Hi Magnus
Chapter 9.9 and Appendix 9.3 in the documentation describes SOM initialization in confusing details.
The numbers in dk--soil.dia are likely estimated with the "Carbon Balance" log.
However, I recommend ignoring the "OrganicMatter" parameter (leave it out), and instead combine 1) a short warmup period (one crop rotation, 2 years at minimum), and 2) an explicit partitioning of the SOM fractions of the Ap horizon with the "SOM_fractions" parameter.
(SOM_fractions 0.3 0.7) for a soil where the organic matter is fresh with a quick turnover and large mineralization, and (SOM_fractions 07.03) for a soil with older organic matter and less N release, or someone in between. Use your plot with the lowest N input for calibration.
Kindly,
Per