Although several lithium (Li) bearing minerals have already been spectrally characterized, there are no current reference spectra for petalite in large and public access spectral libraries. This fact is aggravated by the difficulty in the identification of petalite’s diagnostic features. The study area of this work is the Fregeneda (Spain) – Almendra (Portugal) region, where distinct Li bearing minerals occur in several types of enriched pegmatite dikes. Accordingly, the objectives delineated for this work were: (i) improve the existing knowledge on the spectral signatures of Li bearing minerals (lepidolite, spodumene, petalite); (ii) compare the spectra obtained for petalite and spodumene in the study area; (iii) and compare the spectra of the Li bearing minerals from the Fregeneda-Almendra area with the reference spectra from the United States Geological Survey (USGS), the ECOSTRESS and the Geological Survey of Brazil (CPRM) spectral libraries. For that, spectral measurements were conducted in the laboratory using the SR-6500A (Spectral Evolution, Inc.) spectrometer. The results only allowed to discriminate lepidolite, since that, both, petalite and spodumene, present absorption features typical of montmorillonite and illite, or a combination between these two minerals. This is also verified in samples of corresponding minerals in other spectral libraries. No diagnostic features of these two Li bearing minerals were identified, highlighting the difficulty to spectrally discriminate them from each other and from clay minerals.
The acquisition of field data plays an important role in the calibration of the remotely sensed data, especially when combined with reflectance spectroscopy studies. The study area of this work is the Fregeneda-Almendra pegmatite field (spreading from Portugal to Spain) where different lithium (Li)-pegmatites are known. Several image processing techniques were applied to satellite images for Li exploration in the region, including machine learning classifiers. However, these algorithms identified several zones as Li-pegmatites false positives. Taking this into account, the following objectives were delineated: (i) validate the training areas used in previous studies and collect field data for training area refinement; (ii) assess the reason for the false positives previously obtained through field surveys. For that, various outcropping lithologies (Li-pegmatite, metasediments, granite) were sampled for laboratory spectral analysis. The spectral signature of Li-pegmatite was compared with the remaining outcropping lithologies. Also, the spectral signature of the sampled false positive areas was confronted with the spectra of Li-minerals. It was possible to conclude that these two classes present similar water/hydroxide and Al–OH-related features. The sampled granitic and metasedimentary rocks also presented water and/or hydroxide absorption features that can lead to some spectral confusion. However, Li-pegmatites can be discriminated from the remaining lithologies either in the training areas and the false positive areas due to the absence of iron-related absorption features and to the distinct reflectance magnitudes.
Machine learning algorithms (MLAs) have gained great importance in remote sensing-based applications, and also in mineral prospectivity mapping. Studies show that MLAs can outperform classical classification techniques. So, MLAs can be useful in the exploration of strategical raw materials like lithium (Li), which is used in consumer electronics and in the green-power industry. The study area of this work is the Fregeneda-Almendra region (between Spain and Portugal), where Li occurs in pegmatites. However, their smaller exposition can be regarded as a problem to the application of remote sensing methods. To overcome this, Support Vector Machine (SVM) and Random Forest (RF) algorithms were applied to. This study aims at: (i) comparing the performance accuracy in lithological mapping achieved by SVM and by RF; (ii) evaluating the sensitivity of both classifiers to class imbalance and; (iii) compare the results achieved with previously obtained results. For these, the same Level 1-C Sentinel-2 images (October 2017) were used. SVM showed slightly better accuracy, but RF was able to correctly classify a larger number of mapped Li-bearing pegmatites. The performance of the models was not equal for all classes, having all underperformed in some classes. Also, RF was affected by class imbalanced, while SVM prove to be more insensitive. The potential of this kind of approach in Li-exploration was confirmed since both algorithms correctly identified the presence of Li-bearing pegmatites in the three open-pit mines where they outcrop as well in areas where Li-pegmatites were mapped. Also, some of the areas classified as Li-bearing pegmatites are corroborated by the interest areas delimited in previous studies.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.