[Mitgcm-support] Comparison of adjoint solution w/ TAO data

[mitgcm-support at dev.mitgcm.org]

HyperNews notification of new message.  See: 
  http://escher.JPL.NASA.GOV:2000/HyperNews/get/forums/wkmtg/77/1.html

This figure[1] shows the comparisons of model temperature with TAO
temperature data in terms of rms deviation from (upper) and correction
with (lower) the data for simulation (black) and assimilation (of T/P
data) experiements (red, green, blue). The assimilations are all
10-iterations of optimization each, and differ in preconditioned factors
for the gradients in the minimization algorithm such that the adjustment
(of control variables) favors initial state (red), wind stress (blue),
and somewhere in between (green). One can see that the blue one (mostly
correcting wind) results in the smallest rms and the largest correlation
with TAO data, and is better than the black curve (simulation). Noted
that the cost function (deviation of model SSH anomalies from T/P data)
is the smallest for the "blue", then the "green". So for these cases, a
better agreement with T/P data also results in a better agreement with
TAO data. These statistics are based on daily temperature during
Jan-Mar. 1997, averaged over all the TAO moorings. Points: (1) The
assimilation of T/P sea level anomalies improved the comparison with the
independent subsurface temperature data. (2) Results dependent on
preconditioned factors for gradients. The ones corresponding to the blue
curve will be used in the one-year assimilation.
-----------------------------------------------------------------------
[1] http://escher.jpl.nasa.gov:2000/hosts/escher/escher4/medea/tlee/FIG_FORUM/plot_Tstat_ave.ps