Check out the code for this project here!

As I age, I feel myself sliding into becoming a bird-watcher. As any budding birder would, I started looking into the data available through eBird. As a first test of what it was like to work with the data (accessed through gbif.org), I decided to map the areas in Oakland where there are relatively high observances of crows (which for this analysis, I defined as members of the genus Corvus).

A heatmap showing areas where there is the highest ratio of crow sightings to all bird sightings. Lighter colors show a higher ratio. Only the top 34% of hexagons by crow sighting ratio are visible.

I defined areas with lots of crows as areas where there was a high ratio of Corvus observances relative to the total observances in that area. I bucketed the raw data into hexagonal bins using Uber's H3 library, and calculated that ratio for each bucket. I took the top 34% of hexagons, and used those to create my maps.

Something I wanted to try out with this project was making maps with artistic, custom markers in Python. To try this out, I thought I would visualize areas with higher than average crow observations with flocks of crows flying over the areas.

The final crow flight map

Data source: GBIF.org (21 January 2023) GBIF Occurrence Download https://doi.org/10.15468/dl.u59qez