UW News

April 1, 2019

UW students spearhead efforts to predict peak bloom for cherry trees

UW News


Each spring, thousands of visitors flock to the University of Washington campus to see the iconic cherry trees in the Quad. Class discussions, casual Frisbee tosses, lunchtime picnics and even wedding portraits all take place under the beautiful blossoms, which typically hit their prime in late March or early April. This year, they reached peak bloom on March 29.

But there’s no easy way to predict when peak bloom will occur each year for the trees in the Quad. Every spring, UW Facilities staff diligently check on the blossoms as they progress, but it’s hard to say down to the day, or even week, when the blossoms will be at their best. In contrast, peak bloom periods for Washington, D.C.’s impressive cherry trees can be predicted with reasonable accuracy, thanks to years of meticulous data collection and models linking the historical bloom data with temperature, such as the one developed by UW professor Soo-Hyung Kim.

two data collectors by a tree

Michael Bradshaw, left, works with data collector TJ VanderYacht early in the 2019 bloom season.Kiyomi Taguchi/University of Washington

Now, a team of UW students hopes to make it possible to accurately predict peak bloom timing for the Quad cherry trees. Led by Michael Bradshaw, a doctoral student in the School of Environmental and Forest Sciences, 10 undergraduate students are documenting bloom information on all 118 cherry trees across campus, including the 29 Quad trees.

“We started this project to collect data on trees to predict more accurately when they will bloom, and also to see how climate change is affecting bloom times,” said Bradshaw, who began collecting data on a handful of trees by himself last season. This year, he enlisted the help of other students, who each are responsible for a block of trees around campus.

Collectively, the students visit every cherry tree on campus at least twice a week on average, sometimes swinging by after class or, in the case of one data gatherer, taking a break on his runs. On each visit, the students take two photos — one that shows the full tree, and another close-up shot of a cluster of buds. They also note which of the five bloom stages the tree appears to be in, and how much of that stage is complete. Tree names and dates are filled in automatically in a smartphone app they use to collect the data.

collecting data

TJ VanderYacht collects data from one of the cherry trees in the UW Quad.Kiyomi Taguchi/University of Washington

Favorable temperatures and day length are the two most important factors that determine when plants will reach full bloom, including cherry trees. Cherry tree species bloom at different times, and blossoms usually stay on the trees for two or three weeks if wind is low and temperatures are cool.

For freshman TJ VanderYacht, pausing his run two or three times a week to visit the cherry trees near Husky Stadium is a way to dive into fieldwork early in his time at the UW. VanderYacht, who grew up in Washington, said the cherry trees are famously tied to the UW, a quintessential part of campus.

“I thought it’d be a great way to get involved with my school and the day-to-day research going on here,” he said. “It’s a good opportunity to be part of something that characterizes the UW.”

UW senior Rachel Liu heard about the cherry blossom project and wanted to get involved to practice computer-aided mapping skills, which she is learning simultaneously in a class she is taking this quarter. Liu, who is in charge of the trees along Rainier Vista and half of the Quad, said she stops by after class, usually spending up to an hour three times a week noting details about the trees.

“I thought, wow, this is such a great opportunity to practice these skills,” said Liu, who is majoring in biology. “I’m really interested in plant biology, and I’m hoping to do research after I graduate related to plants or maybe fungi. It’s great to get the experience of collecting data.”

Data collection for the UW trees, while just beginning, is already more robust than the efforts for D.C.’s cherry blossoms, said Bradshaw, who compared notes with the National Park Service that tracks the capitol’s blooms. It might take several years of consistent data gathering for the UW trees, but Bradshaw hopes that patterns will begin to emerge that will help anticipate peak bloom timing. For example, it might become clear that when trees reach 50 percent of a certain stage, peak bloom will follow in a month or less, he explained.

Kim, a UW associate professor of environmental and forest sciences, said he is eager to see the early results of the bloom timing data. The model he developed to help predict the D.C. peak bloom period needs rigorous testing before it can be applied for the trees on the UW campus, mainly because of the differences in regional weather patterns as well as tree genetic backgrounds.

The cherry bloom data collected at UW could prompt additional research and help improve the ability to provide more accurate bloom predictions through advanced observations as well as models, Kim added.