<div>Geological maps are powerful models for visualizing the complex distribution of rock types through space and time. However, the descriptive information that forms the basis for a preferred map interpretation is typically stored in geological map databases as unstructured text data that are difficult to use in practice. Herein we apply natural language processing (NLP) to geoscientific text data from Canada, the U.S., and Australia to address that knowledge gap. First, rock descriptions, geological ages, lithostratigraphic and lithodemic information, and other long-form text data are translated to numerical vectors, i.e., a word embedding, using a geoscience language model. Network analysis of word associations, nearest neighbors, and principal component analysis are then used to extract meaningful semantic relationships between rock types. We further demonstrate using simple Naive Bayes classifiers and the area under receiver operating characteristics plots (AUC) how word vectors can be used to: (1) predict the locations of “pegmatitic” (AUC = 0.962) and “alkalic” (AUC = 0.938) rocks; (2) predict mineral potential for Mississippi-Valley-type (AUC = 0.868) and clastic-dominated (AUC = 0.809) Zn-Pb deposits; and (3) search geoscientific text data for analogues of the giant Mount Isa clastic-dominated Zn-Pb deposit using the cosine similarities between word vectors. This form of semantic search is a promising NLP approach for assessing mineral potential with limited training data. Overall, the results highlight how geoscience language models and NLP can be used to extract new knowledge from unstructured text data and reduce the mineral exploration search space for critical raw materials.</div><div><br></div><div><strong>Citation: </strong>Lawley, C. J. M., Gadd, M. G., Parsa, M., Lederer, G. W., Graham, G. E., and Ford, A., 2023, Applications of Natural Language Processing to Geoscience Text Data and Prospectivity Modeling: Natural Resources Research. https://doi.org/10.1007/s11053-023-10216-1</div>

<div>Maps showing the potential for carbonatite-related rare earth element (REE) mineral systems in Australia. Each of the mineral potential maps is a synthesis of three or four component layers. Model 1 integrates three components: sources of metals, energy drivers, and lithospheric architecture. Model 2 integrates four components: sources of metals, energy drivers, lithospheric architecture, and ore deposition. Both models use a hybrid data-driven and knowledge driven methodology to produce the final mineral potential map for the mineral system. An uncertainty map is provided in conjunction with the mineral potential map for Model 2 that represents the availability of data coverage over Australia for the selected combination of input maps. Uncertainty values range between 0 and 1, with higher uncertainty values being located in areas where more input maps are missing data or have unknown values. An assessment criteria table is provided and contains information on the map creation.</div>

<div>Australian sediment-hosted mineral systems are important sources of base metals and critical minerals that are vital to delivering Australia’s low-carbon economy. In Australia, sediment-hosted resources account for ~82% and ~86% of the total zinc (Zn) and lead (Pb) resources respectively. Given their significance to the Australian economy, four national-scale mineral potential models for sediment-hosted Zn-Pb mineral systems have been developed: clastic-dominated siliciclastic carbonate, clastic-dominated siliciclastic mafic, Mississippi Valley-type and Irish-type. In addition to the potential for Zn-Pb mineralisation, the uncertainty related to data availability has been examined. The mineral potential models were created using a mineral systems-based approach where mappable criteria have been used to assess the prospectivity of each system. Each model has been derived from a large volume of precompetitive geoscience data. The clastic-dominated siliciclastic carbonate mineral potential model predicts 92% of known deposits and occurrences within 15.5% of the area, the clastic-dominated siliciclastic mafic mineral potential model predicts 85% of deposits and occurrences within 27% of the area, and the Mississippi Valley-type mineral potential model predicts 66% of known deposits and occurrences within 31% of the area. Each model successfully predict the location of major sediment-hosted Zn-Pb deposits while highlighting new areas of elevated prospectivity in under-explored regions of Australia, reducing the exploration search space by up to 85% for sediment-hosted Zn-Pb mineral systems.</div>

<div>The mineral potential toolkit (aka minpot-toolkit) provides tools to facilitate mineral potential analysis, from spatial associations to feature engineering and fully integrated mineral potential mapping.</div>

