DYAMOND Summer low-resolution simulation

[sriharshakandala]

Purpose

Add DYAMOND Summer low-resolution simulation.

To-do

Content

---

• I have read and checked the items on the review checklist.

[Sbozzolo]

Could you add documentation before we start looking at the code?

[akshaysridhar]

cc @Sbozzolo

[Sbozzolo]

Isn't this function recompuing some hydrostatic balance? This should be mentioned and explained somewhere, otherwise how could people understand what's the point of these lines

    ᶜ∂lnp∂z = @. -thermo_params.grav /
       (TD.gas_constant_air(thermo_params, TD.PhasePartition(ᶜq_tot)) * ᶜT)
    ᶠlnp_over_psfc = zeros(face_space)
    Operators.column_integral_indefinite!(ᶠlnp_over_psfc, ᶜ∂lnp∂z)
    ᶠp = p_sfc .* exp.(ᶠlnp_over_psfc)

There is no mention anywhere (code, comments, GitHub) of the challanges you were facing and why this code looks like this instead of just reading the pressure.

[akshaysridhar]

Fixed.

[Sbozzolo]

Isn't this function recompuing some hydrostatic balance? This should be mentioned and explained somewhere, otherwise how could people understand what's the point of these lines

    ᶜ∂lnp∂z = @. -thermo_params.grav /
       (TD.gas_constant_air(thermo_params, TD.PhasePartition(ᶜq_tot)) * ᶜT)
    ᶠlnp_over_psfc = zeros(face_space)
    Operators.column_integral_indefinite!(ᶠlnp_over_psfc, ᶜ∂lnp∂z)
    ᶠp = p_sfc .* exp.(ᶠlnp_over_psfc)

There is no mention anywhere (code, comments, GitHub) of the challanges you were facing and why this code looks like this instead of just reading the pressure.

[Sbozzolo]

Could you make this function independent of the parsed_args?

[Sbozzolo]

You can just add DYAMONDSummer to the list below this

[Sbozzolo]

Is the intent for the function added to be general, or is it only for the dyamond case?

[akshaysridhar]

cc @sriharshakandala Do we wish to maintain this overwrite method for all such future initial condition file-reads? Or is there some way this can be done entirely within the LocalState prescription to avoid such an overwrite ?

[sriharshakandala]

With this workflow, the user can define a custom overwrite function for each of the supported initial condition types if, for example, they wish to use data from a file for setting up the initial condition.
Currently, this is primarily intended for the DYAMONDSummer simulation which reads initial condition from file. We are planning to refactor the initial condition prescription to separate out

1. Allocation of prognostic variables and
2. initialization of prognostic variables
   We plan to address this refactoring in a different PR.

[Sbozzolo]

This feature is not being tested anywere, we should add it to buildkite and maybe add a couple of plots too

[Sbozzolo]

Could you add a docstring here with a brief description of this artifact, what it contains, and what's used for?