CG_DELTA not positive, IS4Dvar

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
ckerry
Posts: 10
Joined: Wed Sep 15, 2010 1:59 pm
Location: UNSW Australia

CG_DELTA not positive, IS4Dvar

#1 Post by ckerry » Thu Apr 18, 2019 5:33 am

Hi,
I'm wondering if anyone can help with the following error in the descent algorithm when using IS4DVAR? I get the following error, repeated several times in the run log file.

<<<< Descent Algorithm >>>>

CG_DELTA not positive.
CG_DELTA = -3.936643675943532E+035 , outer = 1 , inner =
1
CG_DELTA not positive.
CG_DELTA = -3.936643675943532E+035 , outer = 1 , inner =
1
CG_DELTA not positive.
CG_DELTA = -3.936643675943532E+035 , outer = 1 , inner =
1
CG_DELTA not positive.
CG_DELTA = -3.936643675943532E+035 , outer = 1 , inner =
1

I have run a VERIFICATION run and none of the observations look out of place/anomalous. I have also run this same 4DVar configuration before for a different time period and got good results.
Any tips for debugging would be much appreciated.

Thanks

jsaynisch
Posts: 10
Joined: Tue Sep 10, 2013 2:00 pm
Location: GFZ Potsdam

Re: CG_DELTA not positive, IS4Dvar

#2 Post by jsaynisch » Tue Apr 23, 2019 5:50 am

Dear ckerry,

have a look here: viewtopic.php?f=16&t=3232

in short: try NRREC=-1

Post Reply