barometric pressure in variable BP_915 not graphed in portal

[emontgomery-usgs]

@kwilcox Several Sandwich, MA files have a variable called BP_915, that isn't plotted in the portal. After being run through convert.py, they have standard_name "air_pressure", long name "Air Pressure" and look like CF, but the variable called Barometric Pressure doesn't get displayed. The Air temperature variable from the same file does. This is a short dataset, so I doubt it 's size that's hanging it up. https://cmgdata.usgsportals.net/#metadata/75379/station

Thanks for looking into this!

[rsignell-usgs]

@emontgomery-usgs can you provide the opendap URLs for an example datasets, both pre and post conversion with convert.py?

I'm thinking something like this, but with the correct datasets: Pre: http://geoport.whoi.edu/thredds/dodsC/silt/usgs/Projects/stellwagen/Data/SANDWICH2016/10691dwvs-cal.nc.html Post: http://geoport.whoi.edu/thredds/dodsC/silt/usgs/Projects/stellwagen/CF-1.6/SANDWICH2016/10691dwvs-cal.nc.html

[emontgomery-usgs]

@rsignell-whoi Here are example URLs of two stations that have BP_915 or BPR_915 in them (neither displays), after the conversion to CF: http://geoport.whoi.edu/thredds/dodsC/usgs/data2/emontgomery/stellwagen/CF-1.6/SANDWICH2016/10601hwlb-a.nc.html http://geoport.whoi.edu/thredds/dodsC/usgs/data2/emontgomery/stellwagen/CF-1.6/SANDWICH2017/10871hwlb-a.nc.html

The URLs of the original form of the same stations are: http://geoport.whoi.edu/thredds/dodsC/usgs/data2/emontgomery/stellwagen/Data/SANDWICH2016/10601hwlb-a.nc.html http://geoport.whoi.edu/thredds/dodsC/usgs/data2/emontgomery/stellwagen/Data/SANDWICH2017/10871hwlb-a.nc.html

On Wed, Mar 21, 2018 at 7:26 AM, Rich Signell notifications@github.com wrote:

@emontgomery-usgs https://github.com/emontgomery-usgs can you provide the opendap URLs for an example datasets, both pre and post conversion with convert.py?

I'm thinking something like this, but with the correct datasets: Pre: http://geoport.whoi.edu/thredds/dodsC/silt/usgs/ Projects/stellwagen/Data/SANDWICH2016/10691dwvs-cal.nc.html Post: http://geoport.whoi.edu/thredds/dodsC/silt/usgs/ Projects/stellwagen/CF-1.6/SANDWICH2016/10691dwvs-cal.nc.html

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/USGS-CMG/usgs-cmg-portal/issues/275#issuecomment-374906302, or mute the thread https://github.com/notifications/unsubscribe-auth/ABsHnsnqdu8wUhU35Dk4G-oifv8b27U-ks5tgjj6gaJpZM4SyUzO .

-- Ellyn Montgomery, Oceanographer and Data Manager U.S. Geological Survey Woods Hole Coastal and Marine Science Center 384 Woods Hole Rd., Woods Hole, MA, 02543-1598 (508) 548-8700 x2356 https://stellwagen.er.usgs.gov; https://woodshole.er.usgs.gov/operations/stg/

[kwilcox]

These variables are being converted to... something else. If you look at the data in the CF files it is below the valid_range value of 800 therefore the portal isn't picking up any valid data. I'll poke around as to why the data is being converted.

[kwilcox]

Solved with https://github.com/USGS-CMG/usgs-cmg-portal/commit/2ad3566837ae1b8445cfcc15c54adda95a81b009. They were being interpreted as sea_water_pressure in PSIA and being converted to dbar

[kwilcox]

@emontgomery-usgs This would have affected all netCDF variables over all projects that had long_name: barometric pressure and no epic_code attribute defined. Have you noticed other projects with missing Air Pressure?

[emontgomery-usgs]

@kwilcox ,We only have a few platforms that made atmospheric measurements. I have not noticed issues with not plotting barometric pressure but will check the others. [checking]

Is the fix implemented and re-harvested? I opened a new portal and 1060, 1070 in sandwich still don't display Barometric pressure, even though the variable is listed. Same for 1075 in West Falmouth Harbor.

The good news is that atmospheric data from 952 in Ogunquit, ME, and 985 off Fire Island both show barometric pressure to be present, and display the data.

Here is the URL to the CF of a file for which the Barometric pressure does plot. The variable name is BP_915, and standard name is 'air_pressure' (which is exactly what 1060 has, so I'm mystified why one plots and the other doesn't). (The URL to 1060 CF is further up in this issue) http://geoport.whoi.edu/thredds/dodsC/usgs/data2/emontgomery/stellwagen/CF-1.6/RCNWR/9521Bhlm-a.nc.html

In collect.py there's this line: for x in ['met', 'hlm', 'alm', 'hwlb']:, it contains all the right types. The ones that aren't plotting now are 'hwlb', so there must be something critical (epic_code attribute maybe?) that's missing from these... FYI if an epiccode attribute is missing, it can always be inferred by reading the numbers following the '' in the variable name. So variables named *_915 map to epic_code 915, air pressure, a variable named u_1205 maps to epic_code 1205, eastward seawater velocity. Maybe this could be used if the absence of epic_code is the problem. Or the EPIC_code attribute can be replaced in the EPIC format files. We may have left it out in some variable processing streams...

[emontgomery-usgs]

@kwilcox I've re-run SANDWICH2016, SANDWICH2017, WFAL_2016 and WFAL_2017, which are the ones for which Air pressure did not plot. I don't think you use my output, but it's there if you want it. I think you'll need to re-run these 4 on your side to get the change you made implemented. Please make sure your source data for Sandwich2016 has 10601hwlb and 10701hwlb.

Thanks!

[kwilcox]

Combining with https://github.com/USGS-CMG/usgs-cmg-portal/issues/277