Search

Data mining is a promising new tool in mineral exploration. Here, we combined data-mining procedures with spatial prediction modeling for gold exploration targeting in the Buhweju area in southwestern Uganda. It was employed in a data-rich context of unavoidably partly redundant and correlated information that offered challenges in extracting significant relationships. Our study utilized a database of co-registered digital maps related to gold mineralization. It comprised Landsat TM, Shuttle Radar Topographic Mission (SRTM), and geophysical (radiometric and magnetic) datasets for geological and structural mapping. The locations of 15 orogenic gold deposits and 87 gold occurrences were obtained from the Geological Survey of Uganda database. These were considered direct evidence of the presence of gold mineralization. The geological and geophysical settings at the gold deposit/occurrences locations were based on geological units as host rocks, contacts, and structural elements, together with continuous field values of geophysics, radiometry, and other remotely sensed imagery. A gold exploration targeting proposition (Tp) was defined as: "That a point p within the study area contains a gold deposit given the presence of spatial evidence." All outstanding combinations of spatial evidence were obtained using empirical likelihood ratios. With a data-mining strategy, the ratios were filtered and modeled to identify stronger spatial associations, to rank the study area according to the likelihood of future discoveries, to represent ranking quality, to estimate associated uncertainty, and to select prospective target areas. The empirical likelihood ratios facilitated a transparent strategy for generating prediction patterns and extracting small prospective target areas with higher likelihood of discovery and lower-ranking uncertainty. Conclusions are provided on the knowledge extraction for prospectivity with further data and the challenges of reducing the arbitrariness of decisional steps. [ABSTRACT FROM AUTHOR]

Deeply dissected mountainous ridges, paucity of exposure, deep weathering, extensive regolith, dense vegetation cover and poor accessibility, are detrimental to conventional field geological mapping. Conditions such as these, characterize our current study area; the Buhweju area in Uganda. The latter is poorly explored and known to host alluvial gold. In this work, an integrated geological mapping approach aiming at updating the geology, building relationships between the lithology and structure and assessing its control on mineralization was tested using Landsat TM, shuttle radar topographic mission (SRTM) and potential filed datasets (radiometrics and magnetics) subsequently validated by field investigations. Several image processing and enhancement techniques were applied that improved the aesthetic quality of the various remotely sensed datasets for visualization and interpretation. Among the image enhancement techniques, image fusion and pan-sharpening of the radiometric composite image with multispectral data and SRTM digital elevation model (DEM) have resulted in better lithological differentiation. In general, the present work, found complementarity in lithology with the legacy geological map. On the basis of similar radiometric signature however, various basement rock units such as pelitic schist, quartzo-feldspathic gneiss and granite were delineated and mapped; some in areas which were not reported before. These were validated during field investigation. Similarly, the aeromagnetic data depicted worthwhile information regarding litho-magnetic domains and tectonics of the Buhweju study area. Three prominent structures trending NNW, ESE and NNE determine the tectonic grain of the area. These orientations are consistent with the two major and important regional tectonic trends known in the area; the NNE to the western Rift while the NNW attest to the regional Aswan shear zone. The control of the Buhweju gold field was also assessed in terms of the improved geological units and interpreted lineaments of the area. Accordingly, most gold occurrences are localized in preferred orientations and host rock. Density analysis carried out on surface and subsurface lineaments reveals the Buhweju group rocks and the undifferentiated schist and amphibolites are highly affected by surface lineaments whereas the basement rocks are affected mainly by subsurface lineaments. Two major clusters of gold occurrences were identified in the Buhweju gold field. The gold cluster found at Kitaka mine is characterised by a low magnetic anomaly and high lineament density, while the gold occurrences found at Katonga is hosted by mudstone and Lubare quartzite and underlain by a high magnetic anomaly. These gold clusters are spatially controlled by the NNW and NNE trending faults and their intersection determines the localization of most of the Buhweju gold occurrences. The two trends of structures in Buhweju could also be interpreted in terms of the Geita and Twangiza structural corridors (gold trends) known from the work of others. The NNW trending Geita structural corridors control gold mineralization in Tanzania while the NNE trending Twangiza corridor controls gold mineralization in the Democratic Republic of Congo (DRC). [ABSTRACT FROM AUTHOR]

Integration of enhanced regional geo-datasets has facilitated new geological interpretation and modelling of prospectivity for orogenic gold in southwestern Uganda. The geo-datasets include historical geological maps, geological field data, digital terrain models, Landsat TM data and airborne geophysical data. The study area, bordered by the western branch of the East African Rift, covers a range of different aged terranes including the Archaean basement gneisses, Palaeoproterozoic volcano-sedimentary Buganda Toro Belt, Mesoproterozoic clastic sedimentary Karagwe Ankolean Belt and several outliers of undeformed Neoproterozoic sediments. The mineral systems approach to practical exploration targeting requires a framework to link conceptual models of mineralisation with available data. A conceptual model requires good understanding of key processes and their timing within the geodynamic history of an area. The challenge is that processes cannot be mapped, only their results or effects. In this study, a district-scale (1:100,000) investigation is considered appropriate given the scarcity of geological information and the absence of world-renowned gold deposits in southwestern Uganda. At this scale of orogenic gold mineral systems understanding, evidence for the source of gold, active pathways and the physical traps are considered critical. Following the mineral system approach, these processes critical to orogenic gold systems are translated into district-scale mappable proxies using available regional-scale datasets. Tectono-stratigraphic domains, mantle indicators and gold occurrences represent the “source of gold” as a critical process. Zones of hydrothermal alteration were extracted from radiometric data, structures involved in the orogenies and terrane contacts were extracted to represent the active pathway as a critical process and finally the physical throttle is represented by rheological contrasts and geological complexity. Then, the knowledge-driven multi-class index overlay method was used to integrate predictor layers representing those processes in order to model orogenic gold prospectivity into a single map. Weighting of the predictor layers, prior to integration, occurs at the level of the critical process and takes into account the relative importance of the critical process mineralisation, the representativeness of a proxy and the accuracy of the proxy. The resultant prospectivity model shows that 83% of all gold occurrences are delineated within predicted prospective areas covering 30% of the study area. Eight sub-areas, covering 2500 km 2 , have been recommended for follow-up exploration. [ABSTRACT FROM AUTHOR]