Modelling of gas hydrates in sedimentary systems : Methane hydrates impact on flow through porous media

Abstract

In nature, gas hydrates exist in areas of permafrost and in shallow subsurface sediments at ocean depths of more than 300-500 metres. In terms of the Sustainable Development Goals (SDG) of the United Nations, better understanding of hydrates in nature can play a role in achieving energy security (SDG7), tackling climate change (SDG13) and increasing sustainability in the use of oceans (SDG14).

Hydrates represent a potential energy resource as one litre of methane hydrate contains 180 litres of methane. However, if heat stress is induced by either artificial or natural causes, its destabilisation can result in the addition of more methane to the ocean-atmosphere system. At the same time, they can trigger geohazards in their natural environments.

The knowledge of the gas hydrate dynamics when changes are imposed either naturally or artificially by drilling and gas exploitation is not sufficiently understood. To contribute to the understanding of gas hydrate dynamics in nature, we used a numerical simulator of hydrates in porous media to reproduce and study hydrate-related processes at different scales.

The TOUGH+HYDRATE (T+H) code was the main tool used in this study. It simulates the behaviour of methane hydrate in sediments and handles both multiphase and multicomponent flow and couples heat and mass flow through porous and fractured media. To streamline the use of T+H, it was necessary to build versatile pre- and post-processing tools. These tools were written in Python and mainly process the input and output data so that the candidate could streamline access to the data, perform analysis, and prepare visualisations. The use of these tools was essential to produce the bulk of the results and accompanying figures presented in this thesis.

The scientific output of this thesis consists of three scientific papers that present numerical modelling of hydrates in porous media in different scenarios. Paper 1 and paper 2 focus on modelling laboratory experiments of hydrate-bearing porous media. Paper 1 focusses on modelling previously acquired measurements of methane relative permeability in hydrate bearing sandstone. Simulations show that the experimental values are difficult to predict by using a homogeneous distribution of hydrates throughout the sample. The experimental magnetic resonance imaging data also showed that the hydrate distribution can be heterogeneous, meaning that the hydrates create patches. Initialising the model with a heterogeneous distribution yielded better results. These results show the impact of heterogeneity on the distribution of hydrate saturation and suggest that small amounts of hydrate can have a disproportionately large effect on the permeability. Therefore, in those cases, the use of simple models will give erroneous results because of the too high permeability.

Paper 2 studies the effect of kaolin clay on the growth of hydrates. Clay minerals are common in subsurface sediments. This study presents both experimental and modelled results. The experiments consist of the growth of methane hydrate in sand mixtures with different amounts of kaolin clay. Experimental results suggest that both clay content and initial fluid phase saturation have a large impact on the hydrate growth rate and final hydrate saturation, respectively. The experiments are simulated using particle size as a proxy for the clay content. The simulations confirm the main two effects inferred from the experiments. However, the discrepancies between the two suggest that additional mechanisms are hindering fluid flow.

Paper 3 switches to mechanisms that occur on a larger temporal and spatial scale. The dynamics of gas hydrates over longer time scales (between 100 and 600 thousand years) were simulated during different sedimentation rates and permeabilities. The results clearly show the connections between all of the detailed physical mechanisms that work during the melting and reforming of the hydrates. Hydrate melting and reformation occur in a stepwise manner, following the continuous sedimentation. The pattern of change is the result of a combination of factors, including the sedimentation rate (heating) and the intrinsic transport properties of the sediments.

The work has been fully theoretical and consisted of careful planning of simulations and the collection of experimental data from both laboratory and field. Cooperation with other research fellows and staff at the Institute for Physics and Technology at the University of Bergen has been important.

The results of this work are relevant for the understanding of natural gas hydrates in the subsurface. They are applicable to the role of hydrates as energy resources and geohazards. They may also be applicable to the ongoing research on the role of hydrates in climate. Real data from nature are hard to collect, and observations other than seismic are also difficult to obtain. There is no evidence that methane in the atmosphere originates from hydrates, but we cannot confirm its fate if the heating of permafrost and oceans continues. What we can tell is that the melting and freezing of the hydrates will probably be very slow processes. This work may provide a tool to calculate the time it will take for melting hydrates to reach the atmosphere. Then, hydrates can be entered into the climate gas budget of the planet.