> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/koala73/worldmonitor/llms.txt
> Use this file to discover all available pages before exploring further.

# Signal Intelligence

> Understanding threat signals, convergence detection, and automated alerts

World Monitor aggregates **real-time signals** from multiple data sources to build a unified geospatial intelligence picture. Signals represent events that may indicate instability, military activity, or infrastructure disruption.

## Signal Types

World Monitor tracks 10+ signal types across geopolitical, military, and infrastructure domains:

<CardGroup cols={2}>
  <Card title="Protests" icon="users">
    **Data sources**: ACLED + GDELT

    Social unrest events with fatality counts, event types (demonstration, riot, violence), and Haversine-based deduplication.

    **Severity factors**:

    * Fatalities
    * Violence level
    * Internet outages during protests
  </Card>

  <Card title="Conflicts" icon="gun">
    **Data sources**: UCDP + ACLED

    Active conflict zones with battle death counts and escalation tracking.

    **Severity factors**:

    * Battle deaths (1,000+ = active war)
    * Civilian casualties
    * Multi-actor involvement
  </Card>

  <Card title="Military Flights" icon="plane">
    **Data source**: ADS-B transponders

    Detected aircraft enriched via Wingbits API with registration, manufacturer, model, and operator.

    **Confidence levels**:

    * **Confirmed**: Operator matches military branch
    * **Likely**: Aircraft type is military-exclusive (tanker, AWACS)
    * **Possible**: Gov't-registered in military area
  </Card>

  <Card title="Naval Vessels" icon="ship">
    **Data sources**: AIS tracking + USNI fleet reports

    Maritime activity with vessel type classification, strike group assignments, and chokepoint monitoring.

    **Notable features**:

    * Dark ship detection (AIS gaps)
    * Carrier strike group tracking
    * 8 strategic chokepoint monitors
  </Card>

  <Card title="Internet Outages" icon="wifi">
    **Data source**: Cloudflare Radar

    AS-level outage detection with severity scoring.

    **CII impact**: Outages during protests increase unrest score by 20%
  </Card>

  <Card title="Earthquakes" icon="earth-americas">
    **Data source**: USGS

    Magnitude 4.5+ earthquakes with population exposure estimation (WorldPop density data).

    **Severity factors**:

    * Magnitude
    * Depth
    * Population within 100km
  </Card>

  <Card title="GPS/GNSS Jamming" icon="satellite-dish">
    **Data source**: gpsjam.org (ADS-B analysis)

    H3 hex cells where aircraft report GPS anomalies.

    **Classification**:

    * **Medium** (amber): >2% interference
    * **High** (red): >10% interference

    **CII impact**: Up to +35 security score
  </Card>

  <Card title="Satellite Fires" icon="fire">
    **Data source**: NASA FIRMS (VIIRS)

    Thermal hotspots indicating wildfires, industrial activity, or conflict-related burns.

    **Filtering**: Only high-confidence detections (>70%) are shown
  </Card>

  <Card title="Displacement Flows" icon="arrows-turn-right">
    **Data source**: UNHCR + HAPI

    Refugee and IDP movements with origin/destination tracking.

    **Dual perspective**:

    * **Origin countries**: CII boost for instability
    * **Host countries**: Infrastructure stress indicator
  </Card>

  <Card title="Climate Anomalies" icon="temperature-high">
    **Data source**: ERA5 reanalysis

    Temperature/precipitation deviations vs 30-day baselines for 15 conflict-prone zones.

    **Severity classification**: Feeds into CII scoring
  </Card>
</CardGroup>

## Geographic Convergence Detection

When **3+ distinct signal types** occur in the same geographic area within a 24-hour window, World Monitor fires a **convergence alert**.

### How It Works

<Steps>
  <Step title="Spatial Binning">
    Events are binned into **1°×1° geographic cells** (approximately 111km × 111km at the equator).
  </Step>

  <Step title="Type Diversity Scoring">
    Each unique signal type in a cell contributes **25 points**:

    * 3 types = 75 points
    * 4 types = 100 points
    * 5+ types = 100 points (capped)
  </Step>

  <Step title="Event Count Bonus">
    Additional **2 points per event** beyond the first of each type:

    * 1 protest + 1 military flight + 1 earthquake = 75 base
    * 3 protests + 2 flights + 1 earthquake = 75 + (2×2 protests) + (1×2 flights) = **85 total**
  </Step>

  <Step title="Reverse Geocoding">
    The cell is reverse-geocoded to a human-readable name using:

    * Conflict zone database
    * Strategic waterway names
    * Hotspot locations
    * Country boundaries
  </Step>

  <Step title="Alert Emission">
    A convergence alert is displayed in the **Intelligence Findings** badge with:

    * Location name
    * Severity score (color-coded)
    * List of contributing signal types
    * Time window (24h)
  </Step>
</Steps>

### Example: Eastern Mediterranean Convergence

```
🟠 CONVERGENCE ALERT
Location: Eastern Mediterranean
Severity: 92/100

Signals detected in 1°×1° cell (35°N, 34°E):
- 4 protests (Cyprus, Lebanon border)
- 3 military flights (Israeli Air Force)
- 2 naval vessels (US Carrier Strike Group 2)
- 1 satellite fire (industrial complex)

Time window: Last 24 hours
```

<Warning>
  Convergence alerts do not imply causation — they simply indicate that multiple event types are co-located. Further investigation is required to determine relationships.
</Warning>

## Regional Convergence Scoring

Convergence detection also operates at the **country and region level**:

* **Per-country clustering** — All signals within a country's borders are aggregated
* **Per-region clustering** — Signals are grouped into 12 operational theaters (Iran-Iraq, Ukraine-Russia, Levant, Baltic, etc.)

