Alpine Atlas
Alpine Atlas turns Zermatt into an interactive 3D mountain explorer grounded in real Swiss terrain, aerial imagery, and dated visual evidence.
Core trick
Constrains the model to call viewpoint-selection tools, then compares two dated camera images and returns observations tied to image regions.
Description
Alpine Atlas turns Zermatt into an interactive 3D mountain explorer grounded in real Swiss terrain, aerial imagery, and dated visual evidence. Visitors explore the Matterhorn, changing sunlight, weather-model conditions, and the relationship between rock, glaciers, and water.
GPT-6 Astra powers an ask → explore → evidence workflow: it selects a viewpoint through constrained tool calls, compares two dated official camera images, and returns structured observations with image regions users can inspect. Camera observations, modeled conditions, and long-term glacier research stay visibly distinct; recorded Astra runs are explicitly labeled.
The custom Three.js landscape uses about one million terrain vertices, georeferenced textures, rock/snow shaders, geographic camera controls, and cinematic orbit. Glacier pins connect sourced change statistics with environmental context and Protect Our Winters resources. Built as a working browser experience during the hackathon.
README excerpt
# Alpine Atlas Explore Zermatt in 3D, then see the visual evidence behind what changed. Built for the OpenAI GPT-6 Astra SF Hackathon, September 8, 2026. **[Open the live demo](https://zermatt-mountain-atlas.oliviameng.chatgpt.site/?scene=reconstruction&view=matterhorn)** · **[Watch the demo video](https://www.loom.com/share/2aeb49305ee84ed38d3304aa39b7d558)** The hosted demo is publicly accessible. Start with the Matterhorn orbit, then choose **Explore with Astra** to inspect dated image evidence. ## What it does - A custom Three.js scene built from measured Swiss terrain, historical aerial imagery, and a modeled winter surface. - A ten-second Matterhorn orbit, geographic compass, Swiss–Italian border, and a rock → glacier → river journey. - Ready-made Astra experiences: choose a viewpoint, compare dated camera frames, then inspect the image regions supporting each observation. - Sun-position visualization and wind/snow particles driven by clearly labeled Open-Meteo weather-model data or separate custom controls. - A Zmuttgletscher pin with sourced historical area change, separate Swiss-wide glacier-volume context, and Protect Our Winters action resources. ## Run locally Node.js 22.13 or newer is required. ```sh npm ci cp .env.example .env.local # Set OPENAI_API_KEY in .env.local. Never commit your key. npm run dev -- --host localhost --port 3000 ``` Open http://localhost:3000/?scene=reconstruction&view=matterhorn. Terrain, orbit, sunlight and research context wor