@hol353 - as part of my work on the error bars in ApsimX, I want to include observed error data in the p/o table, even if there is no predicted error column to match it. This means that not every observed column will necessarily have a matching predicted column, hence this change.
…n don't exist.
@hol353 - as part of my work on the error bars in ApsimX, I want to include observed error data in the p/o table, even if there is no predicted error column to match it. This means that not every observed column will necessarily have a matching predicted column, hence this change.
Edit: working on APSIMInitiative/ApsimX#5341