Dataset for the AquaVIT-4 intercomparison of atmospheric hygrometers

The campaign was divided into two phases, each consisting of four measurement days: an “open intercomparison”, where the simulated conditions were known to the participants, and a “blind intercomparison”, where the conditions were coordinated by independent referees and unknown to the participating teams. Here we present a statistical analysis of the entire dataset, which allows to assess the accuracy and limitations of each instrument.

For the accuracy evaluation, two sets of reference measurements were defined: one for in situ instruments, located inside the AIDA vessel, and one for extractive instruments, sampling the chamber gas through a heated inlet. This distinction accounts for H2O desorption effects, which are most prominent at low pressures and low H2O concentrations. All instruments showed a good agreement with the reference values in the range of H2O > 2 ppm, with mean deviations within ±7 % for H2O > 10 ppm, and ±8 % between 2–10 ppm H2O.

The largest differences were found for H2O < 2 ppm, a rarely observed range in the atmosphere, though most of the instruments still achieved average deviations within ±10 %. Overall, the results of AquaVIT-4 demonstrate the high accuracy and reliability of the four involved sensors for upper atmospheric monitoring and research applications.