**CII Impact**: Countries with active convergence alerts receive a boost to their instability score based on severity.

## Temporal Baseline Anomaly Detection

World Monitor uses **Welford's online algorithm** to detect statistical anomalies in signal frequency:

<Steps>
  <Step title="Streaming Mean/Variance Calculation">
    For each signal type, World Monitor maintains a rolling 90-day window of event counts, computing streaming mean and variance without storing all historical data.
  </Step>

  <Step title="Contextual Grouping">
    Baselines are computed per:

    * **Event type** (protests, flights, conflicts)
    * **Region** (Middle East, Europe, Asia)
    * **Weekday** (Monday vs Sunday patterns differ)
    * **Month** (seasonal variation)
  </Step>

  <Step title="Z-Score Thresholds">
    Current event counts are compared to the baseline:

    * **Z > 1.5** → Low anomaly (yellow)
    * **Z > 2.0** → Medium anomaly (orange)
    * **Z > 3.0** → High anomaly (red)
  </Step>

  <Step title="Example Anomaly">
    ```
    ⚠️ ANOMALY DETECTED
    Signal: Military flights
    Region: Baltic Sea
    Current: 24 flights (Thursday, January)
    Baseline: 7.2 ± 2.1 flights
    Z-score: 3.2 (HIGH)

    Interpretation: Military flights are 3.2x normal for Thursday in January
    ```
  </Step>
</Steps>

<Info>
  **Minimum samples**: 10 historical data points are required before anomalies are reported, preventing false positives during the learning phase.
</Info>

## Strategic Theater Posture Assessment

Nine operational theaters are continuously assessed for **military posture escalation**:

| Theater                   | Key Triggers                                |
| ------------------------- | ------------------------------------------- |
| **Iran / Persian Gulf**   | Carrier groups, tanker activity, AWACS      |
| **Taiwan Strait**         | PLAAF sorties, USN carrier presence         |
| **Baltic / Kaliningrad**  | Russian Western Military District flights   |
| **Korean Peninsula**      | B-52/B-1 deployments, DPRK missile activity |
| **Eastern Mediterranean** | Multi-national naval exercises              |
| **Horn of Africa**        | Anti-piracy patrols, drone activity         |
| **South China Sea**       | Freedom of navigation operations            |
| **Arctic**                | Long-range aviation patrols                 |
| **Black Sea**             | ISR flights, naval movements                |

### Posture Levels

Each theater is assigned a posture level based on a composite of signals:

<Tabs>
  <Tab title="NORMAL">
    **Green** — Routine peacetime operations

    * Low aircraft count
    * No strike packaging (tankers + AWACS + fighters)
    * Minimal naval presence
    * Low CII scores for adjacent countries
  </Tab>

  <Tab title="ELEVATED">
    **Yellow** — Increased activity

    * Moderate aircraft count
    * Partial strike packaging detected
    * Carrier groups in transit
    * Rising CII scores
  </Tab>

  <Tab title="CRITICAL">
    **Red** — High alert

    * High aircraft count
    * Full strike packaging (tankers + AWACS + fighters)
    * Carrier strike groups on station
    * CII scores ≥ 60 for adjacent countries
  </Tab>
</Tabs>

## Focal Point Detection

When the same entity (country, location, organization, person) surfaces across **multiple independent data sources**, World Monitor identifies it as a **focal point** and escalates its prominence:

### Data Source Matching

<CardGroup cols={2}>
  <Card title="News Mentions">
    Entity extraction from headlines (NER, keyword matching)
  </Card>

  <Card title="Military Activity">
    Aircraft callsigns, vessel names, base locations
  </Card>

  <Card title="Protest Feeds">
    Location names, actor entities
  </Card>

  <Card title="Market Signals">
    Prediction market contract titles
  </Card>
</CardGroup>

### Confidence Scoring

Matches are weighted by confidence:

* **Exact name match**: 1.0
* **Alias match**: 0.85–0.95
* **Keyword match**: 0.7

**Example**: If "Iran" appears in:

* 12 news headlines (news signal)
* 3 military flights near Iranian airspace (military signal)
* 2 protests in Tehran (protest signal)
* 1 Polymarket contract about Iran nuclear talks (market signal)

World Monitor identifies **Iran** as a focal point with 4-source convergence and escalates it in the intelligence picture.

<Tip>
  Focal points are displayed with a 🎯 icon in the Intelligence Findings badge and receive priority in the World Brief.
</Tip>

## Intelligence Alert Popups

By default, World Monitor accumulates signals silently in the **Intelligence Findings** badge. You can opt in to **automatic popup notifications** for high-priority alerts:

<Steps>
  <Step title="Click the Intelligence Findings badge">
    The dropdown shows all recent signals and alerts.
  </Step>

  <Step title="Toggle the popup switch">
    Enabled: Popups appear automatically for CRITICAL/HIGH alerts\
    Disabled (default): Badge pulses and counts, but no popups
  </Step>

  <Step title="Popup preference persists">
    Your choice is stored in localStorage and survives page reloads.
  </Step>
</Steps>

<Warning>
  Popups can be disruptive during active monitoring. Only enable them if you need immediate alerts for high-severity events.
</Warning>

## Related Features

* [Country Analysis](/guide/country-analysis) — CII scoring and country-level signals
* [Keyword Monitoring](/guide/keyword-monitoring) — Custom alerts for specific terms
* [World Brief](/guide/world-brief) — AI-synthesized global intelligence summary