Remedy 2: Take steps to better realise the benefits of a system-of-systems approach to observing strategies

Current observational networks are treated as distinct entities, all too frequently meaning that synergies resulting from a system-of-systems approach to observing are not realised. Without means to propagate innovations, practices, and know-how, the benefits of improved understanding from high-quality reference networks are limited. Work is required to develop tools and approaches that allow the effective flow of information from reference quality measurement networks to baseline and comprehensive observing networks, so that the benefits of that improved understanding can be realised. In the first instance, a case study based approach may be advisable that considers a well-defined problem set and allows testing of various approaches, following which are more substantial roll-out would be possible. An obvious candidate may be atmospheric temperature and humidity measurements for which several reference quality measurement techniques exist or are in the advanced stages of preparation and for which assimilation models and other techniques are similarly advanced. Work may include (but not be limited to) aspects such as:

• Use of reference sites to qualify uncertainties in techniques used in remaining networks via intercomparison campaigns. This may benefit from improved management of holdings if the new Copernicus Climate Change Service C3S 311a Lot 3 (access to observations from baseline and reference networks) activity is successfully executed.

• Enhancing observational practices in non-reference networks by taking realisable aspects of best practices from reference techniques. For example, the use of 100%-RH checks on radiosondes to characterise hysteresis effects more explicitly.

• Using data assimilation and statistical techniques to propagate information from reference sites to surrounding locales.

The work would need to involve operators of both reference and baseline / comprehensive networks to be effective and to recognise the realities involved in measurement programs. Cost-effective solutions that were technically and financially achievable should be developed that more effectively integrate information across networks and improve the quality of all observations.