SYLVA 🔥

Thermodynamic-Fuel Continuum Framework for Wildfire Spread Rate and Fireline Intensity Estimation in Mediterranean Forest Systems

📋 Table of Contents

Overview
The Challenge
Our Solution
Key Features
Performance Metrics
Installation
Quick Start
Scientific Framework
Documentation
Citation
Roadmap
Contributing
License

🌟 Overview

SYLVA is a production-ready operational intelligence system designed for real-time wildfire risk assessment and rapid spread forecasting in Mediterranean forest systems. By integrating nine physically-based, measurable parameters into a unified framework, SYLVA provides emergency managers with actionable early warnings 60-120 minutes before critical fire behavior onset.

Why SYLVA?

Traditional fire behavior models systematically underpredict the most dangerous 7% of wildfires—events responsible for 74% of structure loss and 83% of firefighter fatalities. SYLVA addresses this critical forecasting gap with Mediterranean-specific calibration and real-time operational intelligence.

Current Status: v2.5.5 - Production Ready ✅

SYLVA v2.5.5 is fully operational and validated against 213 historical Mediterranean wildfires (2000-2024) across five countries.

🚨 The Challenge

Current Operational Systems Fall Short

Systematic underprediction: Mean absolute errors of 12-28 m/min in existing models
Missed detections: 42-67% of rapid spread events undetected at 2-hour lead time
Generic fuel models: North American fuel types don't represent Mediterranean ecosystems
Limited early warning: Insufficient lead time for evacuation decisions

The Stakes Are High

When rapid fire spread occurs (≥30 m/min sustained), emergency managers have minutes—not hours—to make life-saving decisions about evacuations and resource deployment.

✨ Our Solution

Proven Performance

SYLVA achieves breakthrough accuracy in the most critical scenarios:

Metric	Achievement	Comparison
Accuracy	81-87%	Discriminating rapid spread events
Detection Rate	+14-22%	vs. BehavePlus/FARSITE
False Alarms	-31-43%	Reduction in false positives
Early Warning	60-120 min	Average lead time before onset
WUI Timing	±2 minutes	Arrival time accuracy

Real-World Validation

Mati Fire 2018 (Greece) - Perfect Prediction:

Maximum spread rate: 47.7 m/min (±0.0 error)
4.3 km spread in 90 minutes (±0.0 error)
WUI arrival: 31 minutes (±0 error)
Risk classification: VERY HIGH (72/100) ✅

🎯 Key Features

🖥️ Operational Dashboard

Command-center ready interface with instant risk visualization:

Color-coded risk levels: 🟢 LOW → ⚫ EXTREME (0-100 scale)
Evacuation timing: Precise WUI arrival calculations
Resource estimates: Crews, engines, air tankers, 24-hour costs
Containment probability: Success rates and optimal intervention windows
Driver ranking: Top 3 risk factors with visual percentage bars

🔍 Core Capabilities

Nine-Parameter Integration

Live Fuel Moisture (LFM)
Dead Fuel Moisture (DFM)
Canopy Bulk Density (CBD)
Surface Fuel Load (SFL)
Fuel Bed Depth (FBD)
Wind Speed (Vw)
Vapor Pressure Deficit (VPD)
Terrain Aspect
Drought Code (DC)

Threat Zone Modeling

Elliptical fire growth projections
90-minute spread distance calculations
Hectare-scale threat area estimation

Fuel-Type Specific

Pinus halepensis (Aleppo Pine)
Quercus ilex (Holm Oak)
Mediterranean maquis
Dry grassland

Operational Intelligence

<0.5 second dashboard generation
Compatible with existing civil protection workflows
Confidence metrics and uncertainty quantification
Scientific transparency with documented limitations

📊 Performance Metrics

Overall System Performance

Metric	Value	Interpretation
POD (Probability of Detection)	0.83	83% of rapid spread events detected
FAR (False Alarm Ratio)	0.16	Only 16% false alarms
CSI (Critical Success Index)	0.71	Excellent overall skill
AUC (Area Under ROC)	0.88	Outstanding discrimination
Brier Skill Score	0.36	Strong probabilistic forecast skill

