Data Lamp: Visualizing Wild Horse & Burro Populations 🐎

OVERVIEW

Data Lamp is my bachelor thesis project for my Creative Computing program at the University of Applied Sciences St. Pölten. The project explores data physicalization through the creation of a papercraft lamp that visualizes geospatial data about Wild Horses and Burros in North America.

The lamp’s horse-shaped design incorporates laser-cut or hand-cut patterns that represent population clusters. These clusters are derived from real geospatial datasets and transformed into a visual texture applied to a 3D model in Blender.

PROJECT GOALS

• Combine data visualization and physical design to communicate environmental and animal-related data in a tangible way.
• Use Python to process, filter, and cluster geospatial datasets of wild horse and burro populations.
• Generate a cluster map for UV mapping and texturing a 3D model in Blender.
• Build 3 different printable 3D Models in Blender
• Fabricate a papercraft lamp based on the 3D model, merging digital and physical representation.

DATA PIPELINE

Input

• Datasets from the Bureau of Land Management (BLM) and related sources on wild horse and burro populations
• Filtering relevant U.S. states (only those with active populations).
• Merging multiple datasets into one unified geospatial dataset.
• Clustering population data to identify density regions using a proximity-based algorithm.

Output

• A cluster map representing merged and scaled population clusters.
• This image is then used as a material texture in Blender for UV mapping onto the 3D horse model.

TOOLS & TECHNOLOGIES

• Python – Data processing and clustering (Libraries: pandas, geopandas, matplotlib, shapely)
• Blender – 3D modeling and UV mapping
• Pepakura Designer 6 – Unfolding 3D model for papercraft assembly
• Adobe Illustrator / Inkscape – Vector editing and refinement