Weaving together data from FEMA, NYC’s 311 hotline, and the Census, this scrollable, map-based visualization tackles this common misconception.

Design Strategy

This visualization deliberately moves from the abstract to the concrete. By first unpacking what FEMA’s ‘high risk’ flood prediction means for New York City and then illustrating how reality confirms (and in fact exceeds) this projected risk, I ground FEMA’s complex statistics in a tangible illustration that anybody can grasp.

The shift from choropleth maps to satellite imagery aids in this transition from the theoretical to the accessible. As the imagery of Flatbush zooms in, the viewer can see for themselves what science predicts: that dense neighborhoods, even miles from the coast, are indeed at significant risk.

By connecting nebulous probabilities to indisputable events and relatable neighborhood views, this visualization challenges the defensive ‘flooding won’t ever happen here’ mindset so common in today’s society. As the satellite view of Flatbush’s streets fades into the murky reflection of an average New Yorker trying to cross a waterlogged street, the viewer is forced to reassess. Flatbush, they’re compelled to admit, looks like any other neighborhood in New York. “The person in the water?”, they ask themselves, “Well, I guess that could be me.”

Math

I derived the numbers in the beginning sidecar text from the following probability calculation:

FEMA defines a ‘high risk’ area as one that has a 1% chance of flooding each year. What is the chance that it floods at least once in a century?

You might think that because the chances of it flooding in a given year are 1 in 100, and a century is 100 years, the probability that it floods in a century are 1%. This discounts the fact that this 1% chance is going to be repeated over and over though. The probability actually compounds as follows:

We see this same compounding in more intuitive examples. Imagine one of your friends was going to go sky diving just to try it out once and another was going to take a job as a sky diving coach that would require them to go every day. You’d probably be more worried about the friend taking the job as a coach, right? That’s because the probability of an unlikely outcome happening always goes up if the event is repeated over and over. The same way your friend is more likely to get hurt if they go sky diving every day versus trying it out once, a high risk area is much more likely to flood in 100 years than it is to flood in one.

This is frequently confused by the fact that FEMA often refers to these 1 in 100 annual flood risk areas as 100-year floodplains. This phrase strongly, and incorrectly suggests that ‘high risk’ areas are only likely to flood once in a hundred years. Because so many interpret this designation incorrectly, before showing which areas are high risk in New York City, I found it important to illustrate in my visualization the correct interpretation of this statistic.

Code

Before designing the scrollable narrative in ESRI’s StoryMaps software, I used Python to process the data. This involved pulling hundreds of thousands of 311 records via the NYC Open Portal API and parsing them to remove duplicate incidents (ie. floods reported by multiple citizens) and non-flood related complaints. Using open source administrative boundaries, I then merged these records into shapefiles for mapping. I also used FEMA’s documentation to aggregate granular flood risk codes into the designated high- and mid-risk categories. Using a statistical technique, my Python code also enabled me to choose appropriate bins for the coloring in the final choropleth maps.

Continuing Work

I’m currently working with the Spatial Analysis and Visualization Initiative at Pratt Institute to explore how other environmental issues could benefit from spatial exploration and visual storytelling. The extensions we are exploring include:

1. Who does flooding disproportionately affect and what other factors influence the ability of these groups to react and relocate? (American Community Survey data)

2. How will FEMA’s disputed flood risk maps affect New York City insurance premiums, and where will new mandatory premiums exceed New Yorkers’ ability to pay? (Aggregation of millions of FEMA NFIP insurance plan records + American Community Survey income and mortgage data)

3. Where can New Yorkers who are unable to pay mandatory insurance premiums relocate and what factors stand in their way? (NYC Department of City Planning Zoning data)