A direct result of the fractured governance of observational networks is that instruments that could derive synergistic analysis benefits are very frequently not geographically co-located. That is to say that an instrument may belong to network or operator X and be located 100km distance from a suite of potentially complimentary instruments belonging to network or operator Y. Because the measurements are geographically dispersed, this serves to reduce their value for numerous applications, including, but not limited to, satellite characterisation. This arises either because they measure complementary ECVs that enable fuller understanding, or measure distinct aspects of the same ECV such that, when combined, a fuller understanding of the measurand accrues. This is especially important for certain satellite instruments such as hyperspectral sounders, which, across the sensed channels, are sensitive to a broad range of ECVs such that to adequately characterise them requires quasi-coincident measures of a broad number of ECVs with an overpass.

In a worst-case scenario of a catastrophic space weather event, there remains a risk that multiple satellites are simultaneously unavailable. To bridge such an event from a climate perspective requires the persistence of a set of in-situ sounding capabilities that can measure what is sensed by the satellite instrumentation across the gap. For the more complex instruments, there is value to this being achieved by a set of super-sites that measure multiple ECVs simultaneously and to high quality.

However, in some cases, there may be good reasons to not co-locate measurements: (1) if long time series already exist, it would be counterproductive to climate monitoring to disrupt the time series by re-locating the instrument to another site; (2) the atmospheric variability may be different from one target species to another, justifying their observation at different sites, and (3) the benefits of a site for satellite validation are not necessarily the same as for other research purposes. For example, a mountain site may be very appropriate for stratospheric observations, but is much less appropriate for satellite validation.

Therefore, a careful scientific analysis should be carried out before implementing a new observation site, and before deciding to re-locate an instrument, taking into account the existing data, the existing sites in the neighbourhood, and the main scientific objectives of the (new) observations. Funding authorities and network coordinators should take these scientific analyses into account before taking decisions about the implementation of new observations or moving existing capabilities.