Merging points to linestrings using GeoPandas

Question

I have two GeoPandas DataFrames:

1. points (long,lat) : 60,000 rows
2. Jams segment (LineStrings) : 204,000 rows

I am trying to create a new DataFrame with the points that are located on LineStrings.

I tried using a buffer on the LineStrings and the predicate: contains,intersect but it didn't succeed

len(gdfTraffic_df)
201053
len(gdfAlerts_df)
58844

gdfTraffic_df.buffer(0.1)
merged = gdfTraffic_df.sjoin(gdfAlerts_df, predicate="contains")
len(merged)
474646

Something is not going as expected with the action of spatial join. Any ideas how to solve the problem?

Answer 1

You don't change the lines to polygons so you are trying to contain points within lines.

Try this:

import geopandas as gpd

lines = gpd.read_file(r"C:\GIS\data\lines.shp")
points = gpd.read_file(r"C:\GIS\data\points.shp")

lines["geometry"].buffer(100) #This just return buffered geometries, you dont change the line geometries to polygons.
lines.sjoin(points, how="inner", predicate="contains")
#So you get: Empty GeoDataFrame

lines["oldgeom"] = lines["geometry"] #Save the line geometries, for later use
lines = lines.set_geometry(lines["geometry"].buffer(100)) #Set the geometries of the df to polygons

lines = lines.sjoin(points, how="inner", predicate="contains")
#lines.shape
#Out[18]: (20, 9) Now there are matches

lines = lines.set_geometry(lines["oldgeom"]) #Set the geometries to the lines again.
del(lines["oldgeom"]) #Drop the column or shape export might fail.