Search

Featured Application: Inference of whole rock geochemistry parameters can be inferred from spectral data collected from pulps by combining a dimesionality reduction scheme in the form of non-negative matrix functions and a random forest regression model. The proposed workflow allows the infered geochemical parameters to be used in a smart sampling scheme for both orebody characterisation and/or budgetary constraints. In both cases it can point to areas that do not require intensive sampling while locating areas of mineralisation that are better suited to higher sampling rates. The efficacy of predicting geochemical parameters with a 2-chain workflow using spectral data as the initial input is evaluated. Spectral measurements spanning the approximate 400–25000 nm spectral range are used to train a workflow consisting of a non-negative matrix function (NMF) step, for data reduction, and a random forest regression (RFR) to predict eight geochemical parameters. Approximately 175,000 spectra with their corresponding chemical analysis were available for training, testing and validation purposes. The samples and their spectral and chemical parameters represent 9399 drillcore. Of those, approximately 20,000 spectra and their accompanying analysis were used for training and 5000 for model validation. The remaining pairwise data (150,000 samples) were used for testing of the method. The data are distributed over two large spatial extents (980 km2 and 3025 km2, respectively) and allowed the proposed method to be tested against samples that are spatially distant from the initial training points. Global R2 scores and wt.% RMSE on the 150,000 validation samples are Fe (0.95/3.01), SiO2 (0.96/3.77), Al2O3 (0.92/1.27), TiO (0.68/0.13), CaO (0.89/0.41), MgO (0.87/0.35), K2O (0.65/0.21) and LOI (0.90/1.14), given as Parameter (R2/RMSE), and demonstrate that the proposed method is capable of predicting the eight parameters and is stable enough, in the environment tested, to extend beyond the training sets initial spatial location. [ABSTRACT FROM AUTHOR]

The integration of surface and subsurface geoscience data is critical for efficient and effective mineral exploration and mining. Publicly accessible datasets to evaluate the various geoscience analytical tools and their effectiveness for characterisation of mineral assemblages and lithologies or discrimination of ore from waste are however scarce. The open access Rocklea Dome 3D Mineral Mapping Test Data Set (Laukamp, 2020; https://doi.org/10.25919/5ed83bf55be6a) provides an opportunity for evaluating proximal and remote sensing data, validated and calibrated by independent geochemical and mineralogical analyses, for exploration of channel-iron deposits (CID) through cover. We present hyperspectral airborne, surface and drill core reflectance spectra collected in the visible-near infrared and shortwave infrared wavelength ranges (VNIR-SWIR; 350 to 2500 nm), as well as whole rock geochemistry obtained by means of X-Ray fluorescence analysis and loss on ignition measurements of drill core samples. The integration of surface with subsurface hyperspectral data collected in the frame of previously published Rocklea Dome 3D Mineral Mapping case studies demonstrated that about 30 % of exploration drill holes were sunk into barren ground and could have been of better use, located elsewhere, if airborne hyperspectral imagery had been consulted for drill hole planning. The remote mapping of transported Tertiary detritals (i.e. potential hosts of channel iron ore resources) versus weathered in situ Archaean geology (i.e. barren ground) has significant implications for other areas where "cover" (i.e. regolith and/or sediments covering bedrock hosting mineral deposits) hinders mineral exploration. Hyperspectral remote sensing represents a cost-effective method for regolith landform mapping required for planning drilling programs. In the Rocklea Dome area, vegetation unmixing methods applied to airborne hyperspectral data, integrated with subsurface data, resulted in seamless mapping of ore zones from the weathered surface to the base of the CID - a concept that can be applied to other mineral exploration and mineral deposit studies. Furthermore, the associated, independent calibration data allowed to quantify iron oxide phases and associated mineralogy from hyperspectral data. Using the Rocklea Dome data set, novel geostatistical clustering methods were applied to the drill core data sets for ore body domaining that introduced scientific rigour to a traditionally subjective procedure, resulting in reproducible objective domains that are critical for the mining process. Beyond the already published case studies, the Rocklea Dome 3D Mineral Mapping Test Data Set has the potential to develop new methods for advanced resource characterisation and develop new applications that aid exploration for mineral deposits through cover. The here newly presented white mica and chlorite abundance maps derived from airborne hyperspectral highlight the additional applications of remote sensing for geological mapping and could help to evaluate newly launched hyper- and multispectral spaceborne systems for geoscience and mineral exploration. [ABSTRACT FROM AUTHOR]