Best practices at a GRUAN site aim at obtaining long term climate reference observations and reference observations for validation of other sensors. This requires operational best practices as well as best practices for instruments and data analysis. Establishing measurement uncertainties will be an important element of best practices. This implies that any instrument that is being used needs to be characterized to the best possible extent. For example, ground checks for sounding equipment are highly recommended for each parameter independent from a manufacturers ground check. This will ensure that possible and unexpected instrumental biases may be identified before launch. Instrument preparation needs to be standardized as much as possible to reduce any possible observer influence.
Another element will be establishing measurement uncertainties for every measurement point. This may be done in cooperation with the lead center depending on the instrumentation that is being used. This implies that as part of data analysis all know systematic biases (e.g., time-lag corrections or empirical corrections) are being applied and that all remaining uncertainties are characterized and quantified. Comparison with other reference instrument should be done routinely to verify, whether the uncertainty estimation is consistent and to identify weaknesses that need to be improved. The frequency of reference observations depends on the vertical altitude region as well as the natural variability of the measured parameter. This subject remains under investigation. Metadata need to be complete to describe every aspect of the measurement and every source of uncertainty. The lead centre will provide a common template for metadata collection.