スキル一覧に戻る
DrSatsuma1

verify-dark

by DrSatsuma1

Speleothem-based paleoseismic research - detecting dark earthquakes through cave geochemistry

0🍴 0📅 2026年1月11日
GitHubで見るManusで実行

SKILL.md


name: verify-dark description: Verify dark earthquake claims against fault databases. Use when checking if a candidate earthquake is truly "dark" (unmapped fault) or just pre-historical (known fault). Triggers on "verify dark", "check fault database", "is this dark", "audit earthquake claim".

/verify-dark - Fault Database Verification Skill

Purpose

Verify if a candidate "dark earthquake" claim is truly dark (unmapped source fault) or pre-historical (known fault, no written record). This skill automates the DARK_EARTHQUAKE_AUDIT.md workflow.

Usage

/verify-dark <lat> <lon> <date_CE> [region]

Examples:

/verify-dark 44.2 8.1 1394 italy
/verify-dark 32.7 -117.2 1741 california
/verify-dark -16.5 -44.8 96 brazil
/verify-dark 22.4 -84.0 1400 caribbean

Workflow

Step 1: Identify Region & Databases

Based on coordinates, determine which fault databases to check:

RegionPrimary DatabaseSecondary Databases
ItalyDISS v3.3.1 (INGV)ITHACA, EFSM20, GEM SHARE
CaliforniaSCEC CFM v7.0CGS FER, USGS Quaternary
Caribbean/C. AmericaGEM CCAF-DBUSGS
BrazilGEM SARA(Note: Brazil EXCLUDED from SARA)
Middle EastEMMEGEM, Hessami (2003)
RomaniaRODASEFESHM20, SHARE
TurkeyAFADGEM, Emre et al.

Step 2: Search Fault Databases

For each database, search within 50km and 100km radius:

  1. Use calc_distance MCP tool to compute distances from candidate to known faults
  2. Query database (WFS endpoint if available, or web search for geojson)
  3. Record all faults found with strike, slip type, distance

Example search pattern:

# Check DISS v3.3.1 for Italy
WebFetch: https://diss.ingv.it/diss330/download/DISS330_ISS.geojson
Filter: features within 100km of (lat, lon)

# Check SCEC CFM for California
WebFetch: https://www.scec.org/research/cfm
Search for fault traces near coordinates

Step 3: Check Recent Literature (2010+)

Search for recent fault mapping publications:

  1. WebSearch: "[region] fault mapping [year range]"
  2. WebSearch: "paleoseismic trench [fault name]"
  3. WebSearch: "[coordinates] seismogenic source"

Look for:

  • New fault discoveries
  • Offshore/submarine fault extensions
  • DEM-based lineament studies
  • Paleoseismic trenching results

Step 4: Run DEM Lineament Check (if available)

If DEM data exists for the region:

  • Check dem_tiles/ for existing analysis
  • Reference DEM_LINEAMENT_FINDINGS.md or similar
  • Note any unmapped structures identified

Step 5: Generate Verification Table

Create markdown table summarizing findings:

| Database | Checked | Faults within 50km | Faults within 100km | Notes |
|----------|---------|-------------------|---------------------|-------|
| DISS v3.3.1 | ✅ | [fault names] | [fault names] | [details] |
| ITHACA | ✅ | ... | ... | ... |
| EFSM20 | ✅ | ... | ... | ... |
| GEM | ✅ | ... | ... | ... |
| Recent lit | ✅ | ... | ... | [citations] |

Step 6: Classify Event

Based on findings, classify as:

ClassificationCriteria
TRUE DARKNo mapped fault in ANY database (like Italy 1394, Brazil events)
PRE-HISTORICALKnown fault exists, but earthquake predates written records (like California 1741)
PRE-COLUMBIANEvent in Americas before 1492 (no records possible)
PRE-INSTRUMENTALEvent before seismometer network (varies by region)
VALIDATIONKnown earthquake used to test methodology (like Cuba 1766)
DATABASE ARTIFACT"Missing" due to incomplete database, fault IS mapped elsewhere

Step 7: Update DARK_EARTHQUAKE_AUDIT.md

Add new section to paleoseismic_caves/DARK_EARTHQUAKE_AUDIT.md:

### ✅ [Region]: [Date] - **[CLASSIFICATION]**

**Date verified**: [today's date]
**Classification**: **[CLASSIFICATION]** ([explanation])

**Databases checked**:
- ✅ **[Database 1]** - [findings]
- ✅ **[Database 2]** - [findings]
...

**Key findings**:
[Summary of what was found]

**Conclusion**: [Why this classification]

**Likelihood of database artifact**: **[HIGH/LOW/ZERO]** - [explanation]

Classification Decision Tree

Is there a mapped fault within 100km in ANY database?
├─ YES → Is earthquake in historical catalogs?
│        ├─ YES → VALIDATION EVENT
│        └─ NO → PRE-HISTORICAL (known fault, no record)
│                 └─ If Americas pre-1492 → PRE-COLUMBIAN
│                 └─ If pre-seismometer → PRE-INSTRUMENTAL
└─ NO → Does a fault database exist for this region?
         ├─ NO (e.g., Brazil) → TRUE DARK (no database = genuine gap)
         └─ YES → Did DEM analysis find unmapped structure?
                  ├─ YES → TRUE DARK (candidate)
                  └─ NO → UNATTRIBUTED (needs more research)

Output Format

The skill will output:

  1. Summary box at top with classification
  2. Verification table with all databases checked
  3. Key findings with quotes/evidence
  4. Classification rationale explaining decision
  5. Update to DARK_EARTHQUAKE_AUDIT.md (with approval)

Database URLs Reference

Italy:

California:

Caribbean:

Middle East:

Global:

Important Notes

  1. USGS Quaternary Fault Database has 9-27 year lag - never use as sole source
  2. Brazil is excluded from GEM SARA - stable continental interiors have no fault databases
  3. "Dark" means UNMAPPED FAULT, not just "no written record"
  4. Always check 3+ databases before classifying as TRUE DARK
  5. DEM lineament analysis strengthens case for unmapped structures

スコア

総合スコア

40/100

リポジトリの品質指標に基づく評価

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

レビュー

💬

レビュー機能は近日公開予定です