Usgs earthquake analysis

Skill cxcscmu/SkillLearnBench/skills/b1-one-shot-claude-haiku-4-5/earthquake-plate-calculation/usgs-earthquake-analysis

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install

npx -y skills add cxcscmu/SkillLearnBench --skill usgs-earthquake-analysis

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Load, parse, and process USGS earthquake data in GeoJSON or JSON formats.

SKILL.md

3.4 KB, as published. Nobody here has run it

USGS Earthquake Data Analysis

Overview

USGS earthquake data is typically provided in GeoJSON format or as JSON with earthquake features. Understanding the data structure is essential for filtering, processing, and analysis.

Standard USGS Data Format

GeoJSON Structure

{
  "type": "FeatureCollection",
  "features": [
    {
      "type": "Feature",
      "id": "us1000abc1",
      "geometry": {
        "type": "Point",
        "coordinates": [longitude, latitude, depth]
      },
      "properties": {
        "mag": 4.5,
        "place": "12 km E of somewhere",
        "time": 1632000000000,
        "updated": 1632100000000,
        "url": "https://...",
        "detail": "https://...",
        "felt": null,
        "cdi": null,
        "mmi": null,
        "alert": null,
        "status": "reviewed",
        "tsunami": 0,
        "sig": 350,
        "net": "us",
        "code": "1000abc1",
        "ids": ",us1000abc1,",
        "sources": ",us,",
        "types": ",origin,phase-data,"
      }
    }
  ]
}

Key Fields

geometry.coordinates: [longitude, latitude, depth]
properties.mag: Magnitude
properties.place: Location description
properties.time: Unix timestamp in milliseconds
properties.id: Unique earthquake identifier

Loading and Processing

From GeoJSON

import json
import geopandas as gpd
from datetime import datetime

with open('/root/earthquakes_2024.json', 'r') as f:
    data = json.load(f)

# Convert to GeoDataFrame
gdf = gpd.GeoDataFrame.from_features(data['features'], crs='EPSG:4326')

# Convert timestamp (milliseconds to seconds, then to ISO format)
gdf['time'] = pd.to_datetime(gdf['time'], unit='ms').dt.strftime('%Y-%m-%dT%H:%M:%SZ')
gdf['magnitude'] = gdf['mag']

From Custom JSON Structure

import pandas as pd

# If data is a simple list of earthquakes
earthquakes_list = json.load(open('/root/earthquakes_2024.json'))
df = pd.DataFrame(earthquakes_list)

# Ensure required fields
df['longitude'] = df['lon']
df['latitude'] = df['lat']
df['magnitude'] = df['mag']

Data Validation

Common Issues

Null magnitudes: Some events may not have reliable magnitude estimates
Depth as third coordinate: USGS includes depth in coordinates [lon, lat, depth]
Timestamp format: Always in milliseconds since Unix epoch for USGS data

Validation Checks

# Check for required fields
required_fields = ['id', 'magnitude', 'latitude', 'longitude', 'place', 'time']
for field in required_fields:
    assert field in gdf.columns, f"Missing field: {field}"

# Verify coordinates are in valid range
assert gdf['longitude'].between(-180, 180).all()
assert gdf['latitude'].between(-90, 90).all()

# Check for null values in critical fields
assert not gdf[['id', 'magnitude', 'latitude', 'longitude']].isnull().any().any()

Common Operations

Filter by Region

# Earthquakes within lat/lon bounds
pacific = gdf[(gdf['latitude'] > -60) & (gdf['latitude'] < 70) &
              (gdf['longitude'] > 100) | (gdf['longitude'] < -80)]

Filter by Magnitude

significant = gdf[gdf['magnitude'] >= 4.0]

Convert Time to ISO Format

def unix_ms_to_iso(timestamp_ms):
    return pd.to_datetime(timestamp_ms, unit='ms').strftime('%Y-%m-%dT%H:%M:%SZ')

gdf['iso_time'] = gdf['time'].apply(unix_ms_to_iso)