Year-round flow consistencies of the central Himalayan springs are extremely important for addressing rural water demand. As the prediction of Himalayan spring high-flows is expected to provide better opportunities for the management of excess runoff, this study aims to develop a data-driven model for predicting joint-fracture and depression type spring high-flows of the Kosi watershed of central Himalaya, India. Five machine learning algorithms are used with combinations of predictors, such as standardized anomaly of rainfall, pH, electrical conductivity and water quality index of spring water. The discharge and predictor parameters are used from a total of 06 springs distributed across the watershed, and monitored during January, 2019 to December, 2020 at monthly interval. Due to asymmetric relationships between model predictors and spring discharge, model performances are tested for the predictor time lags of 0–2 (= 60 days). A total of ten experiments are carried out, and model performances during training and testing are evaluated using receiver operator characteristics. The discriminant analysis classifier, in combination with rainfall and electrical conductivity as predictors, is found to be the best model for predicting spring high-flows.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
The spring discharge and chemistry data are available with SM. The data sharing needs approval from the competent authority of GBPNIHE, Almora, India.
The classifier codes are available with SM. The code sharing needs approval from the competent authority of GBPNIHE, Almora, India.
Agarwal A, Bhatnagar N, Nema RK, Agrawal N (2012) Rainfall dependence of springs in the midwestern Himalayan hills of Uttarakhand. Mount Res Dev 32:446. https://doi.org/10.1659/MRD-JOURNAL-D-12-00054.1
Alfaro C, Wallace M (1994) Origin and classification of springs and historical review with current applications. Environ Geol 24(2):112–124. https://doi.org/10.1007/BF00767884
Ballav S, Mukherjee S, Gosavi V, Dimri AP (2021) Projected changes in winter-season wet days over the Himalayan region during 2020–2099. Theor Appl Climatol 146(3–4):883–895. https://doi.org/10.1007/s00704-021-03765-z
Banerjee A, Chen R, Meadows M, Singh RB, Mal S, Sengupta D (2020) An Analysis of long-term rainfall trends and variability in the Uttarakhand Himalaya using Google Earth Engine. Remote Sens 12(4):709. https://doi.org/10.3390/rs12040709
Berrar D (2019) Bayes' theorem and naive Bayes classifier. In: Encyclopedia of bioinformatics and computational biology: ABC of bioinformatics, vol 01. Elsevier, Amsterdam
BIS 10500 (2012) Bureau of Indian Standards drinking water specification, second revision. Manak Bhawan, New Delhi, pp 16
Brown RM, McCleiland MJ, Deininger RA, O'Connor MF (1972) A water quality index—crossing the psychological barrier. In: Proceedings in international conference on water pollution research, pp 787–797
Cheng S, Qiao X, Shi Y, Wang D (2020) Comparison of machine learning methods for predicting Karst spring discharge in North China. ArXiv Preprint. arXiv: 2007.12951
Daniel D, Anandhi A, Sen S (2021) Conceptual model for the vulnerability assessment of springs in the Indian Himalayas. Climate 9(8):121. https://doi.org/10.3390/cli9080121
Dass B, Abhishek SS, Bamola V, Sharma A, Sen D (2021) Assessment of spring flows in Indian Himalayan micro-watersheds—a hydro-geological approach. J Hydrol 598:126354. https://doi.org/10.1016/j.jhydrol.2021.126354
Fawcett T (2006) Introduction to ROC analysis. Pattern Recognit Lett 27:861–874. https://doi.org/10.1016/j.patrec.2005.10.010
Fiorillo F, Doglioni A (2010) The relation between Karst spring discharge and rainfall by the cross-correlation analysis. Hydrogeol J 18:1881–1895. https://doi.org/10.1007/s10040-010-0666-1
