INTRODUCTION
The rise of global sea-levels is one of many worrisome consequences of global warming. Several nations are expected to experience severe damage within the 21st century, with some even predicted to completely submerge under the ocean. New York City is one of the major cities prone to these consequences, given its exposed waterfront location.
This lab report attempts to address the global warming issue by presenting a geospatial visualization of MTA subway entrances. The map will show which stations are most exposed to future floods as a result of the rising sea-level. Subway station floodings could cause severe disruptions, as New York City is heavily reliant on its subway system. The data for this particular visualization corresponds to the estimated sea level in the year 2050.
INSPIRATION
The inspiration for this visualization lab was initially derived from weather maps, such as those presented on news channel weather broadcasts. I wanted to integrate a weather phenomenon into a geospatial visualization of New York City in order to display some consequence of global warming. My second source of inspiration was the MTA subway map, pictured below. I wanted to combine a subway map with data on future sea level. Although I was inspired by the MTA map, I felt that its map projection was too distorted for the purpose of this visualization. Therefore, I used Carto’s voyager base map, which is projected in the Web Mercator format. I ended up plotting subway entrances instead of subway stations, as I believe this led to higher detail. Subway tracks were added to make the map recognizable as a subway map.
MATERIALS
SOFTWARE: Carto, PowerPoint
DATA: Sea Level 2050, NYC Subway Entrances, NYC Borough Boundaries, NYC Subway Lines
METHOD
Preparation
All data sets used in this lab were downloaded from NYC Open Data as shape-files. Thanks to its sufficient data structures, preliminary data manipulation was not required for this project. The shape-files was imported into the freemium GIS software tool Carto, where the visualization was generated.
Carto
I started off by importing geospatial data of the estimated 2050 sea flood risk level. The color of the layer was assigned to match the color of the sea on Carto’s Voyager base map. I used PowerPoint to identify the color of the sea. This was done by importing a screenshot of the map and analyzing its RGB-code. Furthermore, I made the 2050 flood layer slightly transparent. I believe that this color coding and transparency design successfully demonstrates to the observer that it is the sea level rise that is being visualized. However, despite the 2050 flood layer transparency, it was rather difficult to tell where the coastlines are currently located. Hence, I decided to incorporate a borough boundaries dataset. I removed the fill color from the boroughs but kept the outlines colored in black. This resulted in a clear view of the current coastlines. In addition to defining the current sea-level, the borough boundaries data emphasizes the area of interest, i.e. New York City.
Next, I added the subway entrance data layer, which I was required to insert twice. First to map out all stations, then to add a layer of the stations that overlapped with the 2050 flood layer. I added the data and colored all the stations as green dots. Green color represents the physical lights used to identify subway stations. Then I inserted the same data again, except this time, I colored the stations red and ran the intersect and aggregate analysis. This led to an overlapping of red dots for the stations exposed to the 2050 sea level. The color red was used to signify the negative characteristic of being exposed to flooding. The use of contradicting color hues (red vs. green) was the most logical in this case since the information is in binary form, i.e. in the high-risk zone or not in the high-risk zone.
I added subway rail tracks by incorporating another data set. This enhances the fact that the dots are subway entrances and make them stand out in the visual hierarchy. Unfortunately, the Staten Island subway line was not included in the data set, but I decided to keep those data points. Next, I added pop-up labels with the addresses of all stations. I differentiated the popup windows by assigning a black color to exposed stations and white color to low-risk stations. In regard to map options, I decided to leave out the possibility for users to select and remove layers, as this function would not fulfill any purpose. I added the search function, in order to allow users to search for particular city addresses. Lastly, I created a widget that allows users to search for and highlight certain subway routes in order to investigate e.g. how their local train would be affected.
RESULT & INTERPRETATION
I strived to create a rather simple geospatial visualization in order to make it compelling to a general audience, and I believe I achieved a successful end result. The observer can easily see which stations are in the high-risk zone, and I believe that the visual characteristics of the map generate some level of self-explanation, even without written guidelines.
We can see that the lower-most part of Manhattan, the Brighton Beach area and the Rockaway Beach area are heavily exposed to the future sea-level increase. The take-aways from this geospatial visualization could support city planning and preventive decisions in New York, potentially saving the city from heavy disruptions.
CONCLUSION & REFLECTION
I am satisfied with the end result and how simple the map turned out. I believe the most powerful visualizations are those that are simple and clean. Furthermore, I believe that geospatial visualization is a great way to communicate data and its underlying story. Particularly for addressing climate concerns since climate change is itself a geographical phenomenon.
In extensions of this project, it would be interesting to add additional data layers, such as passenger density. The density would be assigned to the size of the data points to generate proportional symbols. It would also be interesting to incorporate temporal characteristics, e.g. by displaying 5-year intervals between 2020 – 2050, along with an animation showing which subway entrances become flooded for every new 5-year period. I tried to incorporate both proportional data points and flood progression in this project, but due to lack of data, I had to abandon that idea.