摘要：Prospective Proterozoic units in the southern Mount Isa Province are concealed by a poorly defined extent of younger basin cover, leading to poor exploration success. Collection of a magnetotelluric (MT) survey in the area containing 809 broadband MT (BBMT) and 855 audiomagnetotelluric (AMT) stations in 2014–2015, offers an opportunity to better model the depth to basement to enable effective exploration. MT inversion models are inherently non-unique, requiring independent geophysical and geological constraint to reduce model uncertainty. Where data are not available to constrain inversion, alternative approaches to dealing in inversion variability are required. This study uses synthetic modelling based on well data combined with two kinds of inversion to generate an interpretation and quantify associated uncertainty. Downhole resistivity logs were obtained from three petroleum wells adjacent to the study area, and 1D resistivity models were generated from the downhole data. A suite of 1D and 2D MT inversion algorithms were tested to determine their ability to resolve basin layering and the basement interface. All inversion algorithms reproduced basin layering, but the basement interface was poorly resolved. A combination of Occam2D and 1D rjMcMC inversions were used to produce interpretation of the base of the Eromanga Basin, an intra-Georgina Basin low-resistivity layer and depth to basement, all of which have associated error estimates. This work highlights the importance of understanding inversion variability during interpretation of geological features, particularly in the absence of constraining information. Distribution of uncertainty between the interpretation features is significantly non-uniform, necessitating careful consideration of inversion results. By quantifying uncertainty rather than ignoring it, we produce an interpretation commensurate with data limitations that still provides valuable new information about the geology of the southern Mount Isa Province.
关键词：Depth to basement; Magnetotelluric inversion; Synthetic modelling; Uncertainty quantification; Mount Isa Province; Georgina Basin; Eromanga Basin
