UCS and Seismic Velocity Correlation

[pokeyjoe17]

Can anyone point me to where I might find correlations between unconfined compressive strength and seismic velocity if such correlations can be made?

We have rock cores that we will test for UCS and seismic refraction profiles in areas close to the core locations. Our goal is to determine the depth of the weathered zone (dolomitic limestone) from info obtained on the profiles.


Thanks

 

[coalman]

This is for underground coal mines but may be of some help. We have been using on a routine basis for the last 20 years.

We began working with sonic data in 1985/86 in connection with Central Colliery, which was the first longwall mine in the Bowen Basin, but it was not until McNally (1987) published his relationship between sonic and UCS that rock strength interpretation became a routine (and accepted) part of the interpretation of ground conditions for underground mining. We use the sonic, usually in conjunction with the gamma log to define and delineate roof strata into geomechanical units, with a strength ascribed from the sonic. These units are used to generate roof sections (maingate sections) for prediction of mining conditions and roof support requirements -prior to mining. They also form the basis for rock mass units (and strength) for the calculation of CMRR.

Crinum uses the average sonic value over the first 0.5m and first 2m of roof strata, which are contoured and plotted over the block, the former gives an indication of potential trouble spots on the longwall face (roof falls in front of the canopies), the latter is used to determine which category of roof support is used where. At German Creek Mines, the sonic interpretations are used to plot the occurrence of particular weaker, more fissile horizons, as they move within the bolting horizon, and also to track the occurrence of thick strong beds that can cause periodic weighting.

In short, sonic logs are a very important, even a key component of roof control in Australia.

In terms of developing techniques there are a couple of pre-requisites

1) you need good, and reliable sonic logs - not all companies are the same; most companies have in fact concentrated on perfecting the other logs, notably the density. The sonic log seems prone to anomalous readings; the most consistent logs in Australia were produced by BPB (now Reeves) and these are still the preferred choice. We have Century, their logs were historically pretty useless and hence non competitive with BPB, but they are now OK but not as good as Reeves. We use only the standard multi channel inverse velocity log for interpretation of rock units, the sonde has receivers at 20cm, 40cm and 60cm spacing, the first being the one used preferentially for interpretation. The preferred units are microsec/ft. The full wave sonic is still essentially a research tool; it is of little value in the routine interpretation of units due to its lack of definition, however it does hold promise for being able to provide other rock mass parameters such as modulus.


2) you need to establish a sonic-UCS relationship. In general these are site specific, in detail they can be lithology specific but normally we would just have one for rock and one for coal. There is no short cut here, you need to do a statistically sufficient number of UCS tests matched against the sonic interval in question. Having done that you can ascribe rock strength, this can be automated to provide a UCS log output if you desire.