something amiss with migration in 6702?

18 views
Skip to first unread message

jaco...@gmail.com

unread,
May 4, 2024, 9:34:29 AM5/4/24
to Atlantis ecosystem model
Hi all,
I am trying to tune migratory whales in 6702 and having quite a bit of difficulty. With the help of others I have run some diagnostics and found some interesting patterns.
What I show below are the results with 73-day output so we can see when migration occurs. 
Both adults and juveniles should migrate and juveniles are limited to cohort 1 ( the plots below say ageclass 1).
Here is the result of for abundance at age. Two strange patters emerge, first that the juveniles migrate and the adults migrate much less (hardly any after cohort 2) second the  very large mortality between cohort 1 and 2. There is no linear mortality on these groups and quadratic mortality is 5.00e-06 for juveniles and 3.83e-07 for adults.

wmw_mig_on_AAA.png
When we inspected the specific mortality file, we found crazy high mortalities at the second time-step (for example, WMW-0-S0-M1 2.37E+19)!
So I reran the model and turned migration off for this group in the groups.csv file. I did not change anything else.
Here is that same abundance at age plot, migration still occurs, although the pattern is quite different. There is still high mortality between cohort 1 and 2, but we no longer have those initial spikes in abundance. Also the mortality for WMW-0-S0-M1 is now 0. wmw_mig_off_AAA.png
So do I have some misspecification here or is there a bug somewhere in 6702? I do not see those crazy high mortality levels in 6665 with 73-day output.
Thanks
Jacob

Fulton, Beth (Environment, Hobart)

unread,
May 6, 2024, 7:25:01 AM5/6/24
to jaco...@gmail.com, Atlantis ecosystem model

G’day Jacob

 

Specific mortality file is not the most trustworthy source (it needs a major rewrite to avoid cases where its doing  what its doing to you which is about timing of its array renewal vs time of year of events) so I’m less worried about that but it would be good to understand what is going on for you.

 

When you turn off migration is vertical or horizontal (trying to figure out which bit of the movement code you are turning off to help track down stuff faster). My fear is its actually the rough approximation we try to do to set up migration and spawning arrays for events that technically begin before the model starts (but we need them otherwise you end up with a hole in the spawning and migration queues).

 

Any way you can share the files for running the model. If you’re happy to be involved in the detective work let me know and I’ll explain where I would put in fprintfs to start figuring it out. If you are up for that it would be faster than waiting on me as I have to go into the field and likely only sporadic access to power outlets for a couple of weeks starting Saturday.

 

Beth

 

From: <atlantis-eco...@googlegroups.com> on behalf of "jaco...@gmail.com" <jaco...@gmail.com>
Date: Saturday 4 May 2024 at 11:34 PM
To: Atlantis ecosystem model <atlantis-eco...@googlegroups.com>
Subject: something amiss with migration in 6702?

 

Hi all,

I am trying to tune migratory whales in 6702 and having quite a bit of difficulty. With the help of others I have run some diagnostics and found some interesting patterns.

What I show below are the results with 73-day output so we can see when migration occurs. 

Both adults and juveniles should migrate and juveniles are limited to cohort 1 ( the plots below say ageclass 1).

Here is the result of for abundance at age. Two strange patters emerge, first that the juveniles migrate and the adults migrate much less (hardly any after cohort 2) second the  very large mortality between cohort 1 and 2. There is no linear mortality on these groups and quadratic mortality is 5.00e-06 for juveniles and 3.83e-07 for adults.


When we inspected the specific mortality file, we found crazy high mortalities at the second time-step (for example, WMW-0-S0-M1 2.37E+19)!

So I reran the model and turned migration off for this group in the groups.csv file. I did not change anything else.

Here is that same abundance at age plot, migration still occurs, although the pattern is quite different. There is still high mortality between cohort 1 and 2, but we no longer have those initial spikes in abundance. Also the mortality for WMW-0-S0-M1 is now 0. 

So do I have some misspecification here or is there a bug somewhere in 6702? I do not see those crazy high mortality levels in 6665 with 73-day output.

Thanks

Jacob

--
You received this message because you are subscribed to the Google Groups "Atlantis ecosystem model" group.
To unsubscribe from this group and stop receiving emails from it, send an email to atlantis-ecosystem...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/atlantis-ecosystem-model/ef32f298-a3c6-4e39-8e57-910b053743c2n%40googlegroups.com.

jaco...@gmail.com

unread,
May 6, 2024, 7:30:33 AM5/6/24
to Atlantis ecosystem model
Hi Beth,
It is horizontal migration.
Yes, I am happy to help troubleshoot, please let me know where to put the fprints.
Here are the migration and biological files. If you were looking for something else, please let me know.

Jacob
B391ES3JK5.5e2.4o10a.6702.prm
GroupsIceland6702.csv

Hem Nalini Morzaria Luna - NOAA Affiliate

unread,
May 10, 2024, 4:04:40 PM5/10/24
to jaco...@gmail.com, Atlantis ecosystem model
Hi, following up if you have found the issue? We see anomalous behavior in the first 5 years of our runs for salmon migrators, which we attributed to the initialization migration array, but those issues disappear after year 5 once the initial array burns through.

Hem



--
Hem Nalini Morzaria-Luna (how to pronounce), Ph.D. | Marine ecologist
Long Live the Kings| NOAA - Affiliate | Pronouns: she/her
Learn how to become a salmon warrior 

+1 (206) 861-7605 | twitter: @hmorzaria | LinkedIn 
_____________________________
** My most consistent working hours are 9am-2pm (PST) **
** I will respond as soon as possible. Thanks for your patience. **
----------------------------------------------

Reply all
Reply to author
Forward
0 new messages