The Conservationist Who Turned 40 Terabytes of Public Data into a Video Game
Data-Driven Conservation: How One Man Turned 40TB of Info into a Game
During a recent video call, a grass fire was visible in the background behind Raffael Hickisch. For most, this would be a cause for alarm, but in Sub-Saharan Africa, such fires are routine—used by locals to clear tall grass to guide cattle toward fresher pastures.
Hickisch is a dedicated conservationist and the 2014 co-founder of the Chinko Nature Reserve. The reserve operates as a public-private venture, blending the efforts of a nonprofit with the government of the Central African Republic.
The Data Struggle
For Hickisch, fire data is a critical metric for tracking human migration and shaping land-use policy. However, the process of utilizing this data was historically efficient grueling. He had to manually synthesize disparate streams:
- Human settlement patterns
- Deforestation metrics
- NASA fire data
He relied on desktop software to visualize these layers, but the computational demand was overwhelming.
"I hardly ever simulate it because it's just a lot of work and my computer doesn’t even support it," Hickisch admitted.
This technical gap has dire consequences. When managing tens of thousands of square kilometers, enforcing "off-limits" zones or seasonal grazing areas is impossible without real-time visibility. Often, deforestation occurs unnoticed simply because government officials lack the technical expertise to access satellite data.
The Evolution of Tooling
| Feature | The "Old Way" (Manual) | The "New Way" (AI-Assisted) |
|---|---|---|
| Budget | Required tens of thousands of dollars | Primarily requires time |
| Development | Professional developers needed | AI Agent (Shelley) |
| Speed | Years of conceptualization | Prototypes in hours |
| Accessibility | Limited to high-end hardware | Cloud-based / Parallel VMs |
Enter the AI Agent: Shelley
Hickisch had a backlog of ideas since 2022, but no capital to realize them. That changed when he utilized exe.dev and its AI agent, Shelley. He provided an expanded version of his notes and issued a specific technical prompt:
Make an app 5mp.globe that displays global nature conservation effort based on
GPS Exchange Format (GPX) files representing tracks of movement of park rangers,
vehicles and aircraft that survey protected areas.
Aggregate this over time (daily, monthly, annually) at 100 sq km pixels globally
and display a zoomable global map, that reveals global conservation effort.
The logic is explained in the text file Five Megapixels of Global Conservation.txt.
Registered protected area managers, governments and non-profits can upload their data.
Monthly and annual reports per protected area as outlined on protectedplanet.org
are prepared, the megapixel map can be downloaded at daily resolution.
The Result: Five Megapixel Conservation
Shelley integrated various public API feeds, producing a prototype in mere hours. While the final polished version required roughly 300 hours of Hickisch's time, the experience was transformative. He no longer needed a massive budget—only his own persistence.
Scaling Up: 40 Terabytes and 100 VMs
Hickisch shifted his focus to his home country, Austria, which releases annual LIDAR scans and high-res aerial photography. These files are massive (up to 15GB each), making them virtually untouchable for the average user.
To solve this, Shelley implemented vsicurl, allowing the software to read small "slices" of data rather than downloading the whole file. Hickisch then deployed a massive parallel processing architecture:
By combining tree height with altitude, Hickisch can estimate the age of the forest using a proxy logic similar to:
Current Progress:
- Develop parallel VM architecture
- Implement ML classification model
- Process remaining 70% of the country
- Store results on Zenodo (CERN's open-access repository)
To date, he has processed 40 terabytes of raw data, covering 30% of Austria.
Gamifying Conservation
To make this staggering amount of data accessible, Hickisch created a "Settlers-style" video game. In this simulation, players:
- Purchase parcels of land.
- Engage in reforestation.
- Hunt for hidden treasures.
The "magic" is that the game is powered by two terabytes of actual LIDAR data. When a player clicks a plot of land, they see the real-world height of the tallest trees on that specific property.
"It’s a stupid game, but the powerful thing is the two terabytes of data that are underlying it," Hickisch notes.
The Bigger Picture
For Hickisch, this isn't about profit—the apps are open source. It is about leverage. Conservationists have historically been outmatched by the technical resources of industrial developers. Now, the playing field is level.
His ultimate goals include:
- Seeing at least two countries adopt
5mp.globefor park monitoring. - Empowering individuals in places like the Central African Republic to build their own web apps.
- Democratizing custom software development for the planet's protectors.