search

noaa-oar-arl / canopy-app

Stand-alone/column canopy codes and parameterizations
MIT License
6 stars 7 forks source link

More explicit M17 way of calculating Uc with Eq. 10 and 14. #102

Closed drnimbusrain closed 10 months ago

drnimbusrain commented 10 months ago

This method simply combines Eq. 10 and 14 from Massman et al. 2017 to calculate the Uc, which may result in changes too. @angehung5 Can you test this?

angehung5 commented 10 months ago

Below is the comparison of wind profiles using original calculation and M17 formulation. M17 does better at Uc estimation and probably better at in-canopy wind, but the above-canopy wind speeds are way higher. Also the discontinuity at the canopy top is more significant when using M17.

I am not sure which approach is better. The original calculation has larger bias in terms of Uc but the overall profile is ok and more smooth. M17 has better performance in Uc but the shape of the overall wind profile is kind of strange.

xTE_wind_profile_0600_20200601_20200930_avg_sensitivity_uc_zh

drnimbusrain commented 10 months ago

Seems new M17 approach is overall better here, and I don't see a greater discontinuity (maybe a bit sharper). What do other sites show, and for those that have above canopy? I agree its definitely higher in new M17 approach.

Thanks

On Wed, Jan 17, 2024, 5:37 PM Wei-Ting Hung @.***> wrote:

Below is the comparison of wind profiles using original calculation and M17 formulation. M17 does better at Uc estimation and probably better at in-canopy wind, but the above-canopy wind speeds are way higher. Also the discontinuity at the canopy top is more significant when using M17.

I am not sure which approach is better. The original calculation has larger bias in terms of Uc but the overall profile is ok and more smooth. M17 has better performance in Uc but the shape of the overall wind profile is kind of strange.

xTE_wind_profile_0600_20200601_20200930_avg_sensitivity_uc_zh.jpg (view on web) https://github.com/noaa-oar-arl/canopy-app/assets/107704243/f04afd1c-adb3-4c6d-a368-3961c9f29668

— Reply to this email directly, view it on GitHub https://github.com/noaa-oar-arl/canopy-app/pull/102#issuecomment-1897201058, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGLFYNSRZTTBLDGUW66UYUDYPBHDVAVCNFSM6AAAAABB7DRWWCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQOJXGIYDCMBVHA . You are receiving this because you authored the thread.Message ID: @.***>

angehung5 commented 10 months ago

Seems the Uc estimation based on M17 works fine at Bartlett forest but is much higher at Chestnut, GoAmazon and Willow Creek, as well as above-canopy wind speeds. I would suggest sticking to current approach.

Presentation1

drnimbusrain commented 10 months ago

@angehung5 OK, I agree, its too high. Sorry, but made another slight change to this M17 approach, which should lower it now. Can you check once more these results at four sites? Thank you!

drnimbusrain commented 10 months ago

Sorry, made another change for the zpd calculation to be consistent with updated M17 method. Please check now after latest commit.

drnimbusrain commented 10 months ago

@angehung5 Closing this for now, as we will remain with original approach for now, and work on this in future for more robust unified stability/RSL approach following Bonan.