Finding the Best Cherry Blossom Streets in Vancouver

Finding the Best Cherry Blossom Streets in Vancouver: Geospatial Analysis, Route Optimization, and Science Communication with R

Every spring, Vancouver's streets burst into pink, but which ones are *actually* worth visiting? And could you see the best of them all in a single route?

This talk walks through an end-to-end open data science project that answers
exactly those questions. Using freely available datasets from the
City of Vancouver Open Data Portal (public tree
inventories, street networks, and neighbourhood boundaries) we filter thousands of
trees down to Japanese flowering cherry varieties, snap each tree to its nearest
street segment using `{sf}`, and compute blossom density per block to rank the top
streets city-wide. We then solve a Traveling Salesman Problem with `{sfnetworks}`,
`{igraph}`, and Kyle Walker's `{spopt}` package to produce the **Vancouver Cherry
Blossom Marathon**: an optimized route through the 10 or 30 densest cherry blossom streets,
exported as GPX files for Garmin/Strava and as a one-click Google Maps link.

The second half of the talk covers how `{Quarto}` ties the analysis into a public-
facing website with interactive maps using `{mapgl}` (MapLibre GL), sortable tables
via `{DT}`, neighbourhood choropleths, and a blog post that made it into local news.
We'll also discuss the reproducibility scaffolding (a `Makefile` pipeline) and what
it means to do science communication with genuinely open tools: MIT-licensed code,
CC-licensed public data, and an R ecosystem that anyone can run from scratch.

**************************************************************
Note: this will be a hybrid event, taking place at the Mt Pleasant branch of the Vancouver Public Library as well as online via Zoom. If you are coming to the event in person, you are welcome to join us for some food and drinks at Main Street Brewing after we have finished.
**************************************************************