Inspiration

There were flash flood warnings in California during winter break.

What it does

Flood Watch analyzes the risk for a region to flood in advance by considering various factors including recent precipitation, presence of drains,

How we built it

We parsed large datasets of geographical data pertaining to Philadelphia using Python and Javascript, and converted this data into a visual geographical format. We roughly implemented the algorithm described in Wicht and Osinska - Skotak "Identifying urban areas prone to flash floods using GIS - preliminary results" to determine the risk of flood.

Challenges we ran into

For some factors we were unable to find sufficient data pertaining to Philadelphia, which potentially compromises the accuracy of the model for Philadelphia. It was also difficult navigating through the various file types used in representing geographical data and converting between them.

Accomplishments that we're proud of

Successfully avoiding floods at least during PennApps

What we learned

Parsing data is hard.

What's next for Flood Watch

Watching out for more floods

Share this project:
×

Updates