Plots to consider

[wwieder]

• [ ] increase organic matter content for dry meadow.
• [ ] masking soil moisture when soil temperature < 0 C
• [ ] grey out soil moisture for non-growing season
• [ ] Compare snow depth and soil moisture time series from saddle grid and sensor nodes, respectively. Could add snow depth from sensor network, but these data are messy.
• [ ] look at GCC data from phenocam data (192) in sensor network
  • start with prepare_obs_for_comparison.R to bring in phenocam data (and eventually aggregated to as soil moisture data are being done).
  • Look at raw data: group by vegetation and plot year of data for each node (do this for soil moisture too...)
  • Compare phenocam data with soil moisture and temperature for each node
  • Aggregate together: assuming look OK, calculate mean and standard deviation for each community
  • Compare GCC with GPP from CLM (two y axes will be needed for this, one panel for each community.
  • [ ] Time series of NEE or NEP or NPB or HR or ER from three communities in CLM, maybe just for a year, or a climatology.

[katyarjay]

Plots after increasing organic matter fraction to 1 in dry and moist meadows and masking values where soil temp <0:

Wet meadow soils are too dry after decreasing d_max, so we should think about making this change only in DM and MM if reasonable:

[katyarjay]

Example time series of snow depth and soil moisture from 2019 from observations and CLM:

Timing of observed vs. CLM snow depth in 2019:

Based on snow and soil moisture plots, it looks like we aren't getting enough late spring snowfall in the dry meadow. Maybe we can find a temperature cutoff to increase snowfall in late spring and better capture the observed peak in soil moisture.

Looking at temperature by DoY only when it snows:

Using a temp threshold somewhere around -4 could allow us to increase snowfall after day 100, though it varies a bit by year.

[wwieder]

@swensosc, in our 2017 paper I'd manually been adjusting precipitation so that wet spring snows would stick around in the dry meadow (instead of blowing away as we're doing now). From @katyarjay's plots above it seems like modifying dry meadow snow using an air temperature based modifier would capture this? e.g.

• if TBOT<=269: SNOW=SNOW*0.05
• else: SNOW = SNOW*0.25.

This should prevent some of the early melt that it seems like we're seeing in the dry meadow simulations?

It seems like to do this we'd want to modify repartition_niwot_snowfall in atm2lndMod.

I tried making these changes here /glade/work/wwieder/ctsm/hillslope/cime/scripts/NIWO_3col_3pft_DMporosity2/SourceMods/src.clm/atm2lndMod

but I must not be doing something correctly, because the case won't build. Do you mind taking a look to see what I may be doing wrong here?

[swensosc]

probably an issue with this not being passed in to the subroutine: forc_t_g

On Fri, Mar 4, 2022 at 2:51 PM will wieder @.***> wrote:

@swensosc https://github.com/swensosc, in our 2017 paper I'd manually been adjusting precipitation so that wet spring snows would stick around in the dry meadow (instead of blowing away as we're doing now). From @katyarjay https://github.com/katyarjay's plots above it seems like modifying dry meadow snow using an air temperature based modifier would capture this? e.g.

else: SNOW = SNOW*0.25.```

This should prevent some of the early melt that it seems like we're seeing in the dry meadow simulations?

It seems like to do this we'd want to modify repartition_niwot_snowfall in atm2lndMod.

I tried making these changes here /glade/work/wwieder/ctsm/hillslope/cime/scripts/NIWO_3col_3pft_DMporosity2/SourceMods/src.clm/atm2lndMod

but I must not be doing something correctly, because the case won't build. Do you mind taking a look to see what I may be doing wrong here?

— Reply to this email directly, view it on GitHub https://github.com/NWTlter/NWT_CLM/issues/38#issuecomment-1059557417, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57GXSRX6YQTSJLPCB4DU6KAVRANCNFSM5PVGWB4A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[swensosc]

previous guess was wrong. when I copy your file into my sourcemods, it builds. where is your build log?

