Ecosystems are complex by their very nature. The importance of the right balance between numerous components is often hard to understand. Builders want more buildings. Homeowners want to be close to the water. Crabbers want more crab. Yet, we clearly can't have more of everything. People often don't understand the relationship among the components of the ecosystem and the delicate balance needed to keep things functioning.
Both developers have a long history building rich visual and interactive simulations, and we were motivated by a Hackathon "consultant" from MDE who suggested building a Bay simulation. We were inspired by the popular games of SimCity and MineCraft as they both show how the richness of the simulation and interactivy is more important than visual realism to convey complexity.
The goal of this project is to help students understand the complexity of the Chesapeake Bay ecosystem, and the nature of the relationships among components. We anticipate that BaySim would work well as a kiosk in an information center, or as an enabling technology in the classroom for middle or high school students.
We think the interactive and visual nature of the simulation makes it unusally engaging and understandable. We proud that we were able to create a fully functional multi-platform (Windows, Mac & Linux) prototype during the hackathon by building on one of the developer's previous educational simulation environment. The code is open source and available on GitHub. It is clearly structured and has parameters that will make it easy for other developers to modify it to create their versions of the simulation and adapt as they see fit. We also used a very flexible model for the bay itself by providing the data in the form of a simple text file with elevations so the simulation could be readily adapted to other regions.
Some technical background:
1) We first had to find a (raster based) dataset that describes the elevations in both land and sea, and then combined and calibrated the separate sea and land dataset. We were surprised that this kind of dataset did not exist. By defining our simulated region by just elevation data, the simulation can work without change to the code simply by changing the data file.
2) The "SimCity" style code that runs the simulation was adapted from a UMD class that one of the developers (Bederson) developed last year to teach Python to non-computer science majors (http://www.cs.umd.edu/~bederson/classes/paths-f13/). That simulation was much simpler and had randomly generated terrain and no sense of pollution, crabs, etc. But the basic simulation mechanism existed which let us focus on adding the details of the Bay ecology to the simulation such as the previously mentioned elevation lab, buildings, pollution run-off, crabs, etc. In addition, by controlling the river input and evaporation rates, the impact on water level can be investigated, thus showing the impact of changing sea levels.
3) The current prototype is written in Python which enables it to run as a downloadable app on Windows, Macs or Linux laptops and desktops. However, we envision that BaySim could be rewritten for the web using standard web technologies (without plugins) so that it could run on any device, including phones and tablets simply by loading a web page. This would enable new modes of collaboration and competition across devices that we think could make BaySim even more exciting!
Log in or sign up for Devpost to join the conversation.