I would like to be able to use the spatial data returned by earthaccess.search_data. (I want to visualize the granule spatial coverage in an ROI without downloading the data. I am also interested in intersecting the granule bounds with more complex shapefiles not supported by the search_data API before download.)
Currently, I'm doing this with
geoms = []
for g in granules:
coords = g['umm']['SpatialExtent']['HorizontalSpatialDomain']['Geometry']['GPolygons'][0]['Boundary']['Points']
geoms.append(shapely.geometry.Polygon([(p['Longitude'], p['Latitude']) for p in coords]))
gdf = gpd.GeoDataFrame(geometry=geoms, crs="EPSG:4326")
This feels like odd/fragile code and it relies on undocumented (afaict?) features of the API, I'm not sure if this format is guaranteed to be stable in future versions.
Would it be possible for the API to return a GeoDataFrame of results? Or it would also be helpful to have a documented result class with data accessors, or to have the polygons returned in a standard (e.g. WKT, geojson) format.
Thanks so much for your work on this!
I would like to be able to use the spatial data returned by
earthaccess.search_data. (I want to visualize the granule spatial coverage in an ROI without downloading the data. I am also interested in intersecting the granule bounds with more complex shapefiles not supported by thesearch_dataAPI before download.)Currently, I'm doing this with
This feels like odd/fragile code and it relies on undocumented (afaict?) features of the API, I'm not sure if this format is guaranteed to be stable in future versions.
Would it be possible for the API to return a GeoDataFrame of results? Or it would also be helpful to have a documented result class with data accessors, or to have the polygons returned in a standard (e.g. WKT, geojson) format.
Thanks so much for your work on this!