TC-RED: Tropical Cyclone Real-time Event Dashboard

A standalone interactive dashboard for tracking, visualising, and analysing tropical cyclones in the Australian region. Built for energy and resources clients who need real-time situational awareness of TC threats to offshore and onshore assets.

Features

Interactive Cyclone Tracking Map

• Multi-system display — track several active cyclones simultaneously
• Analysis (observed) and forecast track lines with position markers colour-coded by Australian TC category (TL through Cat 5)
• BoM forecast confidence cones and confidence areas with time-slider playback
• 250 km danger zone overlay around current position
• Wind radii visualisation (gale, storm, hurricane force) by quadrant
• Four map tile styles: CARTO Dark, CARTO Voyager, Esri Topo, OpenStreetMap

Client Asset Management

• 23 pre-configured energy/resources clients (Santos, Chevron, Woodside, Shell, INPEX, etc.)
• Per-client custom locations (offshore platforms, onshore facilities)
• Category-specific threat distance thresholds
• Configurable range rings for each asset location

Gale Arrival Time Forecasting

• Estimates when gale-force winds (34+ kt) will reach each client location
• Uses linear interpolation between forecast steps for sub-step accuracy
• Considers maximum gale radius across all four quadrants (conservative approach)
• Accounts for client-defined range ring boundaries as trigger distances

Analysis Charts

• Intensity chart — dual-axis time series of maximum sustained wind (kt) and central pressure (hPa), with category boundary lines and analysis/forecast divider
• TC probability chart — development probability over the 7-day outlook, colour-coded by rating
• Movement speed chart — track speed (km/h) over time

Location Impact Assessment

• Distance and bearing from TC to 14 major Australian coastal cities
• Threat level classification (Extreme/High/Moderate/Low/Minimal) based on proximity
• Per-client location impacts with gale arrival countdown

Additional

• 8 Australian timezone options (UTC, AWST, ACST, AEST, and daylight-saving variants)
• Full advisory text display (7-day forecast + track discussion)
• Optional Redis caching for performance
• Google Analytics 4 integration for usage tracking
• Docker deployment with health checks
• /health endpoint for load balancer probes

Architecture

tc-standalone/
├── app.py                  # Dash application entry point
├── config.py               # Environment-driven configuration
├── requirements.txt        # Python dependencies
├── Dockerfile              # Production container
├── docker-compose.yml      # App + Redis stack
├── .env.example            # Environment variable template
├── src/
│   ├── pages/
│   │   └── tc.py           # Main page layout + callbacks (~1600 lines)
│   ├── components/
│   │   ├── map_components.py   # Dash-Leaflet map builder
│   │   └── tc_charts.py       # Plotly chart factories
│   ├── services/
│   │   └── tc_service.py      # Data loading, geospatial analysis, gale arrival (~900 lines)
│   ├── data/
│   │   └── cache.py           # Redis cache wrapper (graceful degradation)
│   └── utils/
│       └── constants.py       # Categories, colours, thresholds, locations, map tiles
├── data/
│   ├── tc_data/            # BoM advisory JSON files (50 files, 13 systems)
│   └── tc_config/          # Client configuration files (23 clients)
└── assets/
    ├── icons/              # TC category marker icons (Cat 1–5, Tropical Low)
    ├── styles/             # CSS (dark theme: base, layout, components)
    └── scripts/            # GA4 event tracking

Data

The app ships with 50 archived BoM tropical cyclone advisories covering 13 systems from the 2024–25 and 2025–26 Australian cyclone seasons:

System	Name	Peak Status	Advisories
AU202425_29U	Errol	Severe Tropical Cyclone	3
AU202526_34U	Narelle	Severe Tropical Cyclone	8
AU202526_12U	(unnamed)	Ex-Tropical Cyclone	19
AU202526_11U	Jenna	Ex-Tropical Cyclone	1
AU202526_02U	(unnamed)	Tropical Cyclone	3
AU202526_03U	(unnamed)	Tropical Low	3
AU202526_15U	(unnamed)	Tropical Low	3
AU202425_30U	(unnamed)	Tropical Low	3
AU202526_16U	(unnamed)	Tropical Low	2
AU202526_14U	(unnamed)	Tropical Low	2
AU202526_13U	(unnamed)	Tropical Low	1
AU202425_31U	(unnamed)	Tropical Low	1
AU202526_17U	(unnamed)	Tropical Low	1

Note: This is a proof-of-concept using archived data. The production version will connect to live BoM advisory feeds.

Quick Start

Local Development

# Clone the repository
git clone https://github.com/metmohammedn/TC-RED-POC.git
cd TC-RED-POC

# Create and activate virtual environment
python3 -m venv venv
source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Copy environment config (optional)
cp .env.example .env

# Run the app
python app.py

Open http://localhost:8050 in your browser.

Docker

# Copy environment config
cp .env.example .env

# Build and run with Docker Compose (includes Redis)
docker compose up --build

The app will be available at http://localhost:8050.

Configuration

All configuration is via environment variables (see .env.example):

Variable	Default	Description
DEBUG	false	Enable Dash debug mode
HOST	0.0.0.0	Server bind address
PORT	8050	Server port
REDIS_URL	redis://localhost:6379/0	Redis connection URL (optional — app runs without it)
GA_MEASUREMENT_ID	(empty)	Google Analytics 4 measurement ID (leave empty to disable)
GUNICORN_WORKERS	4	Number of gunicorn worker processes
GUNICORN_TIMEOUT	120	Gunicorn worker timeout (seconds)

Australian TC Category Scale

The dashboard uses the official Bureau of Meteorology tropical cyclone intensity scale:

Category	Wind Speed (kt)	Colour
Tropical Low	0–33	Yellow
Category 1	34–47	Green
Category 2	48–63	Blue
Category 3	64–85	Orange
Category 4	86–107	Red
Category 5	108+	Purple

Client Configuration

Client locations are defined in JSON files under data/tc_config/. Each client file specifies:

• Asset locations (lat/lon, onshore/offshore type)
• Category-specific threat distance thresholds (km)
• Range rings for map display (nautical miles)

To add a new client:

1. Create data/tc_config/{client}_user.json following the existing format
2. Add an entry to data/tc_config/users_registry.json

Tech Stack

• Framework: Dash 4.0 (Plotly)
• UI Components: Dash Mantine Components 0.14+
• Maps: Dash Leaflet with CARTO/Esri/OSM tiles
• Charts: Plotly
• Data Processing: Pandas, NumPy
• Caching: Redis (optional)
• Server: Gunicorn (production), Dash dev server (debug)
• Container: Docker with multi-service Compose
• Analytics: Google Analytics 4

Production Deployment

The app is Docker-ready for deployment on AWS ECS Fargate or similar container platforms:

• Health check: GET /health returns 200 OK (used by Docker HEALTHCHECK and ALB target groups)
• Gunicorn: 4 workers by default, configurable via GUNICORN_WORKERS
• Redis: Optional sidecar for caching (ElastiCache in production)
• Stateless: No local filesystem writes — safe for horizontal scaling

License

Proprietary. All rights reserved.