The construction of plausible geological models
We had the opportunity to chat with Dr. Roisin Kyne from Teck Resources and Dr. Koen Torremans from the Irish Centre for Research in Applied Geosciences (iCRAG) about the importance of 3D analysis and modelling in their work at iCRAG to understand the formation of the Lisheen and Silvermines deposits in the Irish ore field. Using 3D modelling, the researchers investigated the geometry of normal fault systems and their implications on the origin and nature of mineralization within these systems.
Faults are important structures in the formation of many mineral deposits, often acting as conduits for ore-forming fluids and sometimes providing, or generating, the bounding structures to associated mineralizing sites. Fault segments occur on different scales and have a profound impact on structural evolution, with larger scale segments and intervening relay ramps defining distinct orebodies within deposits and smaller scale segments and relays potentially providing paths for upfault fluid flow. The difference in behavior is attributed to the integrity of associated relay ramps where intact ramps represent orebody-bounding structures, and smaller breached ramps provide enhanced associated hydraulic properties and act as vertical conduits. Hanging-wall deformation along the rheological boundary between host-rock lime- stones and underlying shales has an important control on the localization of earlier dolomitization and/or brecciation and later mineralization adjacent to this contact, and on the migration pathways for basinal brines and mineralizing fluids. Read more – Economic Geology ; 114 (1): 93–116.
Q: We understand that you used GOCAD Mining Suite (GMS) to gain insights into mineralization controls on the Irish Zn-Pb ore field in Central Ireland.
Roisin: Yes. The main purpose of our project was to model the complex fault networks that are present in the Irish ore field. We were particularly interested in how the fault network controlled the Waulsortian limestone [host rocks] as well as its geometry in relation to mineralization. GMS is really powerful for 3D interpolation, analysis, or being able to model a finite fault [only cutting the geological model partially] and interpret that data. Even just data management, importing drillholes, looking at lithologies and classifying these lithologies. So, we chose GMS as one of our main modelling packages because it was probably the most powerful and allowed us to achieve that modelling in a more robust way.
Q: And is it your first use of GMS?
Koen: Yes, it was certainly the first intense use.
Q: How was the GMS structural modelling tool useful in the representation of faults?
Koen: Because the fault systems in Ireland are so complex and very segmented, whichever tool we used had to be able to handle this [complexity and segmentation]. The strength of GMS is that it can deal with finite and complex fault networks. We needed to be able to interactively select and interpret the correct data, and then model a fault, its interaction with other faults, and its interaction with the key horizon, to iteratively mold the horizons to what we needed due to an abundance of complexity along the faults, such as brecciation and fault rocks that needed to be handled.
Roisin: One of the key things GMS allowed us to do was to assign how controlling faults terminate against one another. That’s really key for understanding breaching [between fault segments]. Once we were able to do that, by iteratively going back and reassigning these faults to the various determinations, it allowed us to use this information in our subsequent modelling to rip the [key horizon] surfaces, as a fault would do, and to understand the various relationships. It produced a more accurate, more robust model in the end. GMS allowed us to mold these faults and surfaces in a way that was usable; in a way that we knew was geologically more accurate to then be used in subsequent fault [and horizon] analysis.
Koen: That’s basically it. You can take a horizon surface and rip it to shreds [using the fault network] and the software still works, and that’s amazing.
Glenn: It sounds like you are using the software very effectively. I like the way you used the term molding the faults, or surfaces, or model, which means you are actually using the environment as a 3D interpretation environment, not just necessarily expecting the software to create the model by it- self without a lot of interpretative thought going into it. That’s great to see and hear.
Roisin: That’s absolutely right. We experienced the chaos that can be produced the first time we ran through the model workflows. You need to be able to go in there and decide if we’re in a brecciated zone [or a fault plane] and we need to ignore these points because they’re artificially causing the surface to rip in a non-realistic way. Just being able to do that is helpful. Using our knowledge to mold the model to be more geologically accurate was incredibly powerful.
The Lisheen image shows the horizon model for the base of the Waulsortian Limestone Formation at Lisheen. The orebody is located at or near the base of this unit. The Waulsortian horizon surface is ripped by several fault segments that die out laterally and which structurally control the individual orebodies. The contained distribution of total Fe tonnages in the orebody are also shown, draped on the economic orebody resource. These distributions are calculated from a geostatistically determined block model showing average Fe tonnages in 4x4m vertical columns through the orebody. Surface drillholes are shown, as well as the surface infrastructure and culture surrounding the Lisheen mine, where it is not cut out to show the subsurface geology.
