

Simulation after 10 iterations of an enormous cell (albedo: 0.3, greenHouse: 0.3)

Simulation after 10 iterations (albedo: 0.3, greenHouse: 0.3)

Simulation after 10 iterations (albedo: 0.3, greenHouse: 0.4)

Simulation after 10 iterations (albedo: 0.2, greenHouse: 0.2)

Simulation after 10 iterations (albedo: 0.3, greenHouse: 0.5)

Simulation after 10 iterations (albedo: 0.2, greenHouse: 0.0)
Inspiration
We picked this idea from the project ideas of the hackathon.
What it does
Our application is a simulation of a simple climate model. It computes the average temperature in relation to the latitude with some graphs for visualization.
How we built it
We were in a team of 4 : two programmers and two math/physics guys.
Math/physics side: To make this climate model, we first had to scale the amount of sunlight received by each latitude. After, we had to consider that earth's albedo rises when covered in snow/ice, therefore, if the average temperature of a latitude drops down below 10 C, albedo raises. Finally, we took into account those two constants : the opposite factor and the greenhouse effect.
Programming side: Basically, the application is iteration based, so we can control how many years have passed since the beginning of the simulation. Depending on how many years have passed, we compute the change of temperature using the Earth.py class which stand for an abstraction of all those complicated math/physics guys formulas. After, we put those temperature in some matplotlib graphs for visualization
Challenges we ran into
Understanding physics behind all those climate interactions and figuring out that we had to numerically solve second degree differential equation
What we learned
Math/physics side: Learning the physics aspect of all those climate equations.
Programming side: We never used python or any of the technology we used before, so it was a completely new environment for both of us.
What's next for Snowball Earth Theory
The image visualization of the temperature per latitude didn't work as expected, but we didn't had enough time to correct it, so maybe in the future
Log in or sign up for Devpost to join the conversation.