In this article, I will focus on and challenge the two fundamental assumptions that the current theory for generation and migration of oil and gas rests upon: decreasing TOC in source rocks in the temperature interval between 120°C and 150°C and biodegradation of oil in reservoirs at temperatures below 70°C.
A non-existing problem
Professor Dag Karlsen defends the current theory by asserting that I have misunderstood, and therefore address a “non-existing problem”. He claims that the TOC data I have used when challenging current theories “do not give a representative picture of reality”, a point of view he defends by writing that: “As for the alleged lack of reduction of TOC through the oil window (North Sea), it is very important to understand that there are no samples from the oil-generating deep sub-basins in any shale database from the North Sea. This is because not a single well has been drilled in the actual generating source rocks for the oil and gas on the Norwegian shelf.”
This is not correct. Many of the available TOC samples are collected from what is currently considered to be the oil window. There may of course exist richer source rocks deeper than in today’s oil window, and so deep that we do not have samples from these intervals, but this cannot explain why TOC does not decrease in today’s oil window. A TOC of 5 % or 10 % in the source rock when it entered the oil window is not important for the current theory. What is important, is that according to the present theory half of the TOC, regardless of the TOC value given that we have an effective source rock, will leave the source rock as oil. The TOC data indicate that this does not take place.
Okiongbo, Aplin and Larter (2005) recognized this problem. They noticed that the TOC did not decrease downwards within the oil window. In order to save the conventional theory from being falsified, they assumed that the original TOC in the deepest samples had been twice as high as at present before the source rock entered the oil window. In a second data set where the same problem was present, Okiongbo (2011) solved it by selecting the samples from the base of the oil window with the lowest TOC. This made the remaining data consistent with a 50 % reduction of TOC in the oil window.
Ad hoc assumptions
These are examples of what we within theory of science classify as ad hoc assumptions. Such assumptions and explanations are permissible, and it is always possible to come up with assumptions that save the model. In a situation where an alternative model is lacking, one is almost forced to resort to such additional assumptions. However, in the case of oil generation and migration an alternative theory does exist. The pre-1970 model.
It is important to be aware that the pre-1970 theory has not been refuted. Such a refutation of the old theory is strictly speaking not required when introducing a new theory, but the observations that the old theory was based upon must be included in and be explained by the new theory – like Einstein’s theory of gravitation did when it replaced Newton’s theory. However, such an integration of the old observations into the new theory did not take place during the development of the new theory for oil generation and migration in the 1980s.
This is what my paper documents. To discover what today’s geochemists have overlooked, one has to go back and read the papers I cite. One will then for instance discover that the ability of heavy oils to rotate the plane of polarization of light is strongly reduced as temperature increases and approaches zero at temperatures above 70°C. Earlier authors therefore utilized such measurements to estimate the temperatures oils had been exposed to, a method that at low temperatures is better suited than measurement of vitrinite reflectance.
Overlooking previously published work
As an historian of science, I know that from time to time it is necessary to modify what is considered to be accepted theory within a field, typically because someone has come up with a new and better explanation of the available observations, but I am not aware of cases where practically all workers within a field have overlooked previously published observations concerning the phenomena under discussion. One may disregard other researchers’ explanations, because these explanations depend upon which theory one favours, but not their observations, which is what today’s organic geochemists do (as I have explained in my article).
What makes the situation even more remarkable, is that instead of accepting the observations made by explorationists over a period of more than 100 years, one chooses to construct explanations based upon something that has not been observed: bacteria.
One was forced to introduce the concept of bacterial degradation of oil to explain heavy oil – the most abundant type of oil. Biodegradation of oil is therefore another example of an ad hoc assumption.
The 70°C limit
The history behind the idea of biodegradation is interesting. During the 1970s microbiologists became aware of the existence of bacteria that could withstand temperatures up to 70°C. This was something the geochemists took notice of. Initially it was assumed that these bacteria required oxygen, but around the year 2000 it was accepted that pore waters did not contain enough oxygen, a point that had been known by others for a long time. The problem was solved by the geochemists by postulating that the bacteria were anaerobic and did not require free oxygen. This is an example of yet another ad hoc assumption.
When light oil is transformed to heavy oil, the theory says that the bacteria convert the oil to CH4 and CO2. There is normally no gas above the heavy oil. This is explained by leakage of methane. This is an additional ad hoc assumption.
The proponents of the current theory seem to consider their ability to save their theory in this artificial manner as a strength, whereas it is in fact an expression of the failings of their theory. The biodegradation theory contains a wealth of such ad hoc assumptions. Bacteria are assigned characteristics they must have if the observations are to be explained.
Science starts with observations
«Science starts with observations, and it ends with them” (Einstein). It is therefore not very meaningful to continue the discussion with Karlsen unless he can provide evidence for the existence of live bacteria in the subsurface. Today’s theory for generation and migration of oil stands and falls with bacteria performing the task they are said to perform. If one cannot detect living bacteria, one should be able to find more chemical traces of dead bacteria (necromass) where there is heavy oil, than in shallow pore waters where there is no oil. The published data indicate no such difference.