On Fri, Mar 4, 2022 at 2:51 PM will wieder @.***> wrote:

@swensosc https://github.com/swensosc, in our 2017 paper I'd manually been adjusting precipitation so that wet spring snows would stick around in the dry meadow (instead of blowing away as we're doing now). From @katyarjay https://github.com/katyarjay's plots above it seems like modifying dry meadow snow using an air temperature based modifier would capture this? e.g.

else: SNOW = SNOW*0.25.```

This should prevent some of the early melt that it seems like we're seeing in the dry meadow simulations?

It seems like to do this we'd want to modify repartition_niwot_snowfall in atm2lndMod.

I tried making these changes here /glade/work/wwieder/ctsm/hillslope/cime/scripts/NIWO_3col_3pft_DMporosity2/SourceMods/src.clm/atm2lndMod

but I must not be doing something correctly, because the case won't build. Do you mind taking a look to see what I may be doing wrong here?

— Reply to this email directly, view it on GitHub https://github.com/NWTlter/NWT_CLM/issues/38#issuecomment-1059557417, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57GXSRX6YQTSJLPCB4DU6KAVRANCNFSM5PVGWB4A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[wwieder]

here's the error I get on build /glade/work/wwieder/ctsm/hillslope/cime/src/build_scripts/buildlib.mct FAILED, cat /glade/scratch/wwieder/NEON_NIWO_3col/bld/mct.bldlog.220304-145006

[swensosc]

I think that is not related to your source mods. Try removing your bld directory and re-building (i.e. rm -rf /glade/scratch/wwieder/NEON_NIWO_3col/bld/), then rebuild.

On Mon, Mar 7, 2022 at 8:59 AM will wieder @.***> wrote:

here's the error I get on build /glade/work/wwieder/ctsm/hillslope/cime/src/build_scripts/buildlib.mct FAILED, cat /glade/scratch/wwieder/NEON_NIWO_3col/bld/mct.bldlog.220304-145006

— Reply to this email directly, view it on GitHub https://github.com/NWTlter/NWT_CLM/issues/38#issuecomment-1060843974, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGRN57CB2U73YS4YGRO7J2LU6YRUNANCNFSM5PVGWB4A . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you were mentioned.Message ID: @.***>

[wwieder]

That seemed to do the trick. @katya, I created a new case to test this out, which is running now, NIWO_hh_DMprecip. I just did a transient simulation from an old restart file so we can compare snow depth and soil moisture profiles. If these simulations look OK we should spin up new simulations from bare ground, as the DM growing season length hopefully gets shorter with these simulations.

[wwieder]

See what you think of this quick run, Katya. I'm not sure it does much for the spring snow pack...

[katyarjay]

It looks like early season snow depth increased and spring snow depth decreased in this run (at least in 2018), which doesn't make sense.. Here are some quick snow depth plots from 2018:

The soil moisture profile is very similar to that of DMporosity5:

Maybe I should try this snow modification with one of the newer cases where I also changed dmax and organic matter fraction (ie, DMporosity5_soilmod9b) so we can compare them directly?

[wwieder]

With this modification reduced winter snow (now 5% of default, previously at 10%) and increased snow when temps were > -4 C (now to 50% of measured). Maybe you want to keep the cold snow at 10% as before.

[katyarjay]

I tried the same snow modification (except cold snow at 10% of measured) with the new case and found that soils were somewhat wetter throughout the entire growing season as a result, but still not capturing the early peak:

Here is the snow depth comparison with the old case and the new case with snow modifications:

[katyarjay]

Here are the plots I made yesterday showing how the soil moisture data vary between sensors at each node. There's a lot of variability in some cases (especially moist meadow sites?) but I'm going to start by averaging across a, b, and c at each plot since I don't feel that confident just picking the "best" one. But happy to try that approach too if you think it would work better.