Cloud ANUGA

Professional ANUGA flood simulations with managed data, cloud compute, and published results. No Python environment, no desktop install.

What is Cloud ANUGA?

Hydrata hosts the ANUGA open-source 2D flood solver on AWS cloud infrastructure. You upload terrain data, configure scenarios, and Hydrata handles:

• Mesh generation — automatic triangulation from your DEM and region definitions
• Compute provisioning — AWS Batch allocates resources on demand
• Execution — ANUGA runs on optimized compute instances
• Results publishing — flood outputs published as OGC WMS/WFS layers

How It Works

Your DEM + scenario config
        ↓
    Hydrata API / MCP / Browser
        ↓
    AWS Batch (c7a.xlarge spot, $0.065/hr)
        ↓
    ANUGA solver (unmodified)
        ↓
    Flood depth, velocity, extent → WMS layers

Access Methods

Method	Best for	Documentation
Browser	Interactive setup, visual results	Getting Started
MCP Server	AI-assisted modeling workflows	MCP Server
REST API	Programmatic access, automation	API Reference
Python (local)	Running locally with run_anuga	Standalone Usage

Pricing

Tier	Cost	Includes
Free	$0	3 simulation runs, no credit card required
Pay-as-you-go	$0.25–0.35/hr	Unlimited runs, AWS Batch spot instances
Team	Contact us	Shared workspace, dedicated compute, priority support

Typical costs:

• Merewether benchmark (small urban model): ~$0.017/run
• Medium catchment (1–5 km²): ~$0.50–2.00/run
• Large floodplain (50+ km²): ~$5–20/run

Solver Integrity

Hydrata runs unmodified ANUGA. We do not fork or modify the solver. Your results are identical to running ANUGA locally with the same inputs.

• Solver: anuga_core (GPL, Geoscience Australia + ANU)
• Runner: run_anuga (MIT, GitHub)
• Model files: portable scenario.json format — export and run anywhere

Validated

ANUGA has been validated against 30+ benchmark problems. See Validation for details, including the Merewether benchmark which completes in under 2 minutes on Hydrata cloud.

Get Started

1. Sign up free — no credit card required
2. Create your first project — upload a DEM and run a simulation
3. Or connect your AI assistant — run flood models from Claude Code, Copilot, or Cursor