Aggregate data from source value to mask layer by overlapping area

Question

Let's assume I have a georeferenced data set seed_layer.geojson and a mask layer leech_layer.geojson.

test_leech.geojson:

{
"type": "FeatureCollection",
"features": [
{ "type": "Feature", "properties": { "content_leech": 1000 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 0.0, 1.0 ], [ 0.0, 3.0 ], [ 3.0, 3.0 ], [ 3.0, 1.0 ] ] ] } },
{ "type": "Feature", "properties": { "content_leech": 2000 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 4.0, 1.0 ], [ 4.0, 3.0 ], [ 6.0, 3.0 ], [ 6.0, 1.0 ]] ] } }
]
}

test_seed.geojson:

{
"type": "FeatureCollection",
"features": [
{ "type": "Feature", "properties": { "content_seed": 10 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 0.0, 0.0 ], [ 0.0, 2.0 ], [ 2.0, 2.0 ], [ 2.0, 0.0 ] ] ] } },
{ "type": "Feature", "properties": { "content_seed": 10 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 0.0, 2.0 ], [ 0.0, 4.0 ], [ 2.0, 4.0 ], [ 2.0, 2.0 ]] ] } },
{ "type": "Feature", "properties": { "content_seed": 10 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 2.0, 0.0 ], [ 2.0, 2.0 ], [ 4.0, 2.0 ], [ 4.0, 0.0 ] ] ] } },
{ "type": "Feature", "properties": { "content_seed": 200 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 2.0, 4.0 ], [ 4.0, 4.0 ], [ 4.0, 2.0 ], [ 2.0, 2.0 ] ] ] } },
{ "type": "Feature", "properties": { "content_seed": 10 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 4.0, 0.0 ], [ 6.0, 0.0 ], [ 6.0, 2.0 ], [ 4.0, 2.0 ] ] ] } },
{ "type": "Feature", "properties": { "content_seed": 10 }, "geometry": { "type": "Polygon", "coordinates": [ [ [ 6.0, 2.0 ], [ 6.0, 4 ], [ 4.0, 4.0 ], [ 4.0, 2.0 ] ] ] } }
]
}

I plot both layers ...

import geopandas as gpd

leech_layer = gpd.read_file('../temp/test_leech.geojson')

seed_layer = gpd.read_file('../temp/test_seed.geojson')

base = seed_layer.plot(figsize=(6,3), column='content_seed', legend=True, cmap='Accent', edgecolor='black', alpha=0.75)
base.text(1,1, '0 (10)')
base.text(1,3, '1 (10)')
base.text(3,1, '2 (10)')
base.text(3,3, '3 (200)')
base.text(5,1, '4 (10)')
base.text(5,3, '5 (10)')
leech_layer.plot(ax=base, color='red', alpha=0.5)

... and receive the following figure:

What's the best practice in GeoPandas to aggregate values in a column seed_layer.geojson (e.g. content_seed) to the geometries in leech_layer.geojson as the sumproduct of overlapping area and the value in the content_seed column?

Example: The red shape on the left side would be assigned the following value:

(2*20+2*20+1*10+1*200)/6 = 41.67

Answer 1

One possible way is as follows. I am sure though, there are nicer ways!

def seedtoleech(seed_layer,leech_layer,seed_column):      

    # First, the index is reset.
    # Both indices will be transfered to the gpd.overlay() layer
    # to allow identifying overlaps.

    seed_layer = seed_layer.reset_index()
    leech_layer = leech_layer.reset_index()

    # Add area 
    leech_layer = add_area(leech_layer)
    seed_layer = add_area(seed_layer)

    # Create overlay layer
    overlay_layer  = gpd.overlay(seed_layer,leech_layer)

    # Add area of the overlay area
    overlay_layer = add_area(overlay_layer)

    # Add new columns to the overlay layer
    # The relative overlay
    # The relative_overlay multiplied with the seed value
    overlay_layer['area_rel'] = 0.0
    overlay_layer['seed_value'] = 0.0

    for i in range(len(overlay_layer)):
        overlay_layer.at[i,'area_rel'] = overlay_layer.at[i,'area']/overlay_layer.at[i,'area_2']
        overlay_layer.at[i,'seed_value'] = overlay_layer.at[i,'area_rel']*overlay_layer.at[i,seed_column]

    # seed_value is the area-weighted seed value

    overlay_layer = overlay_layer.drop(['index_1', 'area_1','area_2','area','area_rel','content_seed'], axis=1)

    # Dissolve the overlay layer by index_2 (the geometries of the leech layer) and some the leeched/seeded value
    merged_layer = overlay_layer.dissolve(by='index_2', aggfunc='sum')
    return(merged_layer)