Lithological Mapping Research Articles

Machine Learning Algorithms for Lithological Mapping Using Landsat 9 Data in Central Western Highlands of Yemen

This research designed the lithological units of the Central Western Highlands of Yemen (encompassing parts of Dhamar, Raymah, Sana’a, and northern Ibb) using Landsat 9 imagery. The area's complex geological features, characterized by units of the Yemen Volcanic Group from the Tertiary and Quaternary eras, Tertiary granite intrusions, and limestone, sandstone, metamorphic rocks, and Quaternary deposits, pose challenges for traditional field mapping techniques. By leveraging the spectral resolution of Landsat 9, this study aims to achieve accurate classification and mapping of lithological units. ENVI 5.6 software was used for image processing, applying a supervised classification approach represented by the two most common methods: Support Vector Machine (SVM) and Maximum Likelihood Classifier (MLC), based on training samples for each lithological class. The accuracy assessment of the classification was validated through an error matrix. The overall accuracy of SVM reached 85.3% with a Kappa coefficient of 0.8, while the overall accuracy of MLC reached 83.3% with a Kappa coefficient of 0.8, indicating a high degree of consistency and reliability in the classification process. This signifies a highly reliable classification outcome. The findings of this study highlight the significant advantages of utilizing Landsat 9 for detailed geological mapping of complex terrains, demonstrating a notable improvement in efficiency and accuracy over traditional methodologies. It can be relied upon to classify lithological units in other areas.

Rapid lithological mapping using multi-source remote sensing data fusion and automatic sample generation strategy

ABSTRACT The advancement of remote sensing technology aids geologists in obtaining lithological maps more quickly, comprehensively, and accurately. However, key challenges in lithological mapping include the limited spectral information from individual sensors and the difficulties in visually interpreting lithological samples. In this study, we integrated 241 scenes of optical data and 106 scenes of radar data on the Google Earth Engine (GEE) platform, proposing a rapid lithological identification framework that combines an automatic lithological sample data generation strategy with multi-source data. Using various machine learning algorithms, we evaluated the classification capabilities of heterogeneous predictive factors, feature optimization algorithms, and object-based algorithms. Results indicate that: (1) Combining optical and radar data improves prediction accuracy, with terrain data further enhancing mapping capabilities; (2) Terrain factors contribute most to classification, but SWIR and TIR bands of optical data are critical for lithological identification; (3) The feature optimization algorithm reduces feature redundancy and efficiency issues from multi-source data, achieving 96.51% accuracy with the optimal feature model, an improvement of 0.1%−2.02% over original features; (4) Object-based algorithms show significant potential in mapping areas with large rock outcrops. This study offers new insights for medium- to large-scale lithological maps and provides essential data support for geological work.

Selection of Optimum Enhancement Technique for the Discrimination of Alkali Granites and Syenites around Suryamalai Batholith of Central Tamil Nadu, India

Image processing techniques such as Band Ratio, Principal Component Analysis, Supervised, Unsupervised classification and Reclassification techniques are generally used as an enhancement technique for alteration zone demarcation and lithological discrimination in geosciences environment. In the present study, Landsat OLI satellite imagery has been utilized to produce various combinations of image enhancement to discriminate alkali syenite and alkali granite of Proterozoic age. These techniques were also used to differentiate three phases of granites reported in Suryamalai Batholith which is very difficult to map by conventional method due to its rugged topographic nature. 12 types of enhancement techniques were applied to differentiate rock alkali syenite from granite, however only few methods such as Kaufmanns ratio, ratio of 5/7, 3/1 and 3/5 and supervised classification were found to be suitable for the discrimination of those rocky types. Band ratio 5/7, 3/1 and 3/5, Crosta analysis and thermal reclassification techniques significantly brought to light the difference between the metamorphic rocks such as fissile hornblende biotite gneiss and hornblende biotite gneiss of Archaean age. Similarly, phase I and phase III granites were perfectly discriminated from all the image enhancement techniques. It is differentiated mainly based on its grain size and mineral variation between these two rock types. Study revealed that standard digital image processing technique can be effectively used for lithological mapping.

