How to actually track power outages around Seattle without losing your mind

When the wind comes down off the Cascades in November and takes out a chunk of the peninsula, you don't want to be refreshing a webpage every thirty seconds hoping for an update. The tools available now are better than what we had five years ago, but they still have real gaps that trip people up if you don't know where to look. I started paying close attention to this stuff back when PSE only posted outage maps that refreshed maybe once an hour, and even then the data was sometimes stale. Now with Seattle Outage Map Realtime Power Tracking For Puget Sound Residents you can get reasonably current info, but it still isn't perfect. You just have to understand what it's showing you and what it's not.

Seattle Outage Map Realtime Power Tracking For Puget Sound Residents

The core system pulls from SCADA feeds that PSE and City Light maintain. When a fault happens — tree on a line, transformer blow, anything that trips a breaker — the system logs the event and estimates the affected customers based on feeder topology. That estimated count is what shows up on most public-facing maps. The geographic data comes from GIS layer files that map where each pole and wire segment lives, so the colored zones on a map aren't random. They're tied to actual circuit routes. The refresh rate on the actual map interface varies by provider. PSE's outage map updates roughly every five to ten minutes during an active event. City Light's data tends to lag a bit more, sometimes fifteen to twenty minutes behind real time. If you're watching a storm roll in, that gap matters. I learned that the hard way during the ice storm in January 2023 when my neighborhood lost power at 4 PM and the map didn't reflect it until nearly 4:40. By then I'd already made three separate calls to the utility trying to confirm whether we were really out or just experiencing a dimming issue. Here's the workaround I ended up using: I set up a second screen with PSE's map open alongside a local weather radar app and a couple of neighbor status checks on Facebook or Nextdoor. When the map says zero outages but you can hear the transformer humming irregularly or your lights are flickering, trust your own eyes, not the map. The sensors in the field don't always report partial outages or voltage fluctuations as full outages. They report hard faults. A brownout or intermittent connection might not register at all.

Another thing most people miss is that the customer count on these maps is an estimate, not a headcount. The system matches reported outages to census-adjacent block groups and feeder zones. So if a transformer goes down on a street with thirty homes, the map might say forty-two affected customers because it's also counting vacant lots or commercial units on the same circuit. During the big 2021 wind event, one map showing six thousand outages in South Seattle turned out to be more like four thousand actual residential disruptions once PSE did their final reconciliation. The number looked worse than it was, but not by a huge margin. If you want raw data instead of the prettified map view, PSE has an API endpoint that some third-party tools feed from. It's not officially advertised, but you can find community-maintained scripts on GitHub that pull outage ticket data in JSON format. The tradeoff is that the API data is less polished — no pretty colored zones, just coordinates and timestamps. I use a modified version of one of those scripts myself. It runs locally on a cheap Raspberry Pi and alerts me via text when my zip code shows any new outage tickets. It takes about twenty minutes to set up if you know your way around Python and cron jobs. The alert has been useful exactly twice in two years, but those two times were during emergencies when every minute counted. The biggest limitation everyone ignores is coverage blind spots. Rural areas east of the Sound, parts of Kitsap, and the outer islands often have sparse sensor density. A transformer can blow in a neighborhood near enumeration or Lake Forest Park and the map will show a wide swatch of zeros for half an hour while engineers drive out to confirm it. Meanwhile downtown Seattle, where the monitoring is denser, might look perfectly fine while a whole block is in the dark because the fault is on a lateral service drop that isn't instrumented.

Get the Full Details

Puget Sound-area power outages persist after stormy night | FOX 13 Seattle
Puget Sound-area power outages persist after stormy night | FOX 13 Seattle

Also worth knowing: these maps don't show planned outages the way people expect. Maintenance work gets logged separately from emergency outages, but both appear on the same map interface. If you see a zone marked out during daytime hours on a Tuesday, it might be a scheduled maintenance window that PSE notified affected customers about weeks in advance. Check your email or the utility's notification system if you're unsure. The map itself won't label it as planned versus unplanned. I've also seen people get tripped up by maps that aggregate data across multiple utilities. The Puget Sound area isn't served by a single provider. You've got PSE, Seattle City Light, Eugene Water and Electric Board reaching into some outer areas, and a handful of co-ops. A map that claims to cover the whole region is usually stitching together data from different sources with different update cadences and different accuracy standards. Don't assume consistency across the whole view. If you're preparing for storm season, the practical move is to bookmark PSE's outage center directly and City Light's equivalent page rather than relying on a third-party aggregator. Download neither as an app unless you've verified the data source — some of the popular outage tracker apps on the stores are just pretty wrappers around the same public feeds with added ads. The native utility sites are usually more current and less cluttered anyway.

One more thing that isn't obvious: outage maps don't predict restoration time reliably. The ETA you see on a map is a rolling estimate based on how many crews are assigned and the order of tickets. It changes constantly and is often wrong by several hours, especially during widespread events. I stopped relying on restoration estimates entirely after a 2022 winter storm where my posted ETA went from four hours to twelve hours to "being evaluated" over the course of a single evening. The only timeline that mattered was the crew dispatch log, which wasn't public at all. For most residents, checking the official map once when power goes out, then again in about an hour if nothing has changed, is enough. Refreshing it obsessively won't bring the power back faster and it will just make you more anxious. The systems work well enough for awareness. They just don't work well enough for precision.