Should Geologists Master Math and Statistics to Excel in Modern Science?

In an era where interdisciplinary research is not just beneficial but necessary, the way we present our findings can either foster collaboration or deter it. As a social scientist with qualifications in both science and engineering, I have come to appreciate the wisdom in Shunryu Suzuki's belief that "In the beginner’s mind there are many possibilities, but in the expert’s there are few." This perspective has significantly shaped my approach to research dissemination, particularly in how we write our papers.

Our research group has made a conscious decision to exclude complex mathematical equations from our peer-reviewed publications but we add them in dissertations and workshops. This approach is rooted in the principle that knowledge should be accessible to a broad audience, reflecting Richard Feynman's philosophy that “If you can’t explain something in simple terms, you don’t understand it." Furthermore, as William Shakespeare aptly put it, "Brevity is the soul of wit."

This choice, however, is not without its critics. The purists in Geostats and Mathematical Geology argue that framing ideas mathematically demonstrates a deep understanding and facilitates the replication of techniques. While there is merit to this argument, it often overlooks the fact that significant scientific breakthroughs frequently come from those outside the primary field of study. For instance, my fascination with X-ray computed tomography in 2009/10 allowed me to apply this technique from physics and medical studies to process mineralogy and geometallurgy, enabling real-time metal accounting processes without being an expert in physics and medical sciences.

Yet, challenges remain. I recently encountered a paper rejection where the critique was not on the science but a perceived lack of "mathematical and/or computational equations in the disguise of "lack of mathematical rigour"." Despite referencing my prior publication with the necessary detailed equations, the expectation was to repeat this information like what one typically finds in mathematical geosciences papers - a discipline I practice. Although I later addressed the feedback to get my paper accepted by incorporating more detailed mathematical descriptions, it led me to reflect on the implications of such expectations.

This experience highlighted a broader issue: the barrier to entry for geologists and similar professionals into disciplines such as machine learning and geodata science is often the intimidating requirement of advanced mathematical framing. This is a significant hurdle, considering most geologists receive limited mathematical training. Some of my colleagues believe it's more effective to train statisticians and computer scientists in basic geology concepts to address problems in geodata science. I disagree with this approach.

The question then arises: how can we facilitate the development of computational skills in these professionals without the prerequisite of extensive mathematical and statistical education?

Addressing this challenge requires a shift in academic and practitioner culture to value narrative clarity and application as much as mathematical rigour. We need to find a way to open doors to innovative solutions across disciplines and ensure that scientific advancements are not just the domain of the mathematically proficient. I say this as someone well-trained in statistics and mathematics, but such expertise shouldn't be a prerequisite. Most of our geoscientists are capable of understanding these concepts, even if they can't always express them as first-principle mathematical problems.