Underground » Geology
Seismic surveying, especially by the 3D method, is widely used throughout the Sydney and Bowen basins in order to delineate geological structures with the potential to affect underground coal mining. While use of the technique continues to grow, its performance has not been systematically assessed because structural interpretation of seismic data is generally performed before mining. However, when results are validated against subsequent underground mapping, discrepancies are sometimes observed. These include systematic overestimation of fault throws, missed significant faults and, less commonly, non-validated faults. The underlying causes for these discrepancies are not well understood. This project has attempted to address this issue by undertaking a detailed comparison of two 3D seismic surveys with subsequent geological investigations based on underground mapping and the drilling of surface and in-seam boreholes. The comparison has been supported by computer simulations of seismic reflection results and, in one case, reprocessing of the seismic survey data.
The two 3D seismic data sets are from BMA's Broadmeadow Mine in the Bowen Basin and from Centennial Coal's Mannering Colliery in the Sydney Basin. The Broadmeadow 3D seismic data are of good quality and geological control is available from surface boreholes, an underground exploration adit and longwall gateroads. The reconciliation at Broadmeadow Mine allows the following general observations to be made:
· All faults identified underground and in exploration boreholes with throws greater than 2.5m throw have been identified in the 3D seismic survey results;
· Some of the known faults with throws between 1m and 2.5m have been identified in the seismic data. The faults that were identified with throws in this range tend to be thrust faults. False identification of faults in this range was uncommon; and
· None of the known faults with throws less than 1m were identified in the 3D seismic data.
Computer modelling was used to verify these findings.
In typical coal-related seismic exploration in Australia, where the dominant seismic frequency is about 100 Hz and the seismic velocity of the overburden ranges from 3000m/s to 4000m/s, the dominant wavelength is 30-40m and the 1/8 of wavelength criterion is about 4-5m. Obviously, it is useful to attempt to reduce this (increase the dominant frequency), through use of appropriate survey procedures and computer processing techniques.
A different problem was investigated at Mannering Colliery, which is under Lake Macquarie. Here, a 3D seismic survey failed to map a fault with a throw of 5-6m. Such a fault is in the detectable range for a seismic survey. The results of this survey suggested, however, that a fault with a throw greater than 10m was present and with a throw in the opposite direction to that encountered by the mine. A careful examination of all factors contributing to the development of the final seismic sections was made, including an assessment of the possibility for positional errors in the survey data. No explanation for this failure of the seismic survey was found. Reprocessing of the 3D seismic survey data also led to a similar seismic sections being obtained.
Overall, this project should be viewed as providing encouragement for continued use of seismic surveying, especially by the 3D method. Computer modelling and the reconciliation of seismic survey results from Broadmeadow Mine with actual geology mapped underground and from borehole data gives cause for confidence in the detection of faults with throws even as small as 1/16 of the dominant seismic wavelength. However, issues remain with the application of the seismic method in areas with complex geology. More work is required to improve the success of the seismic method in these situations. To properly understand the results, surveys need to be undertaken with a very good understanding of the geological context.