Royal Dutch Meteorological Institute; Ministery Of Infrastructure And The Environment

Publications, presentations and other activities
Toward Scatterometer Winds Assimilation in the Mesoscale HARMONIE Model
2017
by G.J. Marseille (KNMI), A. Stoffelen (KNMI),

Data assimilation (DA) experiments have been conducted
with the high-resolution limited-area model HARMONIE,
which is operational at most weather centres which are part of
the European HirLAM consortium. Recently, the assimilation
of scatterometer ocean surface winds was introduced, showing
limited forecast skill improvement. Possible explanations are
discussed. These include model bias and the time mismatch
between observation and analysis time which introduces nonnegligible
correlated errors in a three-dimensional variational
(3D-Var) assimilation system. Also, ignoring the time mismatch
increases the innovation, i.e., the observation minus background
(model short-term forecast), by about 20% for scatterometer
winds. The use of observations as point observations in most
DA systems needs reconsideration for mesoscale DA. The introduction
of observation operators, taking into account the
instrument footprint, would improve the innovation by about 5%
for scatterometer winds. Additional directions for improved use
of observations in HARMONIE are discussed based on the notice
that DA is an inherent deterministic concept. Hence, the selection
of the spatial scale for deterministic data assimilation should
depend primarily on the four-dimensional observation coverage
rather than the effective model resolution.

Bibliographic data
Marseille, G.J. and A. Stoffelen, Toward Scatterometer Winds Assimilation in the Mesoscale HARMONIE Model
IEEE Journal of Selected Topics in Applied Earth O, II, 2017, 10, 5, 2383-2393, doi:10.1109/JSTARS.2016.2640339.
Abstract (html)