Ocean Observing Systems

A wide variety of observation data types are used in GODAE assimilation systems. The data include both in situ and remotely sensed measurements from space. As will be discussed, each observing system has its own unique data issues and

Stage 4: Adjoint sensitivities

Analysis Components (QC + 3DVAR)

Forecast Component

Fig. 4.1 Chart showing flow of ocean observations through the different stages of ocean data quality control in the U.S. Navy global HYCOM system. Stage 1 sensibility error checks are performed on the raw data; stage 2 external data checks are performed in the fully automated ocean data QC module; stage 3 internal data checks are performed by the iterative solver in the variational assimilation; and stage 4 adjoint sensitivity calculations are done after the forecast before the next QC data cut (see text for details). Note feedback of the HYCOM forecast model fields and prediction errors into the ocean data QC for use as background fields in the next execution of the stage 2 external data error checks

Stage 4: Adjoint sensitivities

Analysis Components (QC + 3DVAR)

Forecast Component

Fig. 4.1 Chart showing flow of ocean observations through the different stages of ocean data quality control in the U.S. Navy global HYCOM system. Stage 1 sensibility error checks are performed on the raw data; stage 2 external data checks are performed in the fully automated ocean data QC module; stage 3 internal data checks are performed by the iterative solver in the variational assimilation; and stage 4 adjoint sensitivity calculations are done after the forecast before the next QC data cut (see text for details). Note feedback of the HYCOM forecast model fields and prediction errors into the ocean data QC for use as background fields in the next execution of the stage 2 external data error checks quality control requirements. Sources of operational ocean data are described in this section.

Most GODAE assimilation centers receive in situ ocean observations over the Global Telecommunication System (GTS). At the current time, data transmitted via the GTS are coded in specific data type formats which use, at most, two decimal places for measurements of temperature and salinity. Further, the existing formats do not allow for additional information about the data in the form of quality flags. However, observational data on the GTS are moving to a new binary format based on BUFR (Binary Universal Form for the Representation of data—a data format maintained by the World Meteorological Organization). In this format, data values can be transmitted with more precision than the existing text-based data formats, and local tables can be added to the message that contain value added or quality assurance information from the data provider. The move to BUFR on the GTS is a long process, scheduled to be completed for all ocean data types in 2016. In addition to the GTS, Argo float data are also available at two global data assembly centers (GDAC): one in the United States at the Naval Research Laboratory, Monterey, California; and the second in France at the Coriolis Data Center, Brest.

There is no standard way satellite oceanographic observations are distributed to GODAE assimilation systems. In some cases, there is a dedicated push from the data provider to the center. In other cases, data are placed on dedicated servers where the observations are then pulled by the center. For example, the GODAE High Resolution SST pilot project has been instrumental in setting up data servers where satellite SST data providers transmit their SST retrievals in near-real-time in a common format (Donlon et al. 2007). This effort has made the availability of SST data from a wide variety of satellite systems commonplace.

0 0

Post a comment