Top

Hydrogeology Journal

Published in:

17-01-2022 | Paper

An autoregressive machine learning approach to forecast high-resolution groundwater-level anomalies in the Ramotswa/North West/Gauteng dolomite aquifers of Southern Africa

Authors: Zaheed Gaffoor, Ashley Gritzman, Kevin Pietersen, Nebo Jovanovic, Antoine Bagula, Thokozani Kanyerere

Published in: Hydrogeology Journal | Issue 2/2022

Activate our intelligent search to find suitable subject content or patents.

search-config

AI-assisted search

Off

Abstract

A novel approach using machine learning algorithms is used to predict high-resolution groundwater-level changes, both current and in the future. The Southern African Development Community (SADC) is dependent on groundwater resources to meet freshwater demand but knowledge of groundwater resources in the region is hampered by the lack of available data, especially at a local scale. Improving the ability to generate near-real-time and high-resolution groundwater data is imperative to support effective groundwater management in the SADC. In this study, an autoregressive approach is applied to develop a set of generalized gradient-boosting decision tree models that can predict 30-day groundwater-level changes, both current and 1 month ahead. The approach integrates numerous readily available satellite-based, land-surface-model, and hydrogeological datasets to train, calibrate and test the gradient-boosting decision tree models. The models achieve a mean absolute error of 15.27 and 12.78 cm, for the groundwater-level-change predictions that are current and 1 month ahead, respectively. The models were used to predict monthly groundwater-level changes, both current and 1 month ahead, at a 10-km resolution, in the Ramotswa/North West/Gauteng dolomite aquifers of Southern Africa. When compared to observed in-situ groundwater-level changes, the mean absolute error for the current and 1-month-ahead groundwater-level changes were 21.23 and 22.26 cm, respectively. While the results are largely satisfactory, challenges associated with limited target data, and extreme values associated with groundwater abstraction, reduce model performance. Nonetheless, the approach demonstrates the applicability of machine learning to generate local groundwater information to improve groundwater management.

previous article Optimizing river damming and impounding strategies to mitigate seawater intrusion in the coastal aquifer of Dagu River Basin, China

next article An improved Bayesian approach linked to a surrogate model for identifying groundwater pollution sources

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 390 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Maschinenbau + Werkstoffe

Jetzt Wissensvorsprung sichern!

inform now

Altchenko Y, Genco A, Pierce K, Woolf R, Nijsten G-J, Ansems N, Magombeyi M, Ebrahim G, Lautze J, Villholth KG, Lefore N, Modisha RCO, Baqa S, McGill BM, Kenabatho P (2017) Resilience in the Limpopo Basin: the potential role of the transboundary Ramotswa aquifer. United States Agency for International Development, Pretoria, South Africa

Amato F, Guignard F, Robert S, Kanevski M (2020) A novel framework for spatio-temporal prediction of environmental data using deep learning. Sci Rep 10:22243. https://doi.org/10.1038/s41598-020-79148-7CrossRef

Atkinson PM (2013) Downscaling in remote sensing. Int J Appl Earth Obs Geoinform 22:106–114. https://doi.org/10.1016/j.jag.2012.04.012CrossRef

Baqa SS (2017) Groundwater recharge assessment in the Upper Limpopo River Basin: a case study in Ramotswa Dolomitic Aquifer. University of the Witwatersrand, Johannesburg, South Africa

Bazgir O, Zhang R, Dhruba SR, Rahman R, Ghosh S, Pal R (2020) Representation of features as images with neighborhood dependencies for compatibility with convolutional neural networks. Nat Commun 11:4391. https://doi.org/10.1038/s41467-020-18197-yCrossRef

Beck HE, Zimmermann NE, McVicar TR, Vergopolan N, Berg A, Wood EF (2018) Present and future Köppen-Geiger climate classification maps at 1-km resolution. Sci Data 5:180214. https://doi.org/10.1038/sdata.2018.214CrossRef