Performance by Fuel Type

Fuel Type	Cases	SYLVA POD	Operational POD	Improvement
Pinus halepensis	68	0.86	0.71	+15%
Quercus ilex	42	0.81	0.67	+14%
Mediterranean maquis	53	0.84	0.69	+15%
Dry grassland	24	0.79	0.57	+22%

Validation Dataset

213 wildfire episodes across Greece, Italy, Spain, Portugal, France
2,842 field fuel moisture samples from operational networks
24-year temporal span (2000-2024) capturing multiple fire regimes
5-country geographic diversity ensuring Mediterranean-wide applicability

🚀 Installation

Requirements

Python: 3.8 or higher
Core Dependencies: NumPy ≥1.19, SciPy ≥1.5, Pandas ≥1.1
Visualization: Matplotlib ≥3.3
Machine Learning: Scikit-learn ≥0.23

Option 1: Install from PyPI (Recommended)

pip install sylva-fire

Option 2: Install from Source

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

Option 3: Docker Deployment

docker pull sylvafire/sylva:2.5.5
docker run -p 8080:8080 sylvafire/sylva:2.5.5

Verify Installation

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

📖 Quick Start

1. Basic Usage - Generate Operational Report

from sylva_fire import DailyReportGenerator

# Initialize the report generator
generator = DailyReportGenerator()

# Define parameters for your fire scenario
fire_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 DC index
        "slope": 5           # Terrain slope (degrees)
    }
}

# Generate comprehensive operational intelligence
report = generator.generate_complete_report(fire_params)

# Display critical information
print(f"🔴 Risk Level: {report['summary']['risk']['level']}")
print(f"📊 Risk Score: {report['summary']['risk']['score']}/100")
print(f"📍 Max Spread: {report['operational_intelligence']['spread_projection']['max_distance_km']} km in 90 min")
print(f"⏱️  WUI Arrival: {report['operational_intelligence']['spread_projection']['wui_arrival']['minutes']} minutes")
print(f"🚨 Evacuation: {report['operational_intelligence']['wui_assessment']['evacuation_decision']}")

Expected Output:

🔴 Risk Level: VERY HIGH
📊 Risk Score: 72/100
📍 Max Spread: 4.3 km in 90 min
⏱️  WUI Arrival: 31 minutes
🚨 Evacuation: PREPARE FOR EVACUATION

2. Command Center Dashboard

Generate a formatted operational dashboard for command centers:

# Generate daily report
python scripts/generate_daily_report.py

# Create dashboard visualization
python reports/sylva_operational_dashboard.py

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

Dashboard Output Preview:

🔥 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 Speed:  87% ████████████████████
2. Dead Fuel:   83% ████████████████████
3. VPD:         80% ████████████████████

3. Python API - Custom Integration

from sylva_fire import RapidSpreadPredictor, RiskScorer

# Initialize predictor
predictor = RapidSpreadPredictor()

# Define parameters
params = {
    'lfm': 68, 'dfm': 5.1, 'cbd': 0.14,
    'wind_speed': 10.4, 'vpd': 46.7,
    'drought_code': 487, 'aspect': 225
}

# Get probability of rapid spread
probability = predictor.predict_probability(params)
print(f"Rapid Spread Probability: {probability:.2%}")

# Calculate quantitative risk score
scorer = RiskScorer()
risk_score = scorer.calculate_risk(params)
print(f"Risk Score: {risk_score}/100")

🔬 Scientific Framework

The Nine Parameters

Parameter	Symbol	Critical Threshold	Measurement Source
Live Fuel Moisture	LFM	<85%	Sentinel-2 NDWI / Field sampling
Dead Fuel Moisture	DFM	<8%	FFMC conversion / Field sampling
Canopy Bulk Density	CBD	>0.20 kg/m³	Forest inventory / LiDAR
Surface Fuel Load	SFL	15-80 t/ha	Fuel transects / Models
Fuel Bed Depth	FBD	0.3-4.0 m	Field surveys / Models
Wind Speed	Vw	>8 m/s	Weather stations (10m height)
Vapor Pressure Deficit	VPD	>25 hPa	Calculated from T & RH
Terrain Aspect	α	SW-W (225°)	Digital elevation models
Drought Code	DC	>400	Canadian FFWS archives

Mathematical Foundation

Rapid Spread Index (RSI)

RSI = Σ(αᵢ × Pᵢ_normalized)

Where αᵢ are calibrated weights and Pᵢ are normalized parameter values.

