<div>Alluvial sediments have long been used in geochemical surveys as their compositions are assumed to be representative of areas upstream. Overbank and floodplain sediments, in particular, are increasingly used for regional to continental-scale geochemical mapping. However, during downstream transport, sediments from heterogeneous source regions are carried away from their source regions and mixed. Consequently, using alluvial sedimentary geochemical data to generate continuous geochemical maps remains challenging. In this study we demonstrate a technique that numerically unmixes alluvial sediments to make a geochemical map of their upstream catchments. The unmixing approach uses a model that predicts the concentration of elements in downstream sediments, given a map of the drainage network and element concentrations in the source region. To unmix sedimentary chemistry, we seek the upstream geochemical map that, when mixed downstream, best fits geochemical observations downstream. To prevent overfitting we penalise the roughness of the geochemical model. To demonstrate our approach we apply it to alluvial samples gathered as part of the Northern Australia Geochemical Survey. This survey gathered samples collected over a ∼ 500,000 km2 area in northern Australia. We first validate our approach for this sample distribution with synthetic tests, which indicate that we can resolve geochemical variability at scales greater than 0.5 – 1◦ in size. We proceed to invert real geochemical data from the total digestion of fine-grained fraction of alluvial sediments. The resulting geochemical maps for two elements of potential economic interest, Cu and Nd, are evaluated in detail. We find that in both cases, our predicted downstream concentrations match well against a held-out, unseen subset of the data, as well as against data from an independent geochemical survey. By performing principal component analysis on maps generated for all 46 available elements we produce a synthesis map showing the significant geochemical domains of this part of northern Australia. This map shows strong spatial similarities to the underlying lithological map of the area. Finally, we compare the results from our approach to a geochemical map produced by kriging. We find that, unlike the method presented here, kriging generates geochemical maps that are both dampened relative to expected magnitude, as well as being spatially distorted. We argue that the unmixing approach is the most appropriate method for generating geochemical maps from regional-scale alluvial surveys.&nbsp;</div> <b>Citation:</b> Alex G. Lipp, Patrice de Caritat, Gareth G. Roberts, Geochemical mapping by unmixing alluvial sediments: An example from northern Australia, <i>Journal of Geochemical Exploration,</i> Volume 248, <b>2023</b>, 107174, ISSN 0375-6742, https://doi.org/10.1016/j.gexplo.2023.107174. (https://www.sciencedirect.com/science/article/pii/S0375674223000213)

The National Geochemical Survey of Australia (NGSA) is Australia’s only internally consistent, continental-scale geochemical atlas and dataset (<a href="http://dx.doi.org/10.11636/Record.2011.020">http://dx.doi.org/10.11636/Record.2011.020</a>). The present dataset provides additional geochemical data for Li, Be, Cs, and Rb acquired as part of the Australian Government-funded Exploring for the Future (EFTF) program and in support of the Australian Government’s 2023-2030 Critical Minerals Strategy. The dataset fills a knowledge gap about Li distribution in Australia over areas dominated by transported regolith. The main ‘total’ element analysis method deployed for NGSA was based on making a fused bead using lithium-borate flux for XRF then ICP-MS analysis. Consequently, the samples could not be meaningfully analysed for Li. All 1315 NGSA milled coarse-fraction (<2 mm) top (“TOS”) catchment outlet sediment samples, taken from 0 to 10 cm depth in floodplain landforms, were analysed in the current project following the digestion method that provides near-total concentrations of Li, Be, Cs, and Rb. The samples were analysed by the commercial laboratory analysis service provider Bureau Veritas in Perth using low-level mixed acid (a mixture of nitric, perchloric and hydrofluoric acids) digestion with elements determined by ICP-MS (Bureau Veritas methods MA110 and MA112). The data are reported in the same format as the NGSA dataset, allowing for seamless integration with previously released NGSA data. Further details on the QA/QC procedures as well as data interpretation will be reported elsewhere. This data release also includes four continental-scale geochemical maps for Li, Be, Cs, and Rb built from these analytical data. This dataset, in conjunction with previous data published by NGSA, will be of use to mineral exploration and prospectivity modelling around Australia by providing geochemical baselines for Li, Be, Cs, and Rb, as well as identifying regions of anomalism. Additionally, these data also have relevance to other applications in earth and environmental sciences.

<div>The Heavy Mineral Map of Australia (HMMA) project1, part of Geoscience Australia’s Exploring for the Future program, determined the abundance and distribution of heavy minerals (HMs; specific gravity >2.9 g/cm3) in 1315 floodplain sediment samples obtained from Geoscience Australia’s National Geochemical Survey of Australia (NGSA) project2. Archived NGSA samples from floodplain landforms were sub-sampled with the 75-430 µm fraction subjected to dense media separation and automated mineralogy assay using a TESCAN Integrated Mineral Analysis (TIMA) instrument at Curtin University.</div><div><br></div><div>Interpretation of the massive number of mineral observations generated during the project (~150&nbsp;million mineral observations; 166 unique mineral species) required the development of a novel workflow to allow end users to discover, visualise and interpret mineral co-occurrence and spatial relationships. Mineral Network Analysis (MNA) has been shown to be a dynamic and quantitative tool capable of revealing and visualizing complex patterns of abundance, diversity and distribution in large mineralogical data sets3. To facilitate the application of MNA for the interpretation of the HMMA dataset and efficient communication of the project results, we have developed a Mineral Network Analysis for Heavy Minerals (MNA4HM) web application utilising the ‘Shiny’ platform and R package. The MNA4HM application is used to reveal (1) the abundance and co-occurrences of heavy minerals, (2) their spatial distributions, and (3) their relations to first-order geological and geomorphological features. The latter include geological provinces, mineral deposits, topography and major river basins. Visualisation of the mineral network guides parsimonious yet meaningful mapping of minerals typomorphic of particular geological environments or mineral systems. The mineralogical dataset can be filtered or styled based on mineral attributes (e.g., simplified mineralogical classes) and properties (e.g., chemical composition).</div><div><br></div><div>In this talk we will demonstrate an optimised MNA4HM workflow (identification à mapping à interpretation) for exploration targeting selected critical minerals important for the transition to a lower carbon global economy. </div><div><br></div><div>The MNA4HM application is hosted at https://geoscienceaustralia.shinyapps.io/mna4hm and is available for use by the geological community and general public.</div> This Abstract was submitted and presented to the 2023 Goldschmidt Conference Lyon, France (https://conf.goldschmidt.info/goldschmidt/2023/meetingapp.cgi)