mask river cells

[Johannes Ulrich Pein]

Dear all,

in my Setup I would like to let the river start near the sea. Therefor I want to mask river cells that have become obsolete.
So I turned the river cells NaN, which are placed upstream from the point I want to have the freshwater input.
Unfortunately now I get
FATAL GETM ERROR: Called from catch_nan_2d()
FATAL GETM ERROR: NaN values encountered in elevations

Does anybody have any idea what is the problem? Do I have to declare my river cells boundary somewhere?

Cheers
Johannes

[Knut]

Dear Johannes,

> in my Setup I would like to let the river start near the sea. Therefor I want to mask river cells that have become obsolete.
> So I turned the river cells NaN, which are placed upstream from the point I want to have the freshwater input.
>

I have no idea of rivers, but I know that masking cells during a new
hotstart is possible. What do you mean with "turning river cells NaN"?

> Do I have to declare my river cells boundary somewhere?
>

Telling the new positions of the river cells is probably a good idea :-)

Cheers,
Knut

[Karsten Bolding]

multiple reading of 3D boundary data

you and Bianca came up with a solution where you changes .ge. to .gt. -
iyt worked but was it correct?

--
http://www.getm.eu

[Karsten Bolding]

sorry for replying in the wrong thread ...

On Mon, Jan 02, 2012 at 12:59:46 +0100, Knut wrote:

--
http://www.getm.eu

[Johannes Ulrich Pein]

I named the cells in riverinfo.dat. Where else do I have to declare them?

[e f]

Dear Johannes,

nowhere else; the river cell(s), i.e. the cell(s) where the freshwater
is put in, are declared only in the file "riverinfo.dat".

Whatever you meant with "turned the river cells NaN", I suppose you
want to use the file "mask.adjust" and give it the coordinates of the
obsolete grid cells. The format is as follows:
<number X of masked rectangles>
<rectangle 1>
...
<rectangle X>

where a rectangle is specified by the indices of the lower left and
the upper right corner:
<i_low j_low i_high j_high>

Example file "mask.adjust":
2
10 22 49 24
60 1 61 15

The specified rectangle(s) are then masked and ignored in the
simulation. I hope this helped.

Kind regards,
Elisabeth

2012/1/2 Johannes Ulrich Pein <jup...@gmx.de>:

> I named the cells in riverinfo.dat. Where else do I have to declare them?
>
>

2012/1/2 Knut <knut...@gmx.net>:

[Johannes Ulrich Pein]

Hi Knut,

that was very clear.
Cheers
Johannes

[Johannes Ulrich Pein]

I mean Elisabeth:)

[Knut]

Johannes,

> I mean Elisabeth:)
>

sure. nevertheless I reply now:
This is not a chat. Most (?) subscribers do not want to check the
previous mails to put your response into the right context. Please send
your posts with your reply at the bottom of the quoted relevant original
text.

Knut

[Karsten Bolding]

I fully support Knuts message :-)

--
http://www.getm.eu

[Bert Rubash]

Can anyone tell me why the default Charnock constant in GOTM 1400
instead of, say, 0.0185*rho**2, which would be about 18500?

0.0185 is the value suggested by Jin Wu in a 1982 JGR paper. What am I
missing?

-Bert Rubash

[Hans Burchard]

Bert,

this value is used by Burchard (2001), and I took it from Craig & Banner
(1994), not being aware of Wu (1982) (references below).

In my paper, I discuss however on page 3140 that the resulting z0 seems to
be quite low compared to estimates in the literature.

So, if nobody disagrees, we will change the default value to 18500.

Thanks a lot for pointing this out.

Regards, Hans.

References:

Burchard, H., 2001, Simulating the wave-enhanced layer under breaking
surface waves with two-equation turbulence models, J. Phys. Oceanogr., 31,
3133-3145.

Craig, P. D., and M. L. Banner, 1994: Modelling wave-enhanced turbulence
in the ocean surface layer. J. Phys. Oceanogr., 24, 2546�2559.

Wu, J., 1982: Wind-Stress Coefficients Over Sea Surface From Breeze to
Hurricane, J. Geophys. Res., 87(C12), 9704�9706,
doi:10.1029/JC087iC12p09704.

