0

I seek help on:

  • Finding possible errors on projections management
  • Guiding me on which should be the best way to know which "labels - colors" are within the path line (not exposed here)

--

Source 1: Geopandas imported from ArcMap in .shp format - It contains all shapes (4k) and is used as the basic map. It's very populated in terms of ArcMap columns and attributes

Source 2: Geopandas imported from KML coordinates in .csv format - It's a ROV path in UTM 31N zone.

Steps done:

Import and plot ArcMap representation in GeoPandas. Done! "The same" as ArcMap Source 1

Import and plot KML path in GeoPandas. Done!

Source 2

Make them work in the same CRS

# print(gdf.crs) == > epsg:32631

# Source 2 is directly readed from CSV. Need to assign geocode.
# For a non GIS guy, makes sense to apply || epsg: 32631 or 25831

# print(track.crs) == > 32633   Value determined automatically. As the input for source 2 is just KML coordinates
track = track.to_crs(gdf.crs)


print(gdf.crs)     # After conversion, both are the same: epsg:32631 
print(track.crs)

# Follow the convention of plotting both geoPandas by sharing an axes
fig, axes = plt.subplots(figsize=(15,15))
gdf.plot(ax=axes, color=gdf['Color'])
track.plot(ax=axes)

Incorrect output:

Can you see the miniature of SOURCE 1?

enter image description here

I assume this point is my KML path:

enter image description here

What I expected

(picture from ArcMap):

enter image description here

2
  • Plot the converted gpd and check if reference system is the same as that of source1. crs might have changed but not the coordinates. May 29, 2021 at 3:27
  • I'm struggling on these.... it seems it's not been converted. I assumed it was as easy as: track = track.to_crs(gdf.crs) [....]
    – ipesanz
    May 29, 2021 at 23:15

1 Answer 1

0

The only workarround I found was to load the KML to ArcGIS, and then, download the tracks of my interest in .shp format.

Then, both CRS of boths Shapes that were perfectly plotted in ArcGIS, were also properly plotted in Python

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.