Q: How important is the integrated 3D modelling process in building an understanding of data sets, and how does it lead to good insight in understanding the mineralizing systems?
Roisin: For us it’s integral. It allowed us to make a 3D model that was used to better contextualize other data, add the geochemistry on top of it, and understand what was going on with the geochemistry and the patterns we were seeing. Putting [the Lisheen deposit] into a 3D environment allowed us to QA/QC a lot the data and it also helped us to understand some really complex areas. In one of the main orebodies in the south, we have a feeder zone geometry. It’s an incredibly complex area that has been worked on for many years. It was essential to interrogate the data in 3D, if we wanted to unravel the full geological story. We were able to quickly start linking some of the information together realizing a number of things: (1) which faults intersect other faults, (2) where the brecciation was, and (3) in what order these faults interacted with one another. We were actually able to pull out the original horizon geometry from the GMS model. We would have never been able to do that without putting this into 3D, and that original geometry is a completely breached relay ramp.
Koen: The challenge for Silvermines was that it was almost all historical data. We had to do a lot of QA/QC for all the data entries. Once we had that data in 3D and georeferenced, we then could actually use it, as is, into the 3D environment and pick faults on historical underground maps or lithostratigraphic contacts; feeding that into the 3D model that we were building. In terms of insights for the mineralizing system, we saw that certain areas acted as feeder zones and that other areas that were more distal zones had nothing to do with the initial normal fault array. This allowed us to understand the Silvermines fault system as it evolved through time. So that was quite profound.
Q: Have you tested the final stratigraphic model against geophysical data?
Koen: The work was done without having to compare it to geophysical data directly. It was just not needed because there was such a data density in the mine areas. Recently we’ve been using aeromagnetic and electromagnetic (EM) data, as well as gravity data, for regional modelling for the area around the Lisheen and Silvermines deposits. This independent detailed 3D modelling case study actually allowed us to ground truth the geophysics model with the geology that we have in the mine area. We can then apply those insights regionally and that works quite well. Certainly, in terms of aeromagnetic interpretation.
The Silvermines image shows two modelled horizons ripped by finite faults, along with some of the faults in the back. The top horizon is the Waulsortian Limestone, the base of which is the main host for mineralization. Beneath that in grey is a unit called the Lower Dolomite. The topography is shown towards the top of the image but cut out in the model-area (using regions). It also shows drillholes and outline of the orebodies.
Q: Anything else to add to what were the key revelations or realizations regarding the geological model?
Roisin: What is essential for us is the understanding of those breaching faults and how the presence of a breached fault controlled where the fluids were coming up and where the mineralization occurred. The idea with relay ramps has always been prevalent within Irish literature, but by understanding the deposits; we have now really been able to narrow [them] down.
Koen: We compared our 3D model to a block model of the grade distribution in Lisheen as well as to a very simple inverse distance weight grade distribution in Silvermines. That allowed us to study the metal distribution patterns, fluid flow pathways, etc. The 3D model and block model worked together to introduce those insights.
Jean-Philippe: Were the relay ramps more conceptu-al concepts before you guys modelled them in 3D or was this already known and you were generating your model around that?
Roisin: The idea was known, and we knew going in that there were going to be relay ramps. Especially in Lisheen. When we put Silvermines into 3D we showed to geologists that had worked there back when it was opened a long time ago and they were so amazed to see how it physically worked. It was a good visual way of showing people what a relay ramp really is, and how it really works, and that was good.
Q: Has the modelling heightened prospectivity of the area?
Koen: People are looking at near mine exploration and I think our models are definitely helping with that. We’ve had several of our industry partners engage with us on this. The insights that this delivered and the quantitative measurements that we’ve done based on this model can be used in a predictive way. The geometry and the evolution of the systems can be used to better interpret the sparse drillhole data, seismic data, or potential field geophysics. In that sense, yes it does increase prospectivity because it not only helps you to better target new mineralization; it also helps you identify areas that are not of interest. We’ve identified many such fault trends in the Irish midlands and we’re now increasingly working with seismic data as well, which is a true game changer. That really helps with regional modelling.
Roisin: Yes, what we know now, based on our 3D modelling, is that you can almost predict where the next big step [in the system] should be. By looking at relay ramps at all scales, we have a better understanding of the structures [regionally] through the 3D model that we did at the mine scale. You can use that information to search for fault systems and mineralization, both regionally and locally.