Sanity tests for predicted deposition velocities for species other than ozone should also be performed and documented. For the ozone deposition velocity results, large cities have no deposition values shown. This is incorrect and should be rectified. For example, a city such as Atlanta actually has an extensive vegetative canopy. Possibly, a separate urban "canopy" parameterization ultimately needs to be developed, but in the meantime a default "urban" deposition velocity should be provided for each species.

Thanks @rdsaylor-noaa for the great comments! Working now on soil part and will update my plots here also for other species soon for discussion.

Yes, agreed cities with extensive vegetation fraction need to be included in the future. This may be a longer target (outside of this PR), however, until we can use land use fractions instead of dominant land use currently (see Issue #122 ), or running canopy-app at higher resolution to capture vegetation surrounding cities (e.g., 3 km or 1 km even). Also agree though in the meantime we can try and add a default "urban" deposition velocity for each species. Do you have a recommendation on these urban values to follow for the RACM2 species in your ACCESS?

However, this also requires some consideration at this point (at least before we get vegfrac or higher resolution working) because the entire premise of the gridded part of canopy-app is to only handle vegetated grid cells (with defined vegetation and based on satellite data contiguous canopy thresholds, e.g., lai, canopy height, and canopy fraction). This is a must for all calculations in canopy-app to fundamentally treat the vegetative canopy plant distributions. For now though, I can see adding an "urban_opt" somewhat similar to how we can treat crop and shrub/scrub/grass options (e.g., "crop_opt" and "ssg_opt") in canopy-app and set a default urban ddep gas for these dominant vegtype category of urban.

Thanks again!

Now adding gas soil resistance and dry deposition rates for under canopy conditions, they are no longer zero at z = 0. Here is ozone spatial at Level = 0:

Same date/time as above, Timestamp: 2022-07-01-13:00:00.0000 UTC

And the corresponding vertical point profile with non-zero ddep to soil under relevant canopies:

@rdsaylor-noaa Thinking about your comment more on urban vtype deposition, then this would also apply to other vtype categories that canopy-app drydep does not fundamentally treat (e.g., snow/ice, barren, or water, particularly in global applications). I think this then begins to deviate away from what canopy-app was developed to do, i.e., representative a vegetative canopy-only modulating process model aimed at developing relevant parameterizations that interact with host UFS model components (e.g., LSM) to impact such relevant processes (e.g., in-canopy winds, photolysis attenuation, diffusion, biogenic emissions, drydep etc.). We can think about adding these vtypes for deposition in canopy-app stand-alone (using a vtype "other_opt" or "not_canopy_opt" category with simplifications/assumptions), but would envision the host model to more appropriately handle such other vtype categories rather than what canopy-app interacts with thus modulating relevant canopy processes.

Global tests now work OK after some adjustments, and initially needing to run lower vertical canopy model resolution due to memory/time constraints. This satisfies above task 1.

Shown here for ozone dry deposition at soil level 0 (z = 0) and canopy level 1 (z = 10 m) in this case (5 model levels at 10 m vertical resolution), only for forested canopy regions represented by GEDI canopy heights above thresholds and valid vtypes:

canopy-app is highly tunable in the NL and can extend canopy coverage in areas outside of tall forest canopies using lowest canopy thresholds (i.e., lai_thresh = 0.0, cf_thresh = 0.0, and ch_thresh = 0.0) and turning on ssg_opt and crop_opt to get dry deposition for more low-lying vegetation globally. Here turning on only dry deposition option in the NL and selecting only ozone species output we can use more levels (100 levels at 0.5 m resolution), and output in-canopy for different levels to show larger coverage:

Level 0 (z=0 m)

Level 1 (z = 0.5 m)

Level 2 (z = 1.0 m)

However, as discussed above, canopy-app still only handles vegetative canopies and does not currently include urban, snow/ice, barren, or water vtype categories.

