ssshio and nsshio of archv2data2d.f

[Max]

Hi Alan,

Can I get 'ssshio' and 'nsshio' using archv2ncdf2d or archv2data2d.f ?

==

c --- 'ssshio' = steric     SSH   I/O unit (0 no I/O), OPTIONAL, not with atthio

c --- 'nsshio' = non-steric SSH   I/O unit (0 no I/O), OPTIONAL, after ssshio

==

I tried it with archv2ncdf2d and It said it was unable to get them.

Thank you.

-Max

[alan.wa...@hycom.org]

Yes, but only if steric is in the archive file:

field       time step  model day  k  dens        min              max          
montg1   =   30082320  41781.000  2 34.00  -1.2342211E+01   6.8754573E+00
srfhgt   =   30082320  41781.000  0  0.00   2.0770423E+00   1.3685039E+01
steric   =   30082320  41781.000  0  0.00  -5.6709871E+00   1.3621014E+01
surflx   =   30082320  41781.000  0  0.00  -2.6560345E+02   6.3819287E+02

This will be the case for archv and archs files if sshflg=1 on blkdat.input:

   1      'sshflg' = diagnostic SSH flag (0=SSH,1=SSH&stericSSH)

To set this flag you need a relax.ssh.[ab] file that contains mean (depth averaged) potential density, mean SSH, and the bathymetry:

conrad02 162> cat relax.ssh.b
average HYCOM th3d mean:    8.00-  13.00 [09.1H] :   -29.1510        3.77547   
 sea surf. height  mean:    8.00-  13.00 [09.1H] :   -2.14266        1.46927   
min,max depth =     5.00000  7199.97998

If an archive (e.g. a mean archive) does not contain steric SSH, you can get it by running archv2data2d for "atthio" and then using ALL/bin/hycom_stericssh.

Alan.

[Max]

I am using monthly mean data from ftp://ftp.hycom.org/datasets/GLBb0.08/expt_19.1/data/meanstd/

In you shell scrip, data2d_763_0809_steric.csh,

How can I get "${D}/${E}_${y}_ssha.a", "relax_ssh_GDEM42-zssh_11.a", " ${D}/${E}_${y}_sshK.a ", "${D}/${E}_${y}_ssha-K.a"?

It's binary files, so I don't now how to get them.

Thank you

-Max

[alan.wa...@hycom.org]

Download https://drive.google.com/file/d/0BxdbYKKvou9KX2E2SHVJU3cyMGs/view?usp=sharing

It contains relax_ssh_091.[ab], which is what was used for relax.ssh in 19.1.

ajax 123> cat relax_ssh_091.b

average HYCOM th3d mean:    8.00-  13.00 [09.1H] :   -29.1510        3.77547   
 sea surf. height  mean:    8.00-  13.00 [09.1H] :   -2.14266        1.46927   
min,max depth =     5.00000  7199.97998

Note that the script I sent you was looking at steric anomalies (sssha) and the difference between total depth an 1000m steric fields.  You only need one command, something like

hycom_stericssh ${D}/${E}_${y}_ath.a $S/relax_ssh_091.a 4500 3298 ${D}/${E}_${y}_sssh.a >! ${D}/${E}_${y}_sssh.b

Then the non-steric SSH is total SSS - steric SSH, using a hycom_expr like the one in the example script.

Alan.

[Max]

Hi Alan,

Can I get relax_ssh_090.[ab], which is what was used for relax.ssh in 19.0?

Also I am not sure if I got ssh, sssh, and nssh correctly because the values from *ssh.a,  *sssh.a,  *nssh.a are weird.

I set the values in you script as below.

Did I something wrong?

Thank you

-Max

==

  foreach y ( $1 )

    foreach month ( $2 )

    setenv FOR022A  ${OUT}/${E}_archMN.${y}_${month}_EastSea_ath.a

    setenv FOR022   ${OUT}/${E}_archMN.${y}_${month}_EastSea_ath.b

    setenv FOR023A  ${OUT}/${E}_archMN.${y}_${month}_EastSea_athK.a

    setenv FOR023   ${OUT}/${E}_archMN.${y}_${month}_EastSea_athK.b

    setenv FOR024A  ${OUT}/${E}_archMN.${y}_${month}_EastSea_ssh.a

    setenv FOR024   ${OUT}/${E}_archMN.${y}_${month}_EastSea_ssh.b

    /bin/rm $FOR022 $FOR022A $FOR023 $FOR023A $FOR024 $FOR024A  

/data2/mhan/HYCOM/hycom/ALL/archive/src/archv2data2d <<E-o-D

${D}/${E}_archMN.${y}_${month}.a

HYCOM

000    'iexpt ' = experiment number x10 (000=from archive file)

  3     'yrflag' = days in year flag (0=360J16,1=366J16,2=366J01,3-actual)

4500    'idm   ' = longitudinal array size

3298    'jdm   ' = latitudinal  array size

 32     'kdm   ' = number of layers

 34.0   'thbase' = reference density (sigma units)

  0     'smooth' = smooth fields before plotting (0=F,1=T)

  0     'mthin ' = mask thin layers from plots   (0=F,1=T)

665     'iorign' = i-origin of plotted subregion

1972    'jorign' = j-origin of plotted subregion

191     'idmp  ' = i-extent of plotted subregion (<=idm; 0 implies idm)

305     'jdmp  ' = j-extent of plotted subregion (<=jdm; 0 implies jdm)

  0     'areaio' = grid area  I/O unit (0 no I/O) - optional

  0     'botio ' = bathymetry       I/O unit (0 no I/O)

  0     'flxio ' = surf. heat flux  I/O unit (0 no I/O)

  0     'empio ' = surf. evap-pcip  I/O unit (0 no I/O)

  0     'tbfio ' = temp buoyancy flux I/O unit (0 no I/O)

  0     'sbfio ' = saln buoyancy flux I/O unit (0 no I/O)

  0     'abfio ' = tot. buoyancy flux I/O unit (0 no I/O)

  0     'icvio ' = ice coverage     I/O unit (0 no I/O)

  0     'ithio ' = ice thickness    I/O unit (0 no I/O)

  0     'ictio ' = ice temperature  I/O unit (0 no I/O)

 22     'atthio' = average density  I/O unit (0 no I/O), OPTIONAL

 23     'adthio' = ave 0-depth den. I/O unit (0 no I/O), OPTIONAL, after atthio

1000    'adepth' = depth for adthio, OPTIONAL, always after adthio

 24     'sshio ' = sea surf. height I/O unit (0 no I/O)

  0     'bkeio ' = baro. k.e.       I/O unit (0 no I/O)

  0     'bsfio ' = baro. strmfn.    I/O unit (0 no I/O)

  0     'uvmio ' = mix. lay. u-vel. I/O unit (0 no I/O)

  0     'vvmio ' = mix. lay. v-vel. I/O unit (0 no I/O)

  0     'spmio ' = mix. lay. speed  I/O unit (0 no I/O)

  0     'bltio ' = bnd. lay. thick. I/O unit (0 no I/O)

  0     'mltio ' = mix. lay. thick. I/O unit (0 no I/O)

  0     'sstio ' = mix. lay. temp.  I/O unit (0 no I/O)

  0     'sssio ' = mix. lay. saln.  I/O unit (0 no I/O)

  0     'ssdio ' = mix. lay. dens.  I/O unit (0 no I/O)

  0     'mkeio ' = mix. lay. k.e.   I/O unit (0 no I/O)

  1     'kf    ' = first output layer (=0 end output; <0 label with layer #)

 32     'kl    ' = last  output layer

  0     'uvlio ' = layer k   u-vel. I/O unit (0 no I/O)

  0     'vvlio ' = layer k   v-vel. I/O unit (0 no I/O)

  0     'splio ' = layer k   speed. I/O unit (0 no I/O)

  0     'infio ' = layer k   i.dep. I/O unit (0 no I/O)

