navidcy / asc-idealised Goto Github PK

@glwagner, @simone-silvestri, do we have partial cell functionality in Oceananigans? How do I set that up if so?

I also flirted with the idea of using the fft solver as a preconditioned.

In order to have Temp changes, we need a Northern sponge with damping of T, S, u, v, and w. From T+S'14:

"We restore a 50 km wide layer to a potential temperature and salinity profile that is representative of the Weddell Sea, with a timescale of 56 days. The selected profile is based on observations from Thompson and Heywood [2008], but smoothed to remove small-scale noise and interpolated to remove the surface AASW layer and the deep layer of AABW. The restoring timescale varies linearly from to 56 days. We similarly restore the velocity to zero in this region using a shorter restoring timescale of 28 days."

I am thinking 100km though, but everything else will be the same. Pending T and S profiles from Pena-Molino et al., 2016.

Title speaks for itself - store density as a variable in the tracers.nc file.

Need to add a z grid which varies in the vertical from 13m to 100m dz over 53 levels (following TS14).

The sponge layer (denoted by the dashed section) represents a high proportion of the total domain width. This is likely interfering with the dynamics of the system significantly. I propose modifying the lengths such that Lx = 1000 km, Ly = 600 km, and the sponge is only 50 km.

I don't understand exactly how this works:

I think it should be so that stretched_grid(1) = -Lz and stretched_grid(Nz+1) = 0.

There is a strong cooling coincident with the salt flux imposed. Is this due to the EOS? Fluxes will obviously be different with a more realistic salinity profile imposed so this problem may be easily resolved with the other issue.

As in title

From Thompson + Stewart, 2014:

We apply a horizontal Laplacian viscosity of 12 m2s−1 and a vertical Laplacian viscosity of 3×10−4m2s−1. We also employ a horizontal biharmonic viscosity with Courant-Friedrichs-Lewy number 0.1, and biharmonic Leith and modified-Leith viscosities [Fox-Kemper and Menemenlis, 2008] with coefficients both equal to 1. The only explicit mixing of potential temperature and salinity is via a vertical diffusivity of 5×10−6m2s−1. This mixing is supplemented by the K Profile Parametrization (KPP) [Large et al., 1994], which enforces a surface mixed layer of at least 50 m depth. The ocean’s equation of state is computed using the numerically efficient algorithm of McDougall et al.[2003] (25 term EOS).

We should ry to match these parameters as closely as possible

hey @glwagner, we want to make a configuration with @txs156 similar to Stewart and Thompson, GRL 2014. We have a shelf... How do we add bottom drag when we have immersed boundary?

These were from the flat-bottom channel...

The tracer is deposited at some intermediate depth. For now, a surface tracer might be more interesting...

I have gotten a hold of cruise profiles for the 2012 113E cruise. Profiles are selected close to 62S for T and S:

The profiles are provided in the TS_profiles folder as netCDf files.

Below is the horizontally-averaged time evolution of the stored tracers and velocities:

And the globally-averaged (z-grid is unweighted in this avg) tracers + velocities:

A few points:

1. The system is spinning up to unrealistic velocities. I believe this is due to the imposed salinity profile, which is negative. I think this is why when you switch on EOS things blow up because negative salinity doesn't exist. We should prioritise the realistic TS profiles before re-running.

2. The system has clearly reached some velocity threshold where it is about to either blow up or go into an inverse cascade, moving towards large dissipative structures.