Lineament and Lithological Mapping of Meta-Sediments and Granitoids of Goz-Beïda (Eastern Chad) Based on Semi-Automatic Processing of Landsat 8 Oli/Tirs Images

Lineament and Lithological Mapping of Meta-Sediments and Granitoids of Goz-Beïda (Eastern Chad) Based on Semi-Automatic Processing of Landsat 8 Oli/Tirs Images

Enhanced lithological mapping in arid crystalline regions using explainable AI and multi-spectral remote sensing data

Lithological classification is essential for understanding the spatial distribution of rocks, especially in arid crystalline areas. Artificial intelligence (AI) recent advancements with multi-spectral satellite imagery have been utilized to enhance lithological mapping in these areas. Here we employed different AI models namely, Support Vector Machine (SVM), Random Forest Classification (RFC), Logistic Regression, XGBoost, and K-nearest neighbors (KNN) for lithological mapping. This was followed by the application of explainable AI (XAI) for lithological discrimination (LD) which is still not widely explored. Based on the highest accuracy and F1 score of the previously mentioned models, RFC model outperformed all of them, and hence, it was integrated with XAI, using the SHapley Additive exPlanations (SHAP) method.This approach successfully identified critical multi-spectral features for LD in arid crystalline zones when applied on the Landsat-8, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), and SRTM-DEM datasets covering the Hammash and the Wadi Fatimah areas in Egypt and the Kingdom of Saudi Arabia, respectively. Field validation in the Hammash area confirmed the RFC model's efficacy, achieving a satisfactory 94% overall accuracy for 18 features. SHAP was able to identify the top ten features for proper LD over the Hammash area with 90.3% accuracy despite the complex nature of the ophiolitic mélange. For validation purposes, RCF was then utilized in the Wadi Fatimah region, using only the top 10 critical features rendered from the SHAP analysis. It performed well and had 93% accuracy. Notably, XAI/SHAP results indicated that elevation data, Landsat-8's Green Band (B3), and the two ASTER SWIR bands (B5 and B6) were essential and significant for identifying island arc rocks. Moreover, the SHAP model effectively delineated complex mélange matrices, primarily using ASTER SWIR band (B8). Our findings highlight the successful combination of RFC with XAI for LD and its potential utilization in similar arid crystalline environments worldwide.

WITHDRAWN: Improving Lithological Mapping of Tazoult n'Ouzina Massif, Eastern Anti-Atlas, Morocco: Evidences from remote sensing and field Investigations

WITHDRAWN: Improving Lithological Mapping of Tazoult n'Ouzina Massif, Eastern Anti-Atlas, Morocco: Evidences from remote sensing and field Investigations

Integration of Remotely Sensed Data and the Petrographic Analysis for Lithological Mapping of Neoproterozoic Basement Rocks at Um Had Area, Central Eastern Desert, Egypt

Integration of Remotely Sensed Data and the Petrographic Analysis for Lithological Mapping of Neoproterozoic Basement Rocks at Um Had Area, Central Eastern Desert, Egypt

A Novel Sample Generation Method for Deep Learning Lithological Mapping with Airborne TASI Hyperspectral Data in Northern Liuyuan, Gansu, China

