March 27, 2019

UW's Cherry Blossom Timing and Our Winter Weather

One of the singular events each year in Seattle is the stunning display of white/pink blossoms on the Arts Quad at the University of Washington.  At maximum bloom, the flowering is simply stunning.

Interestingly enough, there is quite a bit of variation of the date of peak bloom, the day when the crescendo of color reaches its apex.   A day that I have told follows the beginning of blossoming by about one week.

As shown below, since 2012 the peak bloom dates have varied from mid March through early April.  Talking to some cherry blossom experts today, it appears that this year's peak bloom should be on March 31, give or take a day (and I have plotted it). 

I have done some recreational reading about this important topic, including a very nice paper by Chung et al. that finds that Cherry blossom timing for the famous Washington DC blooms is closely related to winter and early spring temperatures.  Colder winters delay flowering.  And they used some of the UW blossoming data as well.

So can we explain the above variation in blossom date with local temperature variations?  And do the above variations suggest the influence of global warming?

Let me begin by plotting December through February temperatures for the Puget Sound lowlands from the NOAA Climate Division Data. 

The two warmest years were 2015 and 2016 and YES, they had the earliest peak bloom.  Very good.  2017 was the coldest year and its bloom was quite late (well into April).  But 2018 was late, but the temperatures were not exceptionally cold.  In short, there is clearly some kind of correlation, but it isn't perfect.  In fact, the Chung et al paper suggests this, revealing that to get flowering requires sufficient cold early in the season and a certain amount of heating late in the spring.  So the relationship is a bit more complex

One thing is sure, the short period shown above (2012-2019) does not suggest a progressive earlier flowering as would be expected with warming.   And certainly the winter temperatures plot does not suggest a warming trend....if anything temperatures have cooled this decade, with a warm spike in 2015.  But this is simply too short a period to have much confidence in any trends.

I am trying to acquire a longer term blossom date data base to secure a better estimate of a longer-term trend.  Will do another blog on this topic if I get it.  If anyone can help me get it, let me know.


  1. Cliff,what you're talking about is the science of phenology.Very fascinating.I've been charting the dates of first blooms of various flowers and trees for about 45 years now.(It's a little earlier now than in the '70's.)The science also applies to fauna too,not only just flora.

  2. I remember 2011, the year before you chart ends, as a very late blooming year due to the La Nina conditions of the winter. When I visited the Quad on April 2nd, the trees had started flowering but had not peaked yet. That summer also had the latest snowmelt of the past 10 years on the hiking trails of the Cascades (my personal opinion, at least).

  3. Bob Siegfried


  4. I’m sure it’s better than the DC Cherrybloom show, I took that in once
    half the country was there doing the same,stagnated bus routes
    If you have DC on your list of things to do leave your rental car at the airport, by all means

  5. I believe this paper: included ornamental flowering cherries, the phenology dataset goes back to 1959. The data are from Salem, Oregon, however, not from Seattle

  6. Cliff, I think the primary regional expert on this topic of phenology and its relationship to climate change is Connie Harrington at the USFS PNW Research Station in Olympia. She works with conifers primarily, but also other native flowering plants in regards to long-term winter chilling requirement and spring temperature requirements as they relate to budbreak timing.

    Brilliant scientist and nice person to boot. Short of working with you, she could certainly point you in the right direction.

  7. I think it will be very interesting, but there are so many variables. The weather might appear to be similar but the amount of sunshine could be different. Also, The methodology for pruning and fertilization are no doubt important. The changing amount of buildings over time surrounding the cherry blossoms could also have an affect. I would curious also to see how the standard deviation changes over time as well.

    Good luck on the project!

  8. Wouldn't the on campus student newspaper (assuming such a thing exists for UW), have news on the annual blooming that could be used to look back many decades?

  9. Maybe you could try your hand at the Nenana ice classic. Seems like a related skill set.

  10. When I get a chance I’ll look through a few of the photography sites for UW bloom pictures. Should be able to get the dates off of the pictures for peak bloom going back a fairly long way.

  11. Jon Kars, I suspect that most everything except rainfall has been very consistent for the last 40-50 years. The Quad has been just like it is today since before my parents attended the UW back in the early 50’s. A quick check says those buildings where completed in 1950.

  12. Did I miss the date or estimate for this year?


Please make sure your comments are civil. Name calling and personal attacks are not appropriate.

A Drier Than Normal Spring over the Pacific Northwest

I had to water some new plants today because the last few weeks have been relatively dry.   So I was dry has it been in our ...