In addition to the recommendations in Section 10, several individuals were invited to address their specific concerns:
Dr Croom:
GCOS is essentially concerned with long (15 + year) time series, implying requirements not only for RMS accuracy, but also for long-term bias (which cannot safely be assumed to be zero). However, with very few exceptions (e.g. SST, TOA radiation fluxes) specifications for bias have not yet appeared in the requirements. This need for low levels of long-term bias is what in many areas distinguishes climate needs from those of other disciplines. Panels are therefore urged to consider their needs for long-term bias in formulating their requirements.
Dr Cihlar:
1. GOSSP considered the importance of existing data acquisition and analysis programs for building up time series of data. Based on the input of the GCOS/GTOS Terrestrial Observation Panel for Climate, GOSSP endorsed current global AVHRR data acquisition at full (1 km) resolution and its continuation, through the following recommendation:
Recommendation 1. Continue the acquisition of the global 1 km AVHRR data set and ensure ongoing processing from 1992 to beyond the launch of future satellites (SPOT4, EOS-AM, ENVISAT). Provide sufficient overlap between the AVHRR and the subsequent data sets to ensure consistency in the time series and facilitate the transfer of algorithms (such as those based on vegetation indices) from AVHRR to future sensors.
2. GOSSP acknowledged the expanded calibration/validation programme of CEOS as an important step in making space observations a high quality data-gathering tool which will provide critical global information for the observing systems. Two specific recommendations were made to CEOS in the area of calibration/validation. The request for quality calibration information has two elements; 1) obtaining the calibration data and 2) making them available to the user community. Although progress in this area has been made, deficiencies remain. It is recognised that increased effort has been made by many space agencies in ensuring proper calibration of sensors under construction, but increased effort is required for sensors currently operating (e.g., AVHRR and LANDSAT). The second recommendation is a proposal for a pathfinder case in the relatively new area of product validation and its international coordination. CEOS has an important role to play, and its willingness to undertake these activities through the WGCV is acknowledged.
Recommendation 2a. Continue efforts in maintaining accurate knowledge of post-launch sensor calibration, especially for the active AVHRR sensors and for LANDSAT, and make calibration information for present and future satellite sensors accessible to the user community on Internet.
Recommendation 2b. Initiate coordinated global validation efforts for Leaf Area Index (LAI) products and Fraction of Photosynthetically Active Radiation (FPAR) products derived from satellite data.
3. GOSSP discussed the present difficulties in obtaining effective access to high resolution satellite data, principally due to the high costs of most of the data products. This issue is highly relevant to the global observing systems which are designed primarily to provide information for public good. It was pointed out that this is not a new issue and that it has been discussed by various meetings and groups, but it was accepted that this issue is an important one for the initial global observation strategy being considered by CEOS.
Dr. Ryder:
The panel found it difficult to fully represent the requirements of the science panels in its recommendations, in the absence of guidance on the way in which specified measurements would be processed to generate useful products and services.
The difficulty is manifest at the level of individual derived products where the availability of complimenting in-situ measurements and data assimilation by numerical models can increase the value of the combination in a powerful manner, and hence the priority to be attached to the space-based component.
The panel also found the absence of a clear articulation of an agreed overall paradigm (or paradigms) for the research and operational activities which is expected to deliver economic and social benefits, to be a distinct disadvantage and source of confusion for its work.
Recommendation: When expressing requirements for observations, the science panels should make it clear how and for what purpose they will be used in the generation of products and services. Further, the G3OS steering committees should devise and promulgate their vision of the end-to-end system of research and operations by which they expect to deliver the benefits to justify investment in the GOSSP.
Dr Aschbacher:
Information about the availability, accessibility and quality of remote sensing data, products and algorithms should be easily accessible in a timely way to scientific, governmental and inter-governmental organisations concerned with the observation of the Earth's climate, terrestrial and oceanic environment. Emphasis should be put on the further development of earth observation data and information exchange systems, covering both the space and non-space (in-situ) domain.
Recommendation: NASA, NASDA and EC, inter alia, should be encouraged to assure a long-term operational service of their respective Earth observation data and information exchange systems (i.e., NASA's EOSDIS, NASDA's EOIS and EC-CEO's EWSE/ES). Coordination across these systems, and expansion to other remote-sensing data providers should be encouraged in order to develop an integrated, comprehensive and up-to-date global meta-information system including all major Earth observation data and information providers.
