Google services appear throughout the Explorer architecture. The deployment docs describe a serverless Google Cloud Platform setup, with Cloud Run services for the frontend, backend, and AI server, plus Artifact Registry for Docker images and Workload Identity Federation for GitHub-based deployments.
Explorer touchpoints
| Area | Google service |
|---|---|
| Application hosting | Google Cloud Run |
| Container storage | Google Artifact Registry |
| Deployment identity | Workload Identity Federation between GitHub and Google Cloud |
| Hazard overlays | Google Earth Engine metadata and tile URLs |
| Dynamic translation | Google Cloud Translate v3 |
| Maps | Google Maps API key injected into the frontend build |
| Traffic reporting | Google Analytics tag observed on the live Explorer site |
Operational notes
- The backend uses Google Earth Engine for hazard layer metadata and map tiles.
- Dynamic data translations use Google Cloud Translate v3, including acronym-preservation logic before translation.
- The deployment workflow expects Google Cloud project and service-account configuration to be present in repository secrets.
- Google Analytics access depends on identifying the owning Analytics property and admin, not only seeing the measurement ID in the deployed page.
- 2026-06-30 analytics access blocker: the Adapt Action Explorer GA4 and Looker links still require the owning Google Analytics administrator or report owner to grant access and verify the data sources.