High-resolution and thermal infrared hyperspectral data acquired from the Thermal Infrared Airborne Spectrographic Imager (TASI) have been recognized as efficient tools in geology, demonstrating significant potential for rock discernment. Deep learning (DL), as an advanced technology, has driven substantial advancements in lithological mapping by automatically extracting high-level semantic features from images to enhance recognition accuracy. However, gathering sufficient high-quality lithological samples for model training is challenging in many scenarios, posing limitations for data-driven DL approaches. Moreover, existing sample collection approaches are plagued by limited verifiability, subjective bias, and variation in the spectra of the same class at different locations. To tackle these challenges, a novel sample generation method called multi-lithology spectra sample selection (MLS3) is first employed. This method involves multiple steps: multiple spectra extraction, spectra combination and optimization, lithological type identification, and sample selection. In this study, the TASI hyperspectral data collected from the Liuyuan area in Gansu Province, China, were used as experimental data. Samples generated based on MLS3 were fed into five typical DL models, including two-dimensional convolutional neural network (2D-CNN), hybrid spectral CNN (HybridSN), multiscale residual network (MSRN), spectral-spatial residual network (SSRN), and spectral partitioning residual network (SPRN) for lithological mapping. Among these models, the accuracy of the SPRN reaches 84.03%, outperforming the other algorithms. Furthermore, MLS3 demonstrates superior performance, achieving an overall accuracy of 2.25–6.96% higher than other sample collection methods when SPRN is used as the DL framework. In general, MLS3 enables both the quantity and quality of samples, providing inspiration for the application of DL to hyperspectral lithological mapping.

Application of airborne gamma-ray spectrometry in geological lithology mapping—a case study of a region in the Qinling Mountains, China

Application of airborne gamma-ray spectrometry in geological lithology mapping—a case study of a region in the Qinling Mountains, China

Machine learning and remote sensing-based lithological mapping of the Duwi Shear-Belt area, Central Eastern Desert, Egypt

Machine learning and remote sensing techniques are widely accepted as valuable, cost-effective tools in lithological discrimination and mineralogical investigations. The current study represents an attempt to use machine learning classification along with several remote sensing techniques being applied to Landsat-8/9 satellite data to discriminate the various outcropping lithological rock units at the Duwi Shear Belt (DSB) area in the Central Eastern Desert of Egypt. Multi-class machine learning classification, multiple conventional remote sensing mapping techniques, spectral separability analysis based on the Jeffries-Matusita (J-M) distance measure, fieldwork, and petrographic investigations were integrated to enhance the lithological discrimination of the exposed rock units at DSB area. The well-recognized machine learning classifier (Support Vector Machine—SVM) was adopted in this study, with training data determined carefully based on enhancing the lithological discrimination attained from various remote sensing techniques of False Color Composites (FCC), Principal Component Analysis (PCA), and Minimum Noise Fraction (MNF), along with the fieldwork data and the previously published geologic maps. High overall accuracy of the SVM classification was obtained, however, inspection of the individual rock unit classes’ accuracies revealed lower accuracy for certain types of rock units which were also found associated with lower separability scores as well. Among the least separable rock units were; metagabbro rocks that showed high spectral similarity with the volcaniclastic metasediments rocks, and the metaultramafics of the ophiolitic mélange showed spectral attitude of high correlation to that of the Hammamat volcanosedimentary rocks. Target-oriented Color Ratio Composites (CRC) technique was implemented to better discriminate these hardly separable rock units. A final integrated geological map was obtained comprising the various discriminated Neoproterozoic basement rock units of the DSB area. The successfully mapped litho-units include; Meatiq Group (amphibolites, gneissic granitoids, and mylonitized granitoids), ophiolitic mélange (metaultramafics, metagabbro-amphibolites, and volcaniclastic metasediments), Dokhan volcanics, Hammamat sediments, and granites. An adequate description of these rock units was also given in light of the conducted intense fieldwork and petrographic investigations.

Analysis of Devanur and Manamedu Ophiolite Complexes in SGT, India: A detailed examination employing remote sensing techniques and Laboratory Spectral Signature investigations