Gebrehiwot A, Tadesse N, Jigar E (2011) Application of water quality index to assess suitability of groundwater quality for drinking purposes in Hantebet watershed, Tigray, Northern Ethiopia. J Food Agri Sci 1(1):22–30
Granata F, Saroli M, de Marinis G, Gargano R (2018) Machine learning models for spring discharge forecasting. Geofluids. https://doi.org/10.1155/2018/8328167
Gutierrez DD (2015) Machine learning and data science: an introduction to statistical learning methods with R. Technics Publications. https://books.google.co.in/books?id=uAAcjgEACAAJ
Hu C, Hao Y, Yeh T-C, Pang B, Wu Z (2008) Simulation of spring flows from a Karst aquifer with an artificial neural network. Hydrol Proc 22:596–604. https://doi.org/10.1002/hyp.6625
Kresic N, Bonacci O (2010) Spring discharge hydrograph. Groundwater hydrology of springs. Butterworth-Heinemann, Boston, pp 129–163
Kumar K, Tiwari A, Mukherjee S, Agnihotri V, Verma RK (2019) Water at a glance: Uttarakhand. GBPNIHE, Almora
Lambrakis N, Andreou AS, Polydoropoulos P, Georgopoulos E, Bountis T (2000) Nonlinear analysis and forecasting of a brackish Karstic spring. Water Res Res 36(4):875–884. https://doi.org/10.1029/1999WR900353
Mayaud C, Wagner T, Benischke R, Birk S (2014) Single event time series analysis in a binary Karst catchment evaluated using a groundwater model (Lurbach system, Austria). J Hydrol 511:628–639
Meinzer OE (1927) Large springs in the United States. USGS Water Supply Paper: Washington, USA
Moriasi D, Arnold JG, Van Liew MW, Bingner RL, Harmel RD, Veith TL (2007) Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Am Soc Agric Biol Eng 50(3):885–900
Mukherjee S (2021) Nonlinear recurrence quantification of the monsoon-season heavy rainy-days over northwest Himalaya for the baseline and future periods. Sci Tot Environ 789:147754. https://doi.org/10.1016/j.scitotenv.2021.147754
Mukherjee S, Joshi R, Prasad RC, Vishvakarma SCR, Kumar K (2015) Summer monsoon rainfall trends in the Indian Himalayan region. Theor Appl Climatol 121(3–4):789–802. https://doi.org/10.1007/10.1007/s00704-014-1273-1
Mukherjee S, Ballav S, Soni S, Kumar K, De UK (2016) Investigation of dominant modes of monsoon ISO in the northwest and eastern Himalayan region. Theor Appl Climatol 125(3–4):489–498. https://doi.org/10.1007/s00704-015-1512-0
Mukherjee S, Hazra A, Kumar K, Nandi SK, Dhyani PP (2019) Simulated projection of ISMR over Indian Himalayan region: Assessment from CSIRO-CORDEX South Asia experiments. Meteorol Atmos Phys 131(1):63–79. https://doi.org/10.1007/s00703-017-0547-4
Murakami Y, Mizuguchi K (2010) Applying the naïve Bayes classifier with kernel density estimation to the prediction of protein–protein interaction sites. Bioinformatics 26(15):1841–1848. https://doi.org/10.1093/bioinformatics/btq302
Namous M, Hssaisoune M, Pradhan B, Lee C-W, Alamri A, Elaloui A, Edahbi M, Krimissa S, Eloudi H, Ouayah M, Elhimer H, Tagma T (2021) Spatial prediction of groundwater potentiality in large semi-arid and Karstic mountainous region using machine learning models. Water 13(16):2273. https://doi.org/10.3390/w13162273
Negi GCS, Joshi V (1996) Geo-hydrology of springs in a mountain watershed: the need for problem solving research. Curr Sci 71(10):772–776
Negi GCS, Joshi V (2004) Rainfall and spring discharge patterns in two small drainage catchments in the Western Himalayan mountains, India. Environmentalist 24:19–28
Negnevitsky M (2001) Artificial intelligence: a guide to intelligent systems. Pearson Education, Rotherham
NITI Aayog (2018) Report of Working Group II sustainable tourism in the Indian Himalayan region. NITI Aayog, Government of India, Delhi, India