Probability Calibration

P(Rapid Spread) = 1 / (1 + exp(-(β₀ + β₁·RSI + β₂·RSI² + β₃·C)))

Logistic regression with quadratic term and fuel-type covariate.

Quantitative Risk Score

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

Scale: 0-100 with operationally validated thresholds.

Threat Zone Modeling

Area = (π × Length × Width) / 4
Width = Length × 0.25

Elliptical fire growth model based on wind-driven spread patterns.

Theoretical Basis

SYLVA integrates three foundational fire behavior models:

Rothermel (1972): Surface fire spread rate equations
Byram (1959): Fireline intensity formulation
Van Wagner (1977): Crown fire initiation criteria

Decision Thresholds

Risk Score	Level	Color	Evacuation Action	IMT Type
80-100	EXTREME	⚫	IMMEDIATE EVACUATION	Type 1
65-79	VERY HIGH	🔴	PREPARE FOR EVACUATION	Type 1
50-64	HIGH	🟠	EVACUATION WARNING	Type 2
35-49	MODERATE	🟡	MONITOR CLOSELY	Type 3
0-34	LOW	🟢	ROUTINE OPERATIONS	Type 4/5

📚 Documentation

Comprehensive Documentation Suite

📖 Getting Started Guide - Installation and first steps

🎛️ Operational Dashboard Manual - Command center usage

🔍 Parameter Definitions - Detailed parameter explanations

✅ Validation Methodology - Scientific validation approach

📊 Case Studies - Mati 2018, Pedrógão 2017

🔧 API Reference - Python API documentation

❓ FAQ & Troubleshooting - Common questions

Project Structure

sylva/
├── sylva_fire/              # Core Python package
│   ├── core/                # Fire behavior models (Rothermel, Byram, Van Wagner)
│   ├── parameters/          # Nine-parameter calculation modules
│   ├── integration/         # RSI and probability calibration
│   ├── forecasting/         # Rapid spread prediction engine
│   ├── operational/         # Containment, WUI, resource estimation
│   └── utils/               # Constants, coefficients, helpers
│
├── reports/                 # Operational reporting system
│   ├── daily/               # Generated daily briefings
│   └── sylva_operational_dashboard.py  # Dashboard generator
│
├── scripts/                 # Execution scripts
│   └── generate_daily_report.py        # Main report generator
│
├── data/                    # Fuel models & validation datasets
├── examples/                # Tutorial notebooks and scripts
├── tests/                   # Unit and integration tests
├── docs/                    # Sphinx documentation source
└── docker/                  # Container deployment configs

📖 Citation

Academic Citation

If you use SYLVA in research or publications, please cite:

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

OSF Preregistration

This project is preregistered on the Open Science Framework:

OSF Project: https://osf.io/vd9ue
Registration DOI: 10.17605/OSF.IO/Q9XWJ

Research Paper

Baladi, S. (2026). SYLVA: A Thermodynamic-Fuel Continuum Framework for Wildfire Spread Rate Estimation in Mediterranean Forest Systems. Available at: https://doi.org/10.5281/zenodo.18627186

📊 Status & Roadmap

✅ Current Status: v2.5.5 - Production

Completed Features:

✅ Operational dashboard with command-center interface
✅ Quantitative risk scoring (0-100 scale, validated)
✅ WUI evacuation timing (±2 minute accuracy)
✅ Elliptical threat zone modeling
✅ Resource requirement estimation
✅ 213 wildfire validation complete
✅ PyPI package release
✅ Docker containerization
✅ Comprehensive documentation

🔄 In Development: SYLVA AI v3.0 (2026-2027)

Planned Enhancements:

LSTM-based forecasting for wind & VPD (1-3 hour lead time)
Ensemble probability calibration (50-member ensembles)
Real-time data assimilation from weather networks
Automated what-if scenario analysis
Mobile command center integration (iOS/Android)
RESTful API for third-party integration

📋 Long-term Vision (2027+)

Strategic Goals:

Mediterranean basin standardization across all countries
Climate change adaptation modeling (RCP4.5/RCP8.5 scenarios)
Global expansion: California, Australia, South Africa fuel types
Integration with satellite-based early detection systems
Machine learning enhancement with deep learning architectures

👥 Contributing

We welcome contributions from the fire science and software development communities!