This study employs advanced satellite imagery from ASTER and Sentinel-2A to conduct detailed lithological mapping of the Devanur and Manamedu ophiolite complexes in the southern Central Shear Zone (CSZ). The primary focus is on the Manamedu Ophiolite Complex (MOC) and the Devanur Ophiolitic Complex (DOC). Image enhancement techniques such as Color composites, Principal Component Analysis (PCA), and Minimum Noise Fraction (MNF) were utilized to differentiate various rock types. RGB band combinations derived from PCA and MNF outputs demonstrated effective discrimination of rock units. Spectral Angle Mapper (SAM) and Support Vector Machine (SVM) classification methods were employed on ASTER and Sentinel-2A images, yielding classified lithologies that closely matched existing maps from the Geological Survey of India (GSI) and other studies, validating the accuracy of the findings. Additionally, Laboratory Spectral Signature Studies were conducted on 10 rock samples using an ASD FieldSpec Pro® spectroradiometer, providing reflectance spectra from 350 nm to 2500 nm. These spectra, particularly the continuum-removed reflectance, revealed diagnostic absorption features that were corroborated by geochemical analyses. A detailed analysis investigated how elemental compositions and key minerals influenced absorption bands. Major oxide geochemical compositions of DOC and MOC samples were identified using XRF methods. The aim of this research is to characterize DOC and MOC through remote sensing and spectral signature analysis. Sentinel-2A data proved more effective in lithological discrimination compared to ASTER, with spectral signatures indicating the presence of iron (Fe) and magnesium (Mg) contents. Notably, SVM classification of Sentinel-2A MNF + DEM data achieved an overall accuracy of more than 90% when compared with field investigations. This study underscores the efficacy of processing VNIR and SWIR bands from ASTER and Sentinel-2A satellite imagery alongside DEM data and ground surveys for mapping mafic-ultramafic rocks in the DOC and MOC regions of the CSZ.

Lithological Discrimination of Khyber Range Using Remote Sensing and Machine Learning Algorithms

In this study, the satellite data of ASTER and Landsat 8 OLI were used for the discrimination of lithological units covering the Khyber range. Of the 24 tested band combinations, the most suitable include 632 and 468 of ASTER and 754 and 147 of OLI in the RGB sequence. The data were also tested with two conventional machine learning algorithms (MLAs), namely maximum likelihood classification (MLC) and support vector machine (SVM), for lithological mapping. Principal component analysis (PCA), minimum noise fraction (MNF), band ratios, and color composites in combination with available lithological maps and field data were utilized for training sample collection for the MLC and SVM models to classify the lithological units. The accuracy assessment of SVM and MLC was performed using a confusion matrix, which revealed a higher accuracy of 74.8419% and 72.1217% for ASTER and an accuracy of 58.4833% and 60.0257% for OLI, respectively. The results indicate that ASTER imagery is more suitable for lithological discrimination in the study area due to its high spectral resolution in the VNIR to SWIR range. The experiment revealed that the SVM classification offered the highest overall accuracy of nearly 75% and the kappa coefficient value of 0.7 on ASTER data. This demonstrates the effectiveness of SVM classification in exploring lithological mapping in dry to semi-arid regions.

Using Selective Principal Component Analysis (SPCA) for Lithologic Mapping of Different Granitic Phases in South Sinai, Egypt

Using Selective Principal Component Analysis (SPCA) for Lithologic Mapping of Different Granitic Phases in South Sinai, Egypt

Graph convolutional network for lithological classification and mapping using stream sediment geochemical data and geophysical data

Lithological mapping plays a vital role in acquiring a comprehensive understanding of subsurface rocks and geological structures. Traditional lithological mapping relies heavily on manual field surveys, with the accuracy of the mapping greatly constrained by the expertise of mappers, and thus this process is commonly slow and costly. In the present study, the graph convolutional network (GCN) method was proposed for efficient and automatic lithological mapping, using various geological survey data (e.g. geochemical and geophysical data). The GCN is not limited by the spatial distribution of sampling data, and hence it can be directly applied to the integrated resampling data that do not adhere to the regular spatial distribution patterns of Euclidean space. To evaluate the potential application of the GCN method in lithological classification and mapping, a case study was conducted in the Maqin region (China). Geochemical and geophysical data were utilized as the basis for lithological classification, as they can provide valuable information on geochemical compositions and physical properties of geological bodies, respectively. The study process was primarily divided into three parts. First, basic pre-processing was conducted on the geochemical data and the geophysical data, including centred log-ratio transformation and resampling. Subsequently, the resampled points were divided into training and testing sets in the proportion of 5.5:1. In the training set, 20% of the points are randomly selected as the validation set. Ultimately, using each resampled point as a node and the integrated geochemical and geophysical data as features, a GCN model and an undirected graph were constructed based on an adjacency distance of 1 km. The graph was employed by the constructed GCN model for lithological classification. The results reveal that the majority of lithological units can be accurately classified with an overall accuracy of 86%, indicating that the GCN model exhibits good applicability in lithological classification.

