techmatters / soil-id-algorithm

GNU Affero General Public License v3.0
1 stars 0 forks source link

Soil ID: Unexpected ecological site listed for soil #157

Closed DerekCaelin closed 2 weeks ago

DerekCaelin commented 3 months ago

App Version

169

Account

No response

Platforms

Description

Soil scientist encountered Ecological Site Description R084CY168TX, Claypan 37-43 PZfor a soil he expects to be Sandy Loam

Steps to Reproduce

  1. Create site at 32.49718, -97.36264
  2. View Soil ID Tile
  3. See issue

Expected behavior

Top Soil Match: Rader Top ESD: Sandy Loam

Soil Web

Legacy App

Actual behavior

Top Soil Match: Rader Top ESD: R084CY168TX, Claypan 37-43 PZ

Additional context

Full email from george:

Noted that the ecological site information was different than I expected for this site. Also, the LCC is missing the subclass. I was expecting the ecological site R084CY198TX, Sandy Loam 37-43 PZ. The map unit has one component, Rader. Rader component is listed as 100% of the map unit. The new version lists the Top Soil Matches as 100% for Rader. If I select 'View Location on Soil Web', Soil Web lists the ecological site as Sandy Loam for this location. The legacy LandPKS version identifies the ecological site as Sandy Loam. However, the new version is listing the ecological site as R084CY168TX, Claypan 37-43 PZ. Is there a background algorithm that determines the ecological site?

DerekCaelin commented 3 months ago

@jjmaynard A bug report from George for you!

paulschreiber commented 3 months ago

This is correct in the new soilweb: http://soilmap4-1.lawr.ucdavis.edu/dylan/soilweb/api/landPKS.php?q=spn&lon=-97.36264&lat=32.49718

{
"mukey": 368202,
"dist_meters": 0,
"closest_pt_x": -97.36264,
"closest_pt_y": 32.49718,
"which": "a",
"cokey": 24129924,
"compname": "rader",
"localphase": null,
"comppct_r": 100,
"compkind": "Taxadjunct",
"majcompflag": "Yes",
"slope_r": 2,
"elev_r": 160,
"nirrcapcl": 3,
"nirrcapscl": "e",
"nirrcapunit": null,
"irrcapcl": null,
"irrcapscl": null,
"irrcapunit": null,
"taxorder": "Alfisols",
"taxsubgrp": "Aquic Paleustalfs"
},
{
"cokey": 24129924,
"ecoclassid": "R084CY194TX",
"ecoclassname": "Sandy Loam 37-43 PZ",
"edit_url": "https://edit.jornada.nmsu.edu/catalogs/esd/084C/R084CY194TX"
},

But we are grabbing the wrong ecological site:

{
"cokey": 24129501,
"ecoclassid": "R084CY168TX",
"ecoclassname": "Claypan 37-43 PZ",
"edit_url": "https://edit.jornada.nmsu.edu/catalogs/esd/084C/R084CY168TX"
}
jjmaynard commented 1 month ago

I fixed indexing issue when joining ESD data to soil component data. PR #198

DerekCaelin commented 3 weeks ago

@paulschreiber did you say this issue was ready for my review?

DerekCaelin commented 2 weeks ago

This is addressed in v202.

paulschreiber commented 2 weeks ago

@DerekCaelin this was not addressed in 202. This was fixed with a backend change.