# Earthquake Watch

A California earthquake publication that filters, scores, and explains notable events with geographic, fault, and population context.

Public Work / Live

A California earthquake publication that filters, scores, and explains notable events with geographic, fault, and population context.

Direct answer

## What did Devzilla build?

Devzilla built a publication pipeline over the United States Geological Survey event service, including a California polygon filter that rejects out-of-state events returned by the source bounding box.

[Open the live system](https://earthquakewatch.devzilla.co/) [Inspect the primary source ↗](https://earthquake.usgs.gov/fdsnws/event/1/USGS)

Status**Live**

Type**Devzilla-owned public work**

Stack**Python 3.12 / SQLite / Jinja2 / USGS FDSN API / Cloudflare Workers**

Constraint

## The challenge

The United States Geological Survey event API accepts rectangular geographic bounds. Any useful rectangle around California also includes parts of Nevada, Oregon, and Baja California, so a bounding-box query alone makes the publication scope false.

Architecture

## What was built

The source adapter first queries the official event service, then checks every result against a California polygon before scoring it. The editorial pipeline stores records in SQLite, builds static pages with Jinja2, and keeps one earthquake per post with magnitude, depth, nearby population, and fault context.

Engineering decisions

## The choices that shaped the system

-   Reject every event outside the California polygon before it can become an editorial candidate.
-   Use magnitude 3.0 as the primary scoring threshold, with additional weight for shallow events and events at magnitude 5.0 or above.
-   Spike retracted events instead of leaving stale reports live.
-   Keep the public site static even though collection, scoring, and drafting use a Python pipeline.

Current result

## What exists now

The deployed publication has a truthful California-only scope and a reproducible path from official source event to contextual article. The data source requires no private key and remains independently verifiable.

Related Agency work

## Capabilities behind this system

[

### AI Prototype-to-Production Engineering

Devzilla turns promising AI prototypes into evaluated, observable, cost-aware production systems that client teams can operate.

](https://devzilla.co/services/ai-prototype-to-production/)[

### Geospatial Systems

Devzilla builds spatial data pipelines, location intelligence, and mapping products for systems where geography is part of the decision.

](https://devzilla.co/services/geospatial-systems/)

## Bring Devzilla the difficult source.

We will start with what must be true, what can fail, and who needs to own the result.

[Start a project →](https://devzilla.co/contact/)

Canonical URL: https://devzilla.co/work/earthquake-watch/