Mining Contract: MIS Analysis and DINSAR Measurements � Tools for Improving Mine Ground Control Safety

This page is archived for historical purposes and is no longer being maintained or updated.

Contract #	200-2016-90240
Start Date	9/1/2016
End Date	8/31/2020
Research Concept	The objective of the proposed research is to provide an opportunity for qualified individuals to pursue graduate degrees in mining engineering with an emphasis on ground control safety and mine stability. The subtasks identified in this proposal will 1) Improve seismic coverage in the Wyoming trona district, track subsidence by DInSAR for selected coal and trona mines, and compare apparent potential energies with rate and magnitude of detected seismic events. 2) Instrument a yield pillar in advance of encroaching longwall abutment loads, determine strength, elastic properties, and field conditions for the study coal pillar, generate a numerical model, and compare stress change patterns developed by the model and patterns developed by coda interferometry. 3) Develop an extended numerical model for the participating trona mine, calibrate the model using subsidence measured in task 1, look for correlations between model results and calculated positions and relative frequency of events. 4) Evaluate applicability of Bayesian statistical methods for quantifying/improving confidence in event locations, and explore other cutting-edge methods for improving detection and location of MIS. The project will provide opportunities to advance the science of interpreting the cause and potential uses of MIS and subsidence measurements as mine monitoring tools to improve ground control safety in underground coal and trona mines.

Contract Status & Impact

This contract is complete. To receive a copy of the final report, send a request to mining@cdc.gov.

Subsidence plots for a longwall trona mine and a longwall coal mine will be generated using Differential Interferometric Synthetic Aperture Radar (DInSAR). By using DInSAR, the subsidence extent and magnitude can be defined with high spatial and temporal resolution. Corresponding increases in contemporaneous underground mined space will be provided by cooperating mines with the objectives of defining spatial relationships between subsidence and mine advance, and comparing potential energy generated by mining to potential energy consumed by subsidence at one longwall coal mine in central Utah and one longwall trona mine in southwest Wyoming. On-going seismic monitoring in central Utah and southwest Wyoming will be provided for the duration of the project using the existing UUSS regional seismic network. This monitoring will be completed with the objective of developing and maintaining comprehensive catalogs of MIS that ideally have relatively low magnitude detection thresholds and accurate event locations. Rates and magnitudes of MIS will be compared to potential energy with the objective of quantifying how MIS is related to energy dissipation in central Utah and southwest Wyoming.

Methods will be developed to assess the state of stress in gate road and barrier pillars by non-invasive means. Some success has been achieved using seismic tomography based on relative arrival times for P and S waves traveling through a volume of rock under varying stress (Luxbacher et al. 2008). The application of coda interferometry may be more powerful in that the unique signature of heavily scattered seismic waves is used to infer changes in the material. A seismic source in the form of a pendulum hammer will be used periodically to generate a signal which will be recorded by geophones installed in the pillar. Over time, a catalog of seismic data will be archived along with convergence measurements as the abutment loads encroach upon the area. A numerical model representing the pillar and surrounding area will be developed to provide a basis for evaluating the success of interferometry in detecting the changing stress field.

Possible reasons for differences in MIS produced by trona vs. coal mining will be explored. Obvious similarities between Wyoming trona and Utah coal districts are: both contain mines in tabular deposits; both use similar mining methods (room and pillar and longwall) and stratigraphic columns in both districts contain soft strata such as shale and stiff strata such as sandstone. With these similarities, similar rates and magnitudes of MIS would be expected. However, trona mining in southwestern Wyoming produces far fewer events compared to coal mining in central Utah. Comparison in time and space of seismic events, surface subsidence and underground extraction is essential for ultimately discovering the underlying mechanisms that account for differences in MIS between Utah coal mining and Wyoming trona mining. Numerical modeling will allow for study of immediate and deep floor strata that may be of importance, especially when the roof – seam – floor properties contrast is reversed between coal and trona mining.

A new multi-event relocation procedure in studying mining induced seismicity will be investigated. The new method is based on a Bayesian sampling of the model space and is known as Bayesloc (Myers et al. 2007; Myers et al. 2009). Bayesloc is significantly more flexible than previous methods. Populations of solutions are output and true 95% confidence regions for the model parameters can be determined without making the assumption of Gaussian statistics near the optimal model. Bayesloc has been found to outperform more conventional techniques on arrival time data for which ground truth locations were known (Myers, et al. 2007). It has also been used to relocate an anomalously shallow (depths of 1-2 km) sequence of earthquakes in southern Nevada (Pyle et al. 2015). It is believed that Bayesloc would be effective at relocating shallow mining induced seismicity in regions of significant topography, and would provide a more realistic understanding of uncertainties in MIS locations compared to more conventional techniques.