ACC effective diffusivity results


Section A: Links to summary pages

1. Twentieth degree resolution global tracer calculation

2. Quarter degree resolution global tracer calculation

3a. Twentieth degree BIGBIT tracer calculation

3b. Twentieth degree BIGBITII tracer calculation

3c. Twentieth degree bit tracer calculation

4. Twentieth and hundredth degree BIT tracer calculation

5. Twentieth and hundredth degree BIT Keff calculation

6. Peclet vs Nusselt


Section B: List of calculations, links to datasets, and comments

Quarter degree

DiffusivityKeff plotKeff data
K10 (with relax)K10L.dat, pL.dat
K100 (with relax) K100L.dat, pL.dat
K100-NORELAXK100-NORELAXL.dat, pL.dat
lyapftleFTLE,ftlehat

L-kdep (quarter degree K10, K100)

L and ftlehat (quarter degree K100)

Summary

  1. K10 seems to be too small diffusion at this resolution
  2. Relaxation doesn't make too much difference
  3. Keff and Lyap (ie ftlehat) don't have v similar structure but similar maginitude when appropriately scaled (note - peaks at edges of ftlehat are spurious artefacts of averaging method).
Notes
  1. Keff calculations are normalised by Leq from K10000 run
  2. FTLE calcs are after t=200days (ie a couple of eddy turnover times)
  3. Latitudes are: lats=-29.775-0.05*findgen(145)

Twentieth degree

DiffusivityKeff plotKeff data
K10 (with relax)K10L.dat, pL.dat
K10 (without relax)K10L.dat, pL.dat
K50 (without relax)K50L.dat, pL.dat
K50 (no mean)K50L.dat, pL.dat
lyapftleFTLE,ftlehat

L-kdep (twentieth degree K10, K50)

L (twentieth and quarter degree K10)

L (twentieth K50 and quarter K100)

L and ftlehat (twentieth K50)

L*knum (twentieth K50 and K10), knum estimated to be 16.6719 (k=10) and 60.1074 (k=50)

Summary

  1. K10 is indeed too small at quarter degree
  2. K10 and K50 quite similar to eachother and also to K100 at quarter degree - all have lowest values of about 500m2/s and highest values of about 2000m2/s
  3. Keff and Lyap (ie ftlehat) don't have same structure but similar maginitude when appropriately scaled (note - peaks at edges of ftlehat are spurious artefacts of averaging method).

Twentieth degree (BIGBIT)

DiffusivityPlotData
K10pL-K10L.dat, pL.dat
K50pL-K50L.dat, pL.dat
lyapftleFTLE, ftlehat

patch results

L and ftlehat (twentieth K50)

Summary

  1. Some similarity between patch results
  2. Plausible similarity between Keff at K50 (not K10) and lyap exps

Twentieth degree (BIGBITII)

DiffusivityPlotData
K1pL-K1L.dat, pL.dat
K10pL-K10L.dat, pL.dat
K25pL-K25L.dat, pL.dat
K50pL-K50L.dat, pL.dat
K100pL-K100L.dat, pL.dat
K500pL-K500L.dat, pL.dat
lyapftleFTLE, ftlehat

patch results

L and ftlehat (twentieth K50)

Summary

  1. Some similarity between patch results (esp for zoom of tracer and patch; differences in zoom of Keff and patch - warning re interpretation of local features on gloabl Keff picture)
  2. Plausible similarity between Keff at K50 and K10 and lyap exps (also ftlehat)

Twentieth degree (BIT01again and BIT01againNOMEAN )

DiffusivityPlotData
K0.1pL-K0.1L.dat, pL.dat
K1pL-K1L.dat, pL.dat
K10pL-K10L.dat, pL.dat
K100pL-K100L.dat, pL.dat
K10-nomeanpL-K10-nomeanL.dat, pL.dat
lyapftleFTLE, ftlehat

patch results

L and ftlehat (twentieth K10)

Summary

  1. Some similarity between patch results - definitely correct latitude ranges.

Hundredth degree

DiffusivityPlotData
K0.1pL-K0.1L.dat, pL.dat
K1pL-K1L.dat, pL.dat
K2.5pL-K2.5L.dat, pL.dat
K10pL-K10L.dat, pL.dat
K25pL-K25L.dat, pL.dat
K50pL-K50L.dat, pL.dat
K100pL-K100L.dat, pL.dat