You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Assumption that building width (Wb) equals street width can be relaxed and Wb can be derived from morphology #803
There is an implicit assumption in the urban building energy model that building width equals street width. However, this assumption can/should be relaxed and building width can be derived from the morphology dataset.
General bug information
CTSM version you are using: release-clm
Does this bug cause significantly incorrect results in the model's science? Yes for urban (greater than roundoff), No for general climate simulations (same gridcell/regional/global climate).
Configurations affected: All configurations with urban landunits
Details of bug
There is an implicit assumption in the urban building energy model that building width equals street width. However, this assumption can be relaxed and building width can be derived from the Jackson morphology dataset. Specifically, the equations which use H/Ws (building height over street width) should use H/Wb (building height over building width). Building width can be derived from (Ws*froof)/(1-froof), where froof is the roof (building) fraction.
A simulation with the new equations indicate that as expected, for froof=0.5, answers are identical (because H/Ws = H/Wb). Differences in other cases depend on the degree of departure from froof=0.5 and the presence and magnitude of space heating and air conditioning. A full assessment of differences by variable, season, density type, and region using ctsm1.0.dev098 can be found in the attached file. This uses the default CLM5 urban datasets.
I have SourceMods for this currently, but can put together a branch. It would be nice to get this into the release branch, in particular for the CESM2 large ensemble runs, since it's not going to affect climate and would make the urban output more useful. Along with the new urban properties data (see PR#591).
@olyson has this been approved to go on the release branch and for cesm2.1.2? I was thinking urban properties update would only go on master. Is that approved to go on the release branch as well?
Also if you could point me to your case where you have the SourceMods that would be helpful. Thanks.
No, hasn't been approved, I'm simply proposing that these go on the release branch. I'm mainly thinking about the large ensemble that is coming up. I'm not sure whether a decision has been made about whether "bug" fixes will be allowed for the large ensemble tag. I will ask Dave when I get a chance.
The SourceMods are confined to one module:
/glade/work/oleson/release-cesm2.0.1/cime/scripts/clm50_cesm201R_1deg_GSWP3V1_CON_FIXBUILDENERGY_2000/SourceMods/src.clm/UrbBuildTempOleson2015Mod.F90
@olyson I looked at your branch, the changes are pretty small and straightforward. And my understanding is that they won't change answers for the current datasets? They just allow different behavior for the future? Is that right?
Do you want to turn the branch into a PR, and we can start moving it to master?
No, they will change answers for the current datasets. The code changes could come to master either by themselves or as part of PR #591 (the change to nlevurb=10, the updated surface dataset, and the updated urbantv file). It would be nice to have the code changes on master sooner rather than later. So I will turn the branch into a PR if that is ok.
@olyson it looks like PR #591 changes answers, so it would be good to have this separated from that with the bit-for-bit part of it as a separate tag. If there's other bit-for-bit parts of #591, we could lift them out into bit-for-bit PR's as well.
Ahh, OK. Thanks for that clarification. As I read #591 I see it as mostly a new feature for CTSM5_1. The new datasets would be triggered with CTSM5_1, but the old used with CLM5_0. It's good to keep things that are a bug fix that change other versions (like clm4_5, and clm5_0) outside of the new feature for a new version. Does that sound right to you? I may need to discuss this with your more to figure this out...
If this changes answers as a bug fix, do we know that the changes are reasonably small?
Activity
ekluzek commentedon Sep 12, 2019
@olyson has this been approved to go on the release branch and for cesm2.1.2? I was thinking urban properties update would only go on master. Is that approved to go on the release branch as well?
Also if you could point me to your case where you have the SourceMods that would be helpful. Thanks.
olyson commentedon Sep 12, 2019
No, hasn't been approved, I'm simply proposing that these go on the release branch. I'm mainly thinking about the large ensemble that is coming up. I'm not sure whether a decision has been made about whether "bug" fixes will be allowed for the large ensemble tag. I will ask Dave when I get a chance.
The SourceMods are confined to one module:
/glade/work/oleson/release-cesm2.0.1/cime/scripts/clm50_cesm201R_1deg_GSWP3V1_CON_FIXBUILDENERGY_2000/SourceMods/src.clm/UrbBuildTempOleson2015Mod.F90
olyson commentedon Sep 12, 2019
The word from Dave is that these should only go on master.
ekluzek commentedon Sep 12, 2019
OK, the changes are to increase nlevurb to 10. Which is done because of the updated surface dataset that is used in the case...
fsurdat = '/gpfs/fs1/p/cgd/tss/people/oleson/urban_sfcdata/Feddema_urban_data_080410/HIGH_RES_VERSION2/URBAN_PROPERTIES_THESIS_TOOL/JUN_22_2018/surfdata_0.9x1.25_16pfts_Irrig_CMIP6_simyr2000_Control_c190405.nc'
It also uses an update urbantv dataset.
And then this change in the code...
olyson commentedon Sep 12, 2019
Note that the change to nlevurb=10, the updated surface dataset, and the updated urbantv file are part of PR #591.
Building energy fix (issue ESCOMP#803).
olyson commentedon Jun 8, 2020
I created a branch for the code changes:
https://github.com/olyson/ctsm/tree/fixbuildenergy
ekluzek commentedon Jun 8, 2020
@olyson I looked at your branch, the changes are pretty small and straightforward. And my understanding is that they won't change answers for the current datasets? They just allow different behavior for the future? Is that right?
Do you want to turn the branch into a PR, and we can start moving it to master?
olyson commentedon Jun 8, 2020
No, they will change answers for the current datasets. The code changes could come to master either by themselves or as part of PR #591 (the change to nlevurb=10, the updated surface dataset, and the updated urbantv file). It would be nice to have the code changes on master sooner rather than later. So I will turn the branch into a PR if that is ok.
ekluzek commentedon Jun 8, 2020
@olyson it looks like PR #591 changes answers, so it would be good to have this separated from that with the bit-for-bit part of it as a separate tag. If there's other bit-for-bit parts of #591, we could lift them out into bit-for-bit PR's as well.
olyson commentedon Jun 8, 2020
There actually aren't any bit-for-bit parts of #591. And this branch with the code changes isn't bit-for-bit either, it's a bug fix.
ekluzek commentedon Jun 8, 2020
Ahh, OK. Thanks for that clarification. As I read #591 I see it as mostly a new feature for CTSM5_1. The new datasets would be triggered with CTSM5_1, but the old used with CLM5_0. It's good to keep things that are a bug fix that change other versions (like clm4_5, and clm5_0) outside of the new feature for a new version. Does that sound right to you? I may need to discuss this with your more to figure this out...
If this changes answers as a bug fix, do we know that the changes are reasonably small?
olyson commentedon Jun 8, 2020
Let's talk at your convenience (I know you are busy with the release and other things this week).
olyson commentedon Jun 8, 2020
The answer changes due to the bug fix are very small, nothing statistically significant. So it seems like the bug fix itself could go onto master.
http://webext.cgd.ucar.edu/I2000/clm50_ctsm10d098_1deg_GSWP3V1_CON_FIXBUILDENERGY_2000/lnd/clm50_ctsm10d098_1deg_GSWP3V1_CON_FIXBUILDENERGY_2000.1991_2010-clm50_ctsm10d098_1deg_GSWP3V1_CON_2000.1991_2010/setsIndex.html
Merge pull request #1034 from olyson/fixbuildenergy