Findings from hindcast runs 04/30-05/21, 2002 Part II

1
Findings from hindcast runs 04/30-05/21,
2002Part II
2
List of runs
Week18-20 (04/30-05/21)
Grid Hot/cold start Min. mixing length in open sea (m)
RUN1 original cold (from week18) 3
RUN2 Refined grid cold (from week18) 3
RUN3 (week18 only) original cold 2
3
Comparison of RUN1 and 3

• Differences in the plume region are marginal.
• Some differences in the estuary larger mixing
  length in the plume region leads to lower
  salinities in the estuary.

4
Comparison between RUN1 and 2May 13, 2002

• Differences, although still non-trivial, are much
  smaller than previously observed.
• Temperature comparison is not good as expected.
  (RUN12/CR2325_1205a.gif)
• Salinity values are generally good TSG shows
  RUN2 results are slightly better
  (RUN12/CR2325a_tsg_sal.gif).
• This is mainly because of the cruise route being
  essentially east-west (RUN1/CR2325a_tsg_llgif).
• There are some under-estimations of surface
  salinity in RUN2 (e.g., CR2325-1205a.gif), and
  over-estimation in RUN1 (e.g., CR2325-1223a.gif),
  which suggests plume is still somewhat skewed.
• Towards end of this week, RUN1 elevations
  indicate certain instability which magnifies in
  week20, and may partly explain the large
  differences between the two results in that week.
  (COMP-RUN1-RUN2/nf2c5).

5
Comparison between RUN1 and 2May 15, 2002

• This is a better test for the model as the cruise
  route traverses a wider latitudes.
  (CR2327a_tsg_sal_map.gif)
• For casts, RUN2 results capture better vertical
  profile (RUN12/CR2327_1107a.gif).
• Cast locations are generally close to the mouth
  had they been more away, the accuracy of RUN2
  results might deteriorate.
• For TSG, RUN1 results are slightly better
  (surprise!)
• For this period, the wind is upwelling favorable,
  and plume should be moving southward
  (RUN1/anim-sal_plume_min)
• RUN2 results generally indicate a plume bent
  towards north. Cf. animations of plume at min.

6
Comparison between RUN1 and 2May 15, 2002
(contd)

• However, RUN1 elevation doesnt seem to be right
  there is a large set-up in the plume and open-sea
  region (e.g. nf2c5).
• This is believed to be the consequence of a
  larger minimum mixing length (3m) used.
• As a result, the fresh water spreads faster
  towards the boundary and vertically downwards,
  creating larger baroclinic velocity.
  (RUN1/anim-sal_far_min.gif)
• Results indicate the direction in which plume is
  moving is still not right with the refinement,
  especially for upwelling favorable winds.
  (RUN2/anim-sal_far_min.gif)

7
Future work

• Since raising the minimum mixing length (xl)
  didnt seem to make much difference except
  creating instability, wed run the refined grid
  with original xl (2m).