THE EUROGOOS ANALYSIS OF THE NEED FOR OPERATIONAL OCEAN REMOTE SENSING
N C Flemming, EuroGOOS Office, SOC, Empress Dock, Southampton, Hants, SO14 3ZH. (n.flemming@soc.soton.ac.uk)
From: Operational Ocean Observations from Space. EuroGOOS Publication No.16, May 2001, pages 6-12.
SUMMARY
The justification for expenditure on European operational ocean observations from space depends upon the integration of the resulting data into a wide range of models and data processing channels, generating information products which have an enormous range of applications in time and space. EuroGOOS has been involved in many surveys, studies, and projects related to this chain of processing and use of data. Operational remote sensing of the ocean is a necessary component of long term climate monitoring and forecasting, medium term and seasonal analysis of marine and coastal conditions, and a host of short term operational services and products with economic and environmental benefits. The applications range in nature from public good global benefits, to regional and local environmental management benefits, from support for policy to commercial and profit-based activities on all scales. They include every aspect of marine physics and biology. An identified need, even an urgent need, may not be subject to evaluation in direct financial terms. Whenever possible this should be the first line of analysis, but there are many uncertainties regarding risk, the low probability of colossally serious disasters, and the unknown value of irreversible losses or events, which raise questions in the public good domain, and where the risk of taking no action, or being totally ignorant of the risk, is unacceptable. This paper does not seek to itemise the detailed applications and the related products required, which are presented in other papers in this conference, but gives an overview of the EuroGOOS position. Operational ocean observations from space are essential for the coming decades, and Europe needs the capacity to control its own destiny by having direct control of a portion of the world data from this source.
INTRODUCTION
Operational ocean observations can be justified by many different techniques (Flemming, 2000). Broad socio-economic and environmental objectives have been identified by the GOOS project over the last 6-8 years (OECD 1994; Woods et al.1996; IOC- GOOS 1998; Flemming 1999; Sassone and Weiher 1997; Adams et al., 2000). While accurate cost-benefit analysis (CBA) is applicable to commercial or governmental projects which have an obvious cash-based flow of inputs and outputs, combined with a well-documented flow of financial information and identification of customers, component suppliers, etc., the technique is not useful when the information base is limited, the products are uncertain, the risks are uncertain but are thought to be very large, and there is no way to collect the financial information needed before a decision must be made. Standing back, and saying in effect “Since cost benefit analysis cannot be applied in the time available, we will do nothing” is not an option.
Benefits and motives will first be identified in qualitative terms, distinguishing between manifestly different types of objective which cannot be analysed together. These objectives can then be further quantified or graded by various estimates of public demand, political priorities, national security, public health, treaty obligations, and long term estimates of risks. It may be possible to obtain broad financial implications, or an upper and lower bound, at least in some cases.
A general argument developed in the USA considers the large cost of certainly being totally wrong about climate variability and climate change. Most courses of possible action at present are the topic of strong lobbying by conflicting interest groups in addition to the official recommendations of IPCC. Responding directly to any one of these lobby groups is almost certainly very wrong. Doing nothing is probably wrong. Reacting in a purely responsive mode to successive negative events is almost certainly wrong. Since the stakes are very high, any knowledge which reduces uncertainty has a high value. Peck and Teisberg (1993) calculate a present value of $50bn for resolving specific uncertainties about climate change now rather than in 40 years time.
For over two decades it has been apparent that remote sensing of the ocean is vital if we are to understand the processes at work, and build up adequate time series and geographical cover. The remote sensed data are part of the data stream for models which also includes in situ surface and sub-surface data. For ocean observations as such, observing missions in the last 20 years have been conducted almost exclusively as science research projects, with the exception of military operations, and those observations needed for meteorology, namely sea surface temperature and wind fields.
Since the inception of GOOS in the early 1990’s (OECD 1994, IOC 1998) it has been assumed that a point would be reached when the chance sequence of science missions observing the ocean would have to be superseded by an operational programme of dedicated satellites designed to produce data for a continuously functioning ground segment, feeding data to operational models, and then to users. Reviews of missions planned up to the mid-1990’s are given by the regular CEOS Dossiers of those dates. A summary is provided by IOC (1998, p.53-56). The GOOS Prospectus (IOC 1998, p.57) concludes that the assemblage of science oriented missions will provide the necessary flow of data for SST, wind speed and direction, ocean topography, wave spectra, visible imagery, and SAR coverage of ice up to the year 2005. After that…“GOOS will require access on an operational basis to missions that provide (a list of parameters follows)”.
It is now time to start making the decisions which will ensure operational missions to provide the data after 2005 for models and forecasting systems which already exist, or are in the trials and pilot phase.
In this paper I will summarise the needs and justification for the products of operational oceanography, the role of operational remote sensing in generating those products, and link the analysis to the other papers in this Conference. In the present context the term “operational” is used to mean all missions which are planned using established and proven technology to produce a data stream for which there is a planned and identified continuous routine need. The data may be used immediately in real time, or they may be managed off-line and used in applications such as seasonal or multi-annual monitoring and forecasting. Finally, I will refer to the Conference Statement, and the need for a consistent and steady focus on the development of operational ocean observing missions based in Europe.
ESTIMATION OF VALUE, AND CONTRIBUTION FROM REMOTE SENSING.
A cost-effective Global Ocean Observing System (GOOS) is impossible without remote sensed observations by satellite. EuroGOOS has made this statement in its strategy paper submitted to the European Commission ( EuroGOOS, 2000). In the present paper I am not going to try and assess the proportionate value of the data obtained by satellites and by in situ methods. Both are essential. To justify the whole system therefore we have first to evaluate the benefits ( financial or otherwise) generated from all types of ocean observations, after which the role of different technologies can be considered in more detail, as in the OceanObs Conference (1999).
I do not necessarily believe in the reliability of all the methodologies which have been developed for deriving values from environmental information, but let us consider qualitatively different systems of evaluation:
i) Commercial, Economic and Social Value: Many activities which we pursue at sea and on the coast would be much more efficient, or would do less damage, if we had better information on a daily basis. Accidents could be avoided; catastrophic spills could be reduced or managed; scientists could do better research; biological stocks could be protected better; coastal erosion could be controlled at lower cost; ships’ voyages would be more efficient; oil and gas would be produced more efficiently; public health could be improved, etc. On the same basis, though with a longer timescale, improved knowledge and forecastablity of the ocean climate, and the coupled ocean-atmosphere climate, provides benefits to the management of agriculture, fresh water, and power generation. In this category it should be possible to evaluate fairly accurately the value of the industries and services, and the benefits of having better marine forecasts (e.g. WHOI 1993; Sassone and Weiher 1997; Adams et al., 2000). The benefits may accrue to producers, customers, or the community at large. Many of the applications of this kind are being considered by other papers at the Darmstadt Conference.
ii) “True Environmental Value”: On this assumption (Costanza et al., 1997, 2000) the natural environment, the air, forests, rivers, ocean, soil, etc., represent Capital, and are performing valuable functions such as keeping drinking water clean, providing the right balance of oxygen and CO2 in the atmosphere, etc., which we would have to pay for if nature did not do it for us (The Total Natural Capital). If the value of these services were properly appreciated, and if it were included in everyday financial calculations, we would arrive at quite a different way of treating the environment, and would take much more care of it. We need to understand the sea better in order to protect its true value. This argument can be applied to the marine environment as much as to the land.
iii) Sustainable Development: The Framework Climate Convention, the Biodiversity Convention, and other agreements make it obligatory for states to try and manage and control the exploitation of the marine and coastal environment in such a way that the conservation and productivity of the ocean are preserved. On this assumption, some means has to be found to measure and compare costs, benefits, losses, and damage. Although the methods are not well proven, the objective is to convert all values to financial cash equivalent, and to discount for time to net present value, with allowance for the probability of occurrence of various future events.
iv) “Saving the Planet”: The case assumes that the risk of damage to the environment of Planet Earth in the next few decades from loss of biodiversity, global warming, melting glaciers, rising sea levels, thermohaline circulation shut-down, etc., is so high that a Global Ocean Observing System (and presumably a Global Climate Observing System and a Global Terrestrial Observing System) is essential to protect the future of the human race on this planet, to detect and warn of adverse change, and to prevent or avoid it if possible. Given the high stakes on this assumption, there is little point in arguing about the cost of GOOS. On this basis GOOS is a moral imperative. There are however serious questions about uncertainty, risk probability, magnitude of risk, and time horizons. This argument has a political dimension in that it motivates public attitudes, but it is not susceptible even to the approximate quantification which is possible with types (ii) and (iii) above.
Most of the socio-economic justification for GOOS and GCOS has been concerned with arguments of type (i) above. Arguments of type (ii) and (iii) will be developed steadily by environmental research organisations and regulatory authorities in the next few years, and are likely to be very influential within a decade. The sum total of benefits estimated by EuroGOOS for the European region (Woods et al., 1996, p.5, p.21) is of the order of 2-5 bn Euro per year. This figure is obtained by considering mostly components of benefit in type (i) above, with a small component of type (iii).
Adams et al. (2000) present the most sophisticated analysis so far of type (i) benefits, presented in general terms, for the seasonal to inter-annual benefits arising from better marine observations and their effect upon weather and climate forecasting across the USA, as well as for marine exploitation and conservation.. A combination of forecast techniques incorporating ENSO analysis, the Pacific Decadal Oscillation, and the North Atlantic Oscillation, would provide benefits measured in hundreds of millions of dollars per year for each of several management sectors related to fresh water reservoirs, agriculture, oil and gas storage, and agriculture. Medium term to multi-year forecasting in the European region does not have the advantage of a clearly dominant signal like the ENSO cycle, but analysis of Atlantic and Arctic processes is beginning to yield several potential mechanisms and cycles in addition to the NAO, which may permit valuable forecasts. If this is the case, the improved boundary conditions for shelf seas models, and the climate forecasts for the whole continent. could have user values of the same order of magnitude as those derived for the USA.
The largest potential benefits from long-term integrated ocean observations are those which accrue 5-10 years into the future, or beyond, and the largest figures are also those shrouded in the greatest uncertainty given present knowledge. It is therefore prudent to use the observing system as intensively as possible in the short term to generate truly operational products with economic and social benefits. By this tactic the total cumulative discounted cash flow, including expenditure and computed value of benefits, is prevented from dipping too far into deficit (Flemming, 1999), and will show a total net benefit as quickly as possible.
THE GOOS FRAMEWORK AND REGIONAL POLICY: NESTED MODELLING
EuroGOOS was established in 1994, a year after the formation of GOOS itself. GOOS is one of the three global observing systems linked through the Integrated Global Observing Strategy (IGOS). EuroGOOS has 31 Members in 16 countries, and its Members are all government agencies concerned with ocean measurements, forecasting, and management. We maintain close links with the European Ocean Industries Association (EOIA) and the Marine Board of the European Science Foundation, and with other GOOS Regions. There is routine correspondence and discussion of plans with the US national equivalent, the Integrated Ocean Observing System (ISOOS).
Computing power limits operational services to a restricted resolution in time and space for each geographical scale of model. There is no option to increase the time of computing run, since the product has to be generated in real time. Thus there is a natural series of scales and resolutions, with nested models within the larger geographical scales. Very roughly, the scales which have proved useful in recent years are Global, Ocean Basin, Shelf Sea, and Estuary/Coastal. EuroGOOS projects such as Baltic Operational Oceanographic System (BOOS) and the Mediterranean Forecasting System Pilot Project (MFSPP) rely upon multiple models interfaced at these nested scales.
Within the EuroGOOS organisation the contrasting oceanographic conditions and economic needs of the different regional seas provide a natural scale, both for modelling, and for administration. EuroGOOS has established Regional Task Teams for the Baltic, North West Shelf Seas, and the Mediterranean. Additionally we have Task Teams working on the observations and modelling required in the adjacent Oceans, the Atlantic and Arctic. Collectively this results in a pattern of organisations and groups working at the scales needed for nested modelling.
Reports by the various Task Teams ( Le provost 1997; MFSPP 1999; BOOS 2000) and projects in hand (European Shelf Seas Ocean Data Assimilation Experiment (ESODAE), to be completed in 2001) have identified the data types and observations needed in each region. In every case satellite remote sensing is defined as essential at the regional sea level, while the large scale fields for meteorology and marine boundary conditions and fluxes are also needed, again depending upon remote sensing.
The oceanography of the different sea areas could hardly be more contrasted. The shallow Baltic Sea is dominated by fresh water input, winter sea ice in the northern sector, occasional bottom water renewal by inflows of oxygen-rich water from the North Atlantic via the North Sea, and a very large industrialised population on its shores and within its drainage basin. At the opposite extreme the Mediterranean is of full oceanic depth with a maximum trench depth of over 4000m, experiences an excess of evapouration over precipitation, has a higher than usual salinity, has a very small continental shelf, and is half surrounded by very dry countries with low industrialisation. The North West Shelf Seas are dominated by intensive industrialisation, massive exploitation of both fisheries and offshore oil and gas, and the shallow seas are mixed by strong tidal currents, combined with local storminess and steep waves. It follows that each sea area has to develop its own models oriented towards its own customers, but embedded in larger European and Atlantic models. At every scale, and for every purpose, ocean observations from space have a role to play. EuroGOOS Member agencies have already developed some operational models which can utilise remote sensed data from space, and the range of options and demand for data are continuously increasing.
LARGE SCALE EXPERIMENTS AND TRIALS
EuroGOOS projects are taking place against a background of Atlantic scale and global scale experiments and prototype ocean observing systems, within the context of the existing international meteorological data and service system. The Global Ocean Data Assimilation Experiment (GODAE) is designed to test and develop the techniques for gathering a global data set for the upper ocean in real time and provide a quality-controlled data stream for assimilation in real time into numerical models. GODAE includes real time data streams from both in situ and orbiting satellite observing platforms. Within GODAE a special project is being developed known as ARGO, which will utilise up to 3000 automated profiling floats to measure the temperature and salinity profile of the upper 2000m of the ocean on a global scale. EuroGOOS Member agencies are active in conducting trials of the floats for ARGO, and are participating with the USA in a North Atlantic pilot project with several hundred floats, as well as participating in the EU supported GYROSCOPE project to deploy 80 floats in the Atlantic as part of ARGO during the next 3 years.
The development of EuroGOOS projects on all scales, taken within the context of the global and Atlantic projects run by UN bodies and international science programmes such as CLIVAR, demonstrates that oceanography by observations from space is ready to move into a fully operational phase, at least for those parameters which are already best understood.
REQUIREMENTS SURVEYS
The EuroGOOS Data Requirements Survey (Fischer and Flemming 1999) was conducted in six European countries and shows that most organisations and ocean data users require data products which have a spatial horizontal resolution of the order of 1km, temporal resolution of the order of 6hr, data variable accuracy of the order of 1%, with an almost equal interest at all geographical scales from estuarine to oceanic. This is a very brief summary of a complex report, which gives much more detail about the complex relations between these different categories. Products which give the user this degree of resolution are of course the results of high resolution modelling based on the original data.
About 15% of respondents identify remote sensing as a major application of their own work, and the 9 most commonly requested variables by all respondents are characteristics of the surface fields of currents, waves, temperature, wind, and salinity in that order. Not only are these most commonly requested variables all measurable from space, but many of the other commonly requested variables related to sub-surface conditions require at least a component of space observations.
The EuroGOOS Technology Plan (Tziavos,1999) and the EuroGOOS Technology Requirements Survey (Bosman, 1997) both identify remote sensing from space as essential components of an ocean observing systems, and a source of data which is already important and in demand.
The EuroGOOS Regional plans (Atlantic, BOOS, MFSPP, ESODAE) all include frequent reference to the need for space based ocean data. The variables and technical details are provided in the publications, and are spelled out by other papers in this Conference.
EUROPEAN INFRASTRUCTURE
During the last 10 years many marine research projects have been funded during the EC Framework research programmes, FP3, 4, and 5. Many of these projects have recently been directly targeted at the development of operational oceanographic capabilities. Examples are the Mediterranean Forecasting System Pilot Project funded under FP4, and the GYROSCOPE profiling float project in the Atlantic funded under FP5. As EuroGOOS projects become more directly related to fully operational services, the funding must either come from the Member agencies as part of their normal budgets at national level, or some activities may be considered as European operational services, meeting European policy objectives. Additionally, because of the novel and exceptional level of increased co-ordination and integration needed between many agencies to achieve the goals of operational oceanography, some components of a European system might be considered as European Infrastructure.
In May 2000 EuroGOOS submitted a proposal to DG XII, Directorate for Research, proposing that, as a component of European Infrastructure within the new policy of a European Research Area, the Commission should support a three-pronged investment in ocean remote sensing, investment in the development of a petaflops computer system for operational ocean modelling, and investment in an extensive array of in situ observing instruments. The proposal document can be obtained from EuroGOOS.
In summary, EuroGOOS has already proposed at a high level the concept that operational oceanography in Europe should proceed through rapid development of ocean observations from space, super-computers, and advanced in situ systems. High speed interconnectivity between the components is integral within the proposal.
CONCLUSIONS
Europe will benefit to the extent of many hundreds of millions of Euros per year from the application of information based on operational oceanography, both at sea, in the coastal zone; if the landward benefits from medium and long term climate forecasting are added, the potential benefit is of the order of 2-5 bn Euro per year. Regional and international negotiations and regulations concerned with environmental management, response to seasonal and multi-annual climate forecasts, and calculations regarding water quality, public health, and possible anthropogenic climate change, will become increasingly expensive. The cost of compliance with agreed treaties or directives will become increasingly high, while the possible damage resulting from non-compliance, or non-enforcement will be equally costly. Thus Europe must control an independent component of the global ocean observing system in order to form policy. This will give Europe a stake in the global system; provide entitlement to access to all the data and decision-making processes; and guarantee that we can replicate independently all types of models and forecasts upon which political agreements may be based. To fail in this respect would place Europe at a permanent and serious disadvantage. Europe needs a consistent and steady focus on the development of operational ocean observing missions, consistent with its global role and regional requirements. The purpose of the Draft Conference Statement at this meeting is to start this process formally.
References
Adams R, Brown M, Colgan C, Flemming N C, Kite-Powell , McCarl B, Mjelde J, Solow A, Teisberg T, and Weiher, R F, 2000. The economics of sustained ocean observations: Benefits and rationale for public funding. Joint publication of the National Oceanic and Atmospheric Administration and the Office of Naval Research. 53 pp.
Bosman, J, N C Flemming, N Holden and K Tay,or, 1998. The EuroGOOS marine Technology Survey, EuroGOOS Publication No. 4, Southampton Oceanography Centre, Southampton. 47pp. ISBN 0-904175-29-4.
Buch, E, and H Dahlin, 2000. BOOS Plan, Baltic Operational Oceanographic System 1999-2003. EuroGOOS Publication No.14. 51pp. ISBN 0-904175-41-3.
Costanza, R, R D’Arge, R de Groot, S Farber, M Grasso, B Hannon, K Limburg, S Naeem, R V O’Neill, J Paruelo, R G Raskin, P Sutton and M van den Belt, 1997. The value of the world’s ecosystem services and natural capital. Nature 387, p.253-260. 1997.
Costanza et al., 2000. Extensive bibliography and short articles on the web site of the Maryland International Institute for Ecological Economics. (http://www.dieoff.org/ page87.htm).
EuroGOOS, 2000. Towards a European Research Area: The case for investment in the infrastructure of European Operational Oceanography. EuroGOOS document number 00.08. 14pp.
Fischer, J, and N C Flemming, 1999. Operational Oceanogrpahy: Data Requirements Survey. EuroGOOS Publication No.12. 59pp. ISBN 0-904175-36-7.
Flemming, N C, 1999. Dividends from investing in ocean observations: A European perspective. In: Proceedings of OceanObs 99, October 1999, St Raphael, France. (unpaginated).
Flemming, N C, 2000. Evaluating GOOS, original justification, monitoring implementation, and the future use of socio-economic analysis. GOOS Steering Committee, April 2000. Document IOC-GSC-III/17. 8pp.
IOC-IGOOS 1998.
Le Provost, C and N C Flemming (eds.), 1998. The EuroGOOS Atlantic Workshop Report. EuroGOOS Publication 9. Southampton Oceanography Centre. ISBN 0-904175-33-2. 76 pp.
OECD, 1994. Megascience: The OECD Forum. Oceanography. Organisation for Economic Co-operation and Development, Paris, France,. 167 pp. ISBN 92-64 14205-3.
OceanObs Conference Report. Proceedings of OceanObs 99, October 1999, St Raphael, France. (unpaginated). (To be published as a book during 2000).
Pinardi, N and N C Flemming, 1998. The Mediterranean Forecasting System Science Plan. EuroGOOS Publication no. 11. Southampton Oceanography Centre, 48 pp. ISBN 0-904175-35-9.
Sassone, P G, and R F Weiher, 1997. Cost benefit analysis of TOGA and the ENSO Observing System. Proceedings of the first International Conference on EuroGOOS, The Hague, The Netherlands, October 1996. Elsevier, the Netherlands. ISBN 0- 444- 82892-3. p. 36-50.
Tziavos, C and N C Flemming, 1999. The EuroGOOS technology Plan Working Group Report. EuroGOOS Publication No.13. 58pp. ISBN 0-904175-37-5
Woods Hole Oceanographic Institution (WHOI), 1993. Preliminary analysis of benefits and costs of a proposed Coastal Forecast System. Report on workshop held at WHOI, 7-8 June 1993. Marine Policy Centre and coastal Research Centre, Woods Hole, Massachusetts, 35pp.
Woods, J D, H Dahlin, L Droppert, M Glass, S Vallerga, and N C Flemming, 1996. The Strategy for EuroGOOS. EuroGOOS Publication No. 1. Southampton Oceanography Centre. 132pp. + Annexes. ISBN 0-904175-22-7.