Fixes typo in abstract_scalar_biharmonic_diffusivity_closure.jl

[tomchor]

No description provided.

[tomchor]

@navidcy's comment made me realize that AbstractScalarDiffusivity has a similar emphasis on isotropic:

Oceananigans.jl/src/TurbulenceClosures/abstract_scalar_diffusivity_closure.jl

Lines 3 to 8 in 9e44b44

	"""
	abstract type AbstractScalarDiffusivity <: AbstractTurbulenceClosure end
	Abstract type for closures with *isotropic* diffusivities.
	"""
	abstract type AbstractScalarDiffusivity{TD, F} <: AbstractTurbulenceClosure{TD} end

And while that is correct, it may be a bit misleading to users since we can use ScalarDiffusiity to set-up anisotropic diffusivities using Tuples. Should we change to wording there too?

[glwagner]

@navidcy's comment made me realize that AbstractScalarDiffusivity has a similar emphasis on isotropic:

Oceananigans.jl/src/TurbulenceClosures/abstract_scalar_diffusivity_closure.jl

Lines 3 to 8 in 9e44b44

	"""
	abstract type AbstractScalarDiffusivity <: AbstractTurbulenceClosure end
	Abstract type for closures with *isotropic* diffusivities.
	"""
	abstract type AbstractScalarDiffusivity{TD, F} <: AbstractTurbulenceClosure{TD} end

And while that is correct, it may be a bit misleading to users since we can use ScalarDiffusiity to set-up anisotropic diffusivities using Tuples. Should we change to wording there too?

Hm no I don't think it's correct. AbstractScalarDiffusivity is for representing turbulence closures that have a scalar diffusivity, not isotropic diffusion (vertical or horizontal formulations are anisotropic).

[tomchor]

@navidcy's comment made me realize that AbstractScalarDiffusivity has a similar emphasis on isotropic:

Oceananigans.jl/src/TurbulenceClosures/abstract_scalar_diffusivity_closure.jl

Lines 3 to 8 in 9e44b44

	"""
	abstract type AbstractScalarDiffusivity <: AbstractTurbulenceClosure end
	Abstract type for closures with *isotropic* diffusivities.
	"""
	abstract type AbstractScalarDiffusivity{TD, F} <: AbstractTurbulenceClosure{TD} end

And while that is correct, it may be a bit misleading to users since we can use ScalarDiffusiity to set-up anisotropic diffusivities using Tuples. Should we change to wording there too?

Hm no I don't think it's correct. AbstractScalarDiffusivity is for representing turbulence closures that have a scalar diffusivity, not isotropic diffusion (vertical or horizontal formulations are anisotropic).

I changed the text according to your comment. But just to note, I said that is technically correct because the naming given there uses the work isotropic when the ν doesn't change in the directions of the formulation:

Oceananigans.jl/src/TurbulenceClosures/abstract_scalar_diffusivity_closure.jl

Lines 16 to 28 in 75820d6

	"""
	struct ThreeDimensionalFormulation end
	Specifies a three-dimensionally-isotropic `ScalarDiffusivity`.
	"""
	struct ThreeDimensionalFormulation <: AbstractDiffusivityFormulation end
	"""
	struct HorizontalFormulation end
	Specifies a horizontally-isotropic, `VectorInvariant`, `ScalarDiffusivity`.
	"""
	struct HorizontalFormulation <: AbstractDiffusivityFormulation end

Which, again, I think it's a bit misleading, but afaik it's technically correct.

[glwagner]

It's just that "horizontally isotropic" is not the same thing as "isotropic". What is misleading?

[tomchor]

It's just that "horizontally isotropic" is not the same thing as "isotropic". What is misleading?

sorry, nvmd, I'm explaning myself poorly. Nothing is misleading anymore since I removed the word "isotropic" in the last commit. We're good to merge whenever tests pass.