Mapping Groundwater Potential Using a Novel Hybrid Intelligence Approach

Shaghayegh Miraki, Sasan Hedayati Zanganeh, Kamran Chapi (), Vijay P. Singh, Ataollah Shirzadi, Himan Shahabi and Binh Thai Pham
Additional contact information
Shaghayegh Miraki: University of Agricultural Science and Natural Resources of Sari
Sasan Hedayati Zanganeh: University of Tabriz
Kamran Chapi: University of Kurdistan
Vijay P. Singh: Texas A & M University
Ataollah Shirzadi: University of Kurdistan
Himan Shahabi: University of Kurdistan
Binh Thai Pham: University of Transport Technology

Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA), 2019, vol. 33, issue 1, No 16, 302 pages

Abstract: Abstract Identifying areas with high groundwater potential is important for groundwater resources management. The main objective of this study is to propose a novel classifier ensemble method, namely Random Forest Classifier based on Random Subspace Ensemble (RS-RF), for groundwater potential mapping (GWPM) in Qorveh-Dehgolan plain, Kurdistan province, Iran. A total of 12 conditioning factors (slope, aspect, elevation, curvature, stream power index (SPI), topographic wetness index (TWI), rainfall, lithology, land use, normalized difference vegetation index (NDVI), fault density, and river density) were selected for groundwater modeling. The least square support vector machine (LSSVM) feature selection method with a 10-fold cross-validation technique was used to validate the predictive capability of these conditioning factors for training the models. The performance of the RS-RF model was validated using the area under receiver operating characteristic curve (AUROC), success and prediction rate curves, kappa index, and several statistical index-based measures. In addition, Friedman and Wilcoxon signed-rank tests were used to assess statistically significant level among the new model with the state-of-the-art soft computing benchmark models, such as random forest (RF), logistic regression (LR) and naïve Bayes (NB). Results showed that the new hybrid model of RS-RF had a very high predictive capability for groundwater potential mapping and exhibited the best performance among other benchmark models (LR, RF, and NB). Results of the present study might be useful to water managers to make proper decisions on the optimal use of groundwater resources for future planning in the critical study area.

Keywords: Mapping groundwater potential; Least square support vector machine; Random forest random subspace ensemble; Logistic regression; Kurdistan (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (13)

Downloads: (external link)
http://link.springer.com/10.1007/s11269-018-2102-6 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:waterr:v:33:y:2019:i:1:d:10.1007_s11269-018-2102-6

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11269

DOI: 10.1007/s11269-018-2102-6

Access Statistics for this article

Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA) is currently edited by G. Tsakiris

More articles in Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA) from Springer, European Water Resources Association (EWRA)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().