<div>The production of rare earth elements (REEs) is critical to the global transition to a low carbon economy. Carbonatites represent a significant source of REEs, both domestically within Australia, as well as globally. Given their strategic importance for the Australian economy, a national mineral potential assessment has been undertaken as part of the Exploring for the Future program at Geoscience Australia to evaluate the potential for carbonatite-related REE (CREE) mineral systems. Rather than aiming to identify individual carbonatites and/or CREE deposits, the focus of the mineral potential assessment is to delineate prospective belts or districts within Australia that indicate the presence of favourable criteria, particularly in terms of lithospheric architecture, that may lead to the formation of a CREE mineral system.</div><div><br></div><div>This study demonstrates how national-scale multidisciplinary precompetitive geoscience datasets can be integrated using a hybrid methodology that incorporates robust statistical analysis with mineral systems expertise to predictively map areas that have a higher geological potential for the formation of CREE mineral systems and effectively reduce the exploration search space. Statistical evaluation of the relationship between different mappable criteria that represent spatial proxies for mineral system processes and known carbonatites and CREE deposits has been undertaken to test previously published hypotheses on how to target CREE mineral systems at a broad-scale. The results confirm the relevance of most criteria in the Australian context, while several new criteria such as distance to large igneous province margins and distance to magnetic worms have also been shown to have a strong correlation with known carbonatites and CREE deposits. Using a hybrid knowledge- and data-driven mineral potential mapping approach, the mineral potential map predicts the location of known carbonatite and CREE deposits, while also demonstrating additional areas of high prospectivity in regions with no previously identified carbonatites or CREE mineralisation.</div> Presented at the AusIMM Critical Minerals Conference 2023.

<div>Sediment-hosted copper (Cu) mineral systems are important sources of base metals and critical minerals such as cobalt that are vital to delivering Australia’s low-carbon economy. In Australia, sediment-hosted Cu resources account for ~11% of the total Cu resources. Given their significance to the Australian economy, national-scale mineral potential models for sediment-hosted Cu have been developed. In addition to the potential for sediment-hosted Cu mineralisation, the uncertainty related to data availability has been examined. Three mineral potential</div><div>models derived from the combination of two mineral systems have been derived from a large volume of precompetitive geoscience data combined with mineral systems expertise, each using a different combination of input maps to assess the influence of incomplete data on the results. The mineral potential models successfully predict the location of major sediment-hosted stratiform Cu and Mount Isa-type Cu deposits while highlighting new areas of elevated prospectivity in under-explored regions of Australia, reducing the exploration search space</div><div>by up to ~84%.</div>

<div>The production of rare earth elements is critical for the transition to a low carbon economy. Carbonatites (&gt;50% carbonate minerals) are one of the most significant sources of rare earth elements (REEs), both domestically within Australia, as well as globally. Given the strategic importance of critical minerals, including REEs, for the Australian national economy, a mineral potential assessment has been undertaken to evaluate the prospectivity for carbonatite-related REE (CREE) mineralisation in Australia. CREE deposits form as the result of lithospheric- to deposit-scale processes that are spatially and temporally coincident.</div><div><br></div><div>Building on previous research into the formation of carbonatites and their related REE mineralisation, a mineral system model has been developed that incorporates four components: (1) source of metals, fluids, and ligands, (2) energy sources and fluid flow drivers, (3) fluid flow pathways and lithospheric architecture, and (4) ore deposition. This study demonstrates how national-scale datasets and a mineral systems-based approach can be used to map the mineral potential for CREE mineral systems in Australia.</div><div><br></div><div>Using statistical analysis to guide the feature engineering and map weightings, a weighted index overlay method has been used to generate national-scale mineral potential maps that reduce the exploration search space for CREE mineral systems by up to ∼90%. In addition to highlighting regions with known carbonatites and CREE mineralisation, the mineral potential assessment also indicates high potential in parts of Australia that have no previously identified carbonatites or CREE deposits.</div><div><br></div><div><b>Citation: </b>Ford, A., Huston, D., Cloutier, J., Doublier, M., Schofield, A., Cheng, Y., and Beyer, E., 2023. A national-scale mineral potential assessment for carbonatite-related rare earth element mineral systems in Australia, <i>Ore Geology Reviews</i>, V. 161, 105658. https://doi.org/10.1016/j.oregeorev.2023.105658</div>

Maps showing the potential for sediment-hosted base metal mineral systems in Australia. Each of the mineral potential maps is a synthesis of four component layers: sources of metals, energy drivers, lithospheric architecture, and depositional gradients, using a weighted sum to produce the final mineral potential map for the mineral system. Uncertainty maps are provided in conjunction with each of the mineral potential maps that represent the availability of data coverage over Australia for the selected combination of input maps. Uncertainty values range between 0 and 1, with higher uncertainty values being located in areas where more input maps are missing data or have unknown values. The set of input maps used to generate the mineral potential maps is provided along with an assessment criteria table that contains information on the map creation.