Page 436 - Economia Azul - versão: inglês

understanding; and (iii) the implementation in the computational processing capacity models and, thus, produce an accurate ini- time horizons for the future, and making
of the Argo system from the 2000s onwards employed, both to process and make avail- tial condition for the forecasting systems. this information available to those interest-
to measure vertical profiles of temperature able the collected data and to operational- Assimilation produces a more realistic initial ed in these observations and predictions.
(T) and salinity (S) up to 2000 m depth, to- ly integrate complex physics models with condition and, therefore, positive impacts The predictor system is basically composed
day with more than 3,900 profilers, now be- increasingly higher resolutions (e.g., LE on the predictability of models at short, of a numerical model of ocean circula-
ing expanded to collect T/S data throughout SOMMER et al., 2018; CHASSIGNET and extended, seasonal and even longer time- tion, its atmospheric forcings, usually from
the water column, from the surface to the XU, 2017). Despite the importance of the scales (EDWARDS et al., 2015; MARTIN et forecasts offered by atmospheric models
floor, and house sensors for CO , O , and observed data, its coverage is still limited, al., 2015; MOORE et al., 2019). or coupled ocean-atmosphere-land-ice,
2 2
chlorophyll-a, among others (ROEMMICH et considering that most of the data collected In practice, the ocean forecast routine and a data assimilation system, respon-
al. 2019; LE TRAON et al., 2020). today from EOVs are obtained by satellites also required the development of data pro- sible for building the initial condition of
This revolution in observational sys- that only sample the surface of the oceans. cessing systems to collect daily observations the predictor model. In case the predictor
tems has led to a major advance in the This limit allows only a partial understand- in situ and by remote sensing and to carry system considers the biogeochemical and
understanding of physical processes in the ing of subsurface phenomena, which can out a verification and quality control of the ecosystem cycle quantities, the complexity
oceans and ocean-ice-atmosphere interac- only be observed by in situ sensors, such data that could regularly provide informa- increases, because, in addition to physical
tion, demonstrated with the deployment of as those on Argo and gliders, or collected tion for the assimilation and initialization quantities such as velocity, temperature
high-quality weather and climate forecast- by equipment launched directly from ships. of ocean forecasting systems. Without a and salinity, it is necessary to observe,
ing systems and investigations of climate Therefore, the use of numerical models good data quality control system, data with model and assimilate nutrients, oxygen,
change scenarios. It should also be men- with data assimilation, widely used in oper- gross errors could be assimilated and sub- debris, sediments and many other quan-
tioned that these achievements were cru- ational oceanography, is crucial to comple- stantially compromise the predictability of tities. Thus, operational oceanography is
cially supported by the substantial increase ment observational information. the systems. Efforts have been made by the based on the observation-model-assimi-
international community to improve this lation tripod, since without one of these
2. Operational oceanography observation processing chain for operation- components it is not possible to make an
al purposes based on scientific knowledge accurate numerical prediction that is useful
Supported by the success and devel- 1990s, ocean models were able to rep- in calibration and validation of raw obser- for the various sectors of our society that
opment of numerical weather and climate resent the mesoscale ocean circulation vations. For example, sea surface height demand predictions.
forecasting systems carried out by mete- forced by atmospheric fields with the di- (SSH) data collected by satellites are reca-
orological agencies, the oceanographic urnal cycle. In parallel, satellite radiometry librated over time considering new obser- Notes
community started the development of nu- and altimetry provided global coverage of vations with increasingly accurate sensors. It should be mentioned that the ob-
merical forecasting systems for the oceans the SST and dynamic topography that al- This effort is carried out continuously with served data have errors of various natures,
in the late 1990s to implement the first lowed the mapping of large and mesoscale the regular extension of ocean observation so they do not exactly represent reality,
routine and operational activities for fore- global ocean circulation on a weekly basis. systems (PENNY et al., 2019). which is intangible. There is the instru-
casting the ocean circulation. Based on the Combined with the existing in situ ocean mentation error associated with the sen-
atmospheric paradigm, operational ocean- data network, this data allowed monitor- 2.1 What is operational sor, which is often small right after being
ography and ocean forecasting required ing the state of the upper ocean on a large oceanography? The observation- calibrated in the laboratory, but which can
numerical models of the ocean sufficient- scale, creating a critical mass of information model-assimilation tripod increase substantially after some time in
ly reliable to represent oceanic process- that could be used in an operational system. Operational oceanography is the area of use. For example, it was recently identified

es and their evolution over time over the From the meteorology community, data oceanography dedicated to collecting ob- by those responsible for the Argo system
forecast window, along with atmospheric assimilation techniques were adopted and served data in near real time, assimilating that, due to a problem in the construction
forcings and accurate initial conditions of adapted to employ the available ocean ob- the data, producing numerical predictions of SeaBird Scientific salinity sensors from
the “true” state of the ocean. In the late servations to correct the fields of numerical of the ocean environment with various 2016, the instrumentation error after 2

434 BLUE ECONOMIY From Observation to Data Use 435