Beecham S, Rashid M, Chowdhury RK (2014) Statistical downscaling of multi-site daily rainfall in a South Australian catchment using a generalized linear model. Int J Climatol 34:3654–3670. https://doi.org/10.1002/joc.3933CrossRef

Breiman L (1996) Bagging predictors. Mach Learn 24:123–140. https://doi.org/10.1007/BF00058655CrossRef

Bumby AJ, Eriksson PG, Catuneanu O, Nelson DR, Rigby MJ (2012) Meso-Archaean and Palaeo-Proterozoic sedimentary sequence stratigraphy of the Kaapvaal Craton. Mar Pet Geol 33:92–116. https://doi.org/10.1016/j.marpetgeo.2011.09.010CrossRef

Cachay SR, Erickson E, Bucker AFC, Pokropek E, Potosnak W, Bire S, Osei S, Lütjens B (2021) The world as a graph: improving El Nino forecasts with graph neural networks. ArXiv210405089 Phys Stat

Cairncross B (2001) An overview of the Permian (Karoo) coal deposits of southern Africa. J Afr Earth Sci 33:529–562. https://doi.org/10.1016/S0899-5362(01)00088-4CrossRef

Catuneanu O, Wopfner H, Eriksson PG, Cairncross B, Rubidge BS, Smith RMHH, Hancox PJ (2005) The Karoo basins of south-central Africa. J Afr Earth Sci 43:211–253. https://doi.org/10.1016/j.jafrearsci.2005.07.007CrossRef

Center for Space Research (2021) CSR GRACE/GRACE-FO RL06 Mascon Solutions (version 02). http://www2.csr.utexas.edu/grace/RL06_mascons.html. Accessed 10 Dec 2021

CGS (2008) 1:2000000 chronostratigraphic map of South Africa. CGS, Pretoria, South Africa

Chang F-J, Chiang Y-M, Chang L-C (2007) Multi-step-ahead neural networks for flood forecasting. Hydrol Sci J 52:114–130. https://doi.org/10.1623/hysj.52.1.114CrossRef

Climate Data Store (2021a) ERA5-Land hourly data from 1950 to present. https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-land?tab=overview. Accessed 10 Dec 2021

Climate Data Store (2021b) Land cover classification gridded maps from 1992 to present derived from satellite observations. https://cds.climate.copernicus.eu/cdsapp#!/dataset/satellite-land-cover?tab=overview. Accessed 10 Dec 2021

Cobbing J (2018) An updated water balance for the Grootfontein aquifer near Mahikeng. Water SA 44. https://doi.org/10.4314/wsa.v44i1.07

Cobbing J, Eales K, Rossouw T (2016) The path to successful water user associations in the North West Dolomite aquifers. Water Research Commission, Pretoria, South Africa

Cobbing JE, de Wit M (2018) The Grootfontein aquifer: governance of a hydro social system at Nash equilibrium. South Afr J Sci 114:1–7. https://doi.org/10.17159/sajs.2018/20170230CrossRef

CSIR (2003) Protection and Strategic Uses of groundwater resources in the transboundary Limpopo Basin and drought prone areas of the SADC Region: groundwater situation analysis in the Limpopo River Basin

DWAF (2006) A strategy for water allocation reform in South Africa. DWAF, Pretoria, South Africa

EarthData (2021) Data collections. Goddard Earth Sci. Data Inf. Serv. Cent. https://disc.gsfc.nasa.gov/datasets?keywords=GLDAS. Accessed 10 Dec 2021

Elith J, Leathwick JR, Hastie T (2008) A working guide to boosted regression trees. J Anim Ecol 77:802–813. https://doi.org/10.1111/j.1365-2656.2008.01390.xCrossRef

Farr JL, Gumiremhete R, Davies J, Robins NS (2005) Southern African Development Community regional situation analysis. British Geological Survey, Keyworth, UK

Gaffoor Z, Pietersen K, Jovanovic N, Bagula A, Kanyerere T (2020) Big data analytics and its role to support groundwater management in the Southern African Development Community. Water 12:28. https://doi.org/10.3390/w12102796CrossRef

