Interlaboratory comparisons are a basic part of the regular quality controls of laboratories to warranty the adequate performance of test and measurements. The exercise presented in this article is the comparison of indoor radon gas measurements under field conditions performed with passive detectors and active monitors carried out in the Laboratory of Natural Radiation (LNR). The aim is to provide a direct comparison between different methodologies and to identify physical reasons for possible inconsistencies, particularly related to sampling and measurement techniques. The variation of radon concentration during the comparison showed a big range of values, with levels from approximately 0.5 to 30 kBq/m³. The reference values for the two exposure periods have been derived from a weighted average of participants’ results applying an iterative algorithm. The indexes used to analyze the participants’ results were the relative percentage difference D(%), the Zeta score ( $\zeta$ ), and the z-score ( $z$ ). Over 80% of the results for radon in air exposure are within the interval defined by the reference value and 20% and 10% for the first and the second exposure, respectively. Most deviations were detected with the overestimating of the exposure using passive detectors due to the related degassing time of detector holder materials.

Interlaboratory exercises are a good tool to compare the response of different systems to the same quantity and to identify possible inconsistencies between them. One of the main goals of the EMPIR 19ENV01 traceRadon project is to harmonize radon flux measurements based on different systems and methodologies. In the framework of the traceRadon Project, two radon flux intercomparison campaigns were carried out in October 2021 at high and at low radon source areas. Four institutions participated in the field intercomparison exercises with their own systems. Every system was based on a specific radon monitor (diffusion or pump mode) and an accumulation chamber (with manual or automatic opening). Radon fluxes were calculated by each participant using both exponential and linear fittings of the radon activity concentration measured over time within the accumulation chambers. The results of this study show mainly: (i) the exponential approach is not advisable due to the variability of the radon flux and the leakage of the systems during long-time measurements; (ii) the linear approach should be applied to minimize the measurement period in agreement with the time response and sensitivity of the monitors; (iii) radon flux measured at high radon source areas (radium content of about 800 Bq kg⁻¹) risks being underestimated because of the influence of advective effects; (iv) radon flux measured at low radon source areas (radium content of about 30 Bq kg⁻¹) may present large uncertainties if sensitive radon monitors with pump mode are not used.