Regional Mapping of Groundwater Potential in Ar Rub Al Khali, Arabian Peninsula Using the Classification and Regression Trees Model

Samy Elmahdy,Mohamed Mohamed,Tarig Ali

doi:10.3390/rs13122300

Samy Elmahdy, Mohamed Mohamed + Show 1 more

Open Access

https://doi.org/10.3390/rs13122300

Copy DOI

Abstract

Mapping of groundwater potential in remote arid and semi-arid regions underneath sand sheets over a very regional scale is a challenge and requires an accurate classifier. The Classification and Regression Trees (CART) model is a robust machine learning classifier used in groundwater potential mapping over a very regional scale. Ten essential groundwater conditioning factors (GWCFs) were constructed using remote sensing data. The spatial relationship between these conditioning factors and the observed groundwater wells locations was optimized and identified by using the chi-square method. A total of 185 groundwater well locations were randomly divided into 129 (70%) for training the model and 56 (30%) for validation. The model was applied for groundwater potential mapping by using optimal parameters values for additive trees were 186, the value for the learning rate was 0.1, and the maximum size of the tree was five. The validation result demonstrated that the area under the curve (AUC) of the CART was 0.920, which represents a predictive accuracy of 92%. The resulting map demonstrated that the depressions of Mondafan, Khujaymah and Wajid Mutaridah depression and the southern gulf salt basin (SGSB) near Saudi Arabia, Oman and the United Arab Emirates (UAE) borders reserve fresh fossil groundwater as indicated from the observed lakes and recovered paleolakes. The proposed model and the new maps are effective at enhancing the mapping of groundwater potential over a very regional scale obtained using machine learning algorithms, which are used rarely in the literature and can be applied to the Sahara and the Kalahari Desert.

Highlights

Introduction published maps and institutional affilGroundwater is an important source in land development and human activities such as agriculture and industry
The present study focuses on topography, hydrological and geology conditioning factors, which are extracted from the Advanced Land Observing Satellite (ALOS) digital elevation model (DEM) and groundwater wells (Figure 4)
This study presents the classification and regression trees (CART) model to map zones of groundwater potential over a regional scale over a regional scale for the first time with a moderate number of groundwater conditioning factors (GWCFs)

Summary

Introduction

Introduction published maps and institutional affilGroundwater is an important source in land development and human activities such as agriculture and industry. In arid and semi-arid regions, there has been excessive consumption of groundwater and scarce rainfall due to climate change and population growth [1,2]. These have led to sharp depletions in the groundwater table and the quality. About 50% of these countries lay within the largest sand sea in the world, which is known as ARAK It is the largest promising groundwater aquifer in the world and sits on top of a vast amount of fossil water reserves [1,2,3]. The hydrological setting of the ARAK is explored only from oil wells and geophysical reports over a local scale due to its harsh weather, remote location and lack of rock outcrops [2,3]

Objectives

Methods

Results

Discussion

Conclusion