Mohammad Parsa

This paper reviews the application of lake sediment sampling in the recently glaciated terrain (last 100 000 years) of Canada where organic-rich lake sediments have been accumulating in lakes since deglaciation during the last 10 000 or so years. Lake sediment is a useful exploration sample medium for detecting geochemical dispersion from mineralized bedrock and/or metal-rich glacial dispersal trains. The method is used in areas of low to moderate topographic relief and poor to moderately developed drainage systems in the Canadian Shield and some parts of the Appalachian regions of Canada, as well as in low relief areas in the Canadian Cordillera that are covered by thick glacial sediments and have abundant lakes. Government organizations and exploration companies have conducted more than 100 lake sediment surveys across Canada since the 1970s and some of these are used to illustrate key concepts. The paper summarizes protocols for execution and interpretation of lake sediment surveys, serving as a practical guide for both the mineral exploration industry and government agencies. Methods for sample collection, processing, and analysis, quality assurance/quality control procedures, and data reporting are reviewed. Much of the information presented here is based on the methods used by the Geological Survey of Canada, British Columbia Geological Survey, and Ontario Geological Survey to conduct lake sediment surveys in Canada.

The NRCan Open S&T Repository (OSTR) provides free and open access to science publication and research funded by the Canadian government.

The NRCan Open S&T Repository (OSTR) provides free and open access to science publication and research funded by the Canadian government.

Data-driven MPM (mineral prospectivity mapping) of critical minerals, which are elements or minerals with strategic importance and high supply chain risk, is vital for national land use planning. Recently, MPM has been practiced using supervised machine learning classification algorithms. However, applying such algorithms to Canadian critical minerals presents two major challenges. The first stems from the nature of geological knowledge, which is primarily stored in unstructured text. However, most supervised machine learning algorithms struggle to directly incorporate this textual information into predictive models. The second challenge arises from the limited number of known mineral deposits associated with many critical minerals in Canada, resulting in insufficient training labeled data for supervised classification tasks. To address the first challenge, this study employed natural language processing (NLP) techniques and large language models (LLMs) to extract and transform geoscientific knowledge embedded in geoscience text corpora into predictive features for MPM. LLMs operate based on transformer deep learning architectures that use self-attention mechanisms to capture contextual relationships within natural language. A domain-specific LLM, which was fine-tuned in this study and evaluated using geology-related inquiries, was employed for MPM. To address the second challenge, a separate transformer model was developed using a self-supervised learning approach that integrates diverse geophysical, geochronological, and textual data, eliminating the dependency on a substantial number of labeled training samples. The prospectivity model generated using the proposed transformer model significantly reduced the search space—by an average of 87%—for the targeted type of mineral deposits. The findings of this study demonstrate the effectiveness of transformer-based architectures and LLMs in overcoming key limitations of modern MPM approaches for critical mineral exploration.

Mineral prospectivity mapping (MPM) can be deemed a binary classification task, with classifiers trained and validated on labels indicating the presence or absence of the targeted mineralized zones. Using economically viable mineral deposits as positive labels could, in theory, yield prospectivity models with geometallurgical reliability, thereby aiding land management and decision-making. The inherent scarcity of economically viable deposits, however, ultimately affects MPM products. The positive class label, therefore, often requires augmentation with either mineral occurrences (i.e., mineralized sites lacking economic viability) or synthetically generated labels. This paper examines how augmented positive labels and different negative label selection procedures geospatially represent economically viable mineral deposits and affect deep learning-based MPM’s classification performance and its spatial selectivity (i.e., MPM’s capability to efficiently narrow the exploration search space). To achieve this objective, large ensembles of deep learning classifiers were trained and validated with diverse combinations of positive and negative labels. Two positive class label sets were created by augmenting mineral deposits with either synthetic labels, generated using generative adversarial networks, or mineral occurrences, paired with distinct negative label sets selected based on (1) locations distant from known mineral deposits, (2) areas geospatially dissimilar to known mineral deposits, and (3) mineralized areas unrelated to the targeted style of mineralization, resulting in six unique class configurations. This study ultimately provides insights into how different label sets affect MPM's classification performance and spatial selectivity. The results indicate that selecting negative class labels from geospatially different localities enhances classification performance and MPM's spatial selectivity compared to other negative label selection procedures.