Integrating remote sensing and field investigation for lithological mapping of Per-Eonile to Neonile sequences west of Sohag city, Egypt: Impact on urban development

In the domains of geological study, natural resource exploitation, geological hazards, sustainable development, and environmental management, lithological mapping holds significant importance. Conventional approaches to lithological mapping sometimes entail considerable effort and difficulties, especially in geographically isolated or inaccessible regions. Incorporating geological surveys and satellite data is a powerful approach that can be effectively employed for lithological mapping. During this process, contemporary RS-enhancing methodologies demonstrate a remarkable proficiency in identifying complex patterns and attributes within the data, hence facilitating the classification of diverse lithological entities. The primary objective of this study is to ascertain the lithological units present in the western section of the Sohag region. This objective will be achieved by integrated Landsat ETM+ satellite imagery and field observations. To achieve our objectives, we employed many methodologies, including the true and false color composition (FCC&TCC), the minimal noise fraction (MNF), principal component analysis (PCA), decoration stretch (DS), and independent component analysis (ICA). Our findings from the field investigation and the data presented offer compelling evidence that the distinct lithological units can be effectively distinguished. A recently introduced geology map has been incorporated within the research area. The sequence of formations depicted in this map is as follows: Thebes, Drunka, Katkut, Abu Retag, Issawia, Armant, Qena, Abbassia, and Dandara. Implementing this integrated technique enhances our comprehension of geological units and their impacts on urban development in the area. Based on the new geologic map of the study area, geologists can improve urban development in the regions by detecting building materials “aggregates”. This underscores the significance and potential of our research in the context of urban development.

Geological mapping and mineral exploration in Francistown region, north east Botswana from integration of ASTER TIR and aeromagnetic data

To map the region's geology and mineral exploration, this study looked at how remote sensing satellite imaging from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) Thermal Infrared (TIR) sensor and aircraft geophysics may be used. The Francistown region forms part of the Zimbabwe craton, northeast of Botswana. It comprises felsic rock types, including granite and dry river sand, mafic-ultramafic rocks, and gneiss country rock. Based on their spectral emissivity responses, mafic-ultramafic units and quartz-rich units were discovered. These units are important targets for the discovery of mineral resources. The research region can be described as convoluted, fractured, intruded, metamorphosed, and ultimately worn from a geological and geophysical standpoint. Several remote sensing and aeromagnetic approaches were employed. The false color composite (FCC) and mineral indices using ASTER TIR data provided helpful information for delineating the area's geology and concentration of major-rock-forming minerals. The aeromagnetic data extracted lineaments well, and the Center for Exploration Target (CET) grid analysis system was very helpful. The CET Porphyry Analysis technique was applied in mapping near-circular porphyry features, which are good indicators of possible sites with high mineralization potential. The analysis of the lineament map shows that mineral deposits are mostly along the NE-SW direction. The results from the ASTER TIR and aeromagnetic enhancements indicate the occurrence of mafic-ultramafic rocks and felsic-rich units, which have been documented to host mineralization in the region. Porphyry structures are also mapped successfully along the general strike of the geological structure in the study area. The combination of these methods proved their success, and they can be used further in other regions of high mineral potential and lithological mapping for their cost efficiency and reliability.

