There are numerous networks and initiatives con-cerned with the non-satellite-observing segment of Earth ob-servation. These are owned and operated by various enti-ties and organisations often with different practices, norms,data policies, etc. The Horizon 2020 project GAIA–CLIMis working to improve our collective ability to use an ap-propriate subset of these observations to rigorously charac-terise satellite observations. The first fundamental questionis which observations from the mosaic of non-satellite obser-vational capabilities are appropriate for such an application.This requires an assessment of the relevant, quantifiable as-pects of the measurement series which are available. Whilefundamentally poor or incorrect measurements can be rela-tively easily identified, it is metrologically impossible to besure that a measurement series is “correct”. Certain assess-able aspects of the measurement series can, however, buildconfidence in their scientific maturity and appropriateness forgiven applications. These are aspects such as that it is welldocumented, well understood, representative, updated, pub-licly available and maintains rich metadata. Entities such asthe Global Climate Observing System have suggested a hi-erarchy of networks whereby different subsets of the obser-vational capabilities are assigned to different layers basedon such assessable aspects. Herein, we make a first attemptto formalise both such a system-of-systems networks con-cept and a means by which to, as objectively as possible,assess where in this framework different networks may re-side. In this study, we concentrate on networks measuringprimarily a subset of the atmospheric Essential Climate Vari-ables of interest to GAIA–CLIM activities. We show assess-ment results from our application of the guidance and howwe plan to use this in downstream example applications ofthe GAIA–CLIM project. However, the approach laid outshould be more widely applicable across a broad range ofapplication areas. If broadly adopted, the system-of-systemsapproach will have potential benefits in guiding users to themost appropriate set of observations for their needs and inhighlighting to network owners and operators areas for po-tential improvement.