An updated revised MM5 surface layer scheme with multiple options for similarity functions

[prabhakar1508]

Purpose: Inclusion of an updated revised MM5 surface layer scheme with multiple options for similarity functions in stable and convective conditions

TYPE: new feature

KEYWORDS: Updated revised MM5 surface layer scheme, various similarity functions, stable and convective conditions, new namelist flags, new surface layer option

SOURCE: Prabhakar Namdev (Indian Institute of Technology Delhi, India; Karlsruhe Institute of Technology, Germany; Prabhakarnmdv587@gmail.com), Maithili Sharan (Indian Institute of Technology Delhi, India), Piyush Srivastava (Indian Institute of Technology Roorkee, India ), Saroj Kanta Mishra (Indian Institute of Technology Delhi)

DESCRIPTION OF CHANGES:
Problem:
This proposal outlines the scientific advancements, novelty, and community benefits of the updated code. Currently, the following functional forms of similarity or stability correction functions are used in the revised MM5 surface layer scheme for computing the surface fluxes in the WRF model:

1. In stable conditions, the functional forms from Cheng and Brutsaert (2005)
2. In unstable conditions, the functional forms proposed by Fairall et al. (1996)
   We have updated this by incorporating the following functional forms:
   (i) Stable conditions
   In addition to the default one (Cheng and Brutsaert, 2005), the following functions are incorporated:
3. Similarity functions based on Grachev et al. (2007)
4. Similarity functions based on Srivastava et al. (2020)
5. Similarity functions based on Gryanik et al. (2020)
6. Similarity functions based on Holtslag et al. (1990)
7. Similarity functions based on Beljaars and Holtslag (1991)
8. Similarity functions based on Dyunkerke et al. (1991)
   (ii) Unstable conditions
   In addition to the default functions by Fairall et al. (1996), the following functional forms are also incorporated:
9. Similarity functions based on Businger et al. (1971)
10. Similarity functions based on Carl et al. (1973)
11. Similarity functions described by Zeng et al. (1998) based on the three sublayer
    model proposed by Kader and Yaglom (1990)

A separate surface layer module from the current schemes available in the WRF model is designed. It is based on the revised MM5 surface layer scheme (Jimenez et al., 2012) and named as the IITD surface layer scheme (sfclayrev_IITD). It includes a variety of expressions for similarity functions stated above in both stable and unstable conditions, taking into account the fact that their performance is spatially and temporally dependent. Therefore, the surface layer scheme should have a provision for the choice of similarity functions for both stable and unstable conditions. This module has two flags corresponding to stable (psimh_stab_opt) and unstable (psimh_unstab_opt) conditions. The flag "psimh_stab_opt" can take values from 0, 1, 2, 3, 4, 5, and 6, corresponding to the different similarity functions: 0 is default (Cheng and Brutsaert 2005), 1 (Grachev et al., 2007), 2 (Srivastava et al., 2020), 3 (Gryanik et al., 2020), 4 (Holtslag et al., 1990), 5 (Beljaars and Holtslag, 1991), 6 (Dyulkerke et al., 1991). The flag "psimh_unstab_opt" can take values from 0, 1, 2, and 3, corresponding to the different similarity functions in unstable conditions: 0 is default (Fairall et al., 1996), 1 (Businger et al., 1971), 2 (Carl et al., 1973), and 3 (Zeng et al., 1998, based on the three sublayer model by Kader and Yaglom, 1990). In this way, the user will have the flexibility to choose any choice of similarity functions for computing surface fluxes.

The updated module (sfclayrev_IITD) can be used by choosing sf_sfclay_physics=8 in the namelist.input file. The updated module is already evaluated with turbulence measurements over Indian land as well as the CASES99 and ERA5-land reanalysis datasets.
Two of the research articles related to the updated scheme are already published in the Geoscientific Model Development and Atmospheric Research journals. References for the same are provided below:

1. Namdev, P., Sharan, M., Srivastava, P., and Mishra, S. K. (2024): An Updated Parameterization of the Unstable Atmospheric Surface Layer in the Weather Research and Forecasting (WRF) Modeling System, Geosci. Model Dev., 17, 8093-8114, https://doi.org/10.5194/gmd-17-8093-2024
2. Namdev, P., Sharan, M., and Mishra, S. K. (2024): Implementation of various non-linear similarity functions for stable atmospheric surface layer in the WRF modeling system: An evaluation for three contrasting nights in CASES-99 dataset, Atmos. Res., 107825, https://doi.org/10.1016/j.atmosres.2024.107825

Solution:
We wish to point out that some of these functions are capable of dealing with the situation in low wind strong stable/unstable conditions. The incorporated functional forms based on the three sublayer model proposed by Kader and Yaglom (1990) are able to predict the observed non-monotonic behaviour of the drag coefficient (C D ) in unstable conditions reported in various studies (Srivastava and Sharan, 2015; 2021; Maurya et al., 2023; Mosso et al., 2024). Some of these functional forms are recently developed and applicable for a wide range of atmospheric stability conditions.

LIST OF MODIFIED FILES:
M Registry/Registry.EM_COMMON
M dyn_em/module_first_rk_step_part1.F
M dyn_em/start_em.F
M main/depend.common
M phys/Makefile
M phys/module_fddagd_driver.F
M phys/module_fddaobs_driver.F
M phys/module_physics_init.F
A phys/module_sf_sfclayrev_IITD.F
M phys/module_surface_driver.F
M share/module_check_a_mundo.F

TESTS CONDUCTED:

1. The proposed scheme is evaluated against baseline revised MM5 surface layer scheme. The proposed scheme produced physically consistent estimates of surface turbulent fluxes in both stable and convective conditions. Evaluation was performed with case studies using CASES-99 and other benchmark datasets and already published (Namdev et al., 2024; 2025).
2. Jenkins-based tests are not currently configured for this development branch. Code validation was performed through manual compilation with WRF version 4.7.1.

RELEASE NOTE:
This introduces an updated version of the revised MM5 surface layer scheme, providing multiple options for similarity functions under both stable and convective conditions that can be chosen using namelist flags. The implementation is based on the following studies:

1. Namdev, P., Sharan, M., Srivastava, P., and Mishra, S. K. (2024): An Updated Parameterization of the Unstable Atmospheric Surface Layer in the Weather Research and Forecasting (WRF) Modeling System, Geosci. Model Dev., 17, 8093-8114, https://doi.org/10.5194/gmd-17-8093-2024
2. Namdev, P., Sharan, M., and Mishra, S. K. (2025): Implementation of various non-linear similarity functions for stable atmospheric surface layer in the WRF modeling system: An evaluation for three contrasting nights in CASES-99 dataset, Atmos. Res., 107825, https://doi.org/10.1016/j.atmosres.2024.107825

[dudhia]

This should be routed to the develop branch, but also the number of files changed may indicate you did not start with a clean version of our repository. The files changed should be only the few you changed. We can't review it until this is fixed.