Big Data, Machine Learning, and Artificial Intelligence: A Probable Future for Geology

By Paul Joshua Villora

Amidst the COVID-19 pandemic, the future of the field of the geosciences seems to have a vague future along with other fields. Geology in particular, which involves mostly fieldworks and other on-site activities, is hugely affected by this. This is apparent in both working geologists and geology students who, during these trying times, are having a difficult time to adopt alternative ways of “doing geology work”.

Using the famous line, “The earth is constantly changing”, is enough to agree that Geology in particular is relevant. Also, from the line, emphasize on the word “change”. Despite being relevant, how do we, the future generation of professionals, contribute to the “change” or the “necessary advancement” particularly of Geology? One proposal: with the use of Big Data and Machine Learning.

What is Artificial Intelligence?

According to the DOST Project Sparta, a data science training program, big data is generated by the increased analysis and storage of data (i.e. social media, databases, and specialized instruments). It is used to analyze a situation and derive solutions and/or observations from it using statistical methods. This process of synthetic recognition generates patterns that apply autonomous machine work. This synthetic recognition is referred to as Machine Learning. Collectively, this process is called Artificial Intelligence or AI (Chen et al., 2020).

AI sounds very dystopian but it has been present for a long time. Some very common examples include social media advertising where your interests are categorized with your likes and interaction (the data), and used by AI to direct certain products/services ads which are relevant to your interest (machine learning) — the data collected and processed are called “cookies”. Another example is in stock trading where technical indicators use various statistical measurements (machine learning) to evaluate various factors (the data) affecting the stock (i.e. Volume, Price action and Volatility, Buyer/Seller ratio, etc.). Now, how about in the field of Geology?

Why Big Data and Machine Learning (ML)?

More and more people are now in the internet and along with that is the rise of faster- processing technologies. We derive big data from these advancements which we use to teach our existing instruments to become smarter. Many disciplines take advantage of this aside from those in the field of computer science and engineering. Finance, governance, and several science fields make use of these for faster and more accurate prediction of results. So, why not in the geosciences? Well, there are many existing applications practiced already in the industry. The real question is: how?

Using Big Data, ML, and AI in Geology

The essence of asking “how” Big Data and ML is applicable to geology introduces another perspective of why geology is important and relevant to human society. Geology has a wide spectrum of disciplines. Thus, this allows a big opportunity for the application of Big Data and Machine Learning:

On Mining and Mineral Exploration

Finding new mineral deposits is getting more and more difficult as human society continue to progress. Fortunately, this is somehow addressed by increasing the amount and type of suitable data for analysis and synthesis to derive exploration targets (Bergen et al., 2019; Karpatne et al., 2017; Yousefi et al., 2019; Prado et al., 2020). Methods for mineral prospectivity mapping (MPM) involve tools that can be applied with large quantities of geoscientific data. Spatial patterns using the geographic information system (GIS) are helpful in determining areas associated with the occurrence of certain mineralization as well as predicting other areas having similar occurrence of certain types of mineralization.

On Disaster Risk Management and Engineering Geology

Natural hazards have long been affecting human society. One of the most devastating are geologically-induced ones such as earthquakes and earthquake-induced landslides because it’s difficult to predict its occurrence. However, thanks to a more integrated use of historical evidences and geotechnical evaluation, different input parameters are processed into a single output model through the various weighting, calculating and interpolating methods. Susceptibility reports for a certain are produced from this framework which is used to assess it in terms of risks and hazards (Marjanović, Kovačević, Bajat, & Voženílek, 2011). A good example of where this is used is in a hazard assessment site commissioned by different government agencies called HazardHunterPH.

On Space Science

With more geographical data needed in the AI-integrated geosciences, satellite imaging is the best option to attain more of it. This means that geologists, having the knowledge on the context of what data is needed, have a higher chance of becoming scientists to be sent off to space for its collection. Alongside with this opportunity, is the chance to study more objects outside of earth with the advanced tools and knowledge that we have.

Will AI take away my career as a geologist?

To borrow a quote from Dr. Teofilo Abrajano, a Filipino Geologist, non-verbatim: “Succeeding Geologists will replace those that do not adapt to new technology and skills”. So, instead of replacing a geologist’s work, AI will be a complement to it. There are still so many unknowns in the field of Geology. It’s high time we look into another perspective of how we do Geology — integrating it with Artificial Intelligence.


Chen, L et al. (2020). Review of the Application of Big Data and Artificial Intelligence in Geology. J. Phys.: Conf. Ser. 1684 (2020) 012007. doi:10.1088/1742–6596/1684/1/012007

E. Martins Guerra Prado, C. Roberto de Souza Filho, E. John M. Carranza, J. Gabriel Motta, Modeling of Cu-Au Prospectivity in the Carajás mineral province (Brazil) through Machine Learning: Dealing with Imbalanced Training Data, Ore Geology Reviews (2020), doi: 2020.103611

Geological Society of the Philippines (2020). Sustaining Mining in the post-COVID World [Recorded by T. Abrajano].

Marjanović, M., Kovačević, M., Bajat, B., & Voženílek, V. (2011). Landslide susceptibility assessment using SVM machine learning algorithm. Engineering Geology, 123(3), 225–234. doi:10.1016/j.enggeo.2011.09.006



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store