[ecco-support] [EXTERNAL] VT closure North Atlantic

[Fenty, Ian G (US 329B)]

Hi Helene,

Definitely have some pointers.


  1.  Looking at your figure VT_closure_issue I see that you are multiplying NVEL*hFacC.  In fact, instead of the NVEL * hFacC terms you should use the fields called UVELMASS and VVELMASS. In V4, the hFacC field changes through time.  Our UVELMASS and VVELMASS terms track that variation and represent the true mean volume transports.
  2.  In the c-grid system the u points start counting on the ‘west side of the grid cell and the v points start counting on the ‘south’ side of the grid cell.  See the diagram “c-grid horizontal coordinates” here: https://ecco-v4-python-tutorial.readthedocs.io/ECCO_v4_Coordinates_and_Dimensions_of_ECCOv4_NetCDF_files.html
  3.  Consequently, when you combine tiles from ‘face 1’ with tiles from ‘face 5’ you have to shift the indices of the u’s and v’s arrays by 1. I’ve attached a pdf which I hope will make that clear for you.  The same goes for the corresponding DXG and DYG arrays.
  4.  The latest V4 volume flux fields with UVELMASS and VVELMASS can be found via https://search.earthdata.nasa.gov/search  and searching for the following terms:

ECCO_L4_OCEAN_3D_VOLUME_FLUX_LLC0090GRID_MONTHLY_V4R4
ECCO_L4_OCEAN_3D_VOLUME_FLUX_LLC0090GRID_DAILY_V4R4

The corresponding grid geometries (e.g., DXG, DYG) are here:
ECCO_L4_GEOMETRY_LLC0090GRID_V4R4




-Ian




On 3/4/22, 5:19 PM, "ecco-support on behalf of Helene Asbjørnsen" <ecco-support-bounces at mit.edu<mailto:ecco-support-bounces at mit.edu> on behalf of H.Asbjornsen at uib.no<mailto:H.Asbjornsen at uib.no>> wrote:


Hi,



I was wondering if I could get some pointers on how to correctly calculate meridional volume transports across zonal lines on the ECCOv4 native grid.



I’ve been looking at calc_MeridionalTransports.m, which is helpful. But I ultimately want to end up with a full 3D velocity/VT field for the North Atlantic to do further operations on (so only ending up with one total transport value at each latitude is not ideal).





Some details:

I’ve extracted face 1 and 5 and pieced the fields together (recognising that VVEL is -UVEL and dx is dy for face 5). I’ve tried to integrate transports along the x-dimension of the extracted North Atlantic area, but also along near zonal and meridional lines based on output from gcmfaces_lines_zonal.m.



I end up not getting a near zero net transport at around 62N in the North Atlantic. The imbalance is large - around 6 Sv or so, and seems to originate from the part of the section east of Greenland.





All help is much appreciated,

Helene

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mit.edu/pipermail/ecco-support/attachments/20220306/ef58de5b/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: matching atlantic tiles in ECCO V4.pdf
Type: application/pdf
Size: 202897 bytes
Desc: matching atlantic tiles in ECCO V4.pdf
URL: <http://mailman.mit.edu/pipermail/ecco-support/attachments/20220306/ef58de5b/attachment-0001.pdf>