176
IAN ROBINSON
As yet there is no operational ingestion of such data into wave models, but
the improved spectra have raised new interest in doing so.
4.
Preparing satellite SST for assimilation into models
4.1
Introduction
If SST data from space are to be used operationally as part of ocean
observing systems and for creating a reliable, stable climate time series,
there is a need to harmonise and inter-calibrate the SST products already
being produced by several different agencies. Specifically for use in ocean
forecasting models there is also a need to precondition the data to make them
more immediately usable for assimilation into numerical models. This is
true of most types of satellite data required by models although the
processing tasks for preconditioning vary according to the parameter of
interest. The blending and preparation of SST data differs from what has to
be done for altimetric measurements of sea surface height anomaly or for
parameters such as chlorophyll concentration derived from satellite ocean
colour data. The rest of this chapter describes a new international initiative
to perform the intermediate processing tasks needed to generate the best
coherent products from complementary SST data delivered by several
agencies. It serves as a specific illustration of the organisation and initiative
that is desirable in order to enhance the usability in model assimilation of
any observational parameter derived from several independent measurement
programmes, particularly from satellites.
In 2001 the Global Ocean Data Assimilation Experiment (GODAE)
prompted the creation of a working group which has developed into the
GODAE high resolution sea surface temperature pilot project (GHRSSTPP). In the rest of this chapter, section 4.2 identifies the particular issues
relevant to SST data and how these have been rationalised by the GHRSSTPP and formalised in the GHRSST Data Processing Model. Section 4.3
describes how a European project called Medspiration is about to start
producing products conforming to the GHRSST-PP specifications. Section
4.4 concludes with a discussion of what can be learned more generally from
the example of GHRSST-PP and Medspiration.
4.2
The challenge of assimilating SST data from many
sources
4.2.1
Sampling and resolution capability of existing sensors
Section 3.3 noted that satellite sensors for measuring SST may be
differentiated into four classes. The derivation of SST from the top-of-
IAN ROBINSON
As yet there is no operational ingestion of such data into wave models, but
the improved spectra have raised new interest in doing so.
4.
Preparing satellite SST for assimilation into models
4.1
Introduction
If SST data from space are to be used operationally as part of ocean
observing systems and for creating a reliable, stable climate time series,
there is a need to harmonise and inter-calibrate the SST products already
being produced by several different agencies. Specifically for use in ocean
forecasting models there is also a need to precondition the data to make them
more immediately usable for assimilation into numerical models. This is
true of most types of satellite data required by models although the
processing tasks for preconditioning vary according to the parameter of
interest. The blending and preparation of SST data differs from what has to
be done for altimetric measurements of sea surface height anomaly or for
parameters such as chlorophyll concentration derived from satellite ocean
colour data. The rest of this chapter describes a new international initiative
to perform the intermediate processing tasks needed to generate the best
coherent products from complementary SST data delivered by several
agencies. It serves as a specific illustration of the organisation and initiative
that is desirable in order to enhance the usability in model assimilation of
any observational parameter derived from several independent measurement
programmes, particularly from satellites.
In 2001 the Global Ocean Data Assimilation Experiment (GODAE)
prompted the creation of a working group which has developed into the
GODAE high resolution sea surface temperature pilot project (GHRSSTPP). In the rest of this chapter, section 4.2 identifies the particular issues
relevant to SST data and how these have been rationalised by the GHRSSTPP and formalised in the GHRSST Data Processing Model. Section 4.3
describes how a European project called Medspiration is about to start
producing products conforming to the GHRSST-PP specifications. Section
4.4 concludes with a discussion of what can be learned more generally from
the example of GHRSST-PP and Medspiration.
4.2
The challenge of assimilating SST data from many
sources
4.2.1
Sampling and resolution capability of existing sensors
Section 3.3 noted that satellite sensors for measuring SST may be
differentiated into four classes. The derivation of SST from the top-of-