GeoPandas clip produces MultiPolygons instead of Polygons

Question

I have a large polygon shapefile and a small polygon shapefile. Both shapefiles are in the same coordinate system. What I do next is, clip small polygons within large polygon boundaries using GeoPandas's Clip.

And here comes the question, before clipping, small polygon geometry contains only Polygon features and not MultiPolygon, but after using geopandas clip(), some small polygon geometry transforms to MultiPolygons. Is there some specific reason why it happens? I should use the polygons in further analysis where I assume, because of this which will arise some error or problem.

Note: none of the small, neither large polygons overlay.

I did try to transform the MultiPolygons to Polygons with gpd.explode(), which gave me geopandas.geoseries.GeoSeries and the result below, however it can be noted that some polygon "ID" are missing, I also cannot assign an "ID" to each feature with df['ID'] = df.index(). Then arises an error:

TypeError: 'MultiIndex' object is not callable.

Is there some explanation why I get MultiPolygons instead of Polygons in the gpd.clip() result?

import geopandas as gpd

small = gpd.read_file(r'R:\...\small.shp')
large = gpd.read_file(r'R:\...\large.shp')

smallClip = gpd.clip(small, large)
smallClip_explode = smallClip['geometry'].explode()

print(type(smallClip_explode))
Out[50]: geopandas.geoseries.GeoSeries

gdf = gpd.GeoDataFrame(geometry=gpd.GeoSeries(smallClip_explode))

Answer 1

The reason why it happens can be the following:

Input:

Output:

The resulting pink feature is the MULTIPOLYGON(((1)),((2)))

To tackle this circumstances you can either:

• use the GeoDataFrame.explode(), see this thread for more details: Explanding GeoPandas Multipolygon Dataframe To One Poly Per Line

• use the GeoDataFrame.overlay() that contains a keyword keep_geom_type=None, which can be set to True/False:

  If True, return only geometries of the same geometry type the GeoDataFrame has, if False, return all resulting geometries. Default is None, which will set keep_geom_type to True but warn upon dropping geometries.

  Check this issue on GitHub for more details: https://github.com/geopandas/geopandas/issues/1667

• check the geometry type of your input layers, maybe there were initially some MULTIPOLYGONs in it:

  layer['isMulti'] = layer.apply(lambda row: True if row['geometry'].geom_type != 'Polygon' else False, axis=1)
  MultiFeatures = layer.loc[layer['isMulti'] == True]
  print(MultiFeatures)
  

  See these threads for more details:

  • Excluding multi geometry types into a single Polygon type
  • Get geometry type from Feature Collection for every row
  • Check if a polygon is a multipolygon in Shapely

Answer 2

As was written by @g12345kk in his question:

I did find a way how to transform from MultiPolygons to Polygons without error:

import geopandas as gpd

small = gpd.read_file(r'R:\...\small.shp')
large = gpd.read_file(r'R:\...\large.shp')

smallClip = gpd.clip(small, large)
smallClip_explode = smallClip['geometry'].explode()

gdf = gpd.GeoDataFrame(geometry=gpd.GeoSeries(smallClip_explode))
gdf.reset_index(inplace=True)

gdf['FID'] = gdf.index
fid = gdf['FID']
gdf.drop(labels=['FID'], axis=1, inplace=True)
gdf.insert(0, 'FID', fid)