Skip to content

Commit

Permalink
Paper word count reduced
Browse files Browse the repository at this point in the history
  • Loading branch information
nichollsh committed Sep 20, 2024
1 parent 865f739 commit 9d8d41c
Showing 1 changed file with 6 additions and 7 deletions.
13 changes: 6 additions & 7 deletions docs/paper/paper.md
Original file line number Diff line number Diff line change
Expand Up @@ -61,20 +61,19 @@ AGNI is a new radiative-convective atmosphere model developed with the view of b
* solve for an atmospheric temperature structure which conserves energy and allows for convective stability,
* operate with sufficient speed that it may be participate in a wide parameter space.

This is possible due to the method by which AGNI numerically obtains a solution for atmospheric temperature structure and energy transport. Rather than time-stepping each model level according to radiative heating and applying convective adjustment (cf. HELIOS, Exo_k, and various global circulation models), AGNI uses the Newton-Raphson method to find the state which conserves energy fluxes through the column to a required tolerence. This is similar to the method applied by @drummond_effects_2016 and @goyal_library_2020, although with several optimisations. This allows the model to take tens or hundereds of iterations to obtain a solution, in comparison to thousands or tens of thousands with a time-stepping scheme.
This is possible due to the method by which AGNI numerically obtains a solution for atmospheric temperature structure and energy transport. Rather than time-stepping each model level according to radiative heating and applying convective adjustment (cf. HELIOS, Exo_k, and various global circulation models), AGNI uses the Newton-Raphson method to conserve energy fluxes through the column to a required tolerence. This is similar to the method applied by @drummond_effects_2016 and @goyal_library_2020, although with several optimisations. This allows the model to take tens or hundereds of iterations to obtain a solution, in comparison to thousands or tens of thousands.

In AGNI, radiative transfer is performed under the correlated-k and two-stream approximations with up to 4096 spectral bands distributed between 1 and 35000 cm$^{-1}$ [@lacis_corrk_1991; @stamnes_radiative_2017]. This is done by using SOCRATES[^3], a well established FORTRAN code developed by the UK Met Office [@manners_socrates_2024; @amundsen_treatment_2017; @amundsen_radiation_2014]. Convection is parameterised using mixing length theory [@joyce_mlt_2023]. Condensation and sensible heat transport are also modelled.
Radiative transfer is performed under the correlated-k and two-stream approximations using SOCRATES[^3], a well established FORTRAN code developed by the UK Met Office [@manners_socrates_2024; @amundsen_treatment_2017; @amundsen_radiation_2014]. Convection, condensation, and sensible heat transport are also modelled.

Alongside the problem of interior-atmosphere coupling, it is useful to be able to leverage SOCRATES through an interactive interface -- in this case thanks to Julia. This was applied in the recent paper by @hammond_photometric_2024.
AGNI was used in @hammond_photometric_2024.

[^2]: The PROTEUS framework can be found on GitHub [here](https://github.com/FormingWorlds/PROTEUS).
[^3]: Despite SOCRATES being developed under the BSD 3-Clause license, its [main development repository](https://code.metoffice.gov.uk/trac/socrates) is not publically accessible. The code has been re-hosted on GitHub, with additional tools, under the same open-source license [here](https://github.com/nichollsh/SOCRATES). It has also been deposited on Zenodo: @nicholls_socrates_2024.
[^3]: SOCRATES is mirrored on GitHub, with additional tools, under its original open-source license [here](https://github.com/nichollsh/SOCRATES).


# Acknowledgements

The author is supported by the Clarendon Fund and the MT Scholarship Trust.
The author is grateful for the continuing hard work of the Julia developers and those of its many libraries [@julialang].
Additionally, the author thanks James Manners, and others at the Met Office, who have continued to support and develop SOCRATES over the last three decades.
HN is supported by the Clarendon Fund and the MT Scholarship Trust.
We acknowledge the continuing hard work of the Julia developers and those of its many libraries [@julialang].

# References

0 comments on commit 9d8d41c

Please sign in to comment.