Gundogdu KS, Guney I (2007) Spatial analyses of groundwater levels using universal kriging. J Earth Syst Sci 116:49–55. https://doi.org/10.1007/s12040-007-0006-6CrossRef

Ha W, Gowda PH, Howell TA (2013) A review of downscaling methods for remote sensing-based irrigation management: part I. Irrig Sci 31:831–850. https://doi.org/10.1007/s00271-012-0331-7CrossRef

Hengl T, Nussbaum M, Wright MN, Heuvelink GBM, Gräler B (2018) Random forest as a generic framework for predictive modeling of spatial and spatio-temporal variables. PeerJ 6:e5518. https://doi.org/10.7717/peerj.5518CrossRef

Holland M, Wiegmans FE (2009) Gauteng, North West and Ghaap Plateau Dolomite Units Map. DWS, Pretoria, South Africa

Jakeman AJ, Barreteau O, Hunt RJ, Rinaudo J-D, Ross A (eds) (2016) Integrated groundwater management: concepts, approaches and challenges. Springer International, Cham, Switzerland

Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Ye Q, Liu T-Y (2017) LightGBM: a highly efficient gradient boosting decision tree. In: Guyon I, Luxburg U V, Bengio S, Wallach H, Fergus R, Vishwanathan S, Garnett R (eds) Advances in neural information processing systems 30. Curran, pp 3146–3154. proceedings.com. Accessed Dec 2021

Koglin N, Zeh A, Frimmel HE, Gerdes A (2010) New constraints on the auriferous Witwatersrand sediment provenance from combined detrital zircon U–Pb and Lu–Hf isotope data for the Eldorado Reef (Central Rand Group, South Africa). Precambrian Res 183:817–824. https://doi.org/10.1016/j.precamres.2010.09.009CrossRef

Kotsiantis SB (2013) Decision trees: a recent overview. Artif Intell Rev 39:261–283. https://doi.org/10.1007/s10462-011-9272-4CrossRef

Kotu V, Deshpande B (2015) Classification. In: Predictive analytics and data mining. Elsevier, Amsterdam, pp 63–163

Kumar V (2006) Kriging of groundwater levels: a case study. J Spat Hydrol 6:12

Leonard LC (2017) Web-based behavioral modeling for continuous user authentication (CUA). In: Advances in computers. Elsevier, Amsterdam, pp 1–44

Li Z-L, Duan S-B (2018) Land surface temperature. In: Comprehensive remote sensing. Elsevier, Amsterdam, pp 264–283

Liu Z, Xu Z, Charles SP, Fu G, Liu L (2011) Evaluation of two statistical downscaling models for daily precipitation over an arid basin in China: evaluation of two statistical downscaling models. Int J Climatol 31:2006–2020. https://doi.org/10.1002/joc.2211CrossRef

Majumdar S, Smith R, Butler JJ, Lakshmi V (2020) Groundwater withdrawal prediction using integrated multitemporal remote sensing data sets and machine learning. Water Resour Res 56. https://doi.org/10.1029/2020WR028059

Malakar P, Mukherjee A, Bhanja SN, Ray RK, Sarkar S, Zahid A (2021) Machine-learning-based regional-scale groundwater level prediction using GRACE. Hydrogeol J 29:1027–1042. https://doi.org/10.1007/s10040-021-02306-2CrossRef

Manzi M, Hein K, King N, Durrheim R (2013) Neoarchaean tectonic history of the Witwatersrand Basin and Ventersdorp Supergroup: new constraints from high-resolution 3D seismic reflection data. Tectonophysics 590:94–105. https://doi.org/10.1016/j.tecto.2013.01.014CrossRef

Milewski AM, Thomas MB, Seyoum WM, Rasmussen TC (2019) Spatial Downscaling of GRACE TWSA Data to Identify Spatiotemporal Groundwater Level Trends in the Upper Floridan Aquifer, Georgia, USA. Remote Sensing 11:2756. https://doi.org/10.3390/rs11232756

Miro M, Famiglietti J (2018) Downscaling GRACE remote sensing datasets to high-resolution groundwater storage change maps of California’s Central Valley. Remote Sens 10:143. https://doi.org/10.3390/rs10010143CrossRef

Modisha RCO (2017) Investigation of the Ramotswa Trandboundary Aquifer Area: groundwater flow and pollution. University of the Witwatersrand, Johannesburg, South Africa

Moghim S, Bras RL (2017) Bias correction of climate modeled temperature and precipitation using artificial neural networks. J Hydrometeorol 18:1867–1884. https://doi.org/10.1175/JHM-D-16-0247.1CrossRef

Nijsten G-JJ, Christelis G, Villholth KG, Braune E, Gaye CB (2018) Transboundary aquifers of Africa: review of the current state of knowledge and progress towards sustainable development and management. J Hydrol Reg Stud 20:1–14. https://doi.org/10.1016/j.ejrh.2018.03.004CrossRef

Pietersen K, Beekman H (2016) Groundwater management in the Southern African Development Community. Southern African Development Community Groundwater Management Institute, Bloemfontein, South Africa

Pietersen K, Beekman H, Cobbing J, Kanyerere T (2018) Consultancy for capacity needs assessment to determine priority challenges for capacity development initiatives in Member States. Southern African Development Community Groundwater Management Institute, Bloemfontein, South Africa

Pour SH, Shahid S, Chung E-S, Wang X-J (2018) Model output statistics downscaling using support vector machine for the projection of spatial and temporal changes in rainfall of Bangladesh. Atmos Res 213:149–162. https://doi.org/10.1016/j.atmosres.2018.06.006CrossRef

Rahaman M, Thakur B, Kalra A, Li R, Maheshwari P (2019) High-Resolution Groundwater Storage from GRACE: a Random Forest approach. Environments 6:63. https://doi.org/10.3390/environments6060063CrossRef

Reinders C, Ackermann H, Yang MY, Rosenhahn B (2019) Learning convolutional neural networks for object detection with very little training data. In: Multimodal scene understanding. Elsevier, Amsterdam, pp 65–100

Rodell M, Chen J, Kato H, Famiglietti JS, Nigro J, Wilson CR (2007) Estimating groundwater storage changes in the Mississippi River basin (USA) using GRACE. Hydrogeol J 15:159–166. https://doi.org/10.1007/s10040-006-0103-7CrossRef

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-85-3-381CrossRef

SADC-GIP (2021) SADC groundwater information portal. https://sadc-gip.org/. Accessed Dec 2021

SADC-GMI (2010) SADC hydrogeological mapping project: hydrogeological mapping procedures and guidelines. Southern African Development Community Groundwater Management Institute, Bloemfontein, South Africa

Sahoo S, Russo TA, Elliott J, Foster I (2017) Machine learning algorithms for modeling groundwater level changes in agricultural regions of the U.S. Water Resour Res 53:3878–3895. https://doi.org/10.1002/2016WR019933CrossRef

Save H, Bettadpur S, Tapley BD (2016) High-resolution CSR GRACE RL05 mascons. J Geophys Res Solid Earth 121:7547–7569. https://doi.org/10.1002/2016JB013007CrossRef

SAWS (2019) Annual climate summary for South Africa: 2018. South African Weather Service, Pretoria, South Africa

Schapire RE (2003) The boosting approach to machine learning: an overview. In: Denison DD, Hansen MH, Holmes CC, Mallick B, Yu B (eds) Nonlinear estimation and classification. Springer, New York, pp 149–171CrossRef

Seo JY, Lee S (2019) Spatio-Temporal Groundwater Drought Monitoring Using Multi-Satellite Data Based on an Artificial Neural Network. Water 11:1953. https://doi.org/10.3390/w11091953

Seyoum W, Kwon D, Milewski A (2019) Downscaling GRACE TWSA data into high-resolution groundwater level anomaly using machine learning-based models in a glacial aquifer system. Remote Sens 11:824. https://doi.org/10.3390/rs11070824CrossRef

