[ecco-support] missing values in ECCO V4r4 monthly interpolated climatology

[Wang, Ou (US 329B)]

Hello Haakon,

You were looking at an earlier version of ECCO V4r4. There is a newer version available with improved metadata and flags for dry points. You can access it at https://podaac.jpl.nasa.gov/ECCO. Please use this updated version. It's worth noting that the two versions have identical model fields (THETA, in your case) at wet points.

Click the URL above. Scroll down slightly and click on “Data” in the left. This will bring you to a list displaying various V4r4 fields. From there, click on “ECCO Ocean Temperature and Salinity - Monthly Mean 0.5 Degree (Version 4 Release 4)<https://podaac.jpl.nasa.gov/dataset/ECCO_L4_TEMP_SALINITY_05DEG_MONTHLY_V4R4>”. For your convenience, here is the direct URL to the field for monthly mean THETA on the regular 0.5x0.5 degree lat-lon grid:
https://podaac.jpl.nasa.gov/dataset/ECCO_L4_TEMP_SALINITY_05DEG_DAILY_V4R4

Instructions for downloading ECCO V4r4 from PO.DAAC can be found at
https://ecco-group.org/products-ECCO-V4r4.htm. Scroll down slightly and click on the “PO.DAAC Downloading Instructions” button located on the right. As an additional note, the button labeled “ECCO Datasets on PO.DAAC” situated above the “PO.DAAC Downloading Instructions” button provides a more visual way to choose a specific V4r4 field on PO.DAAC.

Best,
Ou Wang

From: ecco-support <ecco-support-bounces at mit.edu> on behalf of Ervik, Haakon L. <haakon at caltech.edu>
Date: Sunday, September 24, 2023 at 7:59 PM
To: ecco-support at mit.edu <ecco-support at mit.edu>
Subject: [EXTERNAL] [ecco-support] missing values in ECCO V4r4 monthly interpolated climatology
Hello,

I am a graduate student at Caltech working on ocean temperature estimation and I have been happily using ECCO data as a reference for many calculations in my research.

A few days ago, I downloaded the monthly interpolated climatology from the ECCO Drive (e.g. https://ecco.jpl.nasa.gov/drive/files/Version4/Release4/interp_monthly/THETA).
Compared to previous times I’ve downloaded ECCO datasets (e.g. V4r4) it appears that the provided netcdf files effectively use the value 0.0 as _FillValue, which is confusing all the time THETA can be both positive and negative.

Using https://ecco.jpl.nasa.gov/drive/files/Version4/Release4/interp_monthly/THETA/1992/THETA_1992_01.nc as an example, I found that for lat= - 66.25 (1-based index 589), lon=114.25 (index 48), the resulting profile is negative near the surface, then positive for a bit, then exactly zero for the remainder of the array (I tried loading the dataset in both Python and Julia). Presumably I reached the sea floor once values are exactly 0.0. I think these should be NaN or Missing (or a really large value) instead. This was done correctly when I’ve downloaded ECCO data in the past.

I’m further confused since, interrogating the global attributes of the dataset, it says it follows CF-1.6 conventions, but the only attribute referring to what should happen with missing data as far as I can tell is a global attribute «no_data» which has value «NaNf».

Am I doing something wrong? Or did something change in how «missing data» is specified in the most recent release of ECCO? Should I direct my questions elsewhere?

I appreciate any help.

Best,
Haakon

--


Haakon Ludvig Langeland Ervik [haw-kon]
PhD Student, Environmental Science and Engineering
California Institute of Technology
www.hlle.no | haakon at caltech.edu

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mit.edu/pipermail/ecco-support/attachments/20231006/fa6b3ec2/attachment-0001.htm>