Swenson S (2009) Dwindling groundwater resources in northern India, from satellite gravity observations. Geophys Res Lett 36:1–5. https://doi.org/10.1029/2009GL039401

    Article  Google Scholar 

  77. Tukey JW (1977) Exploratory data analysis. Addison-Wesley, Reading, UK

    Google Scholar 

  78. Van Buuren S, Groothuis-Oudshoorn K (2011) mice: Multivariate imputation by chained equations in R. J Stat Softw 45:1–67. https://doi.org/10.18637/jss.v045.i03

    Article  Google Scholar 

  79. Vapnik VN (1995) The nature of statistical learning theory. Springer, New York

    Google Scholar 

  80. Vapnik VN (1998) Statistical learning theory. Wiley, New York

    Google Scholar 

  81. Wang WC, Chau KW, Cheng CT, Qiu L (2009) A comparison of performance of several artificial intelligence methods for forecasting monthly discharge time series. J Hydrol 374:294–306. https://doi.org/10.1016/j.jhydrol.2009.06.019

    Article  Google Scholar 

  82. Watkins MM, Wiese DN, Yuan DN, Boening C, Landerer FW (2015) Improved methods for observing Earth’s time variable mass distribution with GRACE using spherical cap mascons. J Geophys Res Solid Earth 120:2648–2671. https://doi.org/10.1002/2014JB011547

    Article  Google Scholar 

  83. Wiese DN, Landerer FW, Watkins MM (2016) Quantifying and reducing leakage errors in the JPL RL05M GRACE mascon solution. Water Resour Res 52:7490–7502. https://doi.org/10.1002/2016WR019344

    Article  Google Scholar 

  84. World Bank (2010) Deep Wells and Prudence: Towards Pragmatic Action for Addressing Groundwater Overexploitation in India. Int Bank Reconstr Dev World Bank 120

  85. Yi S, Wang Q, Sun W (2016) Basin mass dynamic changes in China from GRACE based on a multibasin inversion method. J Geophys Res Solid Earth 121:3782–3803. https://doi.org/10.1002/2015JB012608

    Article  Google Scholar 

  86. Yoon H, Jun SC, Hyun Y, Bae GO, Lee KK (2011) A comparative study of artificial neural networks and support vector machines for predicting groundwater levels in a coastal aquifer. J Hydrol 396:128–138. https://doi.org/10.1016/j.jhydrol.2010.11.002

    Article  Google Scholar 

Download references

Acknowledgements

The authors acknowledge the Central Ground Water Board (Ministry of Jal Shakti, Department of Water Resources, River Development and Ganga Rejuvenation) of the Government of India and Bangladesh Water Development Board of the Government of Bangladesh for data support. The authors also acknowledge the India Meteorological Department (IMD) and Climatic Research Unit (CRU). GRACE land data were processed by Sean Swenson, supported by the NASA MEaSUREs Program. 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). The authors are thankful to Matt Rodell of NASA and Srimanti Duttagupta IIT Kharagpur. The authors acknowledge the use of ArcGIS software (version 10.2.1), Origin software (version 2015), R statistical software, and Ferret program (Pacific Marine Environmental Laboratory NOAA) for analysis.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Abhijit Mukherjee.

Additional information

Publisher’s note

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

Supplementary Information

ESM 1

(PDF 598 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Malakar, P., Mukherjee, A., Bhanja, S.N. et al. Machine-learning-based regional-scale groundwater level prediction using GRACE. Hydrogeol J (2021). https://doi.org/10.1007/s10040-021-02306-2

Download citation

Keywords

  • Groundwater level anomaly prediction
  • Machine learning
  • Satellite imagery
  • Transboundary aquifer
  • Groundwater exploration