lat-lon grid 1/8 global simulations (lle.nb.01)

North Atlantic

Here are some comparisons between the outputs from the ECCO2-MITgcm ll2.nb.01 simulation (lat-lon 1/8 global) and observations from Levitus 94 and the SST from AMSR-E satellite data. We focus on the temperature and salinity fields.

MODEL: monthly climatology from Jun./2002 to Mar./2005 of THETA (monthly averaged available here)
DATA: monthly climatology of SST from the TMI + AMSR-E OI product (limited to Jun./2002 to Mar./2005)
DATA2: monthly climatology of Levitus 94 (THETA)

• Surface THETA, yearly mean

MODEL vs Levitus94

MODEL vs AMSR-E

MODEL

Levitus94

AMSR-E

• Surface THETA, Jan-Feb-Mar mean

MODEL vs Levitus94

MODEL vs AMSR-E

MODEL

Levitus94

AMSR-E

• Monthly surface fraction of each outcrops over the North Atlantic Ocean (%)
  

• Model (lle.nb.01) vs Levitus94, THETA, meridional sections, Jan-Feb-Mar mean:
  

  Over the bassin 290E:10:340E

  focus on 300E (model: red, data: black dashed)

Global

• Monthly climatology

Surface THETA monthly climatology lle.nb.01 vs Levitus94

Surface salinity monthly climatology lle.nb.01 vs Levitus94

Dec->May SST seasonal cycle from MODEL (left) and Levitus94 (right) (monthly clim. minus year mean)

Dec->May SSS seasonal cycle from MODEL (left) and Levitus94 (right) (monthly clim. minus year mean)

Jun->Nov SST seasonal cycle from MODEL (left) and Levitus94 (right) (monthly clim. minus year mean)

Jun->Nov SSS seasonal cycle from MODEL (left) and Levitus94 (right) (monthly clim. minus year mean)

cube-sphere grid 510 global simulations (cube49)

North atlantic

• Monthly volume time series

- Here is a list of plots showing volume time series of each potential temperature classes. Fields analysed are monthly means and the domain goes from:
longitude: [276.0625 359.9375]
latitude: [12.0975 53.2011]
depth: [5 1.0072e+03]

RQ: The volume corresponds to the amount of water having a temperature between T-0.5 and T+0.5.

THETA class (deg. C) = [-2; -1; 0; 1; 2; 3; 4; 5; 6; 7; 8; 9; 10; 11; 12; 13; 14; 15; 16; 17; 18; 19; 20; 21; 22; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37; 38; 39; 40]

- Heat content of the domain: H(t) = sum_T: V(T,t)*RHO(T,S=35)*Cp*T

• Monthly volume time series: CUBE49 vs ECCO-GODDAE

- Comparison with the ECCO-GODAE solution: Adjoint 1x1 grid iteration 199 (real month averages, allow freezing of water at surface)
Legend: blue: the cube49, red: iter199

RQ: To avoid systematic errors due to grid and fields resolutions, we only plot the fraction of the domain volume for each THETA class

THETA class (deg. C) = [-2;-1;0; 1; 2; 3; 4; 5; 6; 7; 8; 9; 10; 11; 12; 13; 14; 15; 16; 17; 18; 19; 20; 21; 22; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37; 38; 39; 40] - Heat content of the domain: H(t) = sum_T: V(T,t)*RHO(T,S=35)*Cp*T
RQ: The red line is the rough computation of the H for the iter199 solution. But differences in grid resolution (1 vs 1/4) makes the total volume in iter199 bigger than in cube49. So the magenta line is the iter199 solution with a rectified total volume to match the cube49 one.

Last, an average temperature difference between years [n-1:n] and [1:2] is computed as: dT = dH / Cp / Vtot / RHO_0

• Surface THETA, yearly mean

CUBE49 vs AMSR-I

CUBE49 vs Levitus94

CUBE49

Levitus94

AMSR-I

• Surface THETA, Jan-Feb-Mar

CUBE49 vs AMSR-I

CUBE49 vs Levitus94

CUBE49

Levitus94

AMSR-I

• CUBE49 vs Levitus94, Meridional sections, JFM mean. (compair with lat-lon 1/8 run here)

cube-sphere grid 510 global simulations (cube54)

North atlantic

• Monthly volume time series

- Here is a list of plots showing volume time series of each potential temperature classes. Fields analysed are monthly means and the domain goes from:
longitude: [276.0625 359.9375]
latitude: [12.0975 53.2011]
depth: [5 1.0072e+03]

RQ: The volume corresponds to the amount of water having a temperature between T-0.5 and T+0.5.

THETA class (deg. C) = [-2; -1; 0; 1; 2; 3; 4; 5; 6; 7; 8; 9; 10; 11; 12; 13; 14; 15; 16; 17; 18; 19; 20; 21; 22; 23; 24; 25; 26; 27; 28; 29; 30; 31; 32; 33; 34; 35; 36; 37; 38; 39; 40]

- Heat content of the domain: H(t) = sum_T: V(T,t)*RHO(T,S=35)*Cp*T



And here is a detailed view of the both Volume and Heat content: