- Email: [email protected]
The Global Ocean Observing System: design and implementation of the coastal module
The Global Ocean Observing System" design and implementation of the coastal module Thomas C. Malone
Horn Point Laboratory, University of Maryland Center for Environmental Science, USA
Abstract The combined effects of climate change, extreme weather and human alterations of the environment are especially pronounced in the coastal zone where people and ecosystem goods and services are most concentrated and inputs of energy and matter from land, sea and air converge. These realities call for a more integrated and adaptive approach to resource management, environmental protection, coastal zone management, coastal engineering, an approach that considers the effects of both human activities and natural variability change in an ecosystem context. Implementing such an approach requires the capability to routinely and rapidly detect and predict changes in the state of the coastal environment. Developing these capabilities is the purpose of the coastal module of GOOS, which is the subject of this presentation.
Keywords: Coastal, monitoring, GOOS, GOOS Regional Alliances 1. Introduction The combined effects of natural hazards, human activities, and climate change are and will continue to be most pronounced in the coastal zone where people and ecosystem goods and services are most concentrated and inputs of energy and matter from land, sea and air converge. Large scale drivers of change (Table 1) that impact social, economic, and ecological systems of the coastal zone include: 1. basin scale changes in the oceans (e.g. the E1 Nifio-Southern Oscillation, the Pacific Decadal Oscillation, and the North Atlantic Oscillation) 2. human alterations of the environment (changing inputs of water, sediments, nutrients, and contaminants from land-based sources; inputs of human pathogens; introductions of non-native species; and extraction of marine resources) 3. extreme weather (e.g., tropical storms) 4. global climate change (global warming, sea level rise, and changes in the hydrological cycle). Associated with these drivers of change there is a variety of phenomena of interest (Table 1) in the coastal marine ecosystem that a coastal observing system needs to consider. Although rapid detection and timely prediction of each phenomenon have unique requirements for data and information, they have many data needs in common. Likewise, the requirements for data communications and management are similar. These realities are the basis for developing the coastal module of GOOS. Consequent changes in coastal marine and estuarine ecosystems affect public health and well being, the safety
* Email: [email protected]
Thomas C. Malone
443
and efficiency of marine operations, ecosystem health, and the sustainability of living marine resources. Table 1 Drivers of change (natural and anthropogenic forcings) and associated phenomena of
interest in coastal marine ecosystems that are the subject of the coastal module of GOOS. Although rapid detection and timely prediction of each phenomenon have unique requirements for data and information, they have many data needs in common. Likewise, the requirements for data communications and management are similar. These realities are the basis for developing the coastal module of GOOS FORCINGS
Natural
9 9 9 9 9 9
Global warming & sea level rise Storms & other extreme weather events Seismic events Ocean scale currents Waves, tides & storm surges River & ground water discharges
Anthropogenic
9 9 9 9 9 9 9 9
Physical restructuring of the environment Alteration of the hydrological cycle Harvesting living & nonliving resources Alteration of nutrient cycles Sediment inputs Chemical contamination Inputs of human pathogens Introductions of non-native species
PHENOMENA OF INTEREST Marine Services, Natural 9 Fluctuations in sea level Hazards & Public Safety 9 Changes in sea state 9 Changes in surface & sub-surface currents 9 Coastal flooding events 9 Changes in shoreline & shallow water bathymetry Public Health
9 Seafood contamination 9 Increasing abundance of pathogens (in water, shellfish)
Ecosystem Health
9 Habitat modification & loss 9 9 9 9 9 9 9 9
Living R e s o u r c e s
Changes in biodiversity Eutrophication Changes in water clarity Harmful algal events Invasive species Biological affects of chemical contaminants Disease & mass mortalities of marine organisms Chemical contamination of the environment
9 Abundance of exploitable living marine resources 9 Harvest of capture fisheries 9 Aquaculture harvest
These realities call for a more integrated, ecosystem-based approach to resource management, environmental protection, coastal zone management and coastal engineering, i.e. the implementation of an integrated coastal management strategy that considers the effects of both environmental variability and of human activities. Implementing such an ecosystem-based strategy depends on the capability to engage in
444
The Global Ocean Observing System: design and implementation of the coastal module
adaptive management, a process that requires routine and rapid detection of changes in the condition of coastal ecosystems, their causes, and their consequences. As indicated by recent attempts to quantify the condition of the world's ecosystems, (e.g. www.wri.org/wr2000/coast page.html, www.heinzctr.org/ecosystems), we do not have this capability today. The current mode of fisheries management exemplifies these problems, i.e., fisheries scientists advise managers based on annual stock assessments and fisheries managers weigh this advice against their socio-political implications to set quotas for the upcoming fishing season. For the most part, science loses out in this process, in part because of two chronic problems that inhibit the development of ecosystem-based fisheries management: 1. the lack of long, high resolution time series and spatially synoptic observations lead to large uncertainties in stock assessments that undermine the credibility of scientific advice, especially when it calls for reductions in catch limits 2. the time taken to make measurements, access and analyse the required data is much too long. The rates of data acquisition and analysis are not well tuned to the time scales on which decisions should be made for the purposes of adaptive management. This puts managers in a difficult position where politics usually wins the day. These problems are magnified when considering the relations between the marine sciences, environmental protection, and fisheries management. In this arena, linkages between science and management are even more uncertain and are rarely institutionalised.
2. The solution A new approach is needed that enables adaptive management through routine, sustained and rapid provision of data and information over the broad spectrum of timespace scales required to link ecosystem scale changes to regional and global scale drivers of change (hours-decades). We are, in fact, on the cusp of a revolution that is making such an approach feasible. The revolution is occurring on two related fronts: 1. advances in environmental sensing and modelling capabilities 2. the emergence of operational oceanography in the form of the Global Ocean Observing System (GOOS). I focus on the latter here. Under the oversight of the GOOS sponsors (IOC, UNEP, WMO, ICSU, FAO), the observing system is being organised in two related and convergent modules: 1. the global ocean module being developed by the Ocean Observations Panel for Climate 2. the coastal module being developed by the Coastal Ocean Observations Panel. The former is primarily concerned with changes in and the effects of the ocean-climate system on physical processes & the global carbon budget. The latter is primarily concerned with the effects of climate and human activities on coastal ecosystems and socio-economic systems of coastal nations.
Thomas C. Malone
445
The purpose GOOS is to continuously provide data and data-products in forms and at rates required to achieve six goals: 9
improve the safety and efficiency of marine operations
9
mitigate the effects of natural hazards on coastal communities and ecosystems more effectively
9
improve predictions of climate changes and their effects on coastal communities and ecosystems
9
minimise public health risks
9
more effectively protect and restore healthy coastal marine ecosystems
9
sustain living marine resources.
Achieving these goals depends on more rapid detection and timely prediction of changes in the marine environment than are currently realised. Clearly, this is a formidable task. However, although each goal has unique requirements for data and information, they have many data needs in common. Likewise, the requirements for data communications management are similar across all six goals. Thus, an integrated approach to the development of a multi-use observing system is both sensible and cost-effective. Monitoring and modelling are mutually dependent, and linking the two processes for more rapid detection and timely prediction requires a managed, two-way flow of data and information among three essential subsystems: 1. an observing (measurement) subsystem for monitoring required variables on specified time and space scales 2. a data communications and management subsystem for serving and archiving data of known quality in real-time or delayed mode as needed 3. a modelling subsystem for data assimilation and analysis. With the exception of marine operations and to some extent natural hazards, such an "end-to-end" system that is routine and sustained is a new concept for both the environmental science and management communities. The World Weather Watch (WWW) provides a model of the kind of operational, end-toend system GOOS is envisioned to be. The first national weather service with a permanent observing network was established in France during the mid-1800s. Its primary purpose was to forecast the weather for farmers. By the early 1900s, a real-time global observing system was in place that consisted of a sparse network of unevenly spaced land-based monitoring sites; by the late 1960s, satellite-borne radiometers were providing global coverage; and by the mid-1990s, an upper air observing network was in place. Today, the global atmospheric observing system of the World Weather Watch (WWW) serves many user groups routinely, including meteorologists and other scientists. In this model, there is a synergy between meteorological research and weather forecasting in which the WWW observing system supplies and manages the data required for numerical weather predictions (NWP) and meteorologists both benefit from the data streams required for weather forecasting and enable improved forecasting skill. This arrangement not only sustains the integrity of meteorological research, it strengthens it.
446
The Global Ocean Observing System: design and implementation of the coastal module
GOOS is intended to perform a function similar to that of the WWW. It is envisioned as a network of national, regional, and global systems that rapidly and systematically acquire and disseminate marine data and data products to serve the needs of many user groups including government agencies, private enterprise, scientists, educators, NGOs and the public that are responsible for or use marine goods and services. Just as weather forecasts are not possible without sustained observations and operational models, implementation of an ecosystem-based approach will not be possible without operational models of marine ecosystems and the uninterrupted provision of oceanographic data required to initialise and up-date model runs. The development of GOOS is not only required to implement adaptive management practices, it will enable advances in marine science and help to maintain the integrity of the scientific method in the face of the growing demand to be more "relevant." The W W W is a good model for an operational system that routinely provides weather forecasts and promotes advances in the science of meteorology. However, it is not an integrated system in that it is not multidisciplinary. The WWW has a singular purpose that depends on a relatively small set of data streams for variables that are relatively easy to measure. Today, programmes that are well integrated in terms of synoptic measurements of physical, chemical, and biological variables are, for the most part, limited to research projects that are finite in duration and are not routine by their very nature. The target of operational oceanography is the development of an observing system for marine ecosystems that is routine, sustained and integrated. To achieve these goals, the movement to establish GOOS is an attempt to network and enhance existing programmes for: 1. more cost-effective use of existing knowledge and infrastructure 2. more rapid detection and timely prediction of changes 3. more rapid access to diverse data from many sources 4. more effective use of environmental data and information. It is an effort that, if successful, will not only significantly increase the value of environmental research, it will enable more integrated approaches to achieving the related goals of environmental protection, resource management, coastal zone management, coastal engineering and marine research.
3. Conceptual design of the coastal module The design plan for the coastal module must consider many factors. These include the need to address a broad diversity of phenomena encompassed by the 6 goals; the fact that the phenomena of interest are globally ubiquitous and tend to be local expressions of regional and global scale processes; and ecosystem theory which posits that the phenomena of interest are related through a hierarchy of interactions that can be modelled. The design must also take into consideration certain important realities: priorities vary among nations and regions; many of the elements required to build the observing system are already in place; those elements of the observing system required to improve marine services and forecast natural hazards are most developed while those required for ecosystem-based environmental protections and management of living
Thomas C. Malone
447
resources are least developed; and most nations do not have the capacity to contribute to and benefit from GOOS at this time. These considerations have important consequences for the design of the coastal module: 9
the design must respect the fact that priorities vary among regions and should leave system design on the regional scale to stakeholders in the regions
9
regional bodies provide the most effective venue for specifying user group requirements for environmental data and information
9
economies of scale can be achieved by establishing a global system that measures variables and manages data streams required by most regions
9
the global coastal module will come into being through a combination of national, regional and global processes
9
the system can be implemented by selectively linking existing elements and can be developed by enhancing and complementing these elements over time
9
high priority must be placed on capacity building in developing countries; on the establishment of the data communications and management infrastructure; and on marine research to develop the sensors and models required to achieve those goals that require biological and chemical data.
Clearly, the coastal module must include both global and regional scale components. This can best be achieved through the establishment of a Global Federation of Regional Systems in which regional observing systems are nested in a Global Coastal Network. The latter establishes a network of reference and sentinel stations; develops international standards and protocols for measurements, data exchange and management; and measures and processes a small set of common variables that are required by most, if not all, regional systems. This provides economies of scale and improves the cost-effectiveness of regional systems by minimising redundancy and optimising data and information exchange. GOOS Regional Alliances (GRAs), guided by national and regional priorities, develop regional system for the provision of data and data-products that are tailored to the requirements of user groups in the region. This will be achieved through national and regional enhancements, i.e., more variables are measured with greater timespace resolution as dictated by national and regional priorities. In this way, GRAs both contribute to and benefit from the global network. It must be emphasised that the global network will not, by itself, provide all of the data and information required to detect or predict changes in the phenomena of interest. There are categories of variables that are important globally, but the variables measured and the time-space scales of measurement change from region to region. These include variables in the categories of fish stock assessments; distribution and condition of essential fish habitats such as sea grasses, kelp beds, tidal wetlands and coral reefs; distribution and abundance of large organisms such as turtles, marine mammals, and seabirds; invasive species; harmful algae; and chemical contaminants. For these categories, decisions concerning exactly what variables to measure, the time-space scales of measurement, and the mix of observing techniques are best made by stakeholders in the regions affected. Thus, regional observing systems are critical building blocks of the coastal module of GOOS, especially for achieving the goals of sustaining and restoring healthy marine ecosystems living marine resources.
448
The Global Ocean Observing System: design and implementation of the coastal module
4. Selecting the common variables Given these considerations, an objective process is needed to select the common variables to be measured as part of the global coastal system. In brief, the global network must measure and manage a set of common variables that are required by most regions to achieve the six goals. The objective is to select the minimum number of variables required to detect or predict changes that are important to a maximum number of users. The process begins with the compilation of a comprehensive list of variables. The variables for detection are then ranked based on the number of phenomena they are relevant to and the number of user groups that are likely to benefit. A similar analysis is done for variables required for prediction based on model requirements. The analysis yielded a provisional list of variables that are recommended for incorporation into the global coastal network: 9
physical variables~temperature, salinity, sea level, vector currents, surface waves spectra, shallow water bathymetry, and shoreline position
9
chemical variables~sediment grain size and organic content, dissolved inorganic nutrients, and dissolved oxygen
9
biological variables~chlorophyll-a, attenuation of downwelling radiation, benthic biomass, and faecal bacteria.
5. Linking observations to models The data management and communications subsystem is the "life-blood" of the observing system and the primary means by which an integrated system will emerge. Thus, the development of an integrated data management and communications subsystem is arguably the highest priority for implementation. Under current conditions, data are often not exchanged freely among nations and, even when data are not proprietary, data management and analysis tend to be programme-specific and analyses that require multi-disciplinary data from many sources take far too much time. The goal is to establish an integrated data management subsystem that serves data in both real-time and delayed mode and allows users to exploit multiple data sets from many different sources through "one-stop-shopping". The integrated plan is based on a hierarchical, distributed network of local-, regional- and global-scale data management activities that build on, link and enhance existing data management centres and programmes.
6. Building the coastal module The development of both the global ocean and coastal components of GOOS are critically dependent on selectively and effectively linking, enhancing and supplementing existing programmes. Although some elements of the system will be global in scale from the beginning (e.g., GLOSS, observations from space), national and regional coastal observing systems will be the building blocks of the global coastal network. GOOS Regional Alliances (GRAs) are planning and implementing regional observing systems that will become the building blocks of the coastal module of GOOS. GRAs are expected to be formed through coalitions among national and regional GOOS programmes, Regional Seas Conventions, Regional Fishery Bodies, Large Marine Ecosystem Programmes, and other bodies and programmes as appropriate.
Thomas C. Malone
449
In this context, implementing the coastal module will require an unprecedented level of cooperation, coordination and collaboration among nations and existing programmes to ensure the emergence of a global network as national and regional systems come on-line. A critical aspect of this process will involve harmonising the need for global coordination with user needs based on national and regional priorities. At present, there is no formal international mechanism in place to promote and guide this process. An intergovernmental commission, such as the Joint Technical Commission on Oceanography and Marine Meteorology (JCOMM with the appropriate advisory bodies), will be needed to facilitate multi-lateral agreements and to address legal issues that will arise from implementation of UNCLOS and other international conventions. The "Integrated Design Plan for the Coastal Module of GOOS" has recently been completed by the COOP and can be found at http://ioc.unesco.org/goos/gsc6/COOPDesign-Plan.doc.
Acknowledgements This work represents the work of the COOP and invited experts including Dagoberto Arcos, Bodo von Bodungen, Alfonso Botello, Robert Bowen, Lauro Julio Calliari, John Cullen, Michael Depledge, Eric Dewailly, Michael J. Fogarty, Juliusz Gajewski, Johannes Guddal, Julie Anne Hall, Hiroshi Kawamura, Anthony Knap (Co-Chair), Coleen Moloney, Nadia Pinardi, Hillel Shuval, Vladimir Smirnov, Keith R. Thompson, MVM Wafar, Rudolph Wu, Robert R. Christian, Chris Crossland, Savithri Narayanan, Worth Nowlin, Shubha Sathyendranath, and Neville Smith.
Horn Point Laboratory, University of Maryland Center for Environmental Science, USA
Abstract The combined effects of climate change, extreme weather and human alterations of the environment are especially pronounced in the coastal zone where people and ecosystem goods and services are most concentrated and inputs of energy and matter from land, sea and air converge. These realities call for a more integrated and adaptive approach to resource management, environmental protection, coastal zone management, coastal engineering, an approach that considers the effects of both human activities and natural variability change in an ecosystem context. Implementing such an approach requires the capability to routinely and rapidly detect and predict changes in the state of the coastal environment. Developing these capabilities is the purpose of the coastal module of GOOS, which is the subject of this presentation.
Keywords: Coastal, monitoring, GOOS, GOOS Regional Alliances 1. Introduction The combined effects of natural hazards, human activities, and climate change are and will continue to be most pronounced in the coastal zone where people and ecosystem goods and services are most concentrated and inputs of energy and matter from land, sea and air converge. Large scale drivers of change (Table 1) that impact social, economic, and ecological systems of the coastal zone include: 1. basin scale changes in the oceans (e.g. the E1 Nifio-Southern Oscillation, the Pacific Decadal Oscillation, and the North Atlantic Oscillation) 2. human alterations of the environment (changing inputs of water, sediments, nutrients, and contaminants from land-based sources; inputs of human pathogens; introductions of non-native species; and extraction of marine resources) 3. extreme weather (e.g., tropical storms) 4. global climate change (global warming, sea level rise, and changes in the hydrological cycle). Associated with these drivers of change there is a variety of phenomena of interest (Table 1) in the coastal marine ecosystem that a coastal observing system needs to consider. Although rapid detection and timely prediction of each phenomenon have unique requirements for data and information, they have many data needs in common. Likewise, the requirements for data communications and management are similar. These realities are the basis for developing the coastal module of GOOS. Consequent changes in coastal marine and estuarine ecosystems affect public health and well being, the safety
* Email: [email protected]
Thomas C. Malone
443
and efficiency of marine operations, ecosystem health, and the sustainability of living marine resources. Table 1 Drivers of change (natural and anthropogenic forcings) and associated phenomena of
interest in coastal marine ecosystems that are the subject of the coastal module of GOOS. Although rapid detection and timely prediction of each phenomenon have unique requirements for data and information, they have many data needs in common. Likewise, the requirements for data communications and management are similar. These realities are the basis for developing the coastal module of GOOS FORCINGS
Natural
9 9 9 9 9 9
Global warming & sea level rise Storms & other extreme weather events Seismic events Ocean scale currents Waves, tides & storm surges River & ground water discharges
Anthropogenic
9 9 9 9 9 9 9 9
Physical restructuring of the environment Alteration of the hydrological cycle Harvesting living & nonliving resources Alteration of nutrient cycles Sediment inputs Chemical contamination Inputs of human pathogens Introductions of non-native species
PHENOMENA OF INTEREST Marine Services, Natural 9 Fluctuations in sea level Hazards & Public Safety 9 Changes in sea state 9 Changes in surface & sub-surface currents 9 Coastal flooding events 9 Changes in shoreline & shallow water bathymetry Public Health
9 Seafood contamination 9 Increasing abundance of pathogens (in water, shellfish)
Ecosystem Health
9 Habitat modification & loss 9 9 9 9 9 9 9 9
Living R e s o u r c e s
Changes in biodiversity Eutrophication Changes in water clarity Harmful algal events Invasive species Biological affects of chemical contaminants Disease & mass mortalities of marine organisms Chemical contamination of the environment
9 Abundance of exploitable living marine resources 9 Harvest of capture fisheries 9 Aquaculture harvest
These realities call for a more integrated, ecosystem-based approach to resource management, environmental protection, coastal zone management and coastal engineering, i.e. the implementation of an integrated coastal management strategy that considers the effects of both environmental variability and of human activities. Implementing such an ecosystem-based strategy depends on the capability to engage in
444
The Global Ocean Observing System: design and implementation of the coastal module
adaptive management, a process that requires routine and rapid detection of changes in the condition of coastal ecosystems, their causes, and their consequences. As indicated by recent attempts to quantify the condition of the world's ecosystems, (e.g. www.wri.org/wr2000/coast page.html, www.heinzctr.org/ecosystems), we do not have this capability today. The current mode of fisheries management exemplifies these problems, i.e., fisheries scientists advise managers based on annual stock assessments and fisheries managers weigh this advice against their socio-political implications to set quotas for the upcoming fishing season. For the most part, science loses out in this process, in part because of two chronic problems that inhibit the development of ecosystem-based fisheries management: 1. the lack of long, high resolution time series and spatially synoptic observations lead to large uncertainties in stock assessments that undermine the credibility of scientific advice, especially when it calls for reductions in catch limits 2. the time taken to make measurements, access and analyse the required data is much too long. The rates of data acquisition and analysis are not well tuned to the time scales on which decisions should be made for the purposes of adaptive management. This puts managers in a difficult position where politics usually wins the day. These problems are magnified when considering the relations between the marine sciences, environmental protection, and fisheries management. In this arena, linkages between science and management are even more uncertain and are rarely institutionalised.
2. The solution A new approach is needed that enables adaptive management through routine, sustained and rapid provision of data and information over the broad spectrum of timespace scales required to link ecosystem scale changes to regional and global scale drivers of change (hours-decades). We are, in fact, on the cusp of a revolution that is making such an approach feasible. The revolution is occurring on two related fronts: 1. advances in environmental sensing and modelling capabilities 2. the emergence of operational oceanography in the form of the Global Ocean Observing System (GOOS). I focus on the latter here. Under the oversight of the GOOS sponsors (IOC, UNEP, WMO, ICSU, FAO), the observing system is being organised in two related and convergent modules: 1. the global ocean module being developed by the Ocean Observations Panel for Climate 2. the coastal module being developed by the Coastal Ocean Observations Panel. The former is primarily concerned with changes in and the effects of the ocean-climate system on physical processes & the global carbon budget. The latter is primarily concerned with the effects of climate and human activities on coastal ecosystems and socio-economic systems of coastal nations.
Thomas C. Malone
445
The purpose GOOS is to continuously provide data and data-products in forms and at rates required to achieve six goals: 9
improve the safety and efficiency of marine operations
9
mitigate the effects of natural hazards on coastal communities and ecosystems more effectively
9
improve predictions of climate changes and their effects on coastal communities and ecosystems
9
minimise public health risks
9
more effectively protect and restore healthy coastal marine ecosystems
9
sustain living marine resources.
Achieving these goals depends on more rapid detection and timely prediction of changes in the marine environment than are currently realised. Clearly, this is a formidable task. However, although each goal has unique requirements for data and information, they have many data needs in common. Likewise, the requirements for data communications management are similar across all six goals. Thus, an integrated approach to the development of a multi-use observing system is both sensible and cost-effective. Monitoring and modelling are mutually dependent, and linking the two processes for more rapid detection and timely prediction requires a managed, two-way flow of data and information among three essential subsystems: 1. an observing (measurement) subsystem for monitoring required variables on specified time and space scales 2. a data communications and management subsystem for serving and archiving data of known quality in real-time or delayed mode as needed 3. a modelling subsystem for data assimilation and analysis. With the exception of marine operations and to some extent natural hazards, such an "end-to-end" system that is routine and sustained is a new concept for both the environmental science and management communities. The World Weather Watch (WWW) provides a model of the kind of operational, end-toend system GOOS is envisioned to be. The first national weather service with a permanent observing network was established in France during the mid-1800s. Its primary purpose was to forecast the weather for farmers. By the early 1900s, a real-time global observing system was in place that consisted of a sparse network of unevenly spaced land-based monitoring sites; by the late 1960s, satellite-borne radiometers were providing global coverage; and by the mid-1990s, an upper air observing network was in place. Today, the global atmospheric observing system of the World Weather Watch (WWW) serves many user groups routinely, including meteorologists and other scientists. In this model, there is a synergy between meteorological research and weather forecasting in which the WWW observing system supplies and manages the data required for numerical weather predictions (NWP) and meteorologists both benefit from the data streams required for weather forecasting and enable improved forecasting skill. This arrangement not only sustains the integrity of meteorological research, it strengthens it.
446
The Global Ocean Observing System: design and implementation of the coastal module
GOOS is intended to perform a function similar to that of the WWW. It is envisioned as a network of national, regional, and global systems that rapidly and systematically acquire and disseminate marine data and data products to serve the needs of many user groups including government agencies, private enterprise, scientists, educators, NGOs and the public that are responsible for or use marine goods and services. Just as weather forecasts are not possible without sustained observations and operational models, implementation of an ecosystem-based approach will not be possible without operational models of marine ecosystems and the uninterrupted provision of oceanographic data required to initialise and up-date model runs. The development of GOOS is not only required to implement adaptive management practices, it will enable advances in marine science and help to maintain the integrity of the scientific method in the face of the growing demand to be more "relevant." The W W W is a good model for an operational system that routinely provides weather forecasts and promotes advances in the science of meteorology. However, it is not an integrated system in that it is not multidisciplinary. The WWW has a singular purpose that depends on a relatively small set of data streams for variables that are relatively easy to measure. Today, programmes that are well integrated in terms of synoptic measurements of physical, chemical, and biological variables are, for the most part, limited to research projects that are finite in duration and are not routine by their very nature. The target of operational oceanography is the development of an observing system for marine ecosystems that is routine, sustained and integrated. To achieve these goals, the movement to establish GOOS is an attempt to network and enhance existing programmes for: 1. more cost-effective use of existing knowledge and infrastructure 2. more rapid detection and timely prediction of changes 3. more rapid access to diverse data from many sources 4. more effective use of environmental data and information. It is an effort that, if successful, will not only significantly increase the value of environmental research, it will enable more integrated approaches to achieving the related goals of environmental protection, resource management, coastal zone management, coastal engineering and marine research.
3. Conceptual design of the coastal module The design plan for the coastal module must consider many factors. These include the need to address a broad diversity of phenomena encompassed by the 6 goals; the fact that the phenomena of interest are globally ubiquitous and tend to be local expressions of regional and global scale processes; and ecosystem theory which posits that the phenomena of interest are related through a hierarchy of interactions that can be modelled. The design must also take into consideration certain important realities: priorities vary among nations and regions; many of the elements required to build the observing system are already in place; those elements of the observing system required to improve marine services and forecast natural hazards are most developed while those required for ecosystem-based environmental protections and management of living
Thomas C. Malone
447
resources are least developed; and most nations do not have the capacity to contribute to and benefit from GOOS at this time. These considerations have important consequences for the design of the coastal module: 9
the design must respect the fact that priorities vary among regions and should leave system design on the regional scale to stakeholders in the regions
9
regional bodies provide the most effective venue for specifying user group requirements for environmental data and information
9
economies of scale can be achieved by establishing a global system that measures variables and manages data streams required by most regions
9
the global coastal module will come into being through a combination of national, regional and global processes
9
the system can be implemented by selectively linking existing elements and can be developed by enhancing and complementing these elements over time
9
high priority must be placed on capacity building in developing countries; on the establishment of the data communications and management infrastructure; and on marine research to develop the sensors and models required to achieve those goals that require biological and chemical data.
Clearly, the coastal module must include both global and regional scale components. This can best be achieved through the establishment of a Global Federation of Regional Systems in which regional observing systems are nested in a Global Coastal Network. The latter establishes a network of reference and sentinel stations; develops international standards and protocols for measurements, data exchange and management; and measures and processes a small set of common variables that are required by most, if not all, regional systems. This provides economies of scale and improves the cost-effectiveness of regional systems by minimising redundancy and optimising data and information exchange. GOOS Regional Alliances (GRAs), guided by national and regional priorities, develop regional system for the provision of data and data-products that are tailored to the requirements of user groups in the region. This will be achieved through national and regional enhancements, i.e., more variables are measured with greater timespace resolution as dictated by national and regional priorities. In this way, GRAs both contribute to and benefit from the global network. It must be emphasised that the global network will not, by itself, provide all of the data and information required to detect or predict changes in the phenomena of interest. There are categories of variables that are important globally, but the variables measured and the time-space scales of measurement change from region to region. These include variables in the categories of fish stock assessments; distribution and condition of essential fish habitats such as sea grasses, kelp beds, tidal wetlands and coral reefs; distribution and abundance of large organisms such as turtles, marine mammals, and seabirds; invasive species; harmful algae; and chemical contaminants. For these categories, decisions concerning exactly what variables to measure, the time-space scales of measurement, and the mix of observing techniques are best made by stakeholders in the regions affected. Thus, regional observing systems are critical building blocks of the coastal module of GOOS, especially for achieving the goals of sustaining and restoring healthy marine ecosystems living marine resources.
448
The Global Ocean Observing System: design and implementation of the coastal module
4. Selecting the common variables Given these considerations, an objective process is needed to select the common variables to be measured as part of the global coastal system. In brief, the global network must measure and manage a set of common variables that are required by most regions to achieve the six goals. The objective is to select the minimum number of variables required to detect or predict changes that are important to a maximum number of users. The process begins with the compilation of a comprehensive list of variables. The variables for detection are then ranked based on the number of phenomena they are relevant to and the number of user groups that are likely to benefit. A similar analysis is done for variables required for prediction based on model requirements. The analysis yielded a provisional list of variables that are recommended for incorporation into the global coastal network: 9
physical variables~temperature, salinity, sea level, vector currents, surface waves spectra, shallow water bathymetry, and shoreline position
9
chemical variables~sediment grain size and organic content, dissolved inorganic nutrients, and dissolved oxygen
9
biological variables~chlorophyll-a, attenuation of downwelling radiation, benthic biomass, and faecal bacteria.
5. Linking observations to models The data management and communications subsystem is the "life-blood" of the observing system and the primary means by which an integrated system will emerge. Thus, the development of an integrated data management and communications subsystem is arguably the highest priority for implementation. Under current conditions, data are often not exchanged freely among nations and, even when data are not proprietary, data management and analysis tend to be programme-specific and analyses that require multi-disciplinary data from many sources take far too much time. The goal is to establish an integrated data management subsystem that serves data in both real-time and delayed mode and allows users to exploit multiple data sets from many different sources through "one-stop-shopping". The integrated plan is based on a hierarchical, distributed network of local-, regional- and global-scale data management activities that build on, link and enhance existing data management centres and programmes.
6. Building the coastal module The development of both the global ocean and coastal components of GOOS are critically dependent on selectively and effectively linking, enhancing and supplementing existing programmes. Although some elements of the system will be global in scale from the beginning (e.g., GLOSS, observations from space), national and regional coastal observing systems will be the building blocks of the global coastal network. GOOS Regional Alliances (GRAs) are planning and implementing regional observing systems that will become the building blocks of the coastal module of GOOS. GRAs are expected to be formed through coalitions among national and regional GOOS programmes, Regional Seas Conventions, Regional Fishery Bodies, Large Marine Ecosystem Programmes, and other bodies and programmes as appropriate.
Thomas C. Malone
449
In this context, implementing the coastal module will require an unprecedented level of cooperation, coordination and collaboration among nations and existing programmes to ensure the emergence of a global network as national and regional systems come on-line. A critical aspect of this process will involve harmonising the need for global coordination with user needs based on national and regional priorities. At present, there is no formal international mechanism in place to promote and guide this process. An intergovernmental commission, such as the Joint Technical Commission on Oceanography and Marine Meteorology (JCOMM with the appropriate advisory bodies), will be needed to facilitate multi-lateral agreements and to address legal issues that will arise from implementation of UNCLOS and other international conventions. The "Integrated Design Plan for the Coastal Module of GOOS" has recently been completed by the COOP and can be found at http://ioc.unesco.org/goos/gsc6/COOPDesign-Plan.doc.
Acknowledgements This work represents the work of the COOP and invited experts including Dagoberto Arcos, Bodo von Bodungen, Alfonso Botello, Robert Bowen, Lauro Julio Calliari, John Cullen, Michael Depledge, Eric Dewailly, Michael J. Fogarty, Juliusz Gajewski, Johannes Guddal, Julie Anne Hall, Hiroshi Kawamura, Anthony Knap (Co-Chair), Coleen Moloney, Nadia Pinardi, Hillel Shuval, Vladimir Smirnov, Keith R. Thompson, MVM Wafar, Rudolph Wu, Robert R. Christian, Chris Crossland, Savithri Narayanan, Worth Nowlin, Shubha Sathyendranath, and Neville Smith.