Hello BRIGHT Run Family,
Hope you are enjoying summer after the nice spring.
Every year, I eagerly await the short-term cherry blossom season in spring. This year was no exception. A marvellous mix of fresh colours invigorates me.
Having a good idea about the precise time of peak-bloom is quite important for nature lovers.
However, it is not easy to predict that time with high precision (almost impossible) before 10 days. Particularly with the ongoing climate change happening globally, it is a tougher task.
This year, a global competition on cherry blossom prediction took place (https://competition.statistics.gmu.edu/).
It was open to students, researchers, and citizen scientists. The competition was organized in a joint effort by scientists from George Mason University, USA and The University of British Columbia.
As you know, relevant data is crucial for making predictions. The organizers provided publicly available data on bloom dates in Washington, D.C. (USA), Kyoto (Japan), and Liestal-Weideli (Switzerland).
The participants could combine other publicly available data with the organizer-provided data and make yearly predictions from 2022 to 2031 for these three locations in addition to a new location, Vancouver, BC. In addition to meteorological data, some participants used heterogeneous data such as population and concentration of greenhouse gases to make predictions (https://news.ubc.ca/2022/03/28/citizen-scientists-predict-cherry-blossom-peak/).
It is fun to find meaningful patterns within any data. It is even more fun to be able to apply those patterns for benefits in the future. What I find even more intriguing is exploring or finding several sources of data each having its own signature pattern to indicate an event.
A recent study1 predicted best times to view cherry blossoms in different prefectures in Japan using ‘geotagged’ (having location information or latitude and longitude) tweets from travellers. Who knew our travel-related social posts have power to predict natural phenomena!
Maybe together we can make good use of technology to get closer to nature and preserve its patterns.
Stay safe and happy.
Ashirbani
Dr. Saha is the first holder of the BRIGHT Run Breast Cancer Learning Health System Chair, a permanent research position established by the BRIGHT Run in partnership with McMaster University.
References:
- Horikawa, Tomonari, et al. “Estimating the Best Time to View Cherry Blossoms Using Time-Series Forecasting Method.” Machine Learning and Knowledge Extraction 4.2 (2022): 418-431.
