Hi Bekir,

The best solution is to divide rock slope into zones or layers having similar mechanical properties. Then, assign appropriate material properties i.e., Hoek-Brown or Mohr-Columb for each zone. You can use 2D or 3D programs of Rocscience, Inc. such as RS2 or RS3 and Slide 2 or Slide 3. If you have any persistent joint system, you can include explicitly joint networks into continuum-based FEM approaches similar to DFN-based DEM approaches.

Best regards,

I think the best approximate method is considered to be an anisotropic continuum. For that many constitutive models have been proposed.

Other method is considered isotopic with properties obtained from field tests.

Regards

Hi!

The anisotropic medium seems a good solution, as well as considering stratified homogeneous layers, as long as the choice is supported by field or laboratory investigations that provide the mechanical parameters. Otherwise it will be just a guess.

Regards

Thank you for your answers. I will check all of your suggestions. In addition, I need a software to simulate such conditions. Any suggestions?

The common practice for a case like this in my country where weathered profiles are everywhere consists of representing the geological-geotechnical profile as "layers" where each one is a weathering level taken from a specific proposal. I normally work with Dearman (1974) which is widely accepted as a methodology for working with weathered rocks.

After defining the stratigraphic profile, you should characterize each level according to the type of material, for instance, using the Mohr-Coulomb criterion for those levels where soil predominates and Hoek-Brown where rock predominates.

Parry et al (2013) also describe a lot of things about geological-geotechnical models, and they mention something about weathering profiles.

References:

Dearman (1974) Article Weathering classification in the characterisation of rock fo...

Parry et al (2013) https://doi.org/10.1007/s10064-014-0576-x

In addition to the anisotropy, you can use the following article to estimate the change of the horizontal and vertical properties: Javashdel, S., Luo, N. and Bathurst, R. J., 2017. Probabilistic analysis of simple slopes with cohesive soil strength using RLEM and RFEM. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 11(3), pp.231-246. But this does not negate the consideration of anisatropy.

Dear Exneyder & Igor, Thank you very much for your responses. I will consider the articles If they are suitable for my problem.

I would recommend a combination of P-wave velocity analysis and Schmidt hammer surveys. This way you could pick up the increasing UCS (and density) and relate Vp increase and reducing gradient (s-1) to increasing Q-values (or vice versa). Note that Q can be anisotropic, Vp also, and the latter is also depth or stress dependent.

Hi Bekir,

The best solution is to divide rock slope into zones or layers having similar mechanical properties. Then, assign appropriate material properties i.e., Hoek-Brown or Mohr-Columb for each zone. You can use 2D or 3D programs of Rocscience, Inc. such as RS2 or RS3 and Slide 2 or Slide 3. If you have any persistent joint system, you can include explicitly joint networks into continuum-based FEM approaches similar to DFN-based DEM approaches.

Best regards,

Concerning Mehmet's comprehensive answer, while I agree that this could be a final approach, one first has to estimate the proposed shear strength properties - for e.g. HB or MC - in the different zones. It is here the big dilemma arises because not everyone agrees that GSI is the way to go to estimate these things, via the truely remarkable HB equations that are hardly empirical in nature and are clothed in so much algebra that extra joint sets or clay on a joint set are likely to be lost in the black box. Being biased it is here that I believe simple, transparent index testing for individual joint sets: JRC, JCS, relevant block size etc come into the running - if you are a believer in truely discontinuum analysis, where explicit joint sets that can dilate and have realistic ratios of Kn/Ks, actually affect bolt forces at joint crossings - a key test of desired rock mass behaviour modelling.

From the figure above it is seems that the rock mass quality increases with depth. You can divide your slope into three distinct zones-layers. GSI seems applicable to this geology. If I recognize it correctly there's a failure at the upper most bench? Good geological mapping (to collect all necessary data) and proper back-analysis would give you an indication of parameters of the weakest rock zone. Moderately weathered and slightly weathered zones seem stable, so FoS above 1.0.

This slope is not truly a candidate for anisotropic modeling (with restriction of having only this figure in front of myself), continuum mechanics principles with proper parameter averaging for every zone would be applicable.

in addition to @berisavljevic, I would suggest to understand how different type of rock masses develop inside, not only at outcrop (suggested seismic survey). Usually weathered rocks from high to low is depth dependent if weathering comes from surface (water etc), some exceptions if there are single and persistent faults etc.

Dear Neil Bar , Mehmet Sari, Nick Ryland Barton and Zoran Berisavljevic, I sincerely thank you for your valuable contributions. As far as I understand from all these answers, I need to divide the rock mass into different zones with different strength criterions or material properties and determine the distribution of heterogeneity with P-wave and Schmidt hammer tests.

However, the slope can be rarely seperated into distinctive layers such as in the photo. Where I took the photo, behind me, slope consists of completly disintegrated materials with various size of bimrocks. So It would be very hard and even maybe nonrepresentative for the overall area.

In this area, the rocks are excavated with controlled blasting so the slope did not fail naturally but it was blasting induced.

Also, I couldn’t spot any discontinuity controlled failure occuring in this area. Even small wedge failures are absent.

In addition, I hesitate very much to use limit equilibrium approach in such an area. Although, some areas consist of individual joints or joint sets which triggers me to use a discrete approach, not every part of the slope contains discontinuity or controlled by one.

-So how can I possible analyse such an area with a numerical method that it consists of various properties? If I use LEM, FEM and DEM whould it be representetive becuase of the transition.?

-Also, does the transition zones between this individual area such as soil, rock mass, bimrocks affect the overall behaviour? Can these trigger a complex or fairly simple failure mechanism? Should I pay attention to these zones carefully?

Thank you very much again,

Best Regards

Bekir SF

Dear Bakir,

You have distinct weathering zones in your bench slope. So, you can define these zones in your model as a starting point. You can use GSI for estimating material properties of these zones. Since you do not have any defined joint sets, as you said, you can apply RS2 or RS3, both are FEM analysis program. Your slope dimension seems greather than REV size of your rock mass. Therefore, you can model your slope using equivalent continuum approach. Your bench slope is more prone to circular (rotational) or non-circular failure mechanisms rather than discontunity controlled slip failures, such as plane, wedge or toppling.

Dear Mehmet Sari thank you very much for you answers. Dear Filippo Forlani, Can you please explain your answer related with the seismic survey? How can I use the data from such a survey? Also, which methods and equipments do I need?

Thank you.

A lot of references you can have on internet.

In my opinion, for your case where you have highly weathered rock in upper region then moderate in the mid section and slightly weathered in lower region you can make the variation by giving different zoning. The higher the discontinuities, the higher the number number of blocks or sub-contacts. On the other hand, the heterogeneity can be represented by making separate block including color.