This post is by Lizzie.
Greetings from sunny Vancouver! I saw a magnolia in bloom today, which feels super early to me. But less early in that cherry and other plum trees have been blooming around the city. We have not had the coldest winter and had a recent warm-ish spell with lots of sun … so maybe that’s the trigger? The usual ‘two bucket’ model for spring flowers is that plants need to fill a bucket of winter cool (cool, but not cold, what this bucket is doing is covering some sort of latent variable related to ‘chilling,’ a possibly mystical biological concept*) and then a bucket of spring warmth (this makes more sense — warmth leads to cell growth and division). A little sun would likely help that second bucket.
Figuring this out of course is the goal of the International Cherry Blossom Prediction competition, which Jonathan Auerbach, David Kepplinger and I run.
Predictions just came in** and they tell me to hold my breath a while on Vancouver but that blossoms are coming this month probably. And I should wait a month for the east coast too — I hear it has been a cold winter there. We’re not far off the Washington Post’s predictions (which just came out) and of course we’re hoping to beat them.
We’re not sure how we compare to the National Park Service’s forecasts. They have had one each of the four years before when we ran the competition but seem quiet this year. I could imagine what may have happened but if anyone actually knows, let me know in the comments.
BTW, the bloom video is of Hibiscus, not cherries (if you want to visualize how far apart they are on the tree of life, go to https://www.onezoom.org/ and then type in ‘Hibiscus’ in the search, pick the first one, then type ‘Prunus’ and pick your favorite from the list, it will zoom you over). And big thanks to Tim Savas for the amazing video.
* I am hopeful we’re getting closer on ‘chilling’ — the best idea so far is that it’s capturing some enzyme or such that breaks down the callose (which I call a sugar, but someone told me is better defined as a polysaccharide) that blocks the plasmodesmata (the little wholes in cells that let them communicate with each other). A new study has new info and suggests this whole thing could connect to bet-hedging in plants … I am dubious about that last part.
**If you have comments on the figures, see also here.
Just picking up on the magnolia v Prunus observation: in the UK there are very few magnolia species planted in gardens (there are lots and lots of specimens, but these are just spread over a small number of species, and some hybrids). Prunus on the other hand, is represented by a much larger number of species and cultivars, which are well known to have reliably different flowering dates (i.e. the sequence of flowering is pretty reliable, even if the first date of any taxon is hard to predict accurately). I think this always makes magnolia blooming seem sudden, as opposed to the “smear” across Prunus taxa (which the average person won’t distinguish).
This leads me to a question about the competition: are folk then always just predicting the first date of the earliest flowering species/variety in an area, or does the competition have an element of species-/variety-level predictive estimation as well that is rewarded with prizes?
Most of the well studies species for phenological modeling are early spring flowering plants, but I would not say that they’re the earliest ones by definition. The competition does have all the species information presented such that someone could include the relatedness or somehow factor this into their models (bit of a mess for the hybrids, but perhaps that makes it a more interesting problem) — though I am not sure anyone has … yet. There has always been conversation about making this an explicit part of the competition in future years.
Sorry for my slow reply; I blame teaching, which slows me down on (it feels like) everything but teaching.
Thanks Lizzie. More specifically I was thinking that, if the competition is about predicting first date flowering in specific locations in Japan or wherever, then if say Prunus “Kanzan” is the earliest flowering cultivar in each area, then it’s just about predicting that taxon, and information on other taxa may or may not be relevant (depending on the shared real world information between taxa and the model used). Probably I should look more at the competition website!
Cool video!