nyc_trees.py

"""
Data sources:
* NYC neighborhoods:
https://data.cityofnewyork.us/City-Government/Neighborhood-Tabulation-Areas-NTA-/cpf4-rkhq
* NYC street trees:
https://data.cityofnewyork.us/Environment/2015-Street-Tree-Census-Tree-Data/pi5s-9p35
"""
import geopandas as gpd
import shapely.ops
from rtree import index


df_neighborhoods = gpd.read_file('Neighborhood Tabulation Areas (NTA).geojson')
df_trees = gpd.read_file('2015 Street Tree Census - Tree Data.geojson')

# 1. First, we create a polygon representing the Upper East Side boundary
# Extract relevant neighborhoods
lenox_hill = df_neighborhoods.loc[df_neighborhoods.ntacode == 'MN31'].geometry
yorkville = df_neighborhoods.loc[df_neighborhoods.ntacode == 'MN32'].geometry
carnegie_hill = df_neighborhoods.loc[df_neighborhoods.ntacode == 'MN40'].geometry

# Both Lenox Hill and Yorkville have islands we want to exclude
lenox_hill = max(lenox_hill.to_numpy()[0], key=lambda x: x.area)
yorkville = max(yorkville.to_numpy()[0], key=lambda x: x.area)
carnegie_hill = max(carnegie_hill.to_numpy()[0], key=lambda x: x.area)

# Merge above neighborhoods to form Upper East Side
geom_upper_east_side = shapely.ops.unary_union([lenox_hill, yorkville, carnegie_hill])

# 2. Next, we create an R-tree index holding all NYC trees
idx = index.Index()
for id, geom in zip(df_trees.index, df_trees.geometry):
    idx.insert(id, geom.bounds)

# 3. Now, we generate a list of 'potential candidates', i.e. all trees that are within the bounding box of the Upper East Side
search_window = geom_upper_east_side.bounds
potential = list(idx.intersection(search_window))

# 4. Finally, we iterate through all potential candidates to extract the ones that are fully within the Upper East Side
trees = []
for p in df_trees.loc[potential, 'geometry']:
    if geom_upper_east_side.contains(p):
        trees.append(p)

# 5. Print results
print('Total number of trees in dataset: %d' % len(df_trees))
print('Total number of trees tested: %d' % len(potential))
print('Number of trees in Upper East Side: %d' % len(trees))