PRISMA vs. Landsat 9 in lithological mapping − a K-fold Cross-Validation implementation with Random Forest

The Egyptian Journal of Remote Sensing and Space Science, Journal Year: 2024, Volume and Issue: 27(3), P. 577 - 596

Published: July 15, 2024

The selection of an optimal dataset is crucial for successful remote sensing analysis. PRISMA hyperspectral sensor (with 240 spectral bands) and Landsat OLI-2 (boasting high dynamic resolution) offer robust data various applications, anticipating their increased demand in the coming years. However, despite potential, we have not identified a rigorous evaluation both datasets geological applications utilizing Machine Learning Algorithms. Consequently, conduct comprehensive analysis using Random Forest, widely-recommended machine learning algorithm, employ K-fold cross-validation K = 2, 5, 10) with grid-search hyperparameter tuning enhanced performance. Toward this aim, diverse image-processing approaches, including Principal Component Analysis (PCA), Minimum Noise Fraction (MNF), Independent (ICA), were applied to enhance feature extraction. Subsequently, ensure better performance RF study utilized well-distributed points instead polygons represent each target, thereby mitigating effects spatial autocorrelation. Our results reveal dataset-hyperparameter dependencies, mainly influenced by max_depth 9 max_features. Employing optimally balances characteristics splitting (folds), generating accurate lithological maps across all values. Notably, significant shift at 10 produces best maps. Fieldwork petrographic investigations validate maps, indicating PRISMA's slight superiority over OLI-2. Despite this, given nature band count difference, still advocate as potent multispectral input future due its superior radiometric resolution.

Language: Английский

---

Modeling Groundwater Resources in Data-Scarce Regions for Sustainable Management: Methodologies and Limits

Hydrology, Journal Year: 2025, Volume and Issue: 12(1), P. 11 - 11

Published: Jan. 9, 2025

Groundwater modeling in data-scarce regions faces significant challenges due to the lack of comprehensive, high-quality data, impacting model accuracy. This systematic review Scopus-indexed papers identifies various approaches address these challenges, including coupled hydrological-groundwater models, machine learning techniques, distributed hydrological water balance 3D groundwater flow modeling, geostatistical remote sensing-based approaches, isotope-based methods, global downscaling, and integrated approaches. Each methodology offers unique advantages for assessment management data-poor environments, often combining multiple data sources techniques overcome limitations. However, face common related quality, scale transferability, representation complex hydrogeological processes. emphasizes importance adapting methodologies specific regional contexts availability. It underscores value provide robust estimates sustainable management. The choice method ultimately depends on objectives, study, available region interest. Future research should focus improving integration diverse sources, enhancing processes simplified developing uncertainty quantification methods tailored conditions.

Language: Английский

---

Delineation of lineaments for groundwater prospecting in hard rocks: inferences from remote sensing and geophysical data

Environmental Earth Sciences, Journal Year: 2024, Volume and Issue: 83(2)

Published: Jan. 1, 2024

Abstract Egypt is grappling with water scarcity challenges, which are exacerbated by extensive urban development in arid coastal regions rugged terrain. Although desalinated an alternative source the remote Halayeib region of Southeast Egypt, its cost increases reliance on groundwater from intricate aquifers. This study aims to accurately delineate hydro-structural features, known as lineaments, and assess their impact conditions this promising region. integrated approach involves assessment various spaceborne sensors, including optical (Landsat 8), Digital Elevation Models (ALOS ASTER-DEMs), radar (Sentinel-1), using geospatial geostatistical techniques within Geographic Information System (GIS). Radar-based particularly Sentinel-1A vertical–vertical (S1A VV) polarization, outperform all other datasets extracting yielding 4883 lineaments that correspond regional geological faults. These trend NE–SW, NNE–SSW, NW–SE, E-W directions. The results also indicated both digital elevation models (DEMs) were less effective, showing different orientations azimuth angles. S1A VV proved be highly effective identifying subsurface fractured hard rock terrains beneath thin sedimentary covers, especially flat area Wadi Serimatai, where they intersected natural drainage pathways. Geoelectrical sections confirmed there orthogonal faults extending basement aquifers near-surface layers. align NE-SW NNE-SSW directions observed lineaments. Geostatistical analysis revealed structural lithological, hydrogeological factors influence occurrence groundwater. emphasizes control over significant flow storage. provides valuable insights for management, guiding decisions related resources.