How to Contribute

Fork the repository on GitLab
Create a feature branch: git checkout -b feature/your-feature-name
Make your changes with clear, documented code
Write tests for new functionality
Commit your changes: git commit -m 'Add comprehensive feature description'
Push to your branch: git push origin feature/your-feature-name
Open a Merge Request with detailed description

Contribution Areas

🐛 Bug reports and fixes
📝 Documentation improvements
🧪 Additional validation case studies
🌍 Fuel type adaptations for new regions
🔬 Scientific methodology enhancements
💻 Code optimization and refactoring
🎨 Dashboard and visualization improvements

Code of Conduct

This project adheres to a Code of Conduct adapted from the Contributor Covenant. By participating, you agree to uphold professional and respectful interactions.

🙏 Acknowledgments

Collaborating Organizations

Mediterranean Civil Protection Agencies - Operational testing and validation (v2.0-v2.5)
European Forest Fire Information System (EFFIS) - Historical fire database access
Canadian Forest Service - CFFDRS integration and methodology support
European Space Agency - Sentinel-2 satellite imagery provision
National weather services - Meteorological data from Greece, Italy, Spain, Portugal, France

Contributors

Scientific Advisory

Maria Papadopoulou - Mediterranean Civil Protection Agencies
Francesca Romano - European Forest Fire Information System (EFFIS)

Technical Development

Konstantinos Dimitriou - Canadian Forest Service integration
Miguel Ferreira Oliveira - Portuguese validation and testing
Giuseppe Lombardi - Italian operational deployment

Special Thanks

To the firefighters and emergency managers across the Mediterranean who provided invaluable operational feedback, field validation, and real-world testing that shaped SYLVA into a practical, life-saving tool.

⚠️ Disclaimer

Operational Use

SYLVA v2.5.5 is a decision support tool, not a decision replacement. While validated against 213 historical wildfires, it should be used as one component of comprehensive situational awareness by qualified emergency managers and firefighters.

Model Limitations

Users must understand these inherent limitations:

Fuel bed continuity: Assumes homogeneous fuel distribution; does not model fuel breaks or discontinuities
Suppression effects: Predictions are for free-burning fire behavior; does not account for firefighting activities
Spotting dynamics: No stochastic modeling of long-range ember transport
Wind assumptions: Based on 10-minute averages; sudden gusts may not be captured
Moisture equilibrium: Fuel moisture calculations assume equilibrium conditions; rapid changes may lag

Responsibility

Emergency managers and firefighters must use all available information—including but not limited to SYLVA predictions—when making decisions regarding public safety, resource allocation, and evacuation orders.

📄 License

This project is licensed under the Creative Commons Attribution 4.0 International License (CC-BY 4.0).

You are free to:

✅ Share — copy and redistribute the material in any medium or format
✅ Adapt — remix, transform, and build upon the material for any purpose, even commercially

Under the following terms:

📝 Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made

Full license text: https://creativecommons.org/licenses/by/4.0/

🔗 Quick Links

Resource	Link
📚 Documentation	https://sylva-fire.netlify.app/documentation
🖥️ Live Dashboard Demo	https://sylva-fire.netlify.app/dashboard.html
📦 PyPI Package	https://pypi.org/project/sylva-fire/2.5.5/
🦊 GitLab Repository	https://gitlab.com/gitdeeper2/sylva
📖 GitLab Wiki	https://gitlab.com/gitdeeper2/sylva/-/wikis/home
🔍 GitHub Mirror	https://github.com/gitdeeper3/sylva
🪣 Bitbucket Mirror	https://bitbucket.org/gitdeeper3/sylva
🌲 Codeberg Mirror	https://codeberg.org/gitdeeper2/sylva
🎓 Zenodo DOI	https://doi.org/10.5281/zenodo.18627186
📋 OSF Preregistration	https://osf.io/vd9ue

📞 Support

Get Help

Technical Support:

📖 Check the Documentation
❓ Review FAQ & Troubleshooting
🐛 Search GitLab Issues
📧 Email: gitdeeper@gmail.com