  0     'thkio ' = layer k   thick. I/O unit (0 no I/O)

  0     'temio ' = layer k   temp   I/O unit (0 no I/O)

  0     'salio ' = layer k   saln.  I/O unit (0 no I/O)

  0     'tthio ' = layer k   dens,  I/O unit (0 no I/O)

  0     'keio  ' = kinetic energy   I/O unit (0 no I/O)

  0     'sfnio ' = layer k  strmfn. I/O unit (0 no I/O)

  0     'kf    ' = first output layer (=0 end output; <0 label with layer #)

E-o-D

  /bin/rm -f      ${OUT}/${E}_archMN.${y}_${month}_EastSea_sssh.a

  hycom_stericssh ${OUT}/${E}_archMN.${y}_${month}_EastSea_ath.a /data2/mhan/HYCOM/data/GLBb0.08/expt_19.1/topo/relax_ssh_091.a 4500 3298 ${OUT}/${E}_archMN.${y}_${month}_EastSea_sssh.a >! ${OUT}/${E}_archMN.${y}_${month}_EastSea_sssh.b

#  /bin/rm -f      ${OUT}/${E}_archMN.${y}_${month}_EastSea_sshK.a

#  hycom_stericssh ${OUT}/${E}_archMN.${y}_${month}_EastSea_athK.a /data2/mhan/HYCOM/data/GLBb0.08/expt_19.1/topo/relax_ssh_091.a 4500 3298 ${OUT}/${E}_archMN.${y}_${month}_EastSea_sshK.a >! ${OUT}/${E}_archMN.${y}_${month}_EastSea_sshK.b

  /bin/rm -f      ${OUT}/${E}_archMN.${y}_${month}_EastSea_nssh.a

  hycom_expr      ${OUT}/${E}_archMN.${y}_${month}_EastSea_sssh.a ${OUT}/${E}_archMN.${y}_${month}_EastSea_ssh.a 4500 3298 1.0 -1.0 ${OUT}/${E}_archMN.${y}_${month}_EastSea_nssh.a >! ${OUT}/${E}_archMN.${y}_${month}_EastSea_nssh.b

  end

end

==

[alan.wa...@hycom.org]

It should be ok to use relax_ssh_091 with 19.0.

If you extract a subset of the domain, that subset's idm,jdm (idmp,jdmp) should be used in the hycom_stericssh and hycom_expr.  However, that means you have to subset relax_ssh_091 as well.  Use the command:

hycom_subset relax_ssh_091.a 4500 3298 665 1972 191 305 relax_ssh_091_EastSea.a >&! relax_ssh_091_EastSea.b

This should give you the correct sssh (and nssh), but ssh should already have been good.  So perhaps there is something else wrong.

Alan.

[Max]

Hi Alan

I calculated ssh, nssh, and sssh by using scripts and relax_ssh_091.ab you gave me.

I think I calculated them in the East Sea (Japan Sea) and the Mediterranean Sea correctly as I have attached.

In the Mediterranean Sea, E-P has maximum in the summer, which means a lot of evaporation in the summer, and the horizontal volume transport convergence between Gibraltar and Dardanelles Straits has maximum also in the summer. So the fluctuation range of non-steric SSH is similar to that of steric SSH there.

In the East Sea (Japan Sea), E-P has minimum in the summer, which means a lot of rain in the summer, and the horizontal volume transport convergence among Korea/Tsushima, Tsugaru, and Soya Straits has maximum in the summer. So I think the non-steric SSH has much larger ranges than steric SSH in the East Sea (Japan Sea) because of a lot or rain and volume transport convergence during the summer.

But the fluctuation range of non-steric SSH is much smaller than that of steric SSH in the East Sea (Japan Sea) from the attached figure.

Can you explain why there is discrepancy between my thought and calculation?

Thank you.

-Max

 

and I have a question about nssh.

