but how do we relate say a rowan tree to a wild black cherry. does every region have its own ogham? and how do we distill that knowledge?

Also perhaps consider the ecology of your local tree population. Does a given native tree hold a position similar - e.g. as growing on riverbanks or as part of the understorey in woodlands - to any of the trees in the traditional collection? Then that tree could be said to be equivalent.

Your post reminds me of this curious observation, Morchú:
Looking at Fionn's Window from the Book of Ballymote, which arranges the Ogham characters in five concentric circles, the trees pertaining to each character fall into the following superorders (or subclasses). I have numbered the groups 1 - 4 starting at the top of the Window and working clockwise. The characters are taken from the outside in.

Group 1: Rosid Rosid Rosid Rosid Asterid
Group 2: Rosid Rosid Asterid Rosid Rosid
Group 3: Rosid Rosid Rosid Rosid Asterid
Group 4: Conifer Rosid Asterid Rosid Conifer

As they lie in the window, the pattern of group 1 is a mirror image of the opposite group, group 3. And except for the fact that the first and last in group 4 are conifers (which itself creates a symmetrical pattern within the group) the pattern of group 2 is a mirror image of the pattern of opposite group 4.

Group 2 also shows a symmetry of orders that reflects the pattern of group 4:
Rosales Fagales Aquifoliales Fagales Rosales

Of course, this could be coincidental, and these are modern taxa unknown to the scholar or scholars who wrote the Book of Ballymote, in which Fionn’s Window first appears, towards the end of the 14th century CE. But they do of course correspond to actual differences in the structure of the plants/trees. There’s no way of showing that the pattern isn’t coincidence, but it does suggest that, just possibly, the order of the trees is not arbitrary.

OBOD seems to have Australian Ogham, but the link isn't working now - I will google that later!