nudging/relaxation time scale field

[jschoi]

I want to apply a relaxation layer for temperature and salinity at only east boundary using a climatology nudging.
Boundary file is used daily HYCOM and climatology file is used monthly HYCOM.

My test cases are three.
common setting: East and West tracer RadNud, TNUD == 5.0, OBCFAC == 5.0 in ocean.in, north and south closed

case 1, control model, this is not used nudging. (LtracerCLM == F F, LnudgeTCLM == F F)

case 2, nudging coefficient field is zero. (LtracerCLM == T T, LnudgeTCLM == T T)

case 3, nudging coefficient field is zero in interior,
however, decrease from 1/30 (day-1) at the only east boundary to zero at the interior,
and, 1/5 (day-1) at the west boundary. (LtracerCLM == T T, LnudgeTCLM == T T)


and, I ran the models just one step (dt == 40 sec) respectively.


My questions are:
1. why the temperatures are difference in his0001 of case 2 and case 1 at both boundaries.
(i.e. case 2 minus case 1 is not zero at both boundaries)

2. why the temperatures are difference at west boundary in his0002 of case 3 and case 1
(i.e. case 3 minus case 1 is not zero at west boundary),
but are not difference in all area in his0001 of case 3 and case 1, unerstandably.

3. Do lateral boundary and climatology only for relaxation share the inverse time scale in nudging nc file?

Salinity is also the same as above.

[kate]

Answers 1, 2 and 3: In t3dbc_im.F, the nudging code has a different nudging constant depending on whether or not LnudgeTCLM is true. You can search for LnudgeTCLM there to see. That is not the behavior I want, so I have modified my code accordingly.