[alan.wa...@hycom.org]

Unless you have set epmass=1 in blkdat.input (which can be fragile, and may cause drift because of E-P imbalance), E-P is a salt flux rather than a water flux.  So it effects steric SSH only, rather than both steric and non-steric SSH.  The steric SSH has the effect both of E-P and of the heat flux, and the latter is larger in most places. All our runs have epmass=0.

Alan.

[Max]

Alan,

If E-P is not a water flux in GLBb0.08, SSH from HYCOM should be different to that from AVISO satellite altimeter data where absolute of E-P is large like the East Sea (Japan Sea).

In the East Sea (Japan Sea), the sea level change from E-P water flux (from observation and HYCOM input) is bigger than that from the horizontal volume transport difference among Korea/Tsushima, Tsugaru, and Soya Straits, and than that from the steric volume change. But the range of SSH fluctuation from HYCOM (without E-P) is similar to that from AVISO data.

Is it because of data assimilation?

Can E-P be neglected in the East Sea?

Do you have SSH data of GLBub.08 before data assimilation?

Thank you.

-Max  

[alan.wa...@hycom.org]

The synthetic profile approach to assimilating SSH assumes the entire SSH anomaly is steric (i.e. is assumes non-steric components have been removed), which is then converted to changes in the T&S vertical profile.  GOFS 3.0 used MODAS and GOFS 3.1 used ISOP; for a summary see:
;
T.L. Townsend, C.N. Barron and R.W. Helber, 2015:
Ocean Prediction with Improved Synthetic Ocean Profiles (ISOP)
NRL Review 105-112

So non-steric SSH that isn't filtered out of the altimeter signal gets aliased to a steric signal, and with MODAS that signal is likely to be entirely in Temperature.  ISOP can spread the signal across T&S.

Alan.

[Max]

I know little about data assimilation.

How often do you do data assimilation in the case of Reanalysis GLBb0.08?

Thank you.

-Max

[alan.wa...@hycom.org]

Once a day, see https://hycom.org/data/glbu0pt08/expt-19pt1

Alan.

[Max]

Alan,

I plotted monthly Mean Sea Level change in the East Sea (Japan Sea) as MSL (VT diff) from volume transport difference among Korea/Thushima, Tsugaru, and Soya Straits using transp_mn3,

and SSH, NSSH, SSSH using archv2data2d, hycom_stericssh etc you gave me.

I think MSL (VT diff) should be similar to Non-Steric (HYCOM) because epmass=0 in blkdat.input but it is not similar. 

Why did it happen? Where can I fix this difference?

Thank you

-Max

[Max]

Hi Alan,

How did you make relax_ssh_091.ab ?

Is it "mean" or "climatology" of depth averaged density, ssh, and depth during 1995~2012 ?

There is only one value in each depth averaged density, ssh, and depth in relax_ssh_091.ab, so I think it's mean.

I plotted SSH, th3d from monthly mean GLBb0.08 data during 1993~2012 and from relax_ssh_091.ab in the East Sea (Japan Sea; 665 1972 191 305),

and 20 year mean SSH and th3d are bigger than (are not same to) SSH and th3d from relax_ssh_091.ab in the East Sea (Japan Sea).

Can I calculate nssh and sssh using 20 year mean th3d and ssh?

For example, the deviation of ssh from 20 year mean ssh can be deviations of nssh and sssh,

and the deviation from th3d can be a sssh part, and the rest can be nssh.

I am trying to calculate the sssh (steric ssh) using temperature and salinity with stmt_fns_SIGMA2_17term.h,

and the fluctuation of thermosteric from that calculation is half of that in sssh from calculation using relax_ssh_091.ab.

So I am wondering how you calculated relax_ssh_091.ab.

Thank you.

-Max

[alan.wa...@hycom.org]

See the attached script.  It calculates (extracts) the vertical average potential density (SSH) from a long term annual mean.  These are the 1st two fields in relax_ssh.[ab], with the third being the bathymetry.

The method used is described in the attached PDF.

Since the calculation involves the mean density, the relax_ssh has to be from a case with the same bathymetry.  If you don't have a suitable long term model mean, in principle you can use the SSH from 19.1 and combine it with the vertical average density from the annual mean climatology over the specified bathymetry.

