Trapping capacity estimates

Several trapping mechanisms can be considered at play at the Utsira Formation:

• Structural trapping: CO₂ migrates as a largely separate plume beneath the sealing caprock.  Parts of this plume get trapped in local domes and pockets, forming an 'inverted lake'. 
• Residual trapping: The migrating plume leaves behind a trail of CO₂ droplets trapped in the pores of the rock by capillary forces.  This CO₂, whose quantity can be considerable, will remain immobilised.
• Dissolution: CO₂ gas is soluble in water (as commonly seen in carbonated drinks), and given enough time, the injected CO₂ could gradually dissolve in the surrounding brine.  The brine with dissolved CO₂ is heavier than brine without, and will thus gradually sink to the bottom of the formation.  The rate at which CO₂ dissolves into brine depends on several factors, and it is not clear how quickly this happens at Utsira.
• Mineral trapping: CO₂ can react chemically with the rock matrix and form minerals.  This is usually considered to be a very slow process and therefore only relevant at very long time scales.

Capacity of structural traps

The structural trapping mechanism can be exploited by injecting into locations where CO₂ is expected to naturally migrate to one or several identified traps.  Since this migration will be largely driven by gravity, a spill-point analysis can be used to identify such locations, as well as estimate the coresponding amounts of CO₂ that will be structurally trapped.

• We obtain a geometric representation of the Utsira caprock by converting data published by the Norwegian Petroleum Directorate.
• We perform a geometric analysis of the caprock shape in order to identify the traps.  These are displayed on the left plot below, colored by size (in terms of pore space).
• The pore space of a structural trap is not sufficient to estimate how much CO₂ it can hold.  It is also necessary to know the local density of CO₂ at the position of the trap. 
• CO₂ density is a function of pressure and temperature.  Conditions at Utsira are such that CO₂ density is expected to vary considerably across the aquifer.  In large parts of the formation, CO₂ is expected to be in gaseous, rather than dense phase.
• Taking estimated local density into account, we produce the plot on the right below, which shows how much CO₂ each trap is expected to retain in mass terms.
• Some of the largest traps (in mass terms) are found in the middle of the southern region.  There is also a very voluminous trap on the south-western border, but CO₂ is expected to be gaseous (very low density) at this location.

• A complete spill-point analysis allows us to trace out the connections ("rivers") between structural traps, and predict where CO₂ would migrate if purely driven by gravity, as  visualised on the left plot below.
• This knowledge allows us to estimate how much structural capacity would ultimately be reached when injecting at a given location.  This is shown on the right plot below.
• According to this analysis, the best place to inject if one wants to maximise the use of structural trapping is at the very south-east region. 
• CO₂ injected in this region will gradually migrate north-west, filling up a number of large traps along the way.

Including residual and dissolution trapping

Although structural trapping is the easiest to aim for, it is also important to benefit as much as possible from the other trapping mechanisms.

• If we consider the Utsira Formation represented as a volume grid consisting of vertical pillars, the total trapping capacity of the aquifer is the sum of the total trapping capacity of each pillar.
• In order to compute the theoretical maximum trapping capacity of a pillar, we sum up:

1. The structural trapping capacity provided by the part of the pillar that is located within a structural trap (if any)
2. The residual trapping capacity of the pillar, which is a property of the rock.  If the rock is homogeneous, it will simply be a multiplier of the pore volume of the part of the pillar not contained within a structural trap.
3. The dissolution trapping capacity can be considered equal to a certain fraction of the pore water remaining in the column after the structural and residual trapping capacities have been realized.

• It is clear that for pillars at a distance from the injection point, it will only be possible to utilize a small fraction of this theoretical capacity.

• When computing the total storage capacity for each pillar in our grid representation of the Utsira Formation, we obtain the figure on the right.  The unit is megatons of CO₂ per pillar, where the lateral extent of each pillar is 500x500 meter (cell size of the grid). 
• The total storage capacity for a pillar is highly dependent on local aquifer thickness, local aquifer depth and presence of local structural traps.
• The sharp line that can be seen cutting into the aquifer from the south-west and exiting at the north-west correspond to the location where CO₂ beneath the caprock passes from dense to gaseous phase.  Locations above this line thus entail considerably lower capacities.
• Altogether, using the above methodology combined with physical parameters from published literature, we estimate the total theoretical storage capacity of the Utsira aqufier to be:

• 1.13 gigatonnes of structural trapping
• 77 gigatonnes of residual trapping
• 34 gigatonnes of dissolution trapping

• Needless to say, these figures remain highly uncertain.

Diagram of aquifer cross section (click on image for a larger version).

_{Color plot showing the structural capacity reachable by gravity-driven migration when injecting at a specific point (click on the image for a larger version).}

_{Diagram showing total trapping capacity of each vertical pillar of the Utsira Formation grid model.  Click on image for a larger version.}