--
Hans Burchard
Leibniz Institute for Baltic Sea Research Warnemuende
Dept. for Physical Oceanography and Instrumentation
Seestrasse 15
D-18119 Rostock-Warnemuende
Germany

Tel. : +49 - 381 - 5197 -140
Fax : +49 - 381 - 5197 -114
web : http://www.io-warnemuende.de/hans-burchard-en.html
E-Mail: hans.b...@io-warnemuende.de
skype : hans.burchard

[adolf...@jrc.ec.europa.eu]

Dear Bert and Hans,

This is actually a quite good point, because indeed the value of 1400 did not produce 

a good simulation of the near surface layer, only when using 14000 satisfactory results were

achieved, see the reference below, where this is discussed.

Therefore indeed the value should be changed to 18500, regards Adolf

"Measurement and simulation of viscous dissipation in the wave affected surface layer."

Adolf Stips, Hans Burchard, Karsten Bolding, Hartmut Prandke, Andre ́ Simone, Alfred Wu ̈ est

Deep Sea Research Part II: Topical Studies in Oceanography, 2005 vol. 52 (9-10) pp. 1133-1155,

doi:10.1016/j.dsr2.2005.01.012

On Jan 9, 2012, at 8:40 AM, Hans Burchard wrote:

Bert,

this value is used by Burchard (2001), and I took it from Craig & Banner
(1994), not being aware of Wu (1982) (references below).

In my paper, I discuss however on page 3140 that the resulting z0 seems to
be quite low compared to estimates in the literature.

So, if nobody disagrees, we will change the default value to 18500.

Thanks a lot for pointing this out.

Regards, Hans.

References:

Burchard, H., 2001, Simulating the wave-enhanced layer under breaking
surface waves with two-equation turbulence models, J. Phys. Oceanogr., 31,
3133-3145.

Craig, P. D., and M. L. Banner, 1994: Modelling wave-enhanced turbulence
in the ocean surface layer. J. Phys. Oceanogr., 24, 2546–2559.

Wu, J., 1982: Wind-Stress Coefficients Over Sea Surface From Breeze to
Hurricane, J. Geophys. Res., 87(C12), 9704–9706,
doi:10.1029/JC087iC12p09704.

[Adolf STIPS]

Hi Hans and Bert,

Adolf,

I do not unterstand from where the factor rho**2 should come.

I would stay therefore with the oceanographic literature and charnock_val=1400 

( This Charnock - function is its own research field..)

The Charnock formulas comes from the atmosphere (and only from there)

Charnock 1955:
charnok = g * z0 / U*^2
U*^2 = Cd * U10^2

In the atmosphere following values are used:
charnok = 0.012     (Charnock 1955)
charnok = 0.0144   (Garatt 1977)
charnok = 0.0185   (Wu 1980)

Fairall ea 1996:
charnok = g * (z0 - 0.11 ny/U*) / U*^2

Foreman & Emeis 2010 plot charnok as a decreasing function of wind speed
(from 0.08 - 0.0012)

In GOTM this concept is applied to the ocean part of the boundary

It must be:
 stress(atmos) == stress(ozean) 

->

rho_w *U_w*^2  = rho_a * U_a*^2

rho_w /charnok_w  * zo_w = rho_a / charnok_a *zo_a 

charnok_w = rho_w/rho_a * zo_w/zo_a  * charnok_a

# there is NO factor rho^2

only with assuming the roughness length:
zo_a ~ 0.0002
zo_w ~ 0.2  

(but that is certainly a to extreme choice) 

charnok_w   ~  1000/1.25 * 1000 * chanok_a

A possible range of charnok_w  is therefor from ca 1000 to ca 20.000.

Bourassa 2000 measured charnock in water and got ca. 850.
(ca 10^5 * charnock_atmosphere according to him)

Waves will complicate the story much more and likely lead to bigger charnok values.

conclusion:

1) A factor of rho^2 does not come from the physics according to my understanding

2) A basic choice of charnok=1400 is okay, as it is also well referenced  

3) If somebody does simulations with wave influence he has to dig the literature 

and might need to adjust the charnok_val (higher values).

So charnok_val in GOTM will not be changed (in agreement with Hans last proposal)

Ciao Adolf

[Bert Rubash]

There is a typo on line 62 of futils/c2x.F90:

"do i=jmin,jmax-1" should be "do j=jmin,jmax-1"

-Bert Rubash

[Karsten Bolding]

Hi Bert

You are right - fixed now.

Fortunately it seems thi routine is not used at all :-)

Please start a new thread when reporting issues.

Karsten

--
http://www.getm.eu