Abstract. Paleovalleys are buried ancient river valleys that often form productive aquifers, especially in the semiarid and arid areas of Australia. Delineating their extent and hydrostratigraphy is however a challenging task in groundwater system characterization. This study developed a methodology based on the deep learning super-resolution convolutional neural network (SRCNN) approach, to convert electrical conductivity (EC) estimates from an airborne electromagnetic (AEM) survey in South Australia to a high-resolution binary paleovalley map. The SRCNN was trained and tested with a synthetic training dataset, where valleys were generated from readily available digital elevation model (DEM) data from the AEM survey area. Electrical conductivities typical of valley sediments were generated by Archie's law, and subsequently blurred by down-sampling and bicubic interpolation to represent noise from the AEM survey, inversion and interpolation. After a model training step, the SRCNN successfully removed such noise, and reclassified the low-resolution, converted unimodal but skewed EC values into a high-resolution paleovalley index following a bimodal distribution. The latter allows us to distinguish valley from non-valley pixels. Furthermore, a realistic spatial connectivity structure of the paleovalley was predicted when compared with borehole lithology logs and a valley bottom flatness indicator. Overall the methodology permitted us to better constrain the three-dimensional paleovalley geometry from AEM images that are becoming more widely available for groundwater prospecting.
The southern McArthur Basin in Australia's Northern Territory is host to some Tier-1 sediment-hosted base metal mineral deposits including the McArthur River Zn-Pb-Ag mine. Airborne electromagnetic (AEM) data sets have been employed as a key exploration technology in the search for these mineral systems. A geological interpretation of results arising from the use of different inversion techniques, including a 1, 2.5 and 3D methods, was undertaken on a helicopter EM data set acquired over a structurally complex sediment package in the Batten Fault Zone north of the McArthur River Mine. The exploration targets were conductive, mineralised units (HYC pyritic shale member) associated with the Barney Creek Formation. Results from this study suggested that although the model fits were good, the derived conductivity models for the 2.5D and 3D inversions appeared to be smooth representations of geological reality, particularly when compared with data from drilling and surface geological mapping. Superficially, the 1D smooth model layered Earth inversions appear to map geological variability and structural complexity in greater detail even though the structures are more 3D in nature. IP effects are observed in the data and influence the modelled structure, but can be accounted for and complement the non IP 1D inversion results. The outcome of this study also indicates that when employing higher order inversion methods in the interpretation of AEM data sets, there may be significant benefit in asking a contractor/consultant for 1D inversion results as well. In the resulting interpretations if conductors appear in one but not the other, it is worth asking the question why?
Saline aquifers in the Murray River or SE Australia are traversed by freshwater rivers, with adjoining riparian and floodplain regions containing freshwater lenses. Bore data and more recent Airborne electromagnetic (AEM) surveys have determined that these lenses are spatially extensive, but have widely varying geometries. The maintenance of these lens systems is important as they support ecologically significant riparian vegetation communities such as Red Gum and Black Box. A more complete understanding of their hydrogeology is required to ascertain how they develop and degrade. Limited ground investigations including 14 C geochemistry have determined that the lens systems contain recent water, indicating that they are dynamic systems with their development defined by the relative rates of recharge from the river and mixing with groundwater. Changes in groundwater gradients and depth, floodplain extent, and topography are believed to control their initial location. The same controls also govern their stability. The potential of AEM systems for defining the geometry of these lens systems in 3D is considered along with an assessment of their value for monitoring variations associated with these ecosystems. The advent of "calibrated" AEM systems and robust inversion tools have given added impetus to their use for environmental monitoring. Spatio-temporal variations are observed in the near surface (top 20m) from a multi-temporal assessment of Clark's Floodplain, adjacent to the Bookpurnong irrigation area, with co-incident AEM surveys acquired between 2008 and 2015. Spatial changes in ground conductivity, attributed to changing groundwater quality have been observed. The freshwater lens systems appear to have contracted significantly over the last decade. This is attributed, in part, to land use patterns and the development of an irrigation-related groundwater mound on the highlands adjacent to the floodplain, and an increased hydraulic gradient towards the river. The results indicate the geometry of the hyporheic zone may have also changed along the river.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.