Towards better prediction of rock thermal conductivity

The ability of rock to transport and distribute heat in the subsurface is of major importance for the formation of oil and gas. Senior research scientist Niels Bo Jensen at IRIS investigates the controls of thermal conductivity in rocks.

Organic matter in petroleum source rocks will mature and form oil and gas when buried for long periods of time at higher temperatures in the subsurface. The heat necessary for this maturation flow from the deeper parts of the earth towards the surface and is a result of a long-term cooling process. In addition, within the earth’s crust, decay of radioactive isotopes, exothermal metamorphic and diagenetic processes contribute as a heat source. In this framework thermal conductivity is important because it is decisive for how the heat will be distributed.

A rock’s thermal conductivity is controlled by its mineral composition, porosity and pore content. Fractures and textural characteristics can also influence thermal conductivity, e.g. the orientation of clay minerals in shales results in a higher conductivity parallel to the bedding and lower across it.

foto elisabeth t%c3%b8nnessen (51)

In his office at IRIS, Niels Bo Jensen investigates the controlling factors of rock thermal properties. Amongst other instruments he uses an optical laser scanner to measure thermal conductivities of core samples from wells in the North Sea and Svalbard.

To be successful there are many details that must be looked into and understood. We have created large data models to describe the (tectonic) geological development through time and space in order to investigate and understand the exact turn of events characterizing the history of the sedimentary basin. In order for them to succeed high demands are made upon the model equipment and the input data. The more detailed information we have, the more important it is that the quality of each and every parameter is satisfactory.

Rock thermal conductivity is measured in a laboratory, however, we may not always have access to sufficient sample material of all the different lithologies within a sedimentary basin. Our solution to this problem has been to make good prediction models for thermal conductivity.

IRIS has  developed a large database of rock mineralogy, porosity and thermal conductivity measurements. By use of multivariate statistics, Jensen, in collaboration with InfoStat AS and Tectonor AS has analysed measurements of rock thermal properties and petrophysical well log information and built a prediction model. The model is capable of predicting the thermal conductivity of a rock in two directions, i.e. horizontally and vertically.

This approach has been a success, says Niels Bo Jensen, and tests have further proved to be both accurate and precise.

For a long time we have thought that knowledge of the mineralogy and its variation in the subsurface is too limited for our basin models. We wanted the models to be based on reliable and objective input and turned our attention towards the use of petrophysical well log rather than making qualified guesses of average mineralogical content and porosity for large rock volumes. Mineralogical information is normally only available from cored reservoir sections in a limited number of wells whereas petrophysical well log information is available from a  much larger number of wells and depth intervals.

In the project collaboration between IRIS, Tectonor AS og InfoStrat AS we have conducted hundreds of thermal conductivity measurements on samples from the North Sea and Svalbard comparing the results with the associated well log information. The preliminary results are very promising and clearly confirm that it is possible to extract latent information from a suite of petrophysical well logs and predict thermal conductivity on a level of detail not seen before.

The preliminary results were presented at Geological Society of Norway Winter Conference in Oslo 8-10. January.