Alan.

[Max]

Thank you, Alan.

I understood how you calculated ssh, nssh, and sssh now.

I compared ath3d from monthly mean GLBb0.08 (1993~2012) data and ath3d from relax_ssh_091.a which you gave me in the East Sea (Japan Sea),

and plotted it as I've attached.

In the first figure, ath3d from 20 year monthly mean in GLBb0.08 (1993~2012) minus ath3d from relax_ssh_091.a are mostly positive,

which means ath3d from relax_ssh_091.a is lower than that from 20 year monthly mean data by 0.05 (kg/m^3) averagely.

So, that's why the mean of SSSH in the second figure is about 0.1 m lower than 0 m.

Can I use ath3d of 20 year monthly mean data in GLBb0.08 (1993~2012) as long term mean ath3d?

The depth in the East Sea (Japan Sea) is same between expt_19.0 and expt_19.1.

Also can I calculate SSSH (steric SSH) using ~/hycom/ALL/include/stmt_fns_SIGMA2_17term.h, and monthly temperature and salinity

with either longterm mean depth averaged density = ath3d 

or dsiglocdt and dsiglocds (stmt_fns_SIGMA2_17term.h), Tf (depth averaged 20 year mean temperature), Sf (depth averaged 20 year mean salinity) from (A3) in Pinardi et al. 2014 (http://journals.ametsoc.org/doi/full/10.1175/JCLI-D-13-00139.1) ?

Thank you,

-Max  

[alan.wa...@hycom.org]

If the bathymetries are the same, you can use 19.1 in place of 09.1.

We always use potential density to calculate HYCOM's Steric SSH.

If you have mean archives, they already contain potential density precisely because mean pot.T and S may not give mean pot.density.  I don't know how big an effect using the equation of state on mean pot.T and S to calculate pot.density would have.

Alan.

[Max]

Alan,

Monthly mean ath3d of 20 year monthly mean data in GLBb0.08 (1993~2012) was calculated using monthly mean potential density in it,

and there are some differences between that and ath3d from relax_ssh_091.a.

You told me that ath3d from relax_ssh_091.a is from a long term annual mean.

Does that mean 20 year monthly mean ath3d is not enough compared to a long term annual mean ath3d ?

Is it because of monthly mean? or is it because 20 year is not long enough?

What is the data interval when you calculated monthly mean data in GLBb0.08 and GLBa0.08?

What is the data interval when you calculated relax_ssh_091.a?

Thank you.

- Max 

[alan.wa...@hycom.org]

The relax_ssh_091 fields are based on a 5-year mean from expt 9.1:

newton 12> cat relax_ssh_091.b

average HYCOM th3d mean:    8.00-  13.00 [09.1H] :   -29.1510        3.77547   
 sea surf. height  mean:    8.00-  13.00 [09.1H] :   -2.14266        1.46927   
min,max depth =     5.00000  7199.97998

This is a non-assimilative case.

We obviously could not use the 19.0+19.1 long term mean for the in-line relax_ssh for 19.0/19.1 - since it had not run yet.  The steric SSH is a diagnostic field in HYCOM (i.e. it does not change the results).

I would say that the 19.0+19.1 long term mean SSH is about the best we have, and its mean density is good too.  So I would use the 19.1 relax_ssh over the 9.1 version.

A viable alternative to the 19.1 mean potential density would be the mean potential density from an annual mean climatology.  This gives you a steric SSH anomaly that is most consistent with how a synthetic profile represents an altimeter SSH anomaly.

Remember that adding a constant to SSH has no dynamical effect (since it does not change the SSH gradients), so it is ok to offset the 19.1 SSH by a constant if you like.

Alan.

[Max]

Hi Alan,

Can I get monthly mean data of 9.1 version for 5 years or more? even the subset of them like in the East Sea (Japan Sea; 4500 3298 665 1972 191 305)?

