mheberger / delineator Goto Github PK

I'm trying to run the delineation on a batch of points, and some of them are outside the merit basins. I tried using the buffered merit basins but I run into a nan error cause it's outside of the catchment basins, so there's no terminal_comid for it. Is there a fix for this? What would you recommend I do so they are delineated?

Some of such points are:
{'id': 217697, 'lat': 49.348897, 'lng': -124.428403}
{'id': 217808, 'lat': 49.23117, 'lng': -123.97092}
{'id': 217714, 'lat': 50.6102, 'lng': -127.2367}

I'm running the delineator tool on over 300 points in my basin csv, and I get this error after 228 watersheds have been delineated:
IndexError: index 0 is out of bounds for axis 0 with size 0

Full snippet of error:

All the data I'm using is as provided in the README except my basins csv:
basins.csv

How different are the standard catchment files to the ones marked 'minor bug fix for coastline pixels'. I tried batch delineating a couple of points using the standard catchment shapefiles, and for a point in basin 78, there seems to be a missing feature / gap in the polygon. I noticed that this is not the case in the bug-fixed file and was wondering which one was better to use in general

Hi Matthew!

I currently try to delineate catchments in Germany (Basin 23).
I followed your instructions from the front page to download the data for Germany and set up everything in config.py.
When I run python delineate.py in the terminal, I get the following output and error:

$ python delineate.py 
/home/alexander/Github/camels/camelsp/merit_hydro/delineator/venv/lib/python3.9/site-packages/geopandas/_compat.py:123: UserWarning: The Shapely GEOS version (3.10.3-CAPI-1.16.1) is incompatible with the GEOS version PyGEOS was compiled with (3.10.4-CAPI-1.16.2). Conversions between both will be slow.
  warnings.warn(
/home/alexander/Github/camels/camelsp/merit_hydro/delineator/venv/lib/python3.9/site-packages/geopandas/_compat.py:123: UserWarning: The Shapely GEOS version (3.10.3-CAPI-1.16.1) is incompatible with the GEOS version PyGEOS was compiled with (3.10.4-CAPI-1.16.2). Conversions between both will be slow.
  warnings.warn(
Reading your outlets data in: /home/alexander/Github/camels/camelsp/merit_hydro/data/outlets.csv
Finding out which Level 2 megabasin(s) your points are in
Your watershed outlets are in 1 basin(s)

Beginning delineation for 10 outlet point(s) in Level 2 Basin #23.
Reading catchment geodata in /home/alexander/Github/camels/camelsp/merit_hydro/data/shp/merit_catchments/cat_pfaf_23_MERIT_Hydro_v07_Basins_v01.shp
Traceback (most recent call last):
  File "/home/alexander/Github/camels/camelsp/merit_hydro/delineator/delineate.py", line 701, in <module>
    delineate()
  File "/home/alexander/Github/camels/camelsp/merit_hydro/delineator/delineate.py", line 449, in delineate
    catchments_gdf.to_crs(crs, inplace=True)
  File "/home/alexander/Github/camels/camelsp/merit_hydro/delineator/venv/lib/python3.9/site-packages/geopandas/geodataframe.py", line 1364, in to_crs
    geom = df.geometry.to_crs(crs=crs, epsg=epsg)
  File "/home/alexander/Github/camels/camelsp/merit_hydro/delineator/venv/lib/python3.9/site-packages/geopandas/geoseries.py", line 1124, in to_crs
    self.values.to_crs(crs=crs, epsg=epsg), index=self.index, name=self.name
  File "/home/alexander/Github/camels/camelsp/merit_hydro/delineator/venv/lib/python3.9/site-packages/geopandas/array.py", line 762, in to_crs
    raise ValueError(
ValueError: Cannot transform naive geometries.  Please set a crs on the object first.

I think this error occurs because there is no .prj file for the merit catchments on Google Drive (linked in your instruction): https://drive.google.com/drive/folders/1owkvZQBMZbvRv3V4Ff3xQPEgmAC48vJo

Your sample data for Iceland includes the file cat_pfaf_27_MERIT_Hydro_v07_Basins_v01.prj which I cannot find for my area.

There are also some incompatibility warnings at the beginning of the output but I do not think that those are causing the problem.

I tried to run python delineate.py in a fresh conda and a fresh venv environemnt, both produce the same error.

Let me know if you need anything else and thank you!

When I run the delineator script, I come across this error:
AttributeError: 'MultiPolygon' object has no attribute 'exterior'
It stems from the line calling the get_subdivided_merit_polygon function in delineator
poly, lat_snap, lng_snap = py.merit_detailed.get_subdivided_merit_polygon(wid, basin, lat, lng, catchment_poly, bSingleCatchment)

and subsequently in this line in the merit_detailed script:
filled_poly = Polygon(poly.exterior.coords)

looks like my catchment_poly is a Multipolygon, so converting it to a polygon and then back to multipolygon doesn't work, but if I skip those steps, it still doesn't work

An ecoregion is "an area with characteristic flora, fauna, and climatic conditions, and related human settlement patterns, and can be comprised of several different watersheds, if the plants and animals are similar." (More on watersheds, ecoregions, and bioregions).

This is useful to organizing and planning in ways that are more ecologically focused (as opposed to relying on administrative demarcations). Also, from the experience of the Cascadia Bioregion in the PNW, many people are interested in seeing their local ecoregions — and don't quite know where to begin or find it, as this may require more technical mapmaking and natural science experience.

Given that ecoregions are a superset of watersheds, and @mheberger's excellent watershed tool , would an ecoregional map project interest you? This could take many different forms. It could simply be an image of a map/shapefiles, which illustrates ecoregions around the world. That shapefile likewise be a layer in online interactive maps, including a layer in the Global Watershed map. It could likewise be an interactive map, which allows you to traverse different scales — from local watersheds, to wider ecoregions, to widest bioregions ("bioregions: the full extent of watersheds within an interconnected area. Bioregional borders tend to be jagged, and hard, such as mountain ranges, peaks, ridges, volcanoes, continental uplifts, tectonic plates and faults, defining how energy flows within a set of boundaries.")

As I write this, it does indeed seem that analytical work to identify ecoregions and bioregions could come first — that analysis could be rendered into shapefiles, which could then easily be integrated into interactive online maps etc.

At the same time, perhaps there could be online tools and data to help bootstrap and assist with that analysis. For example, taking into account the different characteristics relevant to ecoregions:

• Watersheds: (https://mghydro.com/watersheds/)
• Flora: the PlantNet API and AI GeoSpecies Service, which predict local flora based on a given location.
• Fauna: (not sure about tools for this)
• Climactic Conditions, Geology, Soil composition, Rainfall patterns: There are several online datasets and APIs for this data
• Human Settlement Patterns: Perhaps census-related APIs, like the American Community Survey and other US Census APIs? Human cultures and ways of living may also play a role here, which could require a more anthropological/sociological touch.

Is a project/collaboration like this something that would interest you?