The inspiration for this map came from Rebecca Solnit’s Infinite City; an illustrative atlas that features a collection of maps and writings about San Francisco. One of my favorite maps in the book is titled; “Poison/Palate”. The map shows locations of sites designated as either a ‘palate’, ‘poison’, ‘poison/palate’ or ‘EPA Superfund.’ Farmers markets, organic farms and well known eateries are juxtaposed next to nuclear research laboratories, chemical plants, and Silicon Valley’s legacy of tech waste. The map I made shows only locations of NYC’s farmers markets linked to their nearest superfund site (within city limits, their are many more just outside in other counties and New Jersey). With more time I would include other types of ‘palate’ and ‘poison’ sites such as notable NYC eateries and restaurants. Further user interaction with the data would developed as well; such as the ability to search all sites from a certain distance of an address entered by the user.
Data was obtained from the EPA and NY State open-data. CartoDB is being used to host and render the data live; a PostGIS SQL query links the two data-sets and a subtle light-grey base-map I imported from MapBox puts the data in context. To process the data I used GDAL’s ogr2ogr utility to query out the 5 boroughs of farmers markets from all of NY. I used QGIS to perform a spatial query on the EPA data to only select sites within the NYC boroughs (this could probably be done with ogr2ogr but I’m not certain).
*note: when searching for superfund data the EPA Environmental Dataset Gateway is a good place to start. Superfund sites are also known as “Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA)”, a term that comes up a lot in the EPA websites.
**the list of NYC farmers markets I used appears incomplete. The dataset I originally downloaded from nyc open-data contains more records but the address data is not easily geocodable; there are many addresses like: “Crotona Park South & Clinton Ave, in Crotona Park” instead of the typical format of street address, city, state, zipcode. Given more time I could have reconciled the state and city datasets and included more market locations.