[Liwg-core] co-chairs notes 1-3-17

Bill Sacks sacks at ucar.edu
Tue Jan 3 11:37:59 MST 2017


One specific question is whether we want to change the rain-snow partitioning at all at this point. If so, we'll need to decide that soon.

Bill

> On Jan 3, 2017, at 11:21 AM, Bill Sacks <sacks at ucar.edu> wrote:
> 
> Hi all,
> 
> Here are the notes from today's meeting. Many relevant notes so I'm just including everything. Some of these notes will make more sense when accompanied by the slides – and there is some good information on the slides that I didn't duplicate here – specifically about the latest atmosphere developments. I assume JF will send out the slides soon.
> 
> One item I'd like to call your attention to: There was some question as to whether Julio's latest runs – specifically GRNL03 – are acceptable in terms of Greenland precipitation. The general feeling was that there might not be much more we can do if the answer is "no", but they'd still be interested in your answer.
> 
> Bill S
> 
> 
> 
>  January 3, 2017 
> 
> CSL
> 
> JF: These numbers may not be updated correctly. We should see our new Cheyenne allocation reflected.
> 
> Jim E: Nothing official on the timing of Cheyenne, but he heard the week of the 20th.
> 
> ----
> 
> Summary of where we stand
> 
> Angular momentum: Decision has been made to turn off the changes to angular momentum. It is available as an option, but it will be off for CESM2: there was too much uncertainty with this, given everything else going on.
> 
> ----
> 
> Sub-grid orography changes
> 
> Propose just making this change over Greenland.
> 
> Julio: While this feels somewhat arbitrary, it is somewhat justifiable based on the conditions over Greenland.
> 
> Julio did some runs where he also applied this over Antarctica. It didn't improve things much there, and he's concerned that it might have detrimental effects in other parts of the model (e.g., WACCM). So he feels that the best thing is to just apply this over Greenland, admitting that we're making this local adjustment.
> 
> This has NOT been tested with WACCM. We need to do that.
> 
> Joe: Ideally, we would apply this globally; we should just be up-front that we're doing this over Greenland because we care about the evolution of land ice in Greenland. Julio agrees, but also feels that there is at least some physical justification for this.
> 
> There was some discussion of whether / how this will work with an interactive ice sheet. Most of the CMIP6 runs will NOT have an interactive ice sheet. Those that do (for ISMIP6) can (at least in principle) regenerate CAM subgrid orography following the same method as Julio has used. (Note that CISM has topography at 4 km, which is sufficiently high resolution for this.) Note that Julio's scheme doesn't use any information about the ice sheet edge – assuming that some of this effect comes from the rough topography left behind as the ice sheet retreats.
> 
> Are ice sheet modelers happy with the final precip? Unsure... feeling is it's probably barely acceptable, but not totally sure.
> 
> ----
> 
> Supersaturation
> 
> Andrew: The instances of unphysical supersaturations are infrequent. They are going to let these unphysical supersaturations stay, because removing them led to too high climate sensitivity. We don't have time to do this right, so are going to live with these infrequent unphysical events.
> 
> ----
> 
> Angular momentum fixes
> 
> WACCM QBO could probably be improved again via tuning. But this would take time and the WACCM group was uncomfortable with this since they don't have a good physical understanding of why these things would change.
> 
> ----
> 
> CAM - model skill
> 
> SST: Overall Rich feels like this isn't too bad. We've done better, but this isn't bad.
> 
> Too much cloud over tropical land: We can probably tune this. If not, we could live with it as is.
> 
> JF: Do you expect any impact on ENSO? Rich: Unsure... we can look.
> 
> Climate sensitivity: Based on just 6 years: seems we have lower sensitivity than 119, but the question is: is it low enough.... Run has been updated to 12 years, and now there seem to be bigger differences. In addition, RESTOM is dropping in the new run as we'd want it to (which we didn't see in 119). Overall, the new 4xCO2 looks more like what we've seen in the past (as opposed to 119).
> 
> ----
> 
> Some other changes that need to come in
> 
> • CLM:
> Dave L: Can bring in latest tunings once other components are ready
> Working on new transient data sets
> Overall, nothing in the way of us finishing soon
> 
> • Ocean:
> Need to change coupling from 2-hour to 1-hour
> Caspian Sea needs to be brought in
> 
> • Chemistry
> Need to update the emissions
> 
> • BGC
> Keith: There's a lot on the MarBL development side that needs to come onto the trunk. Small amount of development needed for marginal seas
> Needs some tuning, but there's a chain of dependencies: e.g., waiting for dust retuning, which is waiting on land retunings.
> 
> • CISM
> Bill S: New grid in place, including new mappings. Will come in in alpha06a. (New projection, and slight update to glacier cover.)
> There are still some other changes needed, but they will have 0 impact on these runs with a diagnostic ice sheet.
> 
> • CIME
> Mariana: In good shape. The biggest need right now is documentation.
> 
> ----
> 
> Estuary Box Model
> 
> John T brought up the question of whether we want to keep the EBM – which we originally brought in because it fixed the Lab Sea Ice – now that it looks like we have a somewhat warmer climate there, and this may not be needed
> 
> Rich: Shouldn't count on the warmer climate – may be decadal variability
> 
> Gokhan: Asks himself that question, too. But feels we should keep EBM because it's more physically realistic anyway.
> 
> ----
> 
> Integrations
> 
> 125 continuing
> 
> 4xCO2 continuing
> 
> We should do runs with some of the other changes, but save Caspian Sea change for later, since there were some issues with that.
> Rich feels we should do Julio's changes separately, so we can have a run with final atmosphere, and then do other changes on top of that.
> 
> Dave L feels we can shoot for new land with updated parameters around the end of the week.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cgd.ucar.edu/pipermail/liwg-core/attachments/20170103/7fa81f27/attachment.html>


More information about the Liwg-core mailing list