Whitwell Quarry is located in North East Derbyshire in northern England and has an active railway tunnel (Whitwell tunnel) running through it. Whitwell tunnel is situated between the main area of the quarry and the northern extension. The complex legal situation is that whilst the tunnel is owned and operated by Network Rail, the mineral is the property of Lafarge Aggregates. Quarry development plans are to extract mineral by working as closely as possible to the tunnel without damaging. A total of 15 monitoring boreholes were installed, in two separate time periods. Each monitoring borehole had two tri-axial arrays located in each; an upper array aligned with the soffit level of the tunnel (roof), and a lower array corresponding to the invert level (floor). All blasting was been carried out using electronic detonators and pre packaged explosive charges so as to ensure the maximum control possible on the design process. In addition the position of each borehole in the blast was surveyed and the quarry face to be blasted was profiled. Some 46 multi-hole blasts have been monitored at 39 specific locations resulting in 503 blast vibration records being obtained. These results were obtained from bespoke blast monitoring equipment developed by the University of Leeds connected to fixed monitoring points within boreholes at soffit and invert levels; together with data collected from commercial portable seismographs which were deployed on the surface at specific boreholes on the day of each specific blast. The vibration data that resulted from blasting was recorded at both surface and subsurface monitoring location. This data was subsequently analysed by employing a trivariate statistical model that takes into account differing explosive charge weights [E] whilst also respecting the two difference between types of seismic waves [body waves (b) and surface waves (s)] and their attenuation rates with respect to both distance [D] and depth. It was found that in this instance, this model can be said to account for 91% of the differences that were found in the dependant variable [PPV] when taking into account the two independent variables [E/Db] and [(E/Ds)*(1/edepth)], leaving 9% as unexplained variability within the model. This represents a significant advance in statistical blast vibration predictive modelling.

Additional Metadata
Conference 10th International Symposium on Rock Fragmentation by Blasting, FRAGBLAST 10
Birch, W.J., & White, A. (2013). The development of a trivariate statistical blast vibration model that seeks to respect both the difference between types of seismic waves and their attenuation rates. In Rock Fragmentation by Blasting, FRAGBLAST 10 - Proceedings of the 10th International Symposium on Rock Fragmentation by Blasting (pp. 417–424).