16.7.1 Observations and climate change science

• Ongoing observation is required to monitor the rate and magnitude of changes and impacts, over different spatial and temporal scales. In situ observations of sea level should be strengthened to understand the components of relative sea-level change on regional and local scales. While there has been considerable progress in regional projections of sea level since the TAR, such projections have not been fully utilised in small islands because of the greater uncertainty attached to them, as opposed to global projections.
• Since the TAR, it has also been recognised that other climate-change-induced factors will probably have impacts on coastal systems and marine territories of small islands, including rises in sea temperature and changes in ocean chemistry and wave climate. The monitoring of these and other marine variables in the seas adjacent to small islands would need to be expanded and projections developed.
• Although future projections of mean air temperature are rather consistent among climate models, projections for changes in precipitation, tropical cyclones and wind direction and strength, which are critical concerns for small islands, remain uncertain. Projections based on outputs at finer resolution are needed to inform the development of reliable climate change scenarios for small islands. Regional Climate Models (RCMs) and statistical downscaling techniques may prove to be useful tools in this regard, as the outputs are more applicable to countries at the scale of small islands. These approaches could lead to improved vulnerability assessments and the identification of more appropriate adaptation options.
• Supporting efforts by small islands and their partners to arrest the decline of, and expand, observational networks should be continued. The Pacific Islands Global Climate Observing System (PI-GCOS) and the Intergovernmental Oceanographic Commission Sub-Commission for the Caribbean and Adjacent Regions Global Ocean Observing System (IOCARIBE-GOOS) are two examples of regional observing networks whose coverage should be expanded to cover other island regions.
• Hydrological conditions, water supply and water usage on small islands pose quite different research problems from those in continental situations. These need to be investigated and modelled over the range of island types covering different geology, topography and land cover, and in light of the most recent climate change scenarios and projections.