Being modern has not made it marginal. The tree calendar was taken up by modern Druid orders, by Wiccan and broader pagan practice, and by a popular tree astrology in which a birth date yields a tree and a character reading, and it circulates in books, greeting cards, jewellery and websites with the word Celtic attached. Large numbers of people have encountered it, and a fair number keep it.
Historians of Druidry have documented the invention in detail, both the medieval sources and what Graves did with them, and the finding is not disputed among specialists. It sits alongside a good deal of other modern material presented as ancient Celtic religion, an area where the gap between what survives and what circulates is unusually wide. The atlas keeps the entry and states both facts together, because a scheme with a large modern following is a real object of study whatever its pedigree, and because quietly omitting it would leave a reader who meets it elsewhere, as most people will, with nothing to check it against. Recording an invention as an invention is more useful than pretending it is not there. What the calendar is not is evidence for how any ancient Celtic speaking people organised the year. That question is poorly documented, rests on a small number of difficult sources, and is entirely separate from this one.