I want to see the difference between non-assimilative and assimilative GLBb0.08 data, and unfiltered (when assimilating) horizontal volume transport and E-P effect between them.

 

Also, I calculated the differences of sssh from relax_ssh_091.a and from 190+191 (GLBb0.08, 1993~2012).

The mean difference of them is 0.126 m and their fluctuation is very similar (i.e. the difference is less than 0.1 mm after moving one of them vertically) as figures (mean in the East Sea (Japan Sea) and data at a point (131.20E, 38.11N)) I have attached.

I think it is because the annual mean ssh climatology of non-assimilative data is lower than that of assimilative data, probably because of decreasing SSH in the East Sea (Japan Sea) due to model drift in the east of Soya Strait in non-assimilative data.

So the difference between the non-assimilative and assimilative sshs has model drift, horizontal volume transport, and E-P mass flux probably.

How do you think about that?

Thank you.

-Max

[alan.wa...@hycom.org]

See ftp://ftp.hycom.org/datasets/GLBb0.08/expt_10.2/data/meanstd/ this case (10.2) is the non-assimilative twin of 19.0/19.1.

So far I have just put two long term means (1993-2012 and 2003-2012) there.  The difference should give you an idea of trends.

Check that this is ok for what you want to do.  If so, I can put monthly means there too.  Perhaps for 2003-2012 - or suggest a range.

Alan.

[Max]

Thank you, Alan.

I plotted mean ssh, nssh, and sssh of assimilative (191, 1993~2012) and non-assimilative (102, 1993~2012 & 2003~2012) GLBb0.08 data.

Assimilative data are from ftp://ftp.hycom.org/datasets/GLBb0.08/expt_19.1/data/meanstd/191_archMN.1993_00_2012_00.a

and non-assimilative data are from ftp://ftp.hycom.org/datasets/GLBb0.08/expt_10.2/data/meanstd/

In 20 year mean data like ftp://ftp.hycom.org/datasets/GLBb0.08/expt_19.1/data/meanstd/191_archMN.1993_00_2012_00.b

and ftp://ftp.hycom.org/datasets/GLBb0.08/expt_10.2/data/meanstd/102_archMNA.1993_2012.b etc

Why are the mean days of "montg1" and "srfhgt" different to the others?

I though that "montg1" and "srfhgt" are daily data in 20 year mean data but I am not sure if it is correct.

As you see the figure which I have attached, SSSH (dotted black line around -0.09 m) can be explained if you calculated "montg1" and "srfhgt" on January 1st, not as long-term mean. 

Because the density on January 1st is bigger than annual and long-term mean density, and the SSSH which was calculated from it is lower than yearly and long-term mean SSSH.

The data you put are what I want to use and it would be better that the range of monthly mean data is 1993~2012 to write papers.

If it takes time to put all of them, you may put some of them first and then you may put the others later gradually.

If it takes a lot of space there, you may put some of them like 1993~2002 or 2003~2012.

Thank you so much.

-Max

[alan.wa...@hycom.org]

Th 10.2 result is what I expected, it is consistent with a small (about +0.6 cm per year)  steric (T&S) drift in SSH.

I'll put the full monthly 10.2 dataset on the data server, but it might take a while.

Alan.

[Max]

Thank you, Alan.

I know it is a Christmas holiday and you must be busy with your family.

But I can't download the data because of file permissions.

So can you please change the file permissions and upload the rest of them when you are available to access to the server?

Happy holidays.

-Max 

[alan.wa...@hycom.org]

I can access them via anonftp from a web browser at ftp://ftp.hycom.org/datasets/GLBb0.08/expt_10.2/data/meanstd/ and I don't think I have any special privileges in that mode.

Which files can't you access?

I am uploading the rest of the files now.

Alan.

[Max]

Alan,

I could download all of *.b and some of *.a such as *1993*.a, but I couldn't download most of *.a such as *1994*.a, *1995*.a, etc.

Can you please check the permissions of files using ls -al?

Thanks you.

-Max

[alan.wa...@hycom.org]

All the files are there, and they all have world read permission.

