The Adjoint Sensitivity Guidance to Diagnosis and Tuning of Error Covariance Parameters - Data Assimilation for Atmospheric, Oceanic and Hydrologic Applications - page 216

GPS

TPWater

Radiance

TerrPress

Sfcwind

SpecHum

v−wind

u−wind

Temp

0 1 2 3 −40 −20 0 −6 −4 −2 0

obs count obs impact impact/obs

x 10 7 x 10 −5

Fig. 9.6 The total number of observations assimilated in NAVDAS-AR to produce the 00 UTC
analyses over the time period of study, the total observation impact on the 24-h forecast error
reduction (J Kg 1 ), and the average observation impact/observation (J Kg 1 )

by artificially reducing the assigned o D

o;t (optimal estimate). In this context, the evaluation of the forecast error sensitivity
with respect to a selected parameter in the DAS merely provides guidance on the
parameter variations that are necessary to compensate for (unknown) deficiencies in
the error covariance specification associated with other components of the DAS.

o values from the current specification of

9.4 Results with the Adjoint NAVDAS-AR/NOGAPS

The guidance derived from the error covariance sensitivity analysis and its rele-
vance to the observation impact estimates are presented with the NRL NAVDAS-
AR/NOGAPS and their adjoint versions. The forecast error measure ( 9.8 )isdefined
as the moist total energy norm over the global domain and the verification state
x v is the analysis valid at the forecast time. The results are valid for the 24-h
NOGAPS forecasts associated with the 00 UTC NAVDAS-AR analyses produced