Folium-ChoropletMap issues with key_on: does not overlay the choroplet map - geojson

Question

I have a strange issue with the following piece of code:
m10=folium.Map(location=[41.9027835,12.4963655],tiles='openstreetmap',zoom_start=5)
df.reset_index(inplace = True)
folium.Choropleth(
geo_data = df.to_json(),
data = df,
columns=['TERRITORIO', var],
key_on='feature.properties.TERRITORIO',
fill_color='YlGnBu',
fill_opacity=0.6,
line_opacity=1,
nan_fill_color='black',
legend_name= get_title_(file_name),
smooth_factor=0).add_to(m10)
folium.features.GeoJson(df,
name='Labels',
style_function=lambda x: {'color':'transparent','fillColor':'transparent','weight':0},
tooltip=folium.features.GeoJsonTooltip(fields=[var],
aliases = [indicator],
labels=True,
sticky=False
)
).add_to(m10)
I use the same piece of code on two different geodataframes. With the first (smaller) dataframe I have no issues.
However, when I try to do the same with the other one I do not see the choroplet map layer.
This is the first dataset (after the reset of the index):
TERRITORIO ... geometry
0 Nord ... MULTIPOLYGON (((9.85086 44.02340, 9.85063 44.0...
1 Centro ... MULTIPOLYGON (((10.31417 42.35043, 10.31424 42...
2 Mezzogiorno ... MULTIPOLYGON (((8.41112 38.86296, 8.41127 38.8...
This is the second dataset (after the reset of the index)
TERRITORIO ... geometry
0 Abruzzo ... MULTIPOLYGON (((930273.425 4714737.743, 930147...
1 Basilicata ... MULTIPOLYGON (((1073851.435 4445828.604, 10738...
2 Calabria ... MULTIPOLYGON (((1083350.847 4416684.239, 10833...
3 Campania ... MULTIPOLYGON (((1037266.901 4449456.848, 10372...
4 Emilia-Romagna ... MULTIPOLYGON (((618335.211 4893983.160, 618329...
These are insted the json files:
first:
{"type": "FeatureCollection", "features": [{"id": "0", "type": "Feature", "properties": {"INDICATORE": "Densit\\u00e0 di verde storico", "NOTE": null, "Shape_Area": 57926800546.7, "Shape_Leng": 2670893.51269, "TERRITORIO": "Nord", "UNITA_MISURA": "per 100 m2", "V_2004": null, "V_2005": null, "V_2006": null, "V_2007": null, "V_2008": null, "V_2009": null, "V_2010": null, "V_2011": 2.4, "V_2012": 2.4, "V_2013": 2.4, "V_2014": 2.4, "V_2015": 2.4, "V_2016": 2.4, "V_2017": 2.4, "V_2018": 2.4, "V_2019": null, "index": 0}, "geometry": {"type": "MultiPolygon", "coordinates": ...
second:
{"type": "FeatureCollection", "features": [{"id": "0", "type": "Feature", "properties": {"INDICATORE": "Densit\\u00e0 e rilevanza del patrimonio museale", "NOTE": null, "Shape_Area": 10831496151.0, "Shape_Leng": 664538.009079, "TERRITORIO": "Abruzzo", "UNITA_MISURA": "per 100 km2", "V_2004": null, "V_2005": null, "V_2006": null, "V_2007": null, "V_2008": null, "V_2009": null, "V_2010": null, "V_2011": null, "V_2012": null, "V_2013": null, "V_2014": null, "V_2015": 0.22, "V_2016": null, "V_2017": 0.13, "V_2018": 0.11, "V_2019": null}, "geometry": {"type": "MultiPolygon", "coordinates":...
I really do not understand why one works and the other does not.
Do you have any suggestion?
Thank you in advance for your time!

Answer 1

I can't find any geometry for Nord, Centro, Mezzogiorno Italy, so have sythesized by dissolving regions geometry
have setup functions and variables used by your code to make this a MWE
can switch between geometries by # regions==False, north/central/south==True if True: both generate appropriate folium maps
it's clear in your question that your two sets of geometry are using different CRS. First data set looks like EPSG:4326 (hence works). Second looks like a UTM CRS (points in meters not degrees) that would need to be projected to EPSG:4326
import folium
import geopandas as gpd
import numpy as np
# make SO code runnable, get some geometry and set columns / variables used by code
df = gpd.read_file("https://github.com/openpolis/geojson-italy/raw/master/geojson/limits_IT_regions.geojson").sort_values("reg_name")
df["var"] = np.random.randint(1,10,len(df))
df["TERRITORIO"] = df["reg_istat_code_num"]
df["NCM"] = np.where(df["reg_istat_code_num"]<9,"Nord", np.where(df["reg_istat_code_num"]<15,"Centro", "Mezzogiorno"))
var = "var"
file_name="regions"
indicator = "some number"
def get_title_(file_name):
return file_name
# regions==False, north/central/south==True
if True:
df = df.dissolve("NCM")
file_name = "ncm"
# unchanged code
m10 = folium.Map(location=[41.9027835, 12.4963655], tiles="openstreetmap", zoom_start=5)
df.reset_index(inplace=True)
folium.Choropleth(
geo_data=df.to_json(),
data=df,
columns=["TERRITORIO", var],
key_on="feature.properties.TERRITORIO",
fill_color="YlGnBu",
fill_opacity=0.6,
line_opacity=1,
nan_fill_color="black",
legend_name=get_title_(file_name),
smooth_factor=0,
).add_to(m10)
folium.features.GeoJson(
df,
name="Labels",
style_function=lambda x: {
"color": "transparent",
"fillColor": "transparent",
"weight": 0,
},
tooltip=folium.features.GeoJsonTooltip(
fields=[var], aliases=[indicator], labels=True, sticky=False
),
).add_to(m10)
m10

Answer 2

Modifying the crs I was able to overcome the above issiue.
# Create the folium map
m10=folium.Map(location=[41.9027835,12.4963655],tiles='openstreetmap',zoom_start=5)
# Data
df.to_crs(crs = 4326, inplace = True)
df.reset_index(inplace = True)
folium.Choropleth(
geo_data = df.to_json(),
data = df,
columns=['TERRITORIO', var],
key_on='feature.properties.TERRITORIO',
fill_color='YlGnBu',
fill_opacity=0.6,
line_opacity=1,
nan_fill_color='black',
legend_name= get_title_(file_name),
smooth_factor=0).add_to(m10)