309
Figure 9. SSH increment (in m) obtained with the HYCOM model 3 days after
assimilation of vertical T/S profiles.
In order to incorporate analysis increments in a more gradual manner,
a new algorithm based on Incremental Analysis Updates (IAU) has been
proposed by Bloom et al. [1996], which combines aspects of intermittent
and continuous assimilation schemes. Using the regular KF equations,
the IAU algorithm first computes the analysis correction (x a
i+1 x
f
i+1 ).
This correction is then distributed (uniformly or not) over the assimilation window and inserted gradually into the model. The state obtained
at the end of the assimilation window can be used as the initial conditions for the next assimilation cycle, leading to time-continuous filtered
trajectories. The concept of IAU can also be implemented by computing corrections to the atmospheric forcing or by introducing dynamical
relaxation terms in the prognostic equations, according to the analysis
increments. The IAU temporal strategy can be complemented by the
FGAT scheme which computes the innovation “on the flight”.
A possible implementation of the IAU scheme in primitive equation
models may include the following steps: (i) first-guess model integration
and evaluation of the innovation vector using FGAT; (ii) computation
of the analysis increment using the Kalman gain; (iii) second integration
on (t i , t i+1 ) using primitive equations modified by the increment. The
additional cost incurred by this process will be one model integration or
less, depending on the details of the algorithmic implementation. Noting
by T the increment of the temperature field, a possible modification of
OCEAN DATA ASSIMILATION
Figure 9. SSH increment (in m) obtained with the HYCOM model 3 days after
assimilation of vertical T/S profiles.
In order to incorporate analysis increments in a more gradual manner,
a new algorithm based on Incremental Analysis Updates (IAU) has been
proposed by Bloom et al. [1996], which combines aspects of intermittent
and continuous assimilation schemes. Using the regular KF equations,
the IAU algorithm first computes the analysis correction (x a
i+1 x
f
i+1 ).
This correction is then distributed (uniformly or not) over the assimilation window and inserted gradually into the model. The state obtained
at the end of the assimilation window can be used as the initial conditions for the next assimilation cycle, leading to time-continuous filtered
trajectories. The concept of IAU can also be implemented by computing corrections to the atmospheric forcing or by introducing dynamical
relaxation terms in the prognostic equations, according to the analysis
increments. The IAU temporal strategy can be complemented by the
FGAT scheme which computes the innovation “on the flight”.
A possible implementation of the IAU scheme in primitive equation
models may include the following steps: (i) first-guess model integration
and evaluation of the innovation vector using FGAT; (ii) computation
of the analysis increment using the Kalman gain; (iii) second integration
on (t i , t i+1 ) using primitive equations modified by the increment. The
additional cost incurred by this process will be one model integration or
less, depending on the details of the algorithmic implementation. Noting
by T the increment of the temperature field, a possible modification of
OCEAN DATA ASSIMILATION