acsByNTA.py

import pandas as pd
import numpy as np
import geopandas

# ref: https://www1.nyc.gov/assets/planning/download/pdf/data-maps/open-data/nta_acs_2014_2018_datadictionary.pdf?r=1
# Economic variables
columnsToKeepEcon= [
    # "BoroCode",
    # "BoroName",
    "NTACode",
    "NTAName",

    # Commuting to work
    "cw_pbtrnsP", # % commuting to work via public transportation (excluding taxicab)
    # "mntrvtmE", # mean travel time to work (minutes)

    # Occupation
    # "mgbsciartP", # % population employed in Management, business, science and arts
    "srvcP", # % population employed in service occupations
    # "salesoffP", # % population employed in sales and office occupations
    # "infoP", # % population employed information

    # Industry
    # "prfsmgawmP", # % population employed in Professional, scientific, and management, and administrative and waste management services
    "edhlthcsaP", # % population employed in Educational services, and health care and social assistance
    # "artenrafsP", # % population employed in Arts, entertainment, and recreation, and accommodation and food services
    "pubadminP", # % population employed in Public administration
    "gvtwrkrP", # % population government workers


    # Income and benefits
    # "mdhhincE", # median household income ($)
    # "mnhhincE", # mean household income ($)
    "inc_snapP", # % with Food Stamps/SNAP benefits in last 12 months
    "percapincE", # per capita income ($)

    # Health Insurance
    # "hinsP", # % with health insurance
    # "pvhinsP", # % with health insurance who have private insurance
    # "pbhinsP", # % with health insurance who have public insurance
    "nhinsP", # % with no health insurance coverage
    # "nhinsE", # population with no health insurance coverage
    # "emnhinsP", # % employed in labor force with no health insurance
    # "uemnhinsP", # % unemployed in labor force with no health insurance
    # "nlfnhinsP", # % not in labor force with no health insurance

    # Poverty
    "fambwpvP", # % families below poverty line
    "geometry"
]

# Demographic variables
columnsToKeepDemo = [
    "NTACode",
    # Demographics
    # "blnhP", # % Black or African American Alone
    "wtnhP", # % White Alone
    # "hsp1P", # % Hispanic/Latino (of any race) Alone
    ]

def pullInACSGeoJson():
  """
  Pulls in ACS data by Neighborhood Tabulation Area (NTA).
  Selects subset of columns.
  """
  econPath = "data/Neighborhood Tabulation Areas (NTA)/ACS/nta_with_acs_economics.geojson"
  econDf = geopandas.read_file(econPath)
  econDf = econDf[columnsToKeepEcon]

  demoPath = "data/Neighborhood Tabulation Areas (NTA)/ACS/nta_with_acs.geojson"
  demoDf = geopandas.read_file(demoPath)
  demoDf = demoDf[columnsToKeepDemo]
  demoDf["pct_nonwhite"] = 100 - demoDf["wtnhP"]

  return  econDf.set_index("NTACode").join(demoDf.set_index("NTACode")).reset_index()

def pullInStationCSV():
  """
  Pulls in station csv.
  Turns into geopandas so that it can be used in a spatial join.
  """
  path = "data/Stations/MTA_Station_Mapping_from_turnstile_data.csv"
  df = pd.read_csv(path)
  stations = df[~(df.lat.isnull() | df.lat.isna())] # remove empty geometries
  return geopandas.GeoDataFrame(
    stations, geometry=geopandas.points_from_xy(stations["long"], stations["lat"]))

def handleSpatialJoin(ntas, stations):
  """
  Merges stations into their enclosing ntas.
  """
  stationsWithNTA = geopandas.sjoin(stations, ntas, how="left", op="intersects")
  stationColsToKeep = [
    'station_code',
  'station',
  "GTFS_stop_id",
  "C/A",
  "line_name",
  # 'BoroCode',
  # 'BoroName',
  'NTACode',
  'NTAName',
  "geometry", "lat", "long", "unit"]

  return stationsWithNTA[stationColsToKeep]

def main():
  ntas = pullInACSGeoJson()
  stations = pullInStationCSV()
  joined = handleSpatialJoin(ntas, stations)

  ntas.to_file("data/output/acs_nta.geojson", driver='GeoJSON')
  joined.to_csv("data/output/stations_with_ntas.csv", index=False)

if __name__ == "__main__":
  main()