Development of overall strategies for observations of the Earth system has received considerable attention over the last two years. Most attention has been paid to space-based observations, but of comparable importance is the wide variety of in situ observations made at the surface of the Earth and in its atmosphere and oceans. By in situ observations we refer to those observations made as a result of direct measurement at a site: some observations will be made directly, possibly by automated means, whereas many others may be made in the laboratory on samples taken from a site. A joint meeting co-sponsored by the Global Climate, Ocean and Terrestrial Observing Systems (GCOS, GOOS and GTOS) was convened to examine the extent to which observational capabilities match needs and to determine what are the key actions needed to remedy deficiencies in current systems.
In some fields, the last ten years have seen considerable improvements in the acquisition of in situ observations, but other in situ observational systems have suffered serious decline. Some global in situ observational systems are well organized and function very successfully. But there are important in situ observational systems, such as those relating to terrestrial ecology, lacking comprehensive international coordination of global observations.
The meeting concentrated on strategic issues and identification of priorities relating to climate requirements for global in situ observations and products, including those for climate prediction services, climate research and assessment of climate change. It was based on requirements as outlined in the existing plans (e.g., GCOS 1995a, Ocean Observing System Development Panel 1995, ICSU/UNEP/FAO/UNESCO/ WMO 1996) of the three observing systems (GCOS, GOOS and GTOS).
Many global observing systems exist apart from GCOS, GOOS and GTOS. For example the World Weather Watch and Global Atmospheric Watch are global observing systems responsible for the international coordination of many key atmospheric observations. Moreover part of the World Weather Watch is itself called the Global Observing System. But to avoid endless repetition of acronyms and avoidance of such clumsy expressions as the «G3OS's» the term global observing systems will be used within this document to refer to GCOS, GTOS and GOOS.
The three global observing systems already have a number of mechanisms for ensuring coordination of their strategies. There are the two joint design panels, namely the Ocean Observation Panel for Climate (OOPC) and the Terrestrial Observation Panel for Climate (TOPC), the first of which is developing the ocean component of GCOS and the climate component of GOOS and similarly the TOPC is developing the terrestrial component of GCOS and the climate component of GTOS.
Recently it has been decided to expand the scope of the Space-based Observation Panel of GCOS to include the interests of GCOS, GOOS and GTOS and there are proposals in hand similarly to expand the scope of the GCOS Data and Information Management Panel.
The meeting's objectives were to carry out the following:
Provide an overview of the principal internationally coordinated in situ observing systems, mainly by identifying and reviewing critical deficiencies in current in situ observations using the plans of the three global observing systems to identify requirements;
Assess future in situ requirements on the basis of these plans.
Review the status of in situ observations vital for the support of space observations;
Develop a strategy to remedy deficiencies, based on existing mechanisms and networks;
Review the status of data management systems;
Identify deficiencies in the organizational arrangements for the collection assembly and distribution of observational data sets;
Consider the position of in situ observations in relation to other strategies such as those for space in the "integrated global observation strategy".
In situ observations at a global scale are often inadequate for several reasons including the following:
Lack of precision and accuracy, arising through factors such as inadequate instrumentation, lack of skilled personnel and inadequate maintenance;
Inadequate sampling in space and/or time;
Incompatibility of observations, for example by the use of different instruments, different timing of observations and different methods for recording the data;
Inadequate assembly of observations globally often due to a lack of data management but also for some observations, because of national sensitivities about the use of observations relating to national resources.
The reasons for such deficiencies apart from the more obvious technical ones are not difficult to identify. They include inadequate data management locally, nationally and internationally, deficiencies of international coordination mechanisms to ensure that collected observations are assembled into global data sets and of course insufficient financial resources.
Given the heterogeneity of in situ observational systems it is appropriate to question why a strategic oversight for in situ observations is needed. Firstly there is the fact that many in situ observations are irreplaceable in that they are not substitutable by remote sensing observations and have to be maintained for an understanding of climate. However some of these in situ systems are deteriorating. Also many new techniques for in situ observations are becoming available, which may compete with existing systems to the possible detriment of consistent long-term observations. In many areas resources are becoming increasingly constrained, so that choices may have to be made between different observational systems. The latter is one of the main drivers behind the increasing desire of some governments to establish observational priorities with the framework of an overall integrated observational strategy. Finally it is clear, that for globally relevant in situ observations to be collected, many different nations and organizations have to be involved in planning and implementation and a coordinated approach has to be developed.
Strategies are needed to establish requirements in a consistent fashion, to assess priorities and to achieve implementation. This poses considerable challenges. For example prioritization is difficult within observing systems and setting priorities between observing systems may be very difficult indeed. But if the global observing systems do not establish priorities others, outside or peripheral to the global observing systems or charged with implementing systems, certainly will do so.
In some ways it is becoming more difficult to demonstrate the benefits of improving or maintaining observations because of the complex way in which observations are processed to generate the products required by users. For example products are often derived by integration of in situ and remote sensing observations. Four dimensional assimilation techniques are increasingly used to create more consistent data sets, but this may make it difficult to track the contribution of any individual observational component. Also there is the fundamental problem that data quality is highly dependent on the specific application of the data: new applications may often reveal new deficiencies and hence quality and data deficiencies are dynamic attributes of data sets.
Achieving implementation of the plans of the global observing systems whether for in situ or for space observations will prove a difficult task. For this to be possible it is important to ensure that national governments, who provide funding, have a better understanding of the importance of in situ observations through a comprehensive understanding of requirements and capabilities. Of equal importance is that they understand the benefits of these observations and the advantages of international cooperative efforts in their assembly and use. Simply asking for improvements, which demand additional resources, will usually be unsuccessful. There is a need to demonstrate the vital need for improvements, their impacts in terms of use, and their feasibility especially in terms of cost-effectiveness. The latter will be assisted by the strengthening of existing organizational structures and mechanisms, for example through closer collaboration between the global observing systems.
Some mechanisms for implementation already exist. For example, through the World Meteorological Organization (WMO), there exists a process whereby the permanent representatives of member countries can be formally approached seeking their cooperation in collecting new or improved observations following a systematic process of developing, evaluating and vetting proposals from organizations such as GCOS. But for some in situ observations there is no existing international agency responsible for such coordination, for example for the global network of ecological test sites proposed by the TOPC. In such cases it will be important that an appropriate international agency takes on an international coordinating responsibility. Sometimes networks with operational responsibilities emerge from prototype research observational systems, such as the Tropical Atmosphere Ocean (TAO) array, but the transfer from research programme to operational programme may be very lengthy. Improving the quality of in situ observations will be a complex and difficult set of tasks. The report of this meeting is intended to provide an initial overview of requirements, capabilities and the steps needed to improve the most important elements.
