GeoNet Quality Score System - Analysis & Implementation Plan

Date: November 25, 2025 Paper: “A quantitative assessment of GeoNet earthquake location quality in Aotearoa New Zealand” DOI: 10.1080/00288306.2024.2421309 Status: Analysis Complete - Implementation Plan Ready

Executive Summary

The research paper proposes a Quality Score (QS) system ranging from QS0 (unconstrained) to QS6 (best constrained) for earthquake location quality assessment. This is a discrete categorical system based on specific location quality criteria, different from our current continuous 0-100 scoring system.

Key Finding: The two systems are complementary rather than conflicting. We should integrate the GeoNet QS system as an additional quality metric alongside our existing scoring.

Paper’s Quality Score System

QS Scale (QS0 - QS6)

Based on the abstract and search results, the system uses a 7-level categorical scale:

• QS6: Best constrained locations

• QS5-QS1: Progressively less constrained

• QS0: Unconstrained locations

Likely Criteria (Based on Standard Seismological Practice)

While the full paper is behind a paywall, standard earthquake location quality criteria typically include:

1. Azimuthal Gap - Angular distribution of stations around the event - Excellent: < 90° - Good: 90-180° - Poor: > 180° - Critical: > 270°

2. Station Count - Number of seismic stations used - Excellent: ≥ 20 stations - Good: 10-19 stations - Fair: 6-9 stations - Poor: < 6 stations

3. RMS Residual - Root mean square of travel time residuals - Excellent: < 0.3s - Good: 0.3-0.5s - Fair: 0.5-1.0s - Poor: > 1.0s

4. Horizontal Uncertainty - Location precision - Excellent: < 1 km - Good: 1-5 km - Fair: 5-10 km - Poor: > 10 km

5. Depth Uncertainty - Depth precision - Excellent: < 2 km - Good: 2-5 km - Fair: 5-10 km - Poor: > 10 km

6. Minimum Distance - Distance to nearest station - Excellent: < 50 km - Good: 50-100 km - Fair: 100-200 km - Poor: > 200 km

Current Implementation Analysis

Our Existing Quality Scoring System

File: lib/quality-scoring.ts

Approach: Continuous 0-100 scoring with weighted components

Components: 1. Location Quality (35% weight)

• Horizontal uncertainty

• Depth uncertainty

• Time uncertainty

2. Network Geometry (25% weight) - Azimuthal gap - Station count - Phase count

3. Solution Quality (15% weight) - Standard error (RMS)

4. Magnitude Quality (15% weight) - Magnitude uncertainty - Magnitude station count

5. Evaluation Status (10% weight) - Manual vs automatic - Review status

Grades: A+, A, B, C, D, F (based on score ranges)

Strengths of Current System

✅ Comprehensive - Covers all major quality aspects ✅ Granular - 0-100 scale provides fine-grained assessment ✅ Weighted - Prioritizes most important factors ✅ Detailed feedback - Provides strengths, weaknesses, recommendations ✅ Already implemented - Fully functional with tests

Limitations vs GeoNet QS

❌ Not standardized - Custom system, not comparable to published research ❌ Complex - Harder to communicate than simple QS0-QS6 ❌ No discrete categories - Continuous scores less intuitive than QS levels ❌ Missing criteria - Doesn’t include minimum distance to nearest station

Comparison: Our System vs GeoNet QS

Aspect	Our System	GeoNet QS
Scale	0-100 continuous	QS0-QS6 discrete
Grades	A+, A, B, C, D, F	QS6, QS5, …, QS0
Complexity	High (5 components, weighted)	Medium (6-7 criteria)
Standardization	Custom	Published research
Granularity	Very fine (101 levels)	Coarse (7 levels)
Communication	Technical	Simple & clear
Comparability	Internal only	Comparable to GeoNet catalogue
Implementation	Complete	Not implemented

Proposed Integration Strategy

Option 1: Dual System (RECOMMENDED)

Implement both systems side-by-side:

