Regional Mapping of Groundwater Potential Zones in the Saudi Arabia Using Remote Sensing and Machine Learning Algorithms

Samy Elmahdy, Mohamed Mohamed

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The Arabian Peninsula desert is among the driest regions on Earth and climate and rainfall pattern changes might affect groundwater recharge in the entire region. These changes have led to sharp depletion in groundwater level and quality and represent a challenge to developing such regions. The use of machine learning algorithms and remote sensing data allowed the identification of several buried ancient features such as faults, palaeolakes, palaeochannels and mega-basins. The L band (long wavelength) of radar sensors enabled the penetration of the topsoil and revealed several of these hidden features which represent indisputable evidence of the past wet conditions of the region. These features are important input parameters to learn machines and predict zones of a higher probability of groundwater potential. The results show that the integration of remote sensing data and machine learning algorithms is a hopeful approach that can help significantly in groundwater potential mapping in inaccessible and remote arid and semi-arid regions, where groundwater access is important for the developing of the region.

Original languageEnglish
Title of host publicationSpringer Water
PublisherSpringer Nature
Pages311-333
Number of pages23
DOIs
Publication statusPublished - 2022

Publication series

NameSpringer Water
ISSN (Print)2364-6934
ISSN (Electronic)2364-8198

Keywords

  • Arabian Peninsula Palaeochannels
  • DEM
  • GIS
  • Groundwater
  • Radar images
  • Remote sensing
  • Saudi Arabia

ASJC Scopus subject areas

  • Oceanography
  • Aquatic Science
  • Environmental Science (miscellaneous)
  • Water Science and Technology
  • Earth and Planetary Sciences (miscellaneous)

Fingerprint

Dive into the research topics of 'Regional Mapping of Groundwater Potential Zones in the Saudi Arabia Using Remote Sensing and Machine Learning Algorithms'. Together they form a unique fingerprint.

Cite this