Project Nomad

A national fire modeling GUI system for Canadian wildfire management.

Mission

Democratizing fire modeling to save lives through accessible, modern interfaces.

Overview

Project Nomad is a TypeScript React GUI for fire modeling. It provides a MapBox GL-based map interface for fire behavior analysis and prediction modeling in wildfire management operations.

Current Status: Phase 1 MVP Complete. Working SAN deployment with FireSTARR integration, full visualization pipeline, and installer for easy deployment.

Deployment

SAN (Stand Alone Nomad)

Self-hosted application for individual users or small teams:

• Bundled frontend with local backend
• SQLite for data storage
• Simple file-based authentication
• Container or Metal (bare metal) infrastructure modes
• Ideal for field operations or standalone deployments

ACN (Agency Centric Nomad)

Component integrated into existing agency systems:

• Nomad dashboard embedded as a React component
• Agency-managed authentication (FusionAuth, Okta, etc.) — Nomad defers all user auth to the host
• Server-to-server trust key required between host app and Nomad backend
• PostGIS or other agency database
• See EMBEDDING.md for integration guide

Fire Modeling Engine

FireSTARR

Open-source probabilistic fire modeling:

• Monte Carlo simulation for burn probability
• Canadian FBP System implementation
• Fast C++ execution
• Active development

Core Features

Model Setup Wizard

• Ignition input (point drawing on map)
• Temporal configuration (start time, duration)
• Weather data integration (SpotWX)

Model Review

• Fire perimeter visualization on MapBox GL
• Burn probability maps
• Time-stepped animation
• Export to multiple formats

Data Sources

• National fuel type grids
• National DEM
• ECCC weather data
• MODIS/VIIRS hotspots

Technical Stack

Component	Technology
Frontend	TypeScript, React
Map	MapBox GL JS
Backend	Node.js, Express, TypeScript
Database	SQLite
Model Engine	FireSTARR
Infrastructure	Container or Metal (bare metal)

System Requirements

Component	Minimum	Details
Nomad	2 cores, 4 GB RAM, 2 GB disk	Full requirements
FireSTARR	2 cores, 8 GB RAM, 50 GB disk (dataset)	Full requirements

Container mode requires a Docker-compatible API and Docker Compose v2. Bare metal mode requires Node.js >= 20, GDAL, and SQLite. See the requirements documents for complete details.

Documentation

Component Embedding

For embedding the Nomad dashboard in external applications:

• EMBEDDING.md - Consumer entry point for embedding
• openNomad/README.md - Complete adapter implementation guide
• docs/examples/ - Runnable integration examples

SME Documentation

Technical references for fire modeling:

• FireSTARR: Documentation/Research/SME_Data/FireSTARR/

Project Structure

project_nomad/
├── frontend/                  # React TypeScript frontend
├── backend/                   # Express API server
├── scripts/                   # Installer and setup scripts
├── docker/                    # Container build files
├── configuration/             # Agency configuration (generic/ ships with repo; agencies may use submodules)
├── docs/examples/             # Integration examples (React, vanilla JS)
├── Documentation/             # SME knowledge base and project docs
├── NOMAD_REQUIREMENTS.md      # Nomad system requirements
├── FIRESTARR_REQUIREMENTS.md  # FireSTARR system requirements
├── EMBEDDING.md               # Guide for embedding Nomad in other apps
├── install.sh                      # Run this to install
├── docker-compose.yaml        # Container deployment
└── docker-compose.dev.yaml    # Development deployment

Installation

The interactive installer handles all configuration and dependencies:

git clone https://github.com/WISE-Developers/project_nomad.git
cd project_nomad
./install.sh

The installer will guide you through:

1. Infrastructure selection - Container (recommended) or Metal (bare metal)
2. FireSTARR dataset - Download or use existing (~50GB national fuel/DEM data)
3. FireSTARR binary - Download or use existing
4. Configuration - Paths, ports, and environment setup

The installer validates all prerequisites and offers to fix common issues automatically.

Quick Start (Container)

After running the installer:

docker compose up -d

Access the application at http://localhost:3901

Quick Start (Metal)

After running the installer:

npm start

Access the application at http://localhost:4901

Development

npm install
npm run dev            # Watch mode (frontend + backend)
npm test               # Run tests
npm run lint           # ESLint

Contributing

Contributions welcome. See open issues or contact the maintainers.

License

AGPLv3 - See LICENSE for details.

---

Project Nomad - Making fire modeling accessible to those who need it most.