• Keep existing 0-100 system for detailed internal quality assessment

• Add GeoNet QS (QS0-QS6) for standardized comparison and communication

Benefits: - Best of both worlds - Maintains existing functionality - Adds standardization and comparability - Simple communication with QS levels - Detailed analysis with 0-100 scores

Implementation: - Add new calculateGeoNetQS() function - Store both scores in database - Display both in UI - Use QS for filtering, 0-100 for detailed analysis

Option 2: Replace with GeoNet QS

Replace our system entirely with GeoNet QS

Benefits: - Simpler system - Standardized - Easier to communicate

Drawbacks: - Loss of granularity - Loss of existing functionality - Breaking change for users

Recommendation: ❌ Not recommended - too disruptive

Option 3: Map Our Scores to QS Levels

Keep 0-100 system, map to QS levels

Benefits: - No new calculations needed - Adds QS compatibility

Drawbacks: - Mapping may not align with true QS criteria - Not truly implementing the research methodology

Recommendation: ⚠️ Acceptable but not ideal

Implementation Plan (Option 1 - Dual System)

Phase 1: Research & Specification

Tasks: 1. ✅ Analyze paper (COMPLETE) 2. ⏳ Obtain full paper text to confirm exact QS criteria 3. ⏳ Define precise thresholds for QS0-QS6 4. ⏳ Document GeoNet QS algorithm

Estimated Time: 2-3 days (pending paper access)

Phase 2: Core Implementation

Files to Create: - lib/geonet-quality-score.ts - GeoNet QS calculation logic

Files to Modify: - lib/db.ts - Add geonet_qs column to events table - lib/types/earthquake.ts - Add GeoNet QS types - lib/quality-scoring.ts - Integrate with existing system

Database Migration: .. code-block:: sql

ALTER TABLE earthquake_events ADD COLUMN geonet_qs INTEGER CHECK(geonet_qs >= 0 AND geonet_qs <= 6); ALTER TABLE earthquake_events ADD COLUMN geonet_qs_details TEXT; – JSON with criteria breakdown CREATE INDEX idx_events_geonet_qs ON earthquake_events(geonet_qs);

Estimated Time: 3-4 days

Phase 3: UI Integration

Components to Update: - Event detail pages - Show both scores - Event lists - Add QS filter - Quality reports - Include QS distribution - Import validation - Calculate QS on import - Dashboard - QS statistics

Estimated Time: 2-3 days

Phase 4: Testing & Documentation

Testing: - Unit tests for QS calculation - Integration tests with real GeoNet data - Validation against published QS values (if available)

Documentation: - User guide for QS interpretation - API documentation - Developer guide for QS algorithm

Estimated Time: 2 days

Total Estimated Time: 9-12 days

Next Steps

Immediate Actions

1. Obtain Full Paper - Need complete methodology to implement correctly - Try institutional access - Contact authors directly - Check preprint servers (ResearchGate, arXiv)

2. Validate Criteria - Confirm exact thresholds for each QS level - May need to contact GeoNet directly - Review GeoNet documentation

3. Get Stakeholder Approval - Confirm dual system approach - Present this analysis to team - Get buy-in for implementation effort

Questions to Resolve

1. Exact QS Criteria - What are the precise thresholds?

2. Minimum Distance - Should we add this metric to our data model?

3. Backward Compatibility - How to handle existing events without QS?

4. Performance - Impact of calculating two quality scores?

5. UI/UX - How to display both scores without confusion?

Conclusion

The GeoNet Quality Score system is a valuable addition to our platform that would:

✅ Standardize our quality assessment with published research ✅ Improve communication with simpler QS0-QS6 scale ✅ Enable comparison with GeoNet catalogue ✅ Complement our existing detailed scoring system

Recommendation: Implement Option 1 (Dual System) to gain benefits of both approaches.

Next Step: Obtain full paper text to confirm exact QS criteria and thresholds.

Document Status: Analysis Complete - Awaiting Paper Access for Implementation