Language: Английский

---

Mapping Potential Water Resource Areas Using GIS-Based Frequency Ratio and Evidential Belief Function

Water, Journal Year: 2023, Volume and Issue: 15(3), P. 480 - 480

Published: Jan. 25, 2023

Groundwater is a critical freshwater resource that necessary for sustaining life. Thus, targeting prospective groundwater zones crucial the extraction, use, and management of water resources. In this study, we combined remote sensing, GIS-based frequency ratio (FR), evidential belief function (EBF) techniques into model to delineate quantify zones. To accomplish this, processed Shuttle Radar Topography Mission (SRTM), Landsat-8 Operational Land Imager (OLI), Sentinel-2, rainfall data reveal geomorphic, hydrologic, structural elements climatic conditions study area, which downstream Yellow River basin, China. We processed, quantified, twelve factors (the elevation, slope, aspect, drainage density, lineament distance rivers, NDVI, TWI, SPI, TRI, land use/cover, intensity) control infiltration occurrence using FR EBF models produce potential (GWPZs). used natural breaks classifier categorize likelihood at each location as very low, moderate, high, or high. The exhibited better performance than model, evidenced by area under curve (AUC) assessment predictions (FR AUCs 0.707 0.734, 0.665 0.690). Combining FR–EBF increased accuracy (AUC = 0.716 0.747), it areas high moderate potentiality 1.97% entire instead 0.39 0.78% models, respectively. integration sensing GIS-data-driven mapping

Language: Английский

---

Using Remote Sensing and GIS-Based Frequency Ratio Technique for Revealing Groundwater Prospective Areas at Wadi Al Hamdh Watershed, Saudi Arabia

Water, Journal Year: 2023, Volume and Issue: 15(6), P. 1154 - 1154

Published: March 16, 2023

For socioeconomic development in arid regions, there is an increasing need for groundwater resources due to rapid population expansion. It necessary apply innovative approaches managing the sustainability of resources. Thus, remote sensing, geologic, climatic, and hydrologic data are integrated through GIS-based frequency ratio overlay analysis assessing spatial distribution potential zones (GWPZs) Wadi Al Hamdh, Saudi Arabia. Twelve factors controlling groundwater’s existence infiltration were identified, normalized using technique combined GIS techniques. To accomplish this, 313 well locations study area used training (70%) 137 utilized validation (30%). Using receiver operating characteristic (ROC) curves field data, model predictions validated showed very good performance (AUC: 0.893). The five on GWPZs map correspond 2.24, 5.81, 13.39, 53.90, 24.65% entire area. These are: excellent, good, moderate, low, low perspectivity. As a example, applied provided results that significant planning sustainable as regions.

Language: Английский

---

Mapping Groundwater Prospective Areas Using Remote Sensing and GIS-Based Data Driven Frequency Ratio Techniques and Detecting Land Cover Changes in the Yellow River Basin, China

Land, Journal Year: 2023, Volume and Issue: 12(4), P. 771 - 771

Published: March 29, 2023

Groundwater is an essential resource that meets all of humanity’s daily water demands, supports industrial development, influences agricultural output, and maintains ecological equilibrium. Remote sensing data can predict the location potential resources. The current study was conducted in China’s Yellow River region, Ningxia Hui Autonomous Region (NHAR). Through use a GIS-based frequency ratio machine learning technique, nine layers evidence influenced by remote were generated integrated. used are soil characteristics, aspect, roughness index terrain, drainage density, elevation, lineament depressions, rainfall, distance to river from location. Six groundwater prospective zones (GWPZs) found have very low (13%), (30%), moderate (25%), high (16%), (11%), extreme potentiality (5.26%) values. According well validate GWPZs map, approximately 40% wells consistent excellent zones. Information about productivity gathered 150 locations. Using had not been for model training, resulting maps validated using area-under-the-curve (AUC) analysis. FR models accuracy rating 0.759. Landsat characterize area’s changes land cover. spatiotemporal differences cover detected quantified multi-temporal images which revealed water, agricultural, anthropogenic activities. Overall, combining different sets through GIS reveal promising areas resources aid planners managers.

