Core logging: Optimizing best practice (Part Three).

? J.L. Orpen (Resource Exploration & Development Pty. Ltd.)

Diamond drilling and the Earth Sciences at a crossroads.

Diamond drilling is a versatile subsurface investigation technique, but as it is amongst the most expensive it is normally deployed in the targeted stages of an exploration, or site investigation program. Its use is being ramped up, however, as mining and construction industries are motivated in part by government loan facilities set up to encourage exploration for critical minerals (Backing Australia's critical minerals sector | Export Finance Australia ) as well as safe nuclear waste repositories.

Thus, drill rigs regurgitate thousands of meters of core daily, but there is an increasing scarcity of geoscientists with the skills and experience to properly log the core and bore – a situation made worse by globally diminishing support for the profession, with the possible exception of Canada (Geoscience on the chopping block | Nature Reviews Earth & Environment ) and (Australia’s Unfolding Geoscience Malady - Eos ).

Some of the rationale for defunding the Earth Sciences appears to stem from the assertion that Artificial Intelligence and Machine Learning, using data input from various rock scanning techniques such as portable XRF and Hyperspectral scanning, can arrogate many of the functions of a trained observer, specifically core logging. By definition, however, the power of these technologies is only realised if quality, site-specific information is fed into the system. Hence, these hi-tech additions to the geoscientist’s toolbox only work if the AI and ML components are “taught” so as to improve the machine’s “understanding” of the ground explored. Rigorous QA/QC routines must also still be maintained to ensure the AI output is not compromised by some algorithmic quirk of the software.

On balance therefore, these developments should increase demand for Geoscientists, yet the output from universities is declining. Hence, it seems the Earth Sciences are approaching a crossroads and, by default, so is the drilling industry.

Some advocate a quick fix by mining near-earth-asteroids. Others are building nuclear subs to patrol their patch of the seafloor. Hence, there has never been a more imperative time to examine the Geoscientist’s role in society than now. It is crucial for our alumni to show how and why urgent support for Earth Science teaching at all levels, together with encouraging top-flight research, is fundamental to the survival of our species – not only to satisfy our material needs, but also for unravelling climate change and other pressing environmental issues – so as to find solutions to our deteriorating circumstances before we rape the resources Earth has put at our disposal, and crowdfund our own extinction.

Rock-sampling tools.

Geological and geotechnical mapping requires: (i) precise location of the outcrops, (ii) accurate description of the features found, (iii) measurement of the structures, and (iv) the collection of representative rock samples for analysis – knapped from the outcrops using a hammer.

Wherever areas of economic interest are identified, such maps are supplemented by geophysical and geochemical surveys in order to target drilling programs. Drillers then produce rock for loggers to examine and gather data from, by: (i) precisely locating the depth of each sample downhole, (ii) describing the features found in each, (iii) measuring the structures found in oriented core, and (iv) selecting samples for analysis.

Hence, there is little, if any difference between mappers and loggers – except that the former collect their own samples from outcrops using a hammer, whereas the latter examine strings of cylindrical rock produced by a diamond driller.

The disconnect.

Consequently there is a conundrum. The skills required of the mapper and the logger are essentially no different, yet seemingly the former continue to produce good work, but over the last few decades it appears to be accepted that the overall standard of the logger’s output has waned – so that sub-standard data is being logged and used to build sub-standard representations of the ground, which is frustrating the engineers as their projects naturally flounder.

The following comments express the exasperation of the old hands (Q&A from the experts: In conversation with Dr Brett Davis | Coring Magazine ):

• “Geologists usually get thrown into drilling programs but don’t know how to extract the pertinent information from drill core.”
• “Poor results can follow poor geological supervision, even if you have good drill crews. And if you have poor drillers and poor geologists, you have a perfect storm of problems.”
• “However, the digital revolution has also spawned a population of geologists who only want to use computers, sometimes to the extent where they will do geological models without checking the veracity of the data.”
• “Fieldcraft in general is dying at universities…. You only have to look at the popularity of the fantastic videos on Fieldcraft by Nick Tate to see that there is a quest for information not taught at university.”

These sentiments are also echoed in the following paraphrased quotes from (Quantifying reliability in geological prediction (tunneltalk.com) ):

• Improvements in computer methods have coincided with a rise in the number of projects where unexpected problems have occurred, due to a lack of appreciation of how geology impacts construction.
• Projects undertake increased data synthesis of collected geological information but often end up with a diminished understanding of actual site geological structural conditions.
• Geological understanding is becoming divorced from reality because of the delusionary perception of realism and accuracy created in available modelling software, using little to no real data or hard verification checks.

Such a groundswell of criticism is troubling, and a corrective course of action needs to be charted to bring the loggers up to speed with the mappers.

The process starts by answering the question: How did the disconnect occur?

Perceptions.

It is a common perception that the teaching of digital skills has taken precedence over practical training and field skill development in many universities, but surely no one is advocating students should be any less tech-savvy? Besides, as noted, graduates perform well when mapping, so that an apparent reduction in practical tuition alone cannot explain why many of today’s emerging loggers do not log core with the proficiency of their slide-rule wielding predecessors.

It is submitted that a significant factor could be the general change in the business strategy of many companies when, in the 1990s, outsourcing work became increasingly popular.

Prior to this the major mining houses ran on-the-job training in drilling and drill core logging since, at the time, it was impractical for universities to offer such courses. Thus, the training given to the loggers recruited from the annual pool of new graduates was mostly provided by mentors who were well-seasoned in both company practices and their preferred log formats. Rigorous QA/QC routines were applied that ensured consistent quality of work and also made sure that the mentors’ skills were properly transferred before these experienced hands were promoted out of the core shed to higher office. In this way, much of the core was regularly inspected at least twice so that it was rare for important features to be missed.

The system also encouraged good working relationships with the drillers. Geotechnical rig-side logging, in particular, was commonplace, and all parties worked to high standards so that good work was easily recognized and rewarded. Junior companies also benefitted, since their project managers, who were normally recruited from the majors, came with ample experience to run drilling programs with efficiency and confidence.

Outsourcing has all but dismantled this. Service companies tender for logging contracts. Those who are awarded the work are required to use the contractor’s log formats, but usually without much guidance. QC supervision can be irregular and QA audits are generally conducted by external experts, who are nonetheless unfamiliar with the terrane so that it is not unusual for important features to be missed. To compound matters, the core shed has become a convenient locale for the contractor to monitor and poach those with aptitude to fill vacant positions within the company; a counter-productive move as it reduces the subcontractor’s ability to function. The enthusiasm of the remaining loggers is also undermined, and a job that many are not naturally attracted to anyway, becomes harder – so turnover increases. A vicious circle.

In addition, since there are few in the senior ranks of the subcontracting service company that are sufficiently well versed in the drilling process to train loggers appropriately, it is virtually impossible to establish good working relations with the drillers, or work rig-side.

In such circumstances, the key performance metric has become meters drilled and meters logged per shift. There are few meaningful QC procedures, because just about everyone with the necessary experience to properly monitor core quality and logging quality has either retired or moved on up the corporate ladder – and are now concerned with more complex problems than ensuring good data underpins the success of the company.

Towards excellence in logging.

And so it seems we are close to the crossroads, begging the question as to what is to be done?

In the early days of diamond drilling, the mentorship program grew naturally from geologists working closely with the drill manufacturers and contractors, developing it into the superlative tool it has become today. In the process, the science of drill bit design, and that of drilling mud composition, has been researched to the extent that pristine specimens of fresh rock can now be obtained from almost all lithologies, whether they be soft, pliable and prone to swelling, or hard, brittle and susceptible to shattering. Furthermore, where core loss occurs, for those who know how, the reasons can be established and incorporated into a structural model of the ground, especially where zones of water-loss are also encountered.

The list goes on, but, most impressively, sections of core can be precisely oriented so that their in-situ attitude in 3D space can be re-created, allowing the dip and dip direction, or plunge and trend, of any contained structures to be accurately measured and modelled. This ability, combined with the science of borehole survey and downhole geophysics, was a burgeoning field in the last century. A good understanding of the makeup of the ground investigated was thereby consistently achieved, so that its behaviour could be reliably predicted by the engineers before rock was either extracted or loaded, during mining or construction projects.

Thus the challenge is to re-equip core loggers with the knowledge required to confidently capture reliable data at all times.

Ninety percent of the task has already been accomplished. On any project, after the phase of ‘getting-one’s-eye-in’ has passed, graduates are quite capable of picking up any section of core, identifying the rock and its structures, and recording their observations in detail. Confusion arises firstly from weird log formats, often combining interval with point logging, and using quaint look-ups among other foibles that have survived for decades in many companies, because of archaic database structures. Secondly, the inability to understand the drilling process, in particular those aspects that affect core measurements and quality, as outlined in the eight articles posted previously, is massively limiting.

There are a number of possible solutions, some of which will be explored in the next post, but suggestions/comments are welcome.

? J.L. Orpen (Resource Exploration & Development Pty. Ltd.)