An Overview of Global Observing Systems relevant to GODAE

Abstract : A global ocean observing system for the physical climate system, comprising both in situ and satellite components, was conceived largely at the Ocean Observations conference in St. Raphael. France, in October 1999. It was recognized that adequate information was not available on the state of the world ocean or its regional variations to address a range of important societal needs. Subsequent work by the marine carbon community and others in the ocean science and operational communities led to an agreed international plan described in the Global Climate Observing System (GCOS) Implementation Plan (GCOS-92, 2004). This foundation observing system was designed to meet climate requirements, but also supports weather prediction, global and coastal ocean prediction, marine hazard warning systems, transportation, marine environment and ecosystem monitoring, and naval applications. Here, we describe efforts made to reach the goals set out in the international plan. Thanks to these efforts, most of the ice-free ocean above 2000-miles is now being observed systematically for the first time, and a global repeat hydrographic survey and selected transport measurements supplement these networks. The system is both integrated and composite. It depends upon in situ and satellite networks that measure the same variable using different sensors. In this way, optimum use is made of all available platforms and sensors to maximize coverage and attain maximum accuracy. The biggest challenge for the greater oceanographic community -- including both research and operational components -- will be demonstrating impacts and benefits sufficient to justify the funds needed to complete the observing system, as well as to sustain its funding for the long term.

System (GCOS) Implementation Plan (GCOS-92, 2004).This foundation observing system was designed to meet climate requirements, but also supports weather prediction, global and coastal ocean prediction, marine hazard warning systems, transportation, marine environment and ecosystem monitoring, and naval applications.Here, we describe efforts made to reach the goals set out in the international plan.Thanks to these efforts, most of the ice-free ocean above 2000 m is now being observed systematically for the first time, and a global repeat hydrographic survey and selected transport measurements supplement these networks.
The system is both integrated and composite.It depends upon in situ and satellite networks that measure the same variable using different sensors.In this way, optimum use is made of all available platforms and sensors to maximize coverage and attain maximum accuracy.
Wherever feasible, observations are transmitted in real time or near-real time to maximize their utility, from short-term ocean forecasting to estimation of century-long trends.Because our historical knowledge of oceanic variability is limited, we are learning about the sampling requirements and needed accuracies as the system is implemented and exploited.The system will evolve as technology and knowledge improve.
The biggest challenge for the greater oceanographic community-including both research and operational components-will be demonstrating impacts and benefits sufficient to justify the funds needed to complete the observing system, as well as to sustain its funding for the long term.

GlObAl ObSErViNG SyStEMS
The Global Climate Observing System (GCOS) Implementation Plan (GCOS-92, 2004) serves as a useful starting point.It calls for the phased implementation of an integrated and composite satellite and in situ observing system, with related data management and analysis activities. Figure 1 shows the 10-year implementation ramps, the year-by-year progress in reaching the 10-year goals, and the status to date.
in Situ Observing Networks Successful operation of a global in situ observing system requires coordination of activities on a number of levels.
Sensors and best practices need to be agreed on.Deployment opportunities need to be identified and instruments delivered to take advantage of them; where no opportunistic deployment is feasible, timely provision of special deployment efforts needs to be made.

Fixed-Point Observing Networks
The networks of this type are the Global

Satellite Observing Systems
The  # buoys= 1217   well as support seasonal forecasting (Balmaseda et al., 2009).These same data records, when combined with those from complementary altimeters (like that on the European Space Agency's Environmental Satellite [Envisat] and the US Navy's recent Geosat Follow-On [GFO]), enable an approximation of the oceanic mesoscale field-the ocean's weather-and contribute to many applications such as marine safety (Davidson et al., 2009), marine pollution monitoring (Hackett et al., 2009), hurricane intensity forecasting (Goni et al., 2009), and Naval applications (Jacobs et al., 2009).They also provide boundary conditions for nested coastal models (De Mey et al., 2009), support surface wave forecasting, and help characterize the physical context for marine ecosystems (Brasseur et al., 2009).See http:// www.aviso.oceanobs.com/.

Ocean Surface Vector Winds
For more than a decade and a half,

Additional Variables
There are several additional variables that-while important-have been of less direct relevance to GODAE, so they will only be mentioned in passing.A continuous ocean color climate record was initiated in 1997 and is being continued by several satellites (see http:// oceancolor.gsfc.nasa.gov/).Observations of Earth's gravity field have been

SuStAiNiNG thE ObSErViNG SyStEMS
At present, the majority of the in situ  For the satellite observing systems, the issue of transitioning support from research agencies to operational agencies is a critical one, and both the technical feasibility of observing a given variable and the scientific utility of the resulting observations needs to be demonstrated.
In the transition process, the research and operational agencies share the next step, demonstration of the operational utility of an observation, that is, that availability of the observation will have a significant impact on an operational agency's ability to meet a mission need.This justification entails convincing the government supporting that operational agency of the potential impact or value, in terms of societal relevance, of the data collected by a given observing system.Note that some impacts can be expressed in more immediate and quantifiable terms, for example, how much a weather forecast is improved where the impact is realized within hours to days of the time when a given set of observations has been collected.Other values may be expressed in terms of the variable's role in the climate system, for example, the specific impact of the collection of a given set of observations to assess or quantify that variable's role in climate may not be realized for years to decades.

Ensuring Sustained Operations
As the operational agencies collaborate with their research counterparts to ensure sustained operations of the global observing systems, there are particular challenges to be faced.

Societal Relevance
The operational agencies need to make the case that what is proposed to be implemented on a long-continuing basis is worth a corresponding continuing investment of tax dollars over the long term.This rationale is often different from making the case within a research agency.If there were some degree of community consensus based, for example, on compelling issues of societal relevance, it could be used as the basis for prioritization.

Clear, Concise, and Consistent Message
Securing the resources to implement a sustained infrastructure for observing the global ocean will require a clear, concise, and consistent message coming from the community at large that reflects priorities in a progression of successive steps.

lOOkiNG tO thE FuturE
Although moving an observational capability from the point of theoretical possibility to an ongoing, sustained reality is a decades-long process, it is important to note that significant progress has been made.For the first time, most of the ice-free ocean above 2000 m is now being observed systematically.We need to concentrate on the near-term opportunities, as well as engage in a number of international activities that could have significant benefit for promoting G O D A E S p E c i A l i S S u E F E At u r E An Overview of Global Observing Systems relevant to GODAE AbStr Act.A global ocean observing system for the physical climate system, comprising both in situ and satellite components, was conceived largely at the Ocean Observations conference in St. Raphael, France, in October 1999.It was recognized that adequate information was not available on the state of the world ocean or its regional variations to address a range of important societal needs.Subsequent work by the marine carbon community and others in the ocean science and operational communities led to an agreed international plan described in the Global Climate Observing

Figure 1 .
Figure 1.(facing page) Status of the in situ Global Ocean Observing System (GcOS) against targets defined by the GcOS implementation plan and accepted by the World Meteorological Organizationintergovernmental Oceanographic commission Joint technical commission for Oceanography and Marine Meteorology (JcOMM).
b y c A N D y c E c l A r k A N D t h E i N S i t u O b S E r V i N G S y S t E M A u t h O r S , A N D S tA N W i l S O N A N D t h E S At E l l i t E O b S E r V i N G S y S t E M A u t h O r Sbe provided, including that for nearterm operational purposes; and these observations must be integrated with others so that delayed-mode quality control can be done for more exacting research applications.This paper discusses the current status of those global ocean observing systems that are relevant to, and whose data have been used by, the Global Ocean Data Assimilation Experiment (GODAE).It concludes with a discussion of some of the challenges facing these observing systems in our effort to establish sustained funding for them.

iNtrODuctiON
The delivery of ocean services to society depends upon operation of an observing system adequate to support the services desired, analysis systems to integrate all available observations and permit the extraction of ocean information, and appropriate assimilation/analysis/ forecast systems to deliver forecasts of the desired extent into the future.An effective observing system-in situ and/ or satellite-depends upon many data system elements.In particular, measurements must be made with sensors whose characteristics are understood and acceptable; observations from the sensors must be transported to a facility where they can be assembled and given preliminary quality control; access by the wide range of potential users must The system's data coverage needs to be monitored along with sensor lifetimes and provision made to anticipate where gaps will appear so that deployment can be arranged.Successful implementation depends fundamentally upon near-realtime transmission of both observations and relevant metadata.Given that a number of nations participate in each of the observing networks and both "operational" and "research" programs are involved, this monitoring/system management function is critical.There are two different classes of observing activities underway in situthose from fixed points and those whose locations vary with time.Fixed-point observations are made either from moorings or from repeated occupation of stations.Observations whose locations vary with time are made from platforms that move as a result of the ocean's motion or that of a moving vessel.Some moving platforms are thought to follow the motion of water parcels fairly well (i.e., "Lagrangian").

Figure 3 .Figure 4 .
Figure 3. configuration of the Global Sea level Observing System/GcOS core Network in 1999 (left) and 2008 (right).There have been important improvements in the number of tide gauges reporting high-frequency data in real time during the GODAE period.GlOSS = Global Sea level Observing System.
research space agencies have made great progress over the past three decades.Today, spaceborne sensors have a demonstrated capability to collect data for a variety of variables, including altimetry to observe ocean surface topography or sea level, scatterometry to observe ocean surface vector winds, infrared and microwave radiometry to observe sea surface temperature, microwave radiometry to observe sea ice cover, and visible and near-infrared radiometry to observe ocean color.Only several representative satellite systems will be discussed in this article, and only one of those at any length; for

Figure 5 .
Figure 5.The configuration of the surface drifting buoy network on September 13, 1999 (left), and on October 27, 2008 (right).The network reached its initial implementation goal of 1250 in 2005.

Figure 6 .
Figure 6.The global distribution of expendable bathythermograph (Xbt) observations from ships of opportunity in 1999 (left) and 2005 (right).The total number of Xbt profiles from the Ship of Opportunity programme (SOOp) decreased during the GODAE period as the Argo array was implemented.
in particular, the value of ~ 3.1 mm yr -1 from altimeters over the decade beginning in 1993(IPCC, 2007)  is almost twice the estimate of ~ 1.7 mm yr -1 from tide gauges over the past century.To understand and improve the projections of GSLR, it is necessary to collect systematic observations of the two major contributors-thermal expansion due to the warming ocean and the addition of melt water due to the warming of terrestrial ice sheets and glaciers.Thermal expansion estimates-the addition of melt water, for example, by measuring changes in gravity of both the ice sheets and oceanic water masses, as well as changes in the topography and flow rate of glaciers and ice sheets.Together with Jason and Argo observations, these estimates can be used to infer a contribution from melting glaciers and ice sheets as a consistency check for these research efforts, as well as to help assess the performance of climate models projecting sea level rise.Oke et al. (2009) describe how GODAE systems have used observations from different observing systems to meet the needs of a variety of operational oceanographic applications.For example, the climate data record of sea level from Jason-class satellite altimetry-together with Argo float profiles and satellite observations of SST (Donlon et al., 2009)-is required to characterize decadal variability in the ocean and its relation to droughts, floods, and fishery regime shifts, as , was launched in 1999 and is still operating today; the first fully operational scatterometer, the Advanced Scatterometer (ASCAT), on EUMETSAT's Metop-A satellite, was launched in 2006 with units on MetOp-B and -C to follow.Observations of ocean surface vector wind fields are needed for operational forecasting as well as research.For the former, they are needed for early detection, tracking, and characterization of hurricanes and tropical systems; observing and forecasting surface waves and storm surge; detection and characterization of extra-tropical, hurricaneforce winter storms; and observing and forecasting localized wind events and frontal passages (e.g., Figure 8).For research, scatterometer observations provide fundamental characteristics of the wind forcing that drives the oceanic circulation.Moreover, such observations will be key in documenting extreme weather events at sea-events that are thought to become more frequent and intense with our warming climate.See http://winds.jpl.nasa.gov/missions/quikscat/ and http:// www.knmi.nl/scatterometer/.Sea Surface Temperature For a couple of decades, SST observations, to varying degrees of accuracy, have been provided by infrared radiometry (IR).In contrast to IR's relatively fine spatial resolution, which is blocked by the presence of clouds, microwave radiometry (MR) offers all-weather, but relatively coarse, resolution.The interested reader is directed to Donlon et al. (2009), who discuss how the Global High-Resolution SST Project combines the best attributes of IR and MR to develop improved SST products.Sea Ice Cover Continuing observations of sea-ice cover have been collected using MR techniques since 1978; these results have shown in easy-to-understand terms how the Arctic permanent ice cover has visibly declined over the 30-year record of satellite observations (e.g., Figure 9).Moreover, MR has been complemented more recently by scatterometry and synthetic aperture radar to provide information on ice concentration, ice age, and ice temperature.See http://nsidc.org/.

