Data Sources Overview¶
The Outage Detection System aggregates data from multiple sources to provide comprehensive visibility into power and network incidents.
Source Categories¶
Power Outage Sources¶
| Source | Coverage | Update Frequency | Confidence |
|---|---|---|---|
| ODIN | US utilities | Real-time | 0.7 |
| Cal OES | California | Real-time | 0.75 |
| NWS | US (weather-related) | 15 minutes | 0.85 |
Network Incident Sources¶
| Source | Coverage | Update Frequency | Confidence |
|---|---|---|---|
| Cloudflare Radar | Global | Real-time | 0.9 |
| BGP/RIPE RIS | Global | Real-time | 0.5 |
| Kentik | Global | Real-time | 0.85 |
Confidence Scores¶
Each data source has a baseline confidence score that reflects data reliability:
- 0.8 - 1.0: High confidence (curated, verified sources)
- 0.6 - 0.8: Medium confidence (official government/utility data)
- 0.4 - 0.6: Lower confidence (raw, unverified feeds)
Source Identifiers¶
Use these identifiers when filtering by source:
radar - Cloudflare Radar network incidents
odin - ODIN utility power outages
bgpstream - BGP/RIPE RIS routing events
nws - National Weather Service alerts
kentik - Kentik network anomalies
caloes - California OES power outages
Data Flow¶
External APIs → Fetch & Transform → Normalize to Incident[] → API Response
↓
Error Handling
(return [] on failure)
Adding New Sources¶
To add a new data source:
- Create a fetch function in
apps/api/src/sources/ - Transform data to the
Incident[]format - Add to the orchestration in
apps/api/src/index.ts - Update the web app source filter options