Language: Английский

---

An Integrated Remote Sensing and GIS-Based Technique for Mapping Groundwater Recharge Zones: A Case Study of SW Riyadh, Central Saudi Arabia

Hydrology, Journal Year: 2024, Volume and Issue: 11(3), P. 38 - 38

Published: March 3, 2024

It might be difficult to find possible groundwater reservoir zones, especially in arid or hilly regions. In the twenty-first century, remotely sensed satellite imagery may present a new opportunity locate surface and subsurface water resources more quickly affordably. order identify potential current study was conducted Central Saudi Arabia, southwest of Riyadh. The analysis employed multi-criteria approach that relies on remote sensing geographic information systems. variables this technique include geology, rainfall, elevation, slope, aspect, hillshade, drainage density, lineaments Land Use/Land Cover (LULC). Analytical Hierarchical Process (AHP) used for assigning weights parameters, corresponding significance each parameter’s several classes potentiality. Different zones were identified by study: very high (16.8%), (30%), medium (26.7%), low (18.6%), (7.9%). Only two observation wells located “medium” zone, but other ten observed “very high” according validation survey. Consequently, results demonstrate approach, which combines improved conceptualization with AHP define map has greater chance producing accurate can reduce threat drought broader

Language: Английский

---

Optimizing urban water sustainability: Integrating deep learning, genetic algorithm, and CMIP6 GCM for groundwater potential zone prediction within a social-ecological-technological framework

Advances in Space Research, Journal Year: 2024, Volume and Issue: 73(12), P. 5925 - 5948

Published: March 16, 2024

Language: Английский

---

The use of radar-optical remote sensing data and geographic information system–analytical hierarchy process–multicriteria decision analysis techniques for revealing groundwater recharge prospective zones in arid-semi arid lands

Open Geosciences, Journal Year: 2025, Volume and Issue: 17(1)

Published: Jan. 1, 2025

Abstract Arid/semi-arid regions face water challenges much like the Arabian Peninsula, which is primarily caused by continuing shortages and growing activities of reclaimed land, as well industrial domestic activities. Consequently, identifying groundwater prospective zones (GWPZs) has become essential for securing resources. The study aims to delineate predict best areas prospection abstraction implementing analytical hierarchy process-geographic information system (GIS) techniques in a rough terrain that occupies ∼70% fractured hard rocks including ∼34% basaltic flow sheet Wadi Marawani, Saudi Arabia. To investigate combined impact model, 13 input thematic maps, elevation, slope, curvature, depression, drainage density, Topographic Wetness Index, distance river, Stream Power Terrain Roughness geology, lineaments, Normalized Difference Vegetation rainfall factors, were created, employed was subsequently merged through GIS reveal zones. These maps are mainly derived from Shuttle Radar Topography Mission, Sentinel-1, Landsat, Tropical Rainfall Measuring Mission. output map categorized very low, moderate, high, or excellent occupying ∼7%. This promising zone result intersection several criteria control occurrences. results enhanced optical radar remote sensing data, thus, suitable recharge places future governance have been identified using GIS–AHP–multicriteria decision analysis methods. For validation, large numbers well/spring locations reached 415 used total. efficiency model estimated at 79.90% (area under curve) based on receiver operating characteristic curve. Moreover, Interferometry Synthetic Aperture coherence change detection image validated predicted revealed no-coherence marked brown matched vegetated GWPZs. applied methodologies findings this present significant insights resources planning management develop similar worldwide.

Language: Английский

---

Colony Site Selection of Gray Heron (Ardea cinerea) During the Breeding Period at Multiple Spatial Scales

Ecology and Evolution, Journal Year: 2025, Volume and Issue: 15(2)

Published: Feb. 1, 2025

ABSTRACT The colony site selection of birds reflects their adaptability to the ecological environment. As one most common and widely distributed heron species, gray ( Ardea cinerea ) serves as an interesting study case for ornithologists. Researchers often characteristics understand how they adapt different environments these adaptation strategies affect survival reproduction. However, majority research has focused solely on studying in a single region at scales. To maintain model's generalization ability ensure accurate predictions preferences, we avoided using excessively similar landscapes within landscape mode. This utilizes geographic information systems (GIS) random forest (RM) models examine during breeding period herons across various regions spatial scales, providing insights into environmental factors influencing selection. By conducting China, gain valuable environments. results indicate that potentially suitable foraging habitats are primary determinant When habitat requirements met, exhibit degree flexibility choice, highlighting adaptive behaviors offering new widespread distribution. employing this approach, our findings offer wildlife conservation, emphasizing importance interdisciplinary collaboration shaping conservation strategies. Additionally, methods used may be applicable other bird species groups, preferences diverse contexts.

Language: Английский

---