Remote Sensing Analysis and Structural Interpretation of Gold Deposits

Remote Sensing Analysis and Structural Interpretation of Gold Deposits in Magmatic Rock-Alteration-Shear Zones and BIF-Type Gold Deposits in Archean Greenstone Belts: An Overview

Introduction

Gold has been a major contributor to the Australian economy for more than a century, and Western Australia is one of the largest gold-producing regions in the world. The Archean greenstone belts in Western Australia are known for their gold mineralization, including magmatic rock-alteration-shear zones and BIF-type gold deposits.

The gold mineralization in Western Australia is spread across a vast region covering more than 1500 km. The major gold-producing regions include the Yilgarn Craton, Pilbara Craton, and Capricorn Orogen. The Yilgarn Craton, which covers an area of approximately 380,000 square kilometers, is the largest and most productive gold province in Australia, accounting for more than 70% of Australia's gold production. The Pilbara Craton, located in the north-western part of Western Australia, is known for its high-grade gold deposits. The Capricorn Orogen, located in the northern part of Western Australia, is a relatively underexplored gold province.

According to the Western Australian Department of Mines, Industry Regulation and Safety, the state's gold resources were estimated at 10,813 tonnes in 2021. The gold resources in Western Australia are hosted in a variety of geological settings, including greenstone belts, granite-greenstone terranes, and sedimentary basins.

Western Australia is home to many key gold projects, including the Kalgoorlie Super Pit, which is one of the world's largest open-pit gold mines, and the Telfer Mine, which is one of the largest gold-copper mines in the world. Other significant gold projects in Western Australia include the Boddington Gold Mine, Gruyere Gold Project, and the Golden Grove Mine.

In this article, we will discuss the remote sensing analysis and structural interpretation of gold deposits in magmatic rock-alteration-shear zones and BIF-type gold deposits in Archean greenstone belts in Western Australia. We will explore the data acquisition, image pre-processing, image classification, feature extraction, image interpretation, and validation techniques used in the analysis.

Data Acquisition

The first step in remote sensing analysis is to obtain high-quality data for the study area. For this project, we collected Sentinel-2 multispectral imagery with a 10-meter spatial resolution, ASTER multispectral imagery with a 15-meter spatial resolution, and Italian PRISMA hyperspectral imagery. Additionally, we also utilized high-resolution meter-submeter (< 2 m) Google Earth imagery. The collected data were compiled into a remote sensing database for further analysis. The following is the details the acquisition of several types of remote sensing data project involved in this project.

The first type of data is the Sentinel-2 multispectral imagery with a 10-meter spatial resolution, which provides information on the visible, near-infrared, and shortwave infrared regions of the electromagnetic spectrum. This data source is particularly useful for mapping land cover, vegetation, and mineralogy.

The second type of data is the ASTER multispectral imagery with a 15-meter spatial resolution, which also provides information on the visible, near-infrared, and shortwave infrared regions of the electromagnetic spectrum. In addition, this data source provides thermal infrared data, which can be used to map surface temperature and thermal anomalies.

The third type of data is the Italian PRISMA hyperspectral imagery, which provides high spectral resolution data covering a wide range of the electromagnetic spectrum. This data source is particularly useful for mapping mineralogy and identifying subtle changes in surface composition.

The fourth type of data is meter-submeter (<2 m) Google Earth imagery, which provides high-resolution images of the study area that can be used to identify geological features, such as faults, folds, and other structural elements.

Together, these data sources provide a comprehensive view of the study area and can be used to identify potential areas for gold exploration and to map geological structures and mineralogical features associated with gold deposits.

Image Pre-processing

Image pre-processing is an essential step before analyzing remote sensing imagery. The pre-processing steps include image correction, image registration, and image enhancement.

Image correction aims to eliminate any errors caused by atmospheric conditions or sensor characteristics during data acquisition. The corrections can include radiometric calibration, atmospheric correction, and geometric correction. Radiometric calibration is used to convert the digital number values into radiance values, while atmospheric correction corrects for the effects of the atmosphere on the image. Geometric correction aims to correct for any geometric distortions in the image caused by sensor characteristics or terrain.

Image registration involves aligning different images that cover the same area, but acquired at different times or by different sensors. This process ensures that all the images used in the analysis are georeferenced and aligned correctly.

Image enhancement techniques are used to improve the visual quality of the image and highlight features of interest. Enhancement techniques can include contrast stretching, histogram equalization, and filtering. Contrast stretching improves the contrast of the image by expanding the range of pixel values, while histogram equalization adjusts the pixel values to improve the overall brightness and contrast. Filtering can be used to remove noise or highlight specific features in the image.

Overall, the image pre-processing step is critical to ensure that the images used in the analysis are of high quality and free from any errors or distortions that could affect the accuracy of the results.