Moreover, today’s model assumes that oil-consuming bacteria survive in a dormant state for millions of years without nourishment before oil enters the reservoir. There is of course no empirical support for bacteria being able to survive such enormous time spans without nourishment.
Despite these obvious weaknesses, Karlsen claims that there is a general consensus within his field of research. He therefore finds it strange that I, a single individual from outside this field, have challenged them. In reply to this I will firstly adduce that consensus is not a scientific argument, secondly, science is based upon the individual’s right to challenge. I must also add that I do not stand alone – I stand upon the shoulders of those who participated in the discovery of most of the world’s oil reserves more than 50 years ago.
An alternative theory
What I have done, is to suggest an explanation for how oil may leave source rocks early in their burial history due to mobilization by carbon dioxide, and for how the heavy oil may be converted to light oil in the reservoir by addition of H2. The lack of such an explanation in the 1970s made some see the need for an alternative theory. This alternative theory assumed that the oil already had a low viscosity when it left the source rock. This appeared to be an elegant solution to the difficulties inherent in the old theory. However, a physical explanation of how the oil is able to leave the source rock has not been presented.
My starting point for what developed into support for the model of early expulsion at low temperature was an attempt to explain migration from source rocks at temperatures above 120°C, i.e., I assumed that today’s model was correct. This took me a year. It was a physical analysis, and this analysis indicated that expulsion could take place much earlier. The problem that initially held me back was that according to the current model there is not movable oil in source rocks at temperatures less than 120°C.
I therefore searched for data indicating the presence of mobile petroleum/oil shortly after deposition. I discovered such data. This had been observed in shallow samples in the 1950s. Actual hydrocarbons where not abundant, but asphalt and resins (bitumen), which are designated non-hydrocarbons because they also contain nitrogen, oxygen and sulphur, are present. This is where CO2 enters the picture. Large amounts of CO2 are formed from organic matter shortly after deposition, and CO2 gas can mobilize these molecules before they polymerize and become a part of kerogen.
The role of kerogen
It is also worth noting that according to the pre-1970 chemists, the kerogen, that today’s geochemists consider to be the only source of oil (see also Karlsen’s comments), was defined as inert organic matter, in other words organic matter that did not generate oil unless it was exposed to extremely high temperatures. Like when source rocks are invaded by magmatic intrusions.
It is also appropriate to keep in mind that science has a guiding principle saying that one should seek simple solutions. Given the choice between a simple and a complex theory about the same topic, one chooses the simpler alternative.
The pre-1970 model has a simple structure. Early expulsion to adjacent reservoirs, and maturation in the reservoir to light oil when temperatures approach 70°C. Today’s model is far more complicated. It assumes that almost all oil migrates through carrier beds (that we cannot show are present) up through km-thick shales (that lack evidence for massive invasion of oil) and into the upper 2 km (T<70°C) of sediments where most of the oil is consumed by bacteria (that we are unable to detect) and transformed to gas (that we do not find).
Complicated and high speculative
Today’s model is therefore both complicated and highly speculative, whereas the pre-1970 model is simple and based upon what we can see and measure. Moreover, the pre-1970 model explains more than today’s model. The choice should therefore be simple. Due to its simplicity, the pre-1970 model is also easy to use for explorationists in their daily work.
Regarding maturation in the reservoir by external supply of hydrogen Karlsen states that (my translation): “Whereas atomic hydrogen is very reactive, 104 kcal/mol are needed for dissociation of molecular hydrogen. Bjørkum does not explain where this energy comes from.” If Karlsen consults organic chemists, they will tell him that where there is molecular hydrogen (H2) there will always be a small proportion of reactive atomic hydrogen (H).
Karlsen is aware that we do not discover what we say we will find. He explains this by stating that (my translation): «All models, including good ones, can be misused, data may be misinterpreted, and in many cases preconceived opinions regarding exploration models, top-down management and conditions such as excessive bureaucracy instead of flatter structures, prevent a good exploration dialog – without faults being present in the models.”
In other words, when we miss, Karlsen blames the users of the theory. I exonerate the users and blame the theory, a theory that should never have been accepted. It does not fit the facts. One should therefore have continued working with the pre-1970 theory.
No oil to mobilise?
If the researchers within the field had done so, I feel reasonably confident that they would have arrived at the solution that I have suggested regarding early CO2-assisted expulsion of oil and maturation of oil in the reservoir due to external supply of hydrogen. Momper (1978) knew, as mentioned in my article, that CO2 gas could mobilize oil, but shared Karlsen’s misunderstanding that there is no oil to mobilize at such a shallow burial depth.
Anyone interested in studying the scientific basis for the model I have proposed, could start by reading R.H. Dott and M.J. Reynolds impressive “state-of-the-art” publication from 1969 (AAPG). It comprises more than 400 pages, and is, together with the extensive list of references, a gold mine for all empiricists.
PER ARNE BJØRKUM
