Save Blatten & Beyond

Rapid Response to Natural and Man-Made Disasters

An interactive web app to show and predict dangerous locations in the mountains, developed by physics students at UZH for the NASA Hackathon 2025 in Zurich.

---

Table of Contents

1. About
2. Motivation & Goals
3. Features
4. Architecture & Technology Stack
5. Installation & Setup

---

About

“Save Blatten & Beyond” is a web app that helps visualize and predict hazardous areas in mountainous terrain (e.g. avalanche risk zones, landslides) via an interactive map.
The app combines Sentinel-1 and Sentinel-2 radar and imagery data to highlight high-risk zones and inform rescue planning or precautionary measures.

We created this as our project for the NASA Hackathon 2025 in Zurich. We are physics students at the University of Zurich (UZH), bringing together skills in data analysis, geospatial modeling, and web development.

---

Motivation & Goals

• Mountainous terrain poses significant risks (avalanches, rockfalls, landslides).
• Early identification of dangerous areas helps in disaster prevention, planning, and rescue operations.
• We aim to build a tool for rapid response: enabling rescue teams, local authorities, or hikers to visualize risk in real time.
• The app should be intuitive, interactive, and extensible for future enhancements (e.g. adding more sensors, real-time updates).

---

Features

• Interactive map interface with zoom / pan
• Overlay of predicted risk zones (color-coded)
• Ability to click / hover to see risk level, contributing data
• Data ingestion / preprocessing from satellite / remote sensing
• Backend logic to compute predictions
• (Optional / future) Alerts, time series, forecasts

---

Architecture & Technology Stack

Here’s a rough breakdown of the components and technologies used:

Component	Technology / Library
Backend / API	Python (Streamlit, ...)
Predictive model logic	Python, Torch (in development)
Satellite / remote data	Sentinel-2, Sentinel-1
Frontend / UI	Streamlit (Python)
Deployment / scripts	installer.sh, environment setup, etc.
Dependencies	listed in requirements.txt

---

Installation & Setup

Below is a general guide.

Prerequisites

• Python 3.8+
• (Optionally) virtual environment tool (venv, conda)
• Git

Steps

1. Clone the repository

   git clone https://github.com/borystereschenko/Hackthon.git
   cd Hackthon
  

2. Make Installer executable and (optional) give everyone on the system privilege to execute

  chmod +x installer.sh && chmod 755 installer.sh
  

3. Get API keys Get your own api key at sentinel-hub.com

4. Run the installer

   ./installer.sh