Image Classification

Image classification is an important step in the analysis of remote sensing imagery. It involves assigning pixels to specific classes or categories based on their spectral properties. In this project, we used both supervised and unsupervised classification methods to classify the imagery.

Supervised classification involves manually selecting training samples for each class of interest and using these samples to train a classifier algorithm. The classifier then uses the spectral properties of each pixel in the image to assign it to a specific class. This method is more accurate but also more time-consuming.

Unsupervised classification, on the other hand, does not require any prior knowledge of the classes. Instead, the algorithm groups pixels with similar spectral properties into clusters or classes. The user then assigns each class to a specific land cover type based on their knowledge of the study area. This method is faster but may be less accurate than supervised classification.

In this project, we classified the imagery into three classes of interest: alteration zones, lithological units, and structures. Alteration zones were identified based on their spectral properties, such as the presence of specific mineralogical absorption features. Lithological units were classified based on their reflectance properties, such as the difference between the reflectance of different rock types. Structures were identified based on their linear features, such as faults and shear zones.

Overall, image classification was an important step in identifying the different features of interest in the study area and providing a basis for further analysis.

Feature Extraction

After the image classification process, the next step is feature extraction. This involves extracting meaningful information from the classified imagery. In our project, we focused on extracting features that are crucial for identifying potential areas of gold mineralization. These features include ferric and ferrous contamination, hydrothermal alteration, argillic alteration, advanced argillic alteration, propylitic alteration, and chloritization/carbonation.

Ferric and ferrous contamination can indicate the presence of iron oxide minerals, which are often associated with gold mineralization. Hydrothermal alteration refers to the alteration of rocks by hot fluids, and can include silicification, sericitization, and alunitization. This alteration is often associated with gold mineralization in magmatic rock-alteration-shear zones.

Argillic alteration is the alteration of rocks by clay minerals, which can include kaolinite, illite, and smectite. Advanced argillic alteration is characterized by the presence of alunite, which forms in the presence of acidic fluids. Propylitic alteration is the alteration of rocks by minerals such as epidote, chlorite, and calcite. Chloritization/carbonation is the alteration of rocks by chlorite and carbonate minerals. These alteration features are associated with BIF-type gold deposits in Archean greenstone belts.

By extracting these features, we can create maps that highlight potential areas of gold mineralization based on the types of alteration present in the rocks. This information can be used to guide further exploration and development of gold deposits.

Image Interpretation

Image interpretation is a critical step in the remote sensing analysis of gold deposits. In this step, the extracted features are analyzed and interpreted to identify areas of interest. In this project, we used the remote sensing lithologic and alteration information to delineate prospecting areas of gold deposits.

We also mapped alterations caused by gold mineralization in the delineated prospecting areas using hyperspectral imagery. The mapped alterations include ferric and ferrous contamination, hydrothermal alteration, argillic alteration, advanced argillic alteration, propylitic alteration, and chloritization/carbonation. These alterations are crucial for identifying potential areas of gold mineralization.

Furthermore, we delineated hydrothermal alteration zones associated with gold mineralization and pinpointed the alteration center, mineralization center, and check points in the field. These zones are important as they can help to locate areas of high gold concentration.

In addition to identifying potential gold mineralization areas, image interpretation also helps in identifying structural features such as faults, folds, and shear zones that may control gold mineralization. Structural interpretation is an important aspect of the analysis as it provides valuable information on the geometry and orientation of gold mineralization zones, which is critical for exploration and mining activities.

Overall, image interpretation plays a crucial role in the remote sensing analysis of gold deposits as it provides valuable information on potential mineralization areas and structural features that can aid in exploration and mining activities.

Validation

Validation is an essential step in any remote sensing project to ensure the accuracy of the results. In this project, we validated the remote sensing analysis results with geological and geochemical data collected from fieldwork. We used various methods to validate our results, including ground truthing and statistical analysis.

Ground truthing involves visiting the identified areas of interest in the field and verifying the presence or absence of gold mineralization. We also collected rock samples from the identified areas and analyzed them for gold content using geochemical analysis techniques such as X-ray fluorescence and inductively coupled plasma mass spectrometry. We compared the results of these analyses with the remote sensing results to ensure accuracy.

We also used statistical analysis to validate the remote sensing results. We used techniques such as receiver operating characteristic (ROC) analysis and confusion matrix analysis to determine the accuracy of the remote sensing classifications.

Overall, the validation process helped us to confirm the accuracy of our remote sensing analysis and provided us with confidence in the results.

Conclusion

Remote sensing analysis is a valuable tool for exploring gold mineralization in Archean greenstone belts. This project demonstrated the usefulness of remote sensing data for identifying potential areas of gold mineralization and mapping alterations caused by gold mineralization. The results obtained from this study can be used to guide future fieldwork and exploration efforts.