NOAA-OWP / wres

Code and scripts for the Water Resources Evaluation Service
Other
2 stars 1 forks source link

As a user, I want to obtain WRES results in a GeoJSON format #163

Open epag opened 3 weeks ago

epag commented 3 weeks ago

Author Name: Hank (Hank) Original Redmine Issue: 127159, https://vlab.noaa.gov/redmine/issues/127159 Original Date: 2024-02-29


Came up in #123635, but its been talked about at other times, as well. I don't see an existing ticket, but, if one exists, we can reject this as a duplicate.

Specifically, we want to review the WPOD GeoJSON format Jason mentions in #123635-135. If that is a reasonable format, we would like to support it directly. If not, then we would like to support an alternative that would be more reasonable.

Marking urgent, because we may want to support it soon for WPOD.

Hank

epag commented 3 weeks ago

Original Redmine Comment Author Name: Hank (Hank) Original Date: 2024-02-29T12:02:05Z


I related this to the SQLite output ticket, since I see both as being about making our outputs more easily accessible to users.

Hank

epag commented 3 weeks ago

Original Redmine Comment Author Name: Jason (Jason) Original Date: 2024-03-05T13:28:23Z


I attached the GeoJSON output we delivered to WPOD for the v2.1 Retrospective evaluation. Field descriptions can be found in the table here: https://github.com/NOAA-OWP/ISED-Workbench/tree/jregina/retro-refactor#data-description

epag commented 3 weeks ago

Original Redmine Comment Author Name: Jason (Jason) Original Date: 2024-03-05T13:34:59Z


The biggest difference between this GeoJSON output and the WRES GeoCSV is that the GeoJSON is "wide" and the WRES output is "long." The original evaluation output was in GeoPackage format that allowed "non-unique rows" in terms of features. That is the same POINT feature could appear multiple times across different rows (due to the use of multiple thresholds). The result was that when loading these data into a GIS program you saw (or didn't see, technically) multiple overlapping POINT features at each location. The WPOD GIS team found this confusing and difficult to use with ArcGIS online tools. The solution we settled on was a relatively simple GeoJSON file where a particular POINT feature only appeared once in the attribute table. We could do this because all features used a single threshold ("high water").

epag commented 3 weeks ago

Original Redmine Comment Author Name: James (James) Original Date: 2024-03-05T13:48:42Z


Thanks for the additional info. We'll need to take a look at this compromise when implementing for wres - we want something useful to wpod, but also something that is generally useful (which this may be) and can handle the diverse set of shapes produced by wres (e.g., up to N geographic features per pool, many thresholds etc.). Geopackage/sqlite is another thing on our agenda (ticket exists), which is extremely versatile.

epag commented 3 weeks ago

Original Redmine Comment Author Name: James (James) Original Date: 2024-06-25T09:43:46Z


Noting #123635-386. Perhaps the GeoJSON is OBE?