A diverse range of interpolation methods, including Kriging, spline/minimum curvature and radial basis function interpolation exist for interpolating spatially incomplete geoscientific data. Such methods use various spatial properties of the observed data to infer its local and global behaviour. In this study, we exploit the adaptability of locally interacting systems from statistical physics and develop an interpolation framework for numerical geoscientific data called Interacting Immediate Neighbour Interpolation (IINI), which solely relies on local and immediate neighbour correlations. In the IINI method, medium-to-long range correlations are constructed from the collective local interactions of grid centroids. To demonstrate the functionality and strengths of IINI, we apply our methodology to the interpolation of ground gravity, airborne magnetic and airborne radiometric datasets. We further compare the performance of IINI to conventional methods such as minimum curvature surface fitting. Results show that IINI is competitive with conventional interpolation techniques in terms of validation accuracy, while being significantly simpler in terms of algorithmic complexity and data pre-processing requirements. IINI demonstrates the broader applicability of statistical physics concepts within the field of geostatistics, highlighting their potential to enrich and expand traditional geostatistical methods.

The discovery of new lithium resources is essential because lithium plays a vital role in the manufacturing of green technology. Along with brines and volcano–sedimentary deposits, approximately a one-third share of global lithium resources is associated with lithium-cesium-tantalum (LCT) pegmatites, with Canada hosting numerous examples. This research applied generative adversarial networks, natural language processing, and convolutional neural networks to generate mineral prospectivity models and support exploration targeting for Canadian LCT pegmatites. Geoscientific text data included within public bedrock geology maps and natural language processing were used to convert conceptual targeting criteria into evidence layers that complement more traditional, geophysical and geochronological data used for mineral prospectivity modeling (MPM). A multilayer architecture of convolutional neural networks, including an attention mechanism, was designed for data modeling. This architecture was trained and validated using variable synthetically generated class labels, input image sizes, and hyperparameters, resulting in an ensemble of 1000 models. The uncertainty of the ensemble was analyzed using a risk–return analysis, yielding a bivariate choropleth risk–return plot that facilitates the interpretation of prospectivity models for downstream applications. This was further complemented by employing post hoc interpretability algorithms to translate the black-box nature of neural networks into comprehensible content. The low-risk and high return class of our prospectivity models reduces the search space for discovering LCT pegmatites by 88%, delineating 99% of known LCT pegmatites in Canada. The results of this study suggest that our workflow (i.e., combining synthetic data generation, natural language processing, convolutional neural networks, and uncertainty propagation for MPM) facilitates decision-making for regional-scale lithium exploration and could also be applied to other mineral systems.

Carbonatites are the primary geological sources for rare earth elements (REEs) and niobium (Nb). This study applies machine learning techniques to generate national-scale prospectivity models and support mineral exploration targeting of Canadian carbonatite-hosted REE +/− Nb deposits. Extreme target feature label imbalance, diverse geological settings hosting these deposits throughout Canada, selecting negative labels, and issues regarding the interpretability of some machine learning models are major challenges impeding data-driven prospectivity modeling of carbonatite-hosted REE +/− Nb deposits. A multi-stage framework, exploiting global hierarchical tessellation model systems, data-space similarity measures, ensemble modeling, and Shapley additive explanations was coupled with convolutional neural networks (CNN) and random forest to meet the objectives of this work. A risk–return analysis was further implemented to assist with model interpretation and visualization. Multiple models were compared in terms of their predictive ability and their capability of reducing the search space for mineral exploration. The best-performing model, derived using a CNN that incorporates public geoscience datasets, exhibits an area under the curve for receiver operating characteristics plot of 0.96 for the testing labels, reducing the search area by 80%, while predicting all known carbonatite-hosted REE +/− Nb occurrences. The framework used in our study allows for an explicit definition of input vectors and provides a clear interpretation of outcomes generated by prospectivity models.

Using multispectral remote sensing data in cooperation with big data processing and deep fusion learning techniques provides a new approach for mineral and hydrocarbon exploration