Problem with upwelling case

[evridiki]

Dear all,

When I ran the upwelling case (months ago) I got weird characters in my log file. But since the were associated with biological factors and the results seemed fine, I didn't gave much attention. A part of my log file is:

Lateral Boundary Conditions: NLM
============================

Variable Grid West Edge South Edge East Edge North Edge
--------- ---- ---------- ---------- ---------- ----------

zeta 1 Periodic Closed Periodic Closed

ubar 1 Periodic Closed Periodic Closed

vbar 1 Periodic Closed Periodic Closed

u 1 Periodic Closed Periodic Closed

v 1 Periodic Closed Periodic Closed

temp 1 Periodic Closed Periodic Closed

salt 1 Periodic Closed Periodic Closed

NO3 1 Periodic Closed Periodic Closed

NH4 1 Periodic Closed Periodic Closed

chlorophyll 1 Periodic Closed Periodic Closed

phytoplankton 1 Periodic Closed Periodic Closed

zooplankton 1 Periodic Closed Periodic Closed

LdetritusN 1 Periodic Closed Periodic Closed

SdetritusN 1 Periodic Closed Periodic Closed

LdetritusC 1 Periodic Closed Periodic Closed

SdetritusC 1 Periodic Closed Periodic Closed

TIC 1 Periodic Closed Periodic Closed

alkalinity 1 ^@^@^@^@^@^@^@^@ Closed Periodic Closed

oxygen 1 ^@^@^@^@^@^@^@^@ Closed Periodic Closed

Turning off the biology solved the problem but now that I changed machine, the same characters appeared not only in the upwelling case but also in my application (only at the end of log file):

Analytical header files used:

ROMS/Functionals/ana_btflux.h
/mnt/hydradata2/xevri/MYROMS/test/aegean_ydra/ana_fsobc.h
/mnt/hydradata2/xevri/MYROMS/test/aegean_ydra/ana_m2obc.h
ROMS/Functionals/ana_nudgcoef.h
ROMS/Functionals/ana_smflux.h
ROMS/Functionals/ana_stflux.h
^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@^@

ROMS/TOMS: DONE... Thursday - November 14, 2013 - 4:22:38 PM


My question is: What are those characters? What do they mean? Should I worry?
I don't really know what to check. I'm looking forward for any suggestion.

Thanks in advance

[ptimko]

This probably isn't going to help much but I checked on Wikipedia and to answer your first question: "What are those characters?" I found the following:

The at sign, @, normally read aloud as "at", also commonly called the at symbol or commercial at, and less commonly a wide range of other terms,[1][2][3][4] is originally an accounting and commercial invoice abbreviation meaning "at the rate of" (e.g. 7 widgets @ £2 = £14). In recent years, its meaning has grown to include the sense of being "located at" or "directed at", especially in email addresses and social media, particularly Twitter.

It was not included on the keyboard of the earliest commercially successful typewriters, but was on at least one 1889 model[5] and the very successful Underwood models from the "Underwood No. 5" in 1900 onward. It is now universally included on computer keyboards. The mark is encoded at U+0040 @ commercial at (HTML: @).

The fact that there is no single word in English for the symbol has prompted some writers to use the French arobase[6] or Spanish and Portuguese arroba—or to coin new words such as asperand,[3] ampersat[7] or apetail—but none of these has achieved wide currency.

Sorry to say I don't have an answer for your second or third question...

Patrick

[evridiki]

Patrick,

Thanks for your kindly reply. Maybe my question wasn't clear enough so I will try to rephrase.
What I'm trying to find out is not what the @ symbol means generally BUT inside the log file.
As shown above those characters appeared (for the upwelling case) while writing in the log file
the LBC for alkalinity and oxygen and also at the end of the file (for my application).
It seems to me that ROMS couldn't read those biological parameters for some reason.
Am I right?
Maybe it is not important but I would really like to know what went wrong and
if there's a problem with my set up.
Any suggestions?

Thank you all,
Eurydice

[ptimko]

I've never seen output like that before...

I'm wondering if maybe you have a NaN or Inf in your input/output somewhere. Another possibility might be an out of range value that can't be represented by the field descriptor on output. The representation of NaN's, Inf's, or out of range values may be system/compiler dependent when written to a text file. That's my best guess as to what the problem is. I'd be looking at your input files, etc to see if you have any NaN's or Inf's to start with. Next step would be to make sure all the input values are "reasonable" for the variable they represent. I'd also be cautious on using the output from the model with output like that until I figured out what the problem is.

Hope that helps...

Patrick

[ptimko]

Another thought just occured to me, You may be trying to print an unprintable ascii or ebcdic character (such as a tab character, control, EOL, etc) for your system. Not sure why that would happen either...

Patrick

[kate]

There's a thread in here about an ANANAME bug which would explain the one instance, if not the other.

[evridiki]

Kate,

Thanks for your reply. I changed the code in the /Modules/mod_ncparam.F as Hadfield proposed in
viewtopic.php?f=19&t=3154&p=12020&hilit=ANANAME#p12020
and the problem solved for the ana* files.
But I'm still having those characters in the LBC section for the oxygen and alkalinity if BIO_FENNEL is activated.
Do you have any suggestions for that?

Patrick,
I looked as you proposed for unprintable ascii characters but everything seems fine.
I'm still a little confused since I downloaded the ROMS code and ran the upwelling as it was, without any changes.
Thanks for your time.

[kate]

It is indeed worrisome that it doesn't know all your LBC options. I would debug it with your favorite tools, be they print statements or a debugger.

[ptimko]

What compiler are using? what compiler options are you using? and what is your computing platform?

If you have alternative compilers available or alternative platforms available you may want to see if you can replicate the problem. It may help determine if it's compiler or platform specific.

Patrick

[evridiki]

Thank you both for your answers!
I will try again according to your suggestions.

Eurydice