fraction of soil class wth small shapes of a shapefiles

[ShervanGharari]

The soil class fraction for a shape within the provided shapefile Steppler_shapefile.zip is 0.
the output looks like this (example):

COMID, frac_12
123, 1
124, 1
125, 0
126, 1
127, 1

It seems all the shapefiles have positive areas and are valid. I am wondering if you can take a look @kasra-keshavarz or reproduce this issue. It can be very well that the shapefile is smaller than the grid. I did not explore this but might cause an issue for exactextract.
Also sorry, it seems scratch is not working, otherwise I would have copied past the actual output file.

[kasra-keshavarz]

I will have a look, thanks.

[kasra-keshavarz]

I cannot find a COMID with 125 value inside the geometry file. The attached screenshot is all I see:

[ShervanGharari]

yes, they are PolygonId, now I have access to the file from scratch. it looks like this:

PolygonId,lat,lon,frac_12
3,49.3335105341665,-98.3644685359446,1
10,49.3275226774114,-98.370363499065,1
11,49.3289547776079,-98.3906932890098,1
12,49.3282924861091,-98.3818005323894,1
13,49.3277355933317,-98.3666231385368,1
14,49.3271788959891,-98.3906642418226,1
15,49.3266241088553,-98.3885633609083,1
16,49.3269500028119,-98.3802684664347,1
17,49.32618004397,-98.3593108346827,1
18,49.3244355344646,-98.3909474941817,1
19,49.3228138123735,-98.3889258661066,1
21,49.3322670278447,-98.3661465185802,1
24,49.3308761146068,-98.3675766704112,1
26,49.3345258398447,-98.3618156592226,1
27,49.3316590643539,-98.3655269498655,1
28,49.3253373050675,-98.380462708457,1
29,49.3297748037927,-98.3881149420671,1
30,49.3296758726058,-98.3872957837495,1
31,49.3295857594588,-98.3827450445958,1
32,49.3282652171595,-98.3573339741446,0
33,49.3279296238513,-98.3728481827443,1
34,49.3295594163184,-98.3748796615995,1
35,49.329834381477,-98.3704316142598,1
36,49.3310410364066,-98.3665894839823,1
37,49.3307253659404,-98.3655096585408,1
38,49.3284934805845,-98.3616227644936,1
39,49.33140027253,-98.359595091066,1
40,49.3352603866151,-98.3578874015903,1

PolygonId or COMID the output should be the same(?). the frac_12 is 0 for 32. That is the issue.

[kasra-keshavarz]

I will check this and get back to you.

With newer versions of gistool, you can give the ID column of the geometry file as to the --fid argument using which you want the zonal stats be report for.

[ShervanGharari]

it is the latest gistool I assume. We were working on this a few days ago. The config is here:
https://github.com/ShervanGharari/FHIMP/blob/ERA5MODIS/hydro-models/02_model_agnostic/model-agnostic.json#L62
This has another issue one of the files for land cover or soil was not generated or it was empty.
I would suggest the following to see if you can reproduce both:
1- git clone the ERA5MODIS branch and see if you can get the final result from GIS tool. note that the domain is pfaf 28 which is set there.
2- replace the shapefile with the shapefile from here to see also if you can reproduce what we have seen with fract_12
3- there is another issue on top of all for datatool which seems to stop when the shapefile (the small one is given). I had to manually adjust the min and max lat lon.
So to make it short there are three issues. I think the first is to see if you can reproduce each and every one of them. We can then systematically address this.
sorry if this is a bit too much for this issue. We can break it to datatool or FHIMP.

[ShervanGharari]

For the soil issue and the shared shapefile here, I suspect this might be an issue with exactextract for shapefiles that is small that fall in just one grid maybe. Just a speculation, I did not have much time to check this in detail, I might come back to this later this week.

[kasra-keshavarz]

Since the usda_mode_soilclass_250m_ll was problematic, I just changed the default variable to the raster with the global coverage. So, this variable does not exist anymore. Does this resolve the issue you think?

[ShervanGharari]

ok let me try one more time with the new gistool and will let you know. I think, I need to report the issues here more systematically. Thank you.