Lithological discrimination and mapping in part of the upper Benue Trough using Landsat 8

This study focused on evaluating the efficiency and complementarity of image enhancement techniques for lithological discrimination and mapping in the upper parts of Benue Trough using Landsat 8 data. To achieve this, image enhancement techniques such optimum index factor, band ratio, principal component analysis, and minimum noise fraction were applied on Landsat 8 OLI data covering parts of the Upper Benue Trough. False colour composites (FCC) were created from three most informative derivatives of these image enhancements. The FCCs were compared in terms of their capability to differentiate different lithological units in the study area by comparing the FCCs with existing geological map. The results show that the FCC created from band ratios provided the highest amount of information, while those created based on optimum index factor provided the least amount of information with respect to lithological discrimination. Additionally, it was observed that the FCCs created from the derivatives of band ratio, principal component analysis and minimum noise fraction provide complementarity. Thus, the three FCCs from these three enhancement methods were fused into a single image, which further aided lithological discrimination. By combining the outputs from three image enhancement techniques, most of the sedimentary and crystalline rocks in the study area were successfully delineated. The method adopted in this study particularly the fusion of outputs from band ratio, principal component analysis and minimum noise fraction represents a novel approach that is suitable for lithological mapping in remote areas, particularly arid and semi-arid terrains with considerable rock exposure.

Evaluating ASTER data and field spectrometry for lithological discrimination in semi-arid region, Northeast Kohat Plateau, Pakistan

This study employs data of the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensor to delineate and map the distribution of sedimentary lithologies in the semi-arid region of Kohat Plateau, Pakistan. False color composites (FCC) and various image transformation and enhancement techniques including the principal component analysis (PCA), minimum noise fraction (MNF), and band rationing (BR) were used successfully to differentiate four lithological classes. These lithologies include chemically/biochemically formed beds of the marine environment and detrital sequences of marginal marine to the riverine environment. FCC from original reflectance data, PCA, and BR techniques displayed more prominent lithological variation. To map the lithology and show the potential of ASTER data, field spectrometry over the barren lithologies was carried out. The end-member spectra from field spectrometry shows strong agreement with the pixels spectra from ASTER scene. The Spectral Angle Mapper (SAM) mapping method was then used to produce a classified lithological map, where the image pixels spectra proved more suitable reference, in comparison to the end-member spectra. The accuracy of the classified lithological map was evaluated based on field-based point data, which resulted an overall accuracy of 70% and a Kappa coefficient value of 0.679. Carbonates and evaporites showed relatively higher user and producer accuracies which are attributed to their topographic behavior and weathered scree over the adjacent rock unit. The final lithological map provided a clearer picture of surface geology where the existing geological maps lacked lithological continuity.

Fusion of GaoFen-5 and Sentinel-2B data for lithological mapping using vision transformer dynamic graph convolutional network

Lithological identification and mapping using remote sensing (RS) imagery are challenging. Traditional lithological mapping relies mainly on multispectral data and machine learning methods. However, inadequate spectral information and inappropriate classification algorithms are major problems for RS geological applications. Moreover, satellite hyperspectral images (HSI) at low spatial resolution and convolutional neural network (CNN)-based methods with incomplete feature extraction remain challenging because of the limitations of sensor imaging and convolutional kernels for lithological mapping. To address the above issues, in this study, smoothing filter-based intensity modulation (SFIM) fusion technology is first employed to fuse GaoFen-5 hyperspectral images and Sentinel-2B multispectral images. This approach significantly improves spatial details and enriches spectral information. Subsequently, a novel Vision Transformer Dynamic Graph Convolutional Network (ViT-DGCN) is proposed for lithological mapping of the Cuonadong dome, Tibet, China. ViT-DGCN is a joint model consisting of a transformer and a dynamic graph convolution module that enhances feature extraction capabilities by exploring long-range interaction sequence features and dynamic graph structure information in a targeted manner. The proposed algorithm exhibits superior performance compared to the others, achieving an overall accuracy of 97% for the Cuonadong dome using only 1% of the available training samples.

Design and implementation of an Automatic Deep Stacked Sparsely Connected Convolutional Autoencoder (ADSSCCA) neural network for remote sensing lithological mapping using calculated dropout

Design and implementation of an Automatic Deep Stacked Sparsely Connected Convolutional Autoencoder (ADSSCCA) neural network for remote sensing lithological mapping using calculated dropout