I will note compared to Clifton et al. (2020) review paper with big-leaf ozone dry deposition (Fig. 8 latitudinal averages based on an ensemble of global models, mostly using the Wesely, 1989 scheme), the canopy-app values are very reasonable within the peak latitude band of 45-60 N during August (canopy-app plot shown for July):

Mostly all other ddep species look reasonable, e.g., NO2 and HNO3 below:

However, three species have very low, constant in-canopy deposition velocities, e.g., NO (values ~ 6.352326e-07 cm/s) and CO (3.276465e-07 cm/s) and CH4 (~ 4.680663e-07 cm/s) throughout the canopy levels (at z=0, soil ddep velocities look OK):

@rdsaylor-noaa I do notice that for these species NO, CO, CH4, the Henry's Law coefficients are very low, at least an order of magnitude smaller than all other species at 1.9D-03, 9.8D-04, and 1.4D-03, respectively. Likely this is the reason for their much lower in-canopy ddep velocities. Same with NO3, O3P, and O1D, with orders of magnitude smaller Henry's law coefficients.

@drnimbusrain I think this is a general philosophical difference that we have about what canopy-app should be. I understand your reluctance to expand the scope of the app, but what it leads to is a divided treatment of the deposition velocities, with some being calculated in canopy-app and the rest in the "host" model. For water, ice/snow and bare ground this separation is inelegant, but probably not a significant issue. For urban LU with heavy forest coverage (like Atlanta or Baltimore/Washington) the host model's theoretical framework may or may not be consistent with what is used in canopy-app.

Notwithstanding our differences in philosophy about what canopy-app should be, I approve of what has been done in this PR.

Thanks Rick! Let me have another look at treating urban, barren, snow/ice, and water drydep velocities for these RACM2 species. Do you have a quick place I can get the urban, barren, snow/ice, and water surface resistances for these species? I guess for barren I can treat this the same as I am using for soil surface, but need default urban, snow/ice, and water resistances. I can otherwise go in and look at CMAQ etc.

Probably best to use whatever CMAQ uses. -- Rick D. Saylor, PhD Physical Scientist Atmospheric Turbulence and Diffusion Division NOAA Air Resources Laboratory Oak Ridge, TN 37830
…
On Fri, Mar 7, 2025 at 2:30 PM Patrick Campbell @.> wrote: Notwithstanding our differences in philosophy about what canopy-app should be, I approve of what has been done in this PR. Thanks Rick! Let me have another look at treating urban, barren, snow/ice, and water drydep velocities for these RACM2 species. Do you have a quick place I can get the surface resistances for these species/land-use types? I can otherwise go in and look at CMAQ etc. — Reply to this email directly, view it on GitHub <#149 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGCCI52U7PWHFNNKZEKVHC32THXVPAVCNFSM6AAAAABX5RZQE6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMBXGI2DKOBZGA . You are receiving this because you were mentioned.Message ID: @.> [image: drnimbusrain]drnimbusrain left a comment (noaa-oar-arl/canopy-app#149) <#149 (comment)> Notwithstanding our differences in philosophy about what canopy-app should be, I approve of what has been done in this PR. Thanks Rick! Let me have another look at treating urban, barren, snow/ice, and water drydep velocities for these RACM2 species. Do you have a quick place I can get the surface resistances for these species/land-use types? I can otherwise go in and look at CMAQ etc. — Reply to this email directly, view it on GitHub <#149 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGCCI52U7PWHFNNKZEKVHC32THXVPAVCNFSM6AAAAABX5RZQE6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOMBXGI2DKOBZGA . You are receiving this because you were mentioned.Message ID: @.***>

Thanks Rick! OK, I have added barren land/vtype easily using the same soil resistance parametrization, and I have updated canopy ssg and crop thresholds a bit so that it is easy to override and use those grid cells as well. Here is the updated map so far with much greater ddep coverage:

I will now move to adding snow/ice, water, and urban resistance functions/subroutines and allow for canopy-app drydep to those vtypes (even outside of vegetative canopy regions). Will update full global coverage map, hopefully soon.

Added new snow/ice vtype category functionality (based off of CMAQ snow resistance and reactivity relative to HNO3) as well as dynamic control on ground (both beneath vegetative canopies and outside of canopies, e.g., barren ground and snow/ice) to account for if covered by diagnostic snow using the GFS "snowc_ave" variable.

Also added new snow cover threshold in NL for flexibility: