[OC] Hurricane Helene, Animated Map of the Southeast US by Percentage of County Reporting Power Outages
October 3, 2024
[OC] Hurricane Helene, Animated Map of the Southeast US by Percentage of County Reporting Power Outages
Posted by Convillious
8 comments
This was made in python using a variety of packages like Selenium, GeoPandas, and Shapely. It took me many many hours to make this. I saw a nice data source on USAToday’s site (Source 2) so I built a scraper in Selenium that assembled a CSV. I then used Pandas to comb through the CSV and connect the county names to a shapefile of the counties that i found in Source 1. This connection was facilitated by converting the county names to FIPS ID’s which the shapefile used, Source 4. And it was displayed in matplotlib. I now have an automated data collection and data displaying system.
Can anyone explain why there is a horizontal line near the top, i assume a state boundary, above which power outages get worse
Wow look how fast the landfall communities get back up and running. I would’ve thought they’d be the last
Was Duval county (NE FL) less than 20% ?
These aren’t *people* without power, these are *customers* (or *accounts*) without power. My household is 1 customer account that serves 5 people. I don’t know what the people per customer average is (many households are 1 customers but >1 people but a business may be 1 customer but not any actual people). Regardless, 2 million CUSTOMERS without power is surely represents far more than 2 million PEOPLE without power.
thank you for proper use of animation and county resolution data.
As someone who was without electricity and water from 9/27 thru 10/2, thank you for posting this. Its been… tough.
Finally, data actually presented in a beautiful way. Something kind rare in this sub…
8 comments
This was made in python using a variety of packages like Selenium, GeoPandas, and Shapely. It took me many many hours to make this. I saw a nice data source on USAToday’s site (Source 2) so I built a scraper in Selenium that assembled a CSV. I then used Pandas to comb through the CSV and connect the county names to a shapefile of the counties that i found in Source 1. This connection was facilitated by converting the county names to FIPS ID’s which the shapefile used, Source 4. And it was displayed in matplotlib. I now have an automated data collection and data displaying system.
Source:
1.) [https://simplemaps.com/data/us-counties](https://simplemaps.com/data/us-counties)
2.) [https://data.usatoday.com/national-power-outage-map-tracker/](https://data.usatoday.com/national-power-outage-map-tracker/)
3.) [https://github.com/hadley/data-counties/blob/master/county-fips.csv](https://github.com/hadley/data-counties/blob/master/county-fips.csv)
4.) [https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.html](https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.html)
Can anyone explain why there is a horizontal line near the top, i assume a state boundary, above which power outages get worse
Wow look how fast the landfall communities get back up and running. I would’ve thought they’d be the last
Was Duval county (NE FL) less than 20% ?
These aren’t *people* without power, these are *customers* (or *accounts*) without power. My household is 1 customer account that serves 5 people. I don’t know what the people per customer average is (many households are 1 customers but >1 people but a business may be 1 customer but not any actual people). Regardless, 2 million CUSTOMERS without power is surely represents far more than 2 million PEOPLE without power.
thank you for proper use of animation and county resolution data.
As someone who was without electricity and water from 9/27 thru 10/2, thank you for posting this. Its been… tough.
Finally, data actually presented in a beautiful way. Something kind rare in this sub…