SYLVA Documentation

Version 2.5.0 - Production Release

SYLVA (Thermodynamic-Fuel Continuum Framework) is an operational intelligence system for assessing rapid fire spread probability in Mediterranean forest systems by integrating nine physically-based, measurable parameters into a unified command-center ready forecasting platform.

Installation

System Requirements

• Python 3.8+
• NumPy >= 1.19.0
• SciPy >= 1.5.0
• Pandas >= 1.1.0
• Matplotlib >= 3.3.0
• Scikit-learn >= 0.23.0

Installation from GitLab

git clone https://gitlab.com/gitdeeper2/sylva.git
cd sylva
pip install -e .

Verify Installation

python -c "import sylva_fire; print(sylva_fire.__version__)"
# Expected output: 2.5.5

Quick Start

Generate Operational Report

from scripts.generate_daily_report import DailyReportGenerator

# Initialize generator
generator = DailyReportGenerator()

# Define fire scenario parameters
params = {
    "region": "Attica, Greece",
    "wui_distance": 1.5,  # km to Wildland-Urban Interface
    "parameters": {
        "lfm": 68,           # Live Fuel Moisture (%)
        "dfm": 5.1,          # Dead Fuel Moisture (%)
        "cbd": 0.14,         # Canopy Bulk Density (kg/m³)
        "wind_speed": 10.4,  # Wind Speed (m/s)
        "vpd": 46.7,         # Vapor Pressure Deficit (hPa)
        "drought_code": 487, # Canadian Drought Code
        "slope": 5           # Terrain Slope (degrees)
    }
}

# Generate complete report
report = generator.generate_complete_report(params)

# Display key metrics
print(f"Risk Level: {report['summary']['risk']['level']}")
print(f"Risk Score: {report['summary']['risk']['score']}/100")
print(f"Spread Distance: {report['operational_intelligence']['spread_projection']['max_distance_km']} km")

Command Center Dashboard

# Generate dashboard from command line
python scripts/generate_daily_report.py
python reports/sylva_operational_dashboard.py

# View dashboard
cat reports/daily/*_DASHBOARD.txt

The Nine Parameters

SYLVA integrates nine essential parameters to quantify rapid fire spread probability:

Parameter Acquisition

Live Fuel Moisture (LFM)

• Sentinel-2 NDWI inversion at 20m resolution
• 5-day revisit interval
• Field validation campaigns for calibration

Dead Fuel Moisture (DFM)

• Canadian Fine Fuel Moisture Code (FFMC)
• Automatic weather station networks (10-minute observations)
• Equilibrium moisture content calculations

Canopy Bulk Density (CBD)

• Landsat-derived with LiDAR validation
• European Forest Inventory data
• Regional allometric equations

Wind Vector (Vw)

• WindNinja/WindWizard topographic adjustment algorithms
• AWS networks with 10m standardization
• Terrain channeling and acceleration effects

API Reference

Core Modules

DailyReportGenerator

Main class for generating operational fire behavior reports.

from scripts.generate_daily_report import DailyReportGenerator

generator = DailyReportGenerator()

# Generate complete report
report = generator.generate_complete_report(params)

# Generate specific sections
summary = generator.generate_summary(params)
operational = generator.generate_operational_intelligence(params)
resources = generator.generate_resource_requirements(params)

OperationalDashboard

Command center dashboard generator.

from reports.sylva_operational_dashboard import OperationalDashboard

dashboard = OperationalDashboard(report_data)
dashboard.generate()  # Creates *_DASHBOARD.txt file

Core Functions

calculate_rapid_spread_index()

from sylva_fire.integration import calculate_rapid_spread_index

rsi = calculate_rapid_spread_index(
    lfm=68.0,
    dfm=5.1,
    cbd=0.14,
    wind_speed=10.4,
    vpd=46.7,
    drought_code=487
)
# Returns: float (0-100 scale)

calculate_crown_fire_probability()

from sylva_fire.forecasting import calculate_crown_fire_probability

crown_prob = calculate_crown_fire_probability(
    cbd=0.14,
    wind_speed=10.4,
    canopy_moisture=68.0
)
# Returns: float (0-1 probability)

calculate_spread_rate()

from sylva_fire.core import calculate_spread_rate

ros = calculate_spread_rate(
    fuel_type="dry_grassland",
    dfm=5.1,
    wind_speed=10.4,
    slope=5.0
)
# Returns: float (m/min)