Figure 7 .
Figure 7.All satellite altimeters show global mean sea level to be rising.The current estimate of ~ 2.8 mm yr -1 is somewhat lower than the ~ 3.1 mm yr -1 cited by the intergovernmental panel on climate change, most likely due to recent cooling associated with a protracted la Niña. in this figure, the high-precision tOpEX/poseidon and Jason altimeters provided the reference baseline, and results from each of the additional altimeters were adjusted for bias to minimize differences with the baseline.Courtesy of the NOAA Laboratory for Satellite Altimetry and Altimetrics LLC ocean observations are funded by research agencies, and this mode of support is likely to continue for the foreseeable future.At some point, research agencies may look to the operational agencies to assume some responsibility for sustaining at least partial support for routine, systematic observations over the long term.GODAE recognized this need when it was organized a decade ago.One of its basic motivations was to demonstrate in an operational setting the impact of having timely access to data from global ocean observing systems funded by research agencies and, depending on that impact, to develop a rationale to justify the transition of funding for those systems from the research to the operational agencies.Ideally, once the utility of observations had been demonstrated, the operational agencies would incorporate support for at least some of those observing systems into their ongoing programs, thereby providing one avenue to sustain their support.There are many challenges to be addressed in maintaining what has been achieved over the past decade.All programs face nontrivial increases in the cost of hardware and salaries.The VOS program is feeling the impact of cutbacks in national weather services support of the program, particularly reduction in the number of Port Meteorological Officers, and changes in the patterns, staffing, and security concerns of the global merchant shipping fleet.The XBT program also strains to achieve its coverage because of changes in the

Figure 8 .
Figure 8.An extra-tropical cyclone centered just east of Nova Scotia with peak winds of 50-60 knots, as displayed on a forecaster workstation at the NOAA Ocean prediction center.(A) 12:15 GMt, May 13, 2009, Geostationary Operational Environmental Satellite (GOES) visible image showing cloud patterns and surface observations collected by ships (with wind reports) and buoys (without).(b) 13:51 GMt AScAt/MetOp surface vector wind field.(c)09:22 GMt QuikScAt surface vector wind field, with the edge of a 07:42 GMt pass to the east.These AScAt (Advanced Scatterometer) and QuikScAt (Quick Scatterometer) products (developed in Europe and the uS, respectively) provide 12.5-to 25-km-resolution observations of the surface vector wind field, in marked contrast to the relative sparsity of ship and buoy reports, and enable the accurate location of storm centers and associated fronts (such as the one extending to the southwest from the storm center).realizing such improvements for operational forecasting is a prime motivation behind the committee on Earth Observation Satellites (cEOS), encouraging every nation to provide timely access to data from its satellites for the benefit of all.Courtesy of the NOAA Ocean Prediction Center

Fiscal
Operational and research agencies operate, and will continue to operate, in a tight budget environment.For example, in the United States, NOAA is attempting to establish elements of a new (for NOAA) operational ocean capability in a level-funding environment on top of a well-established and growing operational weather forecasting program.

Figure 9 .
Figure 9.The Arctic ice cover has been in decline over the three decades of satellite observation.perennial ice (blue line), defined as the area of minimum ice cover, survives melting and occurs in late summer.Multiyear ice (green), the area of ice at least three years old and generally the thickest, is observed in February; it typically takes several summers for brine to drain from sea ice, leaving it almost salt free and with a distinctive microwave signature.because Multiyear ice is declining faster than perennial ice, the thickness of the Arctic ice cover is, on average, declining as well.This time series demonstrates the value of being able to integrate observations from multiple sources into a single climate record. it is based on data collected by three uS satellite-borne instruments: the Scanning Multi-frequency Microwave radiometer (SMMr) on Nimbus-7, the Special Sensor Microwave imager (SSMi) on the Defense Meteorological Satellite program satellite (DMSp), and the Advanced Microwave Scanning radiometer (AMSr-E) on the Aqua satellite (grey line since 2003).Courtesy of Joey Comiso, NASA Goddard Space Flight Center Focus and PrioritizeOperational agencies typically have little budgetary flexibility, and therefore need to focus and prioritize when attempting to implement operational infrastructure.They need to concentrate on those variables for which there have been successful demonstrations of technical feasibility and scientific utility.

maria hood (Intergovernmental Oceanographic Commission/UNESCO, France), michael mcPhaden (NOAA/Pacific Marine Environmental Laboratory, USA), david meldrum (Scottish Association for Marine Sciences, UK), mark merrifield (University of Hawaii, USA), dean roemmich (Scripps Institution of Oceanography, USA), Chris sabine (NOAA/Pacific Marine Environmental Laboratory,
(European Organisation for the Exploitation of Meteorological Satellites, Germany).collection) and real-time (available in 15 minutes to three hours) GLOSS data are assembled and provided by the University of Hawaii Sea Level Center.The British Oceanographic Data Centre provides final delayed-mode data (see http://www.gloss-sealevel.org/).The status of real-time reporting stations and recently collected time series are available at the Sea Level Station Monitoring Facility maintained by the Flanders Marine Institute (VLIZ).See http://www.vliz.be/gauges/.