Seyoum WM, Milewski AM (2017) Improved methods for estimating local terrestrial water dynamics from GRACE in the Northern High Plains. Adv Water Resour 110:279–290. https://doi.org/10.1016/j.advwatres.2017.10.021CrossRef

Smith RG, Majumdar S (2020) Groundwater storage loss associated with land subsidence in western United States mapped using machine learning. Water Resour Res 56. https://doi.org/10.1029/2019WR026621

South African National Department of Water and Sanitation (2021) CHART. https://www.dws.gov.za/Groundwater/chart.aspx. Accessed Dec 2021

Srivastava PK, Han D, Ramirez MR, Islam T (2013) Machine learning techniques for downscaling SMOS satellite soil moisture using MODIS land surface temperature for hydrological application. Water Resour Manag 27:3127–3144. https://doi.org/10.1007/s11269-013-0337-9CrossRef

Sterckx A, Nijsten G-J, Gomo M, Lukas E, Kukuric N (2019) Capacity building for groundwater data collection and management in SADC member states. International Groundwater Resources Assessment Centre, Delft, The Netherlands

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

Tan L (2015) Code comment analysis for improving software quality. In: The art and science of analyzing software data. Elsevier, Amsterdam, pp 493–517

Wahr J, Molenaar M, Bryan F (1998) Time variability of the Earth’s gravity field: hydrological and oceanic effects and their possible detection using GRACE. J Geophys Res Solid Earth 103:30205–30229. https://doi.org/10.1029/98jb02844CrossRef

Wiegmans FE, Holland M, Janse van Rensburg H (2013) Groundwater resource directed measure for Maloney’s Eye Catchment. Water Research Commission, Pretoria, South Africa

Yin W, Hu L, Zhang M, Wang J, Han S-C (2018) Statistical downscaling of GRACE-derived groundwater storage using ET data in the North China Plain. J Geophys Res Atmos 123:5973–5987. https://doi.org/10.1029/2017JD027468CrossRef

Zaitchik BF, Rodell M, Reichle RH (2008) Assimilation of GRACE terrestrial water storage data into a land surface model: results for the Mississippi River Basin. J Hydrometeorol 9:535–548. https://doi.org/10.1175/2007JHM951.1CrossRef

Zhou J, Cui G, Hu S, Zhang Z, Yang C, Liu Z, Wang L, Li C, Sun M (2020) Graph neural networks: a review of methods and applications. AI Open 1:57–81. https://doi.org/10.1016/j.aiopen.2021.01.001CrossRef

Zink M, Mai J, Cuntz M, Samaniego L (2018) Conditioning a hydrologic model using patterns of remotely sensed land surface temperature. Water Resour Res 54:2976–2998. https://doi.org/10.1002/2017WR021346CrossRef

Title: An autoregressive machine learning approach to forecast high-resolution groundwater-level anomalies in the Ramotswa/North West/Gauteng dolomite aquifers of Southern Africa
Authors: Zaheed Gaffoor
Ashley Gritzman
Kevin Pietersen
Nebo Jovanovic
Antoine Bagula
Thokozani Kanyerere
Publication date: 17-01-2022
Publisher: Springer Berlin Heidelberg
Published in: Hydrogeology Journal / Issue 2/2022
Print ISSN: 1431-2174
Electronic ISSN: 1435-0157
DOI: https://doi.org/10.1007/s10040-021-02439-4

Springer Professional

Abstract

Please log in to get access to your license.

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Springer Professional "Technik"

Other articles of this Issue 2/2022

Correction: Stratigraphic and structural controls on groundwater salinity variations in the Poso Creek Oil Field, Kern County, California, USA

Vadose zone modeling to identify controls on groundwater recharge in an unconfined granular aquifer in a cold and humid environment with different meteorological data sources

The effects of urbanization on the groundwater system of the Kabul shallow aquifers, Afghanistan

An improved Bayesian approach linked to a surrogate model for identifying groundwater pollution sources

Review: Horizontal, directionally drilled and radial collector wells

Isotopic and geochemical tracers of groundwater flow in the Shivwits Plateau, Grand Canyon National Park, USA