Operational Dashboard

Risk Score Calculation

The quantitative risk score (0-100) is calculated from six factors:

RiskScore = DFM(0-25) + Wind(0-25) + VPD(0-15) + 
            DC(0-15) + Crown(0-10) + Containment(0-10)

Decision Thresholds

Dashboard Output Format

🔥 SYLVA OPERATIONAL DASHBOARD 🔴 VERY HIGH RISK
────────────────────────────────────────────────
RISK LEVEL:     🔴 VERY HIGH (Score: 72/100)
WUI ARRIVAL:    31 minutes - 🟠 PREPARE FOR EVACUATION
SPREAD:         4.3km in 90min (Dry Grassland)
CONTAINMENT:    🔴 VERY DIFFICULT (Success: 30%)
CROWN FIRE:     🔴 95% potential - VERY HIGH

🎯 PRIMARY RISK DRIVERS
────────────────────────────────────────────────
1. Wind: 87% ████████
2. DFM: 83% ████████
3. VPD: 80% ████████

Examples

Example 1: Mati Fire 2018

Historical validation case from Greece's deadliest wildfire.

params = {
    "region": "Attica, Greece",
    "date": "2018-07-23",
    "wui_distance": 1.5,
    "parameters": {
        "lfm": 68.0,
        "dfm": 5.1,
        "cbd": 0.14,
        "sfl": 35.0,
        "fbd": 0.8,
        "wind_speed": 10.4,
        "vpd": 46.7,
        "aspect": 225,
        "drought_code": 487,
        "slope": 5.0
    }
}

report = generator.generate_complete_report(params)

# Results:
# Risk Score: 72/100 (VERY HIGH)
# Max ROS: 47.7 m/min
# Spread Distance: 4.3 km in 90 minutes
# WUI Arrival: 31 minutes
# Crown Fire: 95% probability

Example 2: Custom Scenario

params = {
    "region": "Custom Location",
    "wui_distance": 2.0,
    "parameters": {
        "lfm": 75.0,
        "dfm": 8.0,
        "cbd": 0.18,
        "wind_speed": 12.0,
        "vpd": 38.0,
        "drought_code": 420
    }
}

report = generator.generate_complete_report(params)

Validation Results

Overall Performance

Fuel Type Performance

Deployment

Docker Deployment

# Build Docker image
docker build -t sylva:2.5.5 .

# Run container
docker run -d \
  --name sylva-fire \
  -v $(pwd)/reports:/app/reports \
  sylva:2.5.5

# Generate report
docker exec sylva-fire python scripts/generate_daily_report.py

Production Environment

Cron Job Example

# Run SYLVA report every 6 hours
0 */6 * * * cd /opt/sylva && python scripts/generate_daily_report.py

Troubleshooting

Common Issues

Issue: Import Error

pip install -e .

Issue: Missing Dependencies

pip install -r requirements.txt

Issue: Parameter Validation

Support

For additional support:

• GitLab Issues: https://gitlab.com/gitdeeper3/sylva/-/issues
• Email: [email protected]
• Documentation: https://sylva-fire.readthedocs.io

Author & Contributors

🤝 Scientific Advisory Board

💻 Development Team

🔬 Validation & Testing

🙏 Acknowledgments

This project received funding and support from:

• European Forest Fire Information System (EFFIS)
• Mediterranean Climate Outreach Programme
• Open Earth Observation for Wildfire Monitoring Initiative
• GitLab Open Source Program

Special thanks to all wildfire firefighters and emergency managers who provided invaluable feedback during the development and validation phases.

📄 How to Cite

@software{baladi2026sylva,
  author       = {Baladi, Samir},
  title        = {SYLVA: Operational Intelligence System for 
                  Mediterranean Wildfire Rapid Spread Forecasting},
  year         = 2026,
  version      = {2.5.5},
  doi          = {10.5281/zenodo.18627186},
  url          = {https://doi.org/10.5281/zenodo.18627186}
}