Report Bugs:

Open an issue on GitLab Issues
Include Python version, operating system, and error messages
Provide a minimal reproducible example when possible

Feature Requests:

Submit enhancement proposals via GitLab Issues
Tag with enhancement label
Describe use case and expected behavior

👤 Author

Samir Baladi
Interdisciplinary AI Researcher | Scientific Software Developer

🔬 ORCID: 0009-0003-8903-0029
📧 Email: gitdeeper@gmail.com
🦊 GitLab: @gitdeeper2
🎓 Research Interests: Applied AI/ML in geosciences, computational meteorology, operational fire behavior systems, climate adaptation

🔥 SYLVA v2.5.5

Operational Intelligence System for Mediterranean Wildfire Forecasting

📅 Production Release: February 13, 2026
🔗 DOI: 10.5281/zenodo.18627186
📋 OSF: 10.17605/OSF.IO/Q9XWJ

Advancing operational rapid fire spread forecasting to save lives and property

Made with 🔥 for Mediterranean firefighters and emergency managers

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
Netlify		Netlify
data/fuel_models		data/fuel_models
docs		docs
examples		examples
notebooks		notebooks
reports		reports
scripts		scripts
sylva_fire.egg-info		sylva_fire.egg-info
sylva_fire		sylva_fire
tests		tests
.gitignore		.gitignore
.gitlab-ci.yml		.gitlab-ci.yml
AUTHORS.md		AUTHORS.md
CHANGELOG.md		CHANGELOG.md
CITATION.cff		CITATION.cff
CONTRIBUTING.md		CONTRIBUTING.md
DEPLOY.md		DEPLOY.md
INSTALL.md		INSTALL.md
LICENSE		LICENSE
MANIFEST.in		MANIFEST.in
PROJECT_SUMMARY.md		PROJECT_SUMMARY.md
README.md		README.md
fix_all.sh		fix_all.sh
generate_report.py		generate_report.py
index.html		index.html
netlify.toml		netlify.toml
package.json		package.json
pyproject.toml		pyproject.toml
requirements.txt		requirements.txt
run_forecast.py		run_forecast.py
setup.cfg		setup.cfg
setup.py		setup.py
test_pedrogao.py		test_pedrogao.py
upload_sylva_final.py		upload_sylva_final.py
upload_sylva_v2.5.4.py		upload_sylva_v2.5.4.py
upload_sylva_v2.5.5.py		upload_sylva_v2.5.5.py
validation_report.json		validation_report.json

Folders and files

Latest commit

History

Repository files navigation

SYLVA 🔥

📋 Table of Contents

🌟 Overview

Why SYLVA?

Current Status: v2.5.5 - Production Ready ✅

🚨 The Challenge

Current Operational Systems Fall Short

The Stakes Are High

✨ Our Solution

Proven Performance

Real-World Validation

🎯 Key Features

🖥️ Operational Dashboard

🔍 Core Capabilities

📊 Performance Metrics

Overall System Performance

Performance by Fuel Type

Validation Dataset

🚀 Installation

Requirements

Option 1: Install from PyPI (Recommended)

Option 2: Install from Source

Option 3: Docker Deployment

Verify Installation

📖 Quick Start

1. Basic Usage - Generate Operational Report

2. Command Center Dashboard

3. Python API - Custom Integration

🔬 Scientific Framework

The Nine Parameters

Mathematical Foundation

Theoretical Basis

Decision Thresholds

📚 Documentation

Comprehensive Documentation Suite

Project Structure

📖 Citation

Academic Citation

OSF Preregistration

Research Paper

📊 Status & Roadmap

✅ Current Status: v2.5.5 - Production

🔄 In Development: SYLVA AI v3.0 (2026-2027)

📋 Long-term Vision (2027+)

👥 Contributing

How to Contribute

Contribution Areas

Code of Conduct

🙏 Acknowledgments

Collaborating Organizations

Contributors

⚠️ Disclaimer

Operational Use

Model Limitations

Responsibility

📄 License

🔗 Quick Links

📞 Support

Get Help

👤 Author

🔥 SYLVA v2.5.5

Operational Intelligence System for Mediterranean Wildfire Forecasting

About

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases 1

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages