Scenic quality allowing line vector structures

[dcdenu4]

@jesssilver mentioned

The scenic quality model - it used to require points for the input called 'Features Impacting Scenic Quality', and the user's guide still indicates that's the case. A person who I'm working with in Alaska input a polyline network and the model seems to have taken it and produced outputs, but I'm trying to replicate their work and getting an error that would lead me to believe that I need to convert to points.

She's right, currently the model does not validate for Point Vector type and will run on a polyline. I reproduced this by creating a polyline in QGIS around the point vector provided in the sample data. The model ran and completed successfully.

Looking at the model, it very much assumes point vector type and doesn't do any fancy line handling that I could see.

[phargogh]

@dcdenu4 could you attach your polyline vector here? I agree, the model is only intended to run on points, so the fact that the model is somehow extracting points from a polyline is a little bit surprising. I'm curious which points the model is extracting in this case.

[jesssilver]

I can upload the inputs from the user. J.

…

On Mon, Mar 13, 2023 at 10:43 AM James Douglass ***@***.***> wrote: @dcdenu4 <https://github.com/dcdenu4> could you attach your polyline vector here? I agree, the model is only intended to run on points, so the fact that the model is somehow extracting points from a polyline is a little bit surprising. I'm curious which points the model is extracting in this case. — Reply to this email directly, view it on GitHub <#1245 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEX6RCWJTPNA4LNBGZIDRJTW35MCVANCNFSM6AAAAAAVYDASTY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

-- Jess Silver The Natural Capital Project, Stanford University c/o School of Environmental and Forest Sciences University of Washington, Box 352100 Seattle, WA 98195 *206-616-8730* (ph)

[dcdenu4]

@jesssilver shared the data with me, I'll upload here.

I'm curious which points the model is extracting in this case.

Good question. The model is just iterating over those line features and extracting a point from the geometry. I do get a nice list of points, but I'm not sure if it's just returning the first point of the line geometry...

from osgeo import ogr, gdal
structure_path ='NAC_Anthropogenic_12012022l.shp'
vector = gdal.OpenEx(structure_path)
layer = vector.GetLayer(0)
for point in layer:
    geom = point.GetGeometryRef()
    viewpoint = (geom.GetX(), geom.GetY())
    print(viewpoint)
...
(435752.73330000043, 1432924.1162999999)
(469712.45109999925, 1304662.856899999)
(465154.60639999993, 1298408.638699999)
(437119.62209999934, 1407142.2459999993)
(438037.83999999985, 1399165.9740999993)
(463563.5987999998, 1335260.3095999993)
(462564.66080000065, 1327714.1539999992)

invest/src/natcap/invest/scenic_quality/scenic_quality.py

Line 494 in 0a4ec9b

viewpoint = (geometry.GetX(), geometry.GetY())

NAC_Anthropogenic_12012022l.zip

[phargogh]

I just cracked the vector open and yes, it looks like GDAL is pulling the coordinates of the first point as its X and Y. Kind of an arbitrary choice if you ask me.