Alan.

[bullabbay s]

Hi,

Just asking out of curiosity, instead of generating relax_ssh from long model simulations,

are there any tools to generated relax_ssh from observations like WOA13, AVISO's latest MDT?

[alan.wa...@hycom.org]

One approach we have used is to calculate the mean potential density from annual-mean climatology and use the global Reanalysis long term mean for SSH.

The advantage of this approach is that the model Steric SSH anomaly is then exactly the SSH anomaly from altimeters that we assimilate.  The disadvantage is that the long term mean SSH is not necessarily dynamically consistent with the mean potential density.  So you might get some "steric" signal aliased into the non-steric SSH.

Since Steric SSH is a diagnostic quantity in HYCOM, you can start out using the climatology approach and then switch to a model mean approach when you have enough model years for a mean.

Alan.

[Max]

Hi Alan,

According to my calculation, SSH rate of 10.2 is +0.47 cm/year and steric (potential density) drift is +0.50 cm/year in the East Sea (Japan Sea) for 20 years.

If SSH can be explained only by steric (potential density or T&S) drift in SSH, is there any effect of mass in SSH like volume transport difference (divergence) among Korea/Tsushima, Tsugaru, and Soya Straits?

The effect (drift) of volume transport difference on the SSH in the East Sea (Japan Sea) is -17.4 cm/year in my calculation and SSH should be decreasing if SSH calculation included the volume transport difference.

I know you tried not to change the global SSH and did not include E-P in 19.0/19.1 and it's data assimilative version. So you could control SSH.

But 10.2 is a non assimilative and I don't know if you tried not to change the global SSH also.

So my question is the volume transport difference (divergence) in a marginal basin was included when you calculated SSH there.

Thank you

-Max  

On Wednesday, December 20, 2017 at 10:50:37 PM UTC+9, alan.wa...@hycom.org wrote:

[alan.wa...@hycom.org]

There should not be any long term mass storage (difference between inflow and outflow) in a semi-enclosed sea.  This storage will be in the non-steric SSH and their might be a seasonal signal and perhaps some multi-year wiggling but I would not expect a 20 year consistent drift.  So your non-steric drift is -0.03 cm per year is consistent with my expectations.  Note that it is mass, not volume, transport that matters although I would not expect this to make much difference.  I can't explain your -17.4 cm/year number except to say that storage of this size isn't likely - if you create a pressure head between the East Sea and the Pacific you will get a counterbalancing current.

Alan.

[Max]

Alan,

Can I have blkdat.input_102?

Thanks

-Max

[alan.wa...@hycom.org]

The 10.2 blkdat.input is the same as for 19.1 but with 'incflg'=0.

Alan.

[Max]

Hi Alan,

I've attached time series of monthly mean volume transport difference in the East Sea (Japan Sea) like volume transport in the Korea/Tsushima Strait minus volume transports in the Tsugaru and Soya Straits.

I don't know why there is a long term volume transport difference.

Total monthly mean volume transport difference and mean sea level change due to volume transport difference are

-0.005 Sv averaged per month [ (-1.29 Sv)/(20 year*12 month) = -0.005 Sv per month ]

and -13 mm averaged per month [ -0.005 * (10^6 m^3) * (60*60*24*365/12 s)/(10^12 m^2; area of the East Sea) = -13 mm ]

,which mean -3.12 m mean sea level decrease in the East Sea [ -13*12*20 = -3.12 m during 20 years ].

Did I miscalculate it?

Thank you.

-Max

[Max]

Alan,

I know that HYCOM's data assimilative SSH is constrained to be "close" to annual mean climatology.

Is it same to HYCOM's non-data assimilative SSH (GLBb0.08 expt_10.2) ?

Thank you.

-Max

[alan.wa...@hycom.org]

The non-assimulative cases are not constrained, so what you see is model climatology.  It often shows sea level rise over time due to the ocean getting warmer and fresher on average, but it would be largely accidental if this agreed with observed sea level change.

Alan.