Thought of the day: Can someone be a good engineering geologist without having a solid grasp of “general” geology? Personally I think not, but I know that others disagree. What do you think? Also, what areas of geology do you think are of the most importance to engineering geologists?
The NGI-Q system is widely used internationally for assessing temporary support for tunnels and caverns in civil engineering.
However, it often seems that geologists and engineers are deliberately conservative when assessing the Q value in excavation faces. This caution can take a number of forms including:
1) “Making” a low Q value by selecting the worst examples of features when assessing the individual parameters, even where these are related to seperate sets of discontinuities and/or are not related to potential instability mechanisms
2) Assessing the Q on the basis of the most decomposed discontinuity set, even if this set is not critical to stability
3) Underestimation of RQD
4) Underestimation of discontinuity roughness (Jr)
5) Overestimation of discontinuity alteration (Ja)
6) Application of stress reduction factors for even the most modest and favourably orientated of “weakness zones”
7) The geologist/engineer gets used to using low and “safe” parameters and in a sense stops mapping the excavation and simply goes through the motions, happy to keep using the low parameters they are used to and comfortable with
For some reason Jw does not seem to be an issue in my experience.
I would be interested to hear if the above points chime with the experience of others – also do you have any other examples?
Also, assuming that the design of the tunnel/cavern/adit/shaft etc. already incorporates an appropriate degree of conservatism / factor of safety, do you think it is appropriate to err on the side of caution when assessing the rock mass for Q (or RMR, RMi, GSI etc.) or should the geologist/engineer assess the the rock mass as accurately as possible? Personally I think the assessment should be as close to the actual conditions as possible, as otherwise I believe we are heaping conservatism upon conservatism. This leads to greater costs for the client (and usually the taxpayer), lower sustainability and possibly lower safety (the excavation is arguably no safer as it is over-supported, but there is more chance of an accident as the construction takes longer as more support needs to be installed).
Perhaps you think that classification systems should not even be used for estimating support during construction (their use to be confined to feasibility assessments as has been suggested by several authors)? I am not too sure about this; rock mass classification systems seem to be a bit like democracy – the least bad system we have.
Finally, to what extent to you think the issues related to Points 1 to 7 above come down to inexperience and/or lack of training and how much do they come down to personality and differences in local practice?
As always it would be great to hear your views!
A quick comment on one of my regular bugbears:
Discontinuity poles or great circle intersections plotting within kinematic envelopes on stereonets does not necessarily mean there is a possibility of rock fall on the slope you are assessing. Please don’t make this mistake – there are a quite a few of other conditions that need to be satisfied before rock fall can happen (have a look at Rock Slope Engineering by Wyllie and Mah for these). Also please, please do not install rock support on this basis alone!
Recently I came across a possible misinterpretation in a drill hole log. This involved a feature that might possibly be a buried alluvial channel of Pleistocene age, but which had been logged as weathered rock. The setting is a valley located in an area where buried channels are possible, due to former lower sea levels and subsequent infill by alluvium and colluvium when sea levels rose. The uncertainty is partly due to the fact that the portion of the drill hole in question was wash bored. Without going into the details, the potential consequences are significant.
So, what can you do when logging to reduce the possibility of such misinterpretations and their consequences? I suggest that, wherever possible, you:
1) Find out what the project is for, where the site is located and what the geo-team (Brunsden, 2001) consider to be the project’s key areas of interest, uncertainty and risk.
2) Familiarise yourself with the local geological setting before you log. What are the expected materials, geomorphological setting, structure, geological history and processes?
3) Go to site often during the drilling. This allows you to see the setting, see the materials come out of the ground, speak to the drillers and generally develop a much better understanding of the geological setting.
4) Consider any drilling anomalies such as loss of flush carefully. To further develop the current example, if a section of a drill hole has been wash bored ask yourself why this is the case – better still – ask the drillers.
5) Make sure that you clearly communicate any important features and uncertainties (such as tens of metres of wash boring) to the other members of the geo-team. This can be done in several ways, including on the logs, in the report text, on figures, in uncertainty and risk registers and of course – face to face or over the phone (although make sure you document these issues as well).
6) Make every effort to achieve good communication – be proactive and reach out to other members of the geo-team. Do not assume that others will recognise the important features and problems – no one else will have as good a ‘feel’ for the ground as you, the person who has logged it.
7) Seek out experienced mentors to train and advise you and check your work.
All of the above seems elementary, but it is rarely the case that all aspects are satisfactorily carried out. In my experience items 1, 2, 5 and 6 are generally not carried out well, which can be frustrating as these issues could be easily resolved.
As always, I would be interested to hear what others think – do you agree or disagree? What do you think could be done to improve logging?
I have been reading ‘Structural Geology of Rocks and Regions’ by Davis and Reynolds, the preface of which contains a quote I really like:
‘By [primary sources] we mean the first-hand things, the authentic ground of facts and ideas, the original wells and springs out of which all the rest either is drawn or flows…. Regard for the primary sources makes one forever the enemy of preconceptions, of manipulated data, raw opinion, and guesswork – of all the sleek shortcuts to wisdom in ten easy lessons…. Exclusive reliance on second-hand things makes second-hand men and women. It deludes us into thinking we are wiser than we are…. Breadth of knowledge, even knowing a little about a lot, has its obvious value. But breadth that perpetually sends down no clean, strong roots in the primary sources – into the deep earth and “the hidden rivers murmuring in the dark” of the rocks – such breadth clarifies very little. It merely puts our bewilderment on a broader basis. It leads us into incredible naivete and gullibility. It makes us to quick to believe all we read.’ (From College Talks by H. F. Lowry, edited by J. R. Blackwood, pp. 86-87. Oxford University Press, New York, copyright, 1969).
While Lowry was writing as a Professor of English Literature, this seems very relevant to modern engineering geological practice and struck a strong chord with me, given that so much of the approach to work in our industry (and I’m not excluding myself here) falls within the scope of Lowry’s criticism. As always, I would be interested to hear the views of others.
Something often seen in ground engineering reports are values quoted with an inappropriate degree of accuracy (and/or precision). For example:
Too many decimal points are misleading, as they suggest a higher degree of accuracy than is justifiable. Users of values quoted in this way could assume that these values are derived from lots of reliable, representative and carefully assessed data. However, often this is not the case. Furthermore, the nature of typical geological materials means that there is a level of variability and uncertainty that cannot be reduced.
Why then do some geologists and engineers quote so many decimal points? Some possible reasons:
Therefore, ground engineering practitioners should carefully consider how many decimal places they quote, so that a false impression of the degree of accuracy is not given.
I am currently writing a short guidance document on behalf of the AGS-HK, of only six pages or so, on rock mass classification systems.
Given the short length of the document (relative to the breadth of the topic), I don’t intend to go into the details of any particular system, as there are many good references that do this already, obviously including the original papers. Also, it would be impossible to do justice to any of the systems in such a short space.
Rather, I plan to focus on the benefits, limitations and misapplications of what are probably the three main systems: Q, RMR and RMi. I think this is the correct focus as it seems that anyone can – and do – have a go at using these systems. However, the results of using them without a proper understanding is typically, to say the least, problematic.
Therefore, some of the key messages of the guidance will be:
I have my own views on the pros and cons of these three systems and of classification schemes in general, but it would be great to hear what others think these are.
P.S. The document is being prepared as a collaborative effort and it is intended that it is widely circulated for comment before publication. At the moment I am only preparing the draft version of the document.
