The application is designed for ease of use. The user begins by choosing a geological formation from a list of available alternatives. Then he/she specifies boundary conditions, injection sites, injection rates, durations and physical effects. After launching the simulation, progress can be tracked in a separate visualisation window. When simulation is completed, the resulting CO₂ inventory is presented. The visualization window can also be used to plot the value of a number of variables (CO₂ saturation, pressure, etc.) for any timestep of the simulation. The outcome can also be saved to disk.

A number of simulations with different input parameters can be launched from the scenario specification window, each being assigned its own visualization window.

Despite its simplicity and low compuational requirements, the simulator is capable of taking a significant range of physical effects into account, including realistic fluid properties (based on accurate equation of states), rock properties (relative permeability, capillary effects, compressibility), two-phase flow, CO₂ dissolution into brine, residual saturation, and structural trapping on macro and subscale levels. There are three different types of boundary conditions, all of which can be applied on different parts of the boundary for any given simulation.

First, the user chooses the desired formation model from the drop-down list at the lower left side of the interface window. The default formation is Utsira ("Utsirafm"), but there are sixteen different formations to choose from. All models represent formations located below the Norwegian North Sea.

The chosen formation will immediately be displayed in the graphical window on the left side of the scenario specification interface. The model can be rotated and visually inspected as long as the "Rotate model" radio button box is selected (located at the upper right corner of the interface). The user has the option to highlight the location of structural traps by ticking the "Outline traps" checkbox found in the lower right corner of the interface.

When rotating the model, it will appear to have no thickness. In fact, only the caprock shape is displayed. The full aquifer thickness will however be taken into account during simulation.

Each part of the lateral boundaries of the aquifer must be one of three types: open, semi-open, or closed:

• open: fluids flow freely across this boundary, without any resistance.
• semi-open: fluids flow across this boundary, but with an additional resistance equivalent to flow through an additional distance outside the simulation domain, using the same permeability value as in the aquifer (default distance is 100 km, but this is configurable).
• closed: no flow occurs across this boundary.

In order to specify boundaries, the user must first activate the "Edit boundary" radio button found in the upper right panel of the interface. The default is open boundaries everywhere (indicated as a thick green aquifer outline in the display window). If a completely closed or semi-open boundary is desired instead, the user may select the appropriate radio button in the upper middle panel, and then click "Set all". Semi-open and closed boundaries are drawn with yellow and red lines, respectively.

Boundary conditions may also be defined piecemeal by clicking with the mouse pointer inside the aquifer model. The first click will set a marker (indicated as a red star) at the given position. A second click away from the first will define a line. The part of the boundary closest to this line will change status to whatever condition is active in the upper middle display (open, semi-open or closed). This way, detailed boundary conditions can be set up in an intuitive way even for aquifers with complex boundaries.

It is possible to define up to ten injection locations for a given simulation scenario. Injection locations are chosen by first activating the "Select wellsite" radiobutton in the upper right panel, and then clicking at the desired locations on the model. The location of the well will be indicated with a black dot.

Figure 2: The well table

The injectors will be named W1 to W10. Each injector will be assigned a row in the well table, located in the middle right part of the user interface. When a injector is placed, its row in the well table will become active. Its latitude/longitude coordinates will appear in the second field, and its annual injection rate can be adjusted with the slider (from 0 to 10 megatons per year). An erroneously placed injector can be deleted by clicking the button labeled X in its corresponding row.

On the right hand side of the user interface, there is a small box which allows specifying the duration of the injection (default: 50 years), the duration of the migration phase (default 3000 years), and the number of discrete timesteps used to represent each of these periods. The user is free to play around with these numbers, but should keep in mind that a too small timestep size will just lead to a waste of computing resources, and too large steps may lead to 'timestep cutting', which is also computationally wasteful. In general, the migration phase admits longer timesteps than the injection phase.

On the right of the green 'Launch' button, there is a small box where the user is allowed to select which additional physical effects to include in the simulation. These options are:

• Capillary fringe - The interface between CO₂ and brine will not be modeled as a sharp line, but as a diffuse transition from one phase to the other. This may have a significant impact on plume migration.
• Subscale trapping - Take into account the structural trapping that occurs below the resolution of the simulation grid. Further information can be found in publications 2, 3 and 9 from the literature list at the bottom of this page.
• Use dissolution - Take into account the dissolution of CO₂ into brine (if left unchecked, the phases will be considered immiscible).

When the above steps have been taken care of, the scenario is completely specified, and the simulation can be launched by clicking on the green button labeled "Launch new simulation!".

Once a simulation is launched, a new window opens, presenting user with the result visualization interface. 

Figure 3: Tool for visual inspection of simulation result

As soon as the simulation is finished, the user can use the slider to move to any timestep in the simulation. Different visualization variables can be chosen from the drop-down menu, including:

• s: CO₂ saturation (default)
• pressure: fluid pressure in the aquifer
• sGmax: Historical maximum CO₂ saturation
• h: CO₂ plume height (closely related to saturation in a vertical-equilibrium setting)
• plume_s_avg: The average saturation within the free plume, as a fraction of maximum achievable saturation. For a simulation without capillary effect (sharp-interface), this value will remain constant, but if a capillary fringe is included, this number yield insight on the local significance of the fringe.
• overpressure: fluid overpressure in the aquifer, defined as the difference between current pressure and initial pressure.

For a better view, the result visualization window can be resized, and the model can be rotated using the rotation button found in the toolbar.

The CO₂ inventory display also opens when a simulation has completed. The inventory display tracks the trapping state of CO₂ over the course of the simulated (injection and migration) period. For any given year, the inventory shows how much CO₂ is found in each of the following states:

• Dissolved : CO₂ that has dissolved into brine. (Only available if "Use dissolution" was selected when the scenario was specified).
• Structural residual : CO₂ that is both structurally and residually trapped (i.e. residually trapped CO₂ within a structural trap).
• Residual : CO₂ that is residually trapped, outside structural traps, and not in the free plume.
• Residual in plume: CO₂ that is currently part of the free plume, but is destined to be left behind as residual saturation after the imbibition process has been fully completed.
• Structural subscale : CO₂ that is caught in structural traps too small to be captured by the grid's resolution. (Only available if "Use subscale trapping" was selected when the scenario was specified).
• Structural plume : Structurally trapped CO₂ other than "Structural residual".
• Free plume : CO₂ that is free to migrate.
• Exited : CO₂ that has left the simulation domain (through an open or semi-open boundary).

Figure 4: CO₂ inventory