Add physical tests for surface fluxes

[szy21]

Purpose

Add physical tests for surface fluxes based on Monin-Obukhov similarity theory

Cost/Benefits/Risks

cost: Time of a few (2) atmos team members benefits: Surface fluxes needs some substantial refactoring, mostly for EDMF and also for the land. Having physical tests can ensure we don't break things (as we did a few times before!) risks: We are designing our own tests as there are not many existing tests that we can follow

People and Personnel

(@ the personnel required/requested to implement the software solution)

• Lead: @akshaysridhar
• Collaborators: @LenkaNovak
• Reviewers: @tapios @szy21

Components

see task breakdown

Inputs

N/A

Results and deliverables

Tests that are run in ci

Task Breakdown And Tentative Due Date

• [ ] Test recover_profile yields logarithmic velocity and thermodynamic profiles under neutral condition, or in the stable/unstable regime but close to the neutral condition
• [x] Test the sign of surface fluxes
• • [x] obukhov_length should be negative for unstable regime and positive for stable regime
• • [x] compute_bstar should be negative for unstable regime and positive for stable regime
• • [x] Momentum flux should be negative for positive velocity and positive for negative velocity
• • [x] For dry case, sensible heat flux should be positive for unstable regime and negative for stable regime, compute_buoyancy_flux should be positive for unstable regime and negative for stable regime.
• • [x] For moist case, evaporation should be positive if q_sfc > q_int and negative if q_sfc < q_int (Any other signs that we can check for moist case?)
• [ ] Compare the magnitude of surface fluxes with Figures 5 and 7 in Nashizawa and Kitamura 2018. The results won't be exactly the same as we don't have the input values, but the magnitude should be comparable.

Proposed Delivery Date

Nov 18, 2022 (We can prioritize it if needed by EDMF)

[tapios]

LGTM. However, please note that the momentum flux can have either sign (depending on sign of velocity);

[szy21]

Thanks for adding in some of the tests. @akshaysridhar @LenkaNovak What is the timeline for the rest of the tests? No hurry, I just want to keep the delivery date up-to-date. I know that instead of following the paper you plan to use era data - please feel free to change the SDI too!

[LenkaNovak]

Akshay already has a qualitative comparison with ERA-5, but we're currently exploring whether we can easily extend this to use surface data instead of data on a pressure level, for an even better agreement. I'll defer to @akshaysridhar for the timelines, since he's leading the effort.