Petroleum Geoogy is a broad subject and one could write books about it - indeed, many books have been written about it, as well as huge numbers of papers in the scientific literature.
Why it is relevant to the conversation is that it is a historical science (firmly grounded in empirical data and the known processes of physics, chemistry and biology, with at times a smattering of astronomy as well) where our hypotheses are regularly tested.
You donât find hydrocarbons by throwing darts at the map. In the words of a famous early Petroleum Geologist, Wallace Pratt, âoil is found in the minds of menâ. In his days they turned their ideas into analog models, paper maps, correlation panels, cross sections, burial graphs etc. Nowadays everything is of course digital so we construct computer models that are vastly more intricate and complex and can perform orders of magnitude more, and more accurate, calculations. But even then, the seed of the discovery resides in the mind of the geologist.
Our models need to be in four dimensions, since time is an essential element in geology. It takes time to deposit the sediments that eventually wil generate hydrocarbons, reservoirs and cap rocks. It takes time to deform strata to create traps and it takes time for hydrocarbons to migrate from their source area into the reservoirs.None of these processes are instantaneous or fast, in human terms. Using computer models we can speed up time and simulate the entire process, always within the constraints of known physics, to understand if and where we may encounter oil or gas.
As input data we uses virtually all geological, geophysical and geochemical data that we can lay our hands on, the more the better. Unfortunately, it will never be enough because or models will always be vastly undersampled and not all processes are understood as well as we would like. Even so, the success rate in the industry has long been more than enough to make it profitable, and vastly more so than if we would just throw darts at a map.
Oil is found in geologically favourable places. The task of the geologist is to understand what factors are favourable and predict where these places may be. In addition, almost equally important is to understand and quantify the risk of getting it wrong. This is one reason why we try to be as quantitative as we can, so that we can use statistical methods for quantifying uncertainty. This is actually not at all straightforward, and to minimise personal bias such evaluations are normally done by a review-and-challenge panel rather than by individuals.
Reconstructing events aims to tell us where the hydrocarbon source may be, when it may have started generating and expelling oil and/or gas. It aims to tell us where suitable reservoir rocks may be, and if and when they may have been deformed to create traps, or if they have perhaps been adversely been affected by diagenetic alterations. It aims to tell us if there is a chance of an impermeable seal overlying the reservoir so that the hydrocarbons canât escape to the surface and be lost. It aims to inform us about potentially dangerous or even catastrophic overpressures deep in the subsurface that may be encountered during drilling.
It is a plain and simple fact that no wells would get drilled if we didnât have at least some model of the geological history of an area. The risks are simply to great, both technical and economical. It is also a plain and simple fact that nobody has ever made a useful petroleum geological model starting from YEC assumptions. Just saying.
Every time we drill a well we test our model, and whatever the results we update and refine them. Even dry holes will yield large amounts of important data that will be used in future models and proposals. Clearly, a dry hole will require more substantial model revisions than a successful well. Parts of the model will be falsified and cannot be taken forward unchanged. In this way our assumptions are tested and our knowledge keeps evolving and growing.
There is one unfortunate factor in all of this, which is that due to the highly competitive nature of the industry a great amount of data and models reside in corporate archives and will only very slowly be released into the public domain. There is a lot to find in the published literature, but even that only skims the surface of what would be available if everything were to be disclosed. Who knows, once the time comes when oil and gas have lost their importance (and eventally, such a time will come), all that data may be released to the benefit of all of geological science. And beyond, I think.
