4. Generate the comparison figures#

4.1. Goal#

Locate the figures that compare the YOURS result with the experimental and published FluidFlower benchmark data.

4.2. Configuration#

The original command used -f all. This produces the basic benchmark figures and the Wasserstein-distance comparisons. The complete set requires evaluation times of 24, 48, 72, 96, and 120 hours.

4.3. Command#

The figures were already generated by the combined tutorial command. To regenerate figures from existing data without rerunning OPM Flow, use:

pofff -o YOURS -m none -t 24,48,72,96,120 -f all

4.4. How it works#

In none mode, pofff skips model generation and simulation. It reads the existing benchmark-format CSV files and recreates spatial overlays, time-series comparisons, sparse-data comparisons, Wasserstein-distance products, and the error table.

4.5. Result#

One of the principal outputs is:

FluidFlower time-series comparison including the tutorial simulation — Published FluidFlower results with the new simulation labeled by its output directory name.#

The figure should include a series labeled YOURS. If it does not, verify that the output directory and benchmark CSV files are in the locations expected by the postprocessing workflow.

4.6. Next#

Continue with Interpret and improve the results.