climateprediction.net home page
fortran runtime errors and a question about the model

fortran runtime errors and a question about the model

Message boards : climateprediction.net Science : fortran runtime errors and a question about the model
Message board moderation

To post messages, you must log in.

AuthorMessage
darknightcl

Send message
Joined: 7 Mar 05
Posts: 7
Credit: 441,905
RAC: 0
Message 23346 - Posted: 22 Jun 2006, 21:47:38 UTC
Last modified: 22 Jun 2006, 21:51:46 UTC

Hi,
A few days ago the climate model I\'m running (coupled experiment) generated a fortran runtime error and lost a year of calculation. Right now it is recalculating that year and is almost to the point it failed at before. Since the failure it has started trickling more frequently, usually every time the model starts. The graphics also perpetually display \"calculating yearly data\". Unfortunately I did not take a close look at the error message that was displayed, but I was hoping that someone might have some insight into the problem.

Now for the question about the model itself. If the climate warms enough to cause polar ice caps, like Greenland, to melt, will the model adjust the sea levels to take into account this melting? Will the graphics reflect that change by redrawing the coastlines?

Thanks.
ID: 23346 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 23351 - Posted: 23 Jun 2006, 0:20:49 UTC
Last modified: 23 Jun 2006, 0:23:04 UTC

If the Fortran error mentioned something like /tmp/pipe_error, then it\'s something to do with your hd.
It may be developing bad sectors, or it may just be that the program was doing a read or write when there was a hiccup somewhere.

The persistance of the \"calculating yearly data\" message is just an oddity. It will go away at the start of the next model year.

\"If the climate warms enough ...\"
You need to remember that all these millions of models are just \"what ifs\". They don\'t reflect real life, and the coastlines in the graphics won\'t change, either for climate changes in the real world, or in the model world.
(Apparently it took Tolu a lot of time to get the graphics set up in the first place.)

The melting of large ice sheets will, apparently, take a few hundred years, if at all, and this project will only last, (wild guess), 4-5 years.

ID: 23351 · Report as offensive     Reply Quote
darknightcl

Send message
Joined: 7 Mar 05
Posts: 7
Credit: 441,905
RAC: 0
Message 23355 - Posted: 23 Jun 2006, 4:42:10 UTC - in response to Message 23351.  


\"If the climate warms enough ...\"
You need to remember that all these millions of models are just \"what ifs\". They don\'t reflect real life, and the coastlines in the graphics won\'t change, either for climate changes in the real world, or in the model world.
(Apparently it took Tolu a lot of time to get the graphics set up in the first place.)

The melting of large ice sheets will, apparently, take a few hundred years, if at all, and this project will only last, (wild guess), 4-5 years.



I\'m sorry I guess I wasn\'t clear enough. I\'m not worried about our real ice caps in this case (I think my Conservation biology prof said they could take up to 1000 years to melt - even then not completely, and it all depends on how much carbon dioxide we dump into the atmosphere), I was wondering about the ice caps in the model. I was wondering if the sea levels can rise in the modelled world. I realise that the change is small, perhaps only 1.5m (if anything) over the course of the model simulation, but it still needs to be accounted for. So, can the model allow for a change in the sea levels in the simulation (even if the graphics can\'t show it)?

On a happier note, my model passed the point at which it failed last time, with no problems, and seems to be doing well.
ID: 23355 · Report as offensive     Reply Quote

Message boards : climateprediction.net Science : fortran runtime errors and a question about the model

©2024 climateprediction.net