Panwar S (2020) Vulnerability of Himalayan springs to climate change and anthropogenic impact: a review. J Mt Sci 17(1):117–132. https://doi.org/10.1007/s11629-018-5308-4
Rani M, Joshi H, Kumar K, Bhatt DK, Kumar P (2021) Climate change scenario of hydro-chemical analysis and mapping spatio-temporal changes in water chemistry of water springs in Kumaun Himalaya. Environ Dev Sustain 23(3):4659–4674. https://doi.org/10.1007/s10668-020-00793-z
Rathod R, Kumar M, Mukherji A, Sikka A, Satapathy KK, Mishra A, Goel S, Khan M (2021) Resource book on springshed management in the Indian Himalayan region: Guidelines for policy makers and development practitioners. International Water Management Institute (IWMI), India, New Delhi
Savas C, Dovis F (2019) The impact of different kernel functions on the performance of scintillation detection based on support vector machines. Sensors 19(23):5219. https://doi.org/10.3390/s19235219
Sifaou H, Kammoun A, Alouini MS (2020) High-dimensional linear discriminant analysis classifier for spiked covariance model. J Mach Learn Res 21:1–24
Springer AE, Stevens LE, Anderson DE, Parnel RA, Kreamer DA, Levin L, Flora S (2008) A comprehensive springs classification system: integrating geomorphic, hydrogeochemical, and ecological criteria. In: Stevens LE, Meretsky VJ (eds) Arid land springs in North America: ecology and conservation. University of Arizon Press, Tucson
Tambe S, Kharel G, Arrawatia ML, Kulkarni H, Mahamuni K, Ganeriwala AK (2012) Reviving dying springs: climate change adaptation experiments from the Sikkim Himalaya. Mt Res Dev 32(1):62–72. https://doi.org/10.1659/MRD-JOURNAL-D-11-00079.1
Tarafdar S (2013) Understanding the dynamics of high and low spring flow: a key to managing the water resources in a small urbanized hill-slope of Lesser Himalaya, India. Environ Earth Sci 70(5):2107–2114. https://doi.org/10.1659/10.1007/s12665-011-1493-y
Valdiya K, Bartarya S (1989) Diminishing discharges of mountain springs in a part of Kumaun Himalaya. Curr Sci 58:417–426
White W (2010) Springwater geochemistry. In: Groundwater hydrology of springs, pp 231–268. https://doi.org/10.1016/B978-1-85617-502-9.00006-2
Wunsch A, Liesch T, Cinkus G, Ravbar N, Chen Z, Mazzilli N, Jourde H, Goldscheider N (2021) Karst spring discharge modeling based on deep learning using spatially distributed input data. Hydrol Earth Syst Sci. https://doi.org/10.5194/hess-2021-403
Zhang Z (2016) Introduction to machine learning: K-nearest neighbors. Ann Transl Med 4(11):218–218. https://doi.org/10.21037/atm.2016.03.37
Zhang YK, Bai EW, Libra R, Rowden R, Liu H (1996) Simulation of spring discharge from a limestone aquifer in Iowa, USA. Hydrogeol J 4(4):41–54
Research funding of NMHS, MoEFCC, GoI, is acknowledged (NMHS-2017-18/MG-02/478) for completing this work. Mr. Vinod Kanwal is highly acknowledged for collecting monthly spring discharge data and water samples. The Central Laboratory of GBPNIHE, Uttarakhand, India, is acknowledged for chemical analysis of the water samples. Mr. Sourab Singh and Kunal Joshi are acknowledged for preparing the spring land cover maps and geological information. The Director of GBPNIHE, Uttarakhand, India, is acknowledged for providing the computational facilities.
Research funding of NMHS, MoEFCC, GoI, is acknowledged (NMHS-2017–18/MG-02/478).
Conflict of interest
No competing interests are present among the authors.
Consent for publication
All the authors have their consents for publication.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Mukherjee, S., Sen, S. & Kumar, K. Multifactor prediction of the central Himalayan spring high-flows using machine learning classifiers. Environ Earth Sci 82, 85 (2023). https://doi.org/10.1007/s12665-023-10775-9