Inter-comparison study of atmospheric 222 Rn and 222 Rn 1 progeny monitors 2

20 The use of the noble gas radon (222Rn) as tracer for different research studies, for example observation21 based estimation of greenhouse gas (GHG) fluxes, has led to the need of high-quality 222Rn activity 22 concentration observations with high spatial and temporal resolution. So far a robust metrology chain for 23 these measurements is not yet available. 24 A 3-month inter-comparison campaign of atmospheric 222Rn and 222Rn progeny monitors based on 25 different measurement techniques was realized during the fall and winter of 2016-2017 to evaluate: i) 26 calibration and correction factors between monitors necessary to harmonize the atmospheric radon 27 observations; and ii) the dependence of each monitor’s response in relation to the sampling height, 28 meteorological and atmospheric aerosol conditions. 29 Results of this study have shown that: i) all monitors were able to reproduce the atmospheric radon 30 variability on daily basis; ii) linear regression fits between the monitors exhibited slopes between 0.62 31 https://doi.org/10.5194/amt-2019-378 Preprint. Discussion started: 6 November 2019 c © Author(s) 2019. CC BY 4.0 License.


e fo
most applications (Schery and Huang, 2004).

In recent decades the atmospheric scientific community has been addressing different research topics using 222 Rn as a tracer.Examples of such applications include: the improvement of inverse transport models (Hirao et al., 2010), the improvement of chemical transport models (Jacob and Prather, 1990;Chambers et al. 2019a), the study

two
most commonly employed measurement systems at European 222 Rn monitoring stations are: the dual-flow-loop two-filter monitor (Whittlestone and Zahorowski, 1998;Zahorowski et al. 2004;Chambers et al., 2011Chambers et al., , 2014Chambers et al., , 2018;;Griffith et al., 2016), which samples and measures radon directly, and the one-filter monitors, of which several kinds are in use (e.g.Stockburger and Sittkus, 1966;Polian, 1986;Paatero et al., 1998;Levin et al., 2002), which sample and measure aerosol-bound radon progeny.Finally, a third method is being used at several Spanish atmospheric stations (Vargas et al., 2015;Hernández-Ceballos et al., 2015;Grossi et al., 2016;Frank et al., 2016;Grossi et al., 2018;Gutiérrez-Álvarez et al., 2019).This type of instrument performs a direct measurement of 222 Rn and 220 Rn (thoron) activity concentrations using the already existent method based on the electrostatic deposition of 218 Po and 216 Po, respectively (Hopke, 1989;Tositti et al., 2002;Grossi et al., 2012).

The diversity of these three aforementioned measurement techniques could introduce biases or compatibility issues that would limit the comparability of the results obtained by independent studies and the subsequent application of atmospheric radon data for regional-to-global investigations (e.g.Schmithüs

et
l., 2017).Thus, a comparative assessment of all the experimental techniques applied for atmospheric 222 Rn activity concentration measurements and a harmonization of their datasets is needed, as suggested by the International Atomic Energy Agency (IAEA, 2012).Xia et al. (2010) carried out a comparison of the response of a dual-f

alia
Nuclear Science and Technology Organisation (ANSTO, Whittlestone and Zahorowski 1998) and a one-filter monitor (α/β Monitor P3) manufactured by the Bundesamt für Strahlenschutz, Germany (BfS) (Stockburger and Sittkus, 1966), for atmospheric 222 Rn measurements under various meteorological conditions at 2.5 m above ground level (a.g.l.) over one year.Their results showed that both systems followed the same patterns and produced very similar results most of the time, except under specific

eteo
ological conditions such as when precipitation or the proximity of the forest canopy could remove short-lived progeny from the air mass to be measured by the one-filter monitor.However, Xia et al. (2010) did not find a clear relationship between precipitation intensity and the ratio between progenyderived 222 Rn and 222 Rn activity concentration to convert the progeny signal to 222 Rn activity concentration.Gr

si e
al. (2016) presented results from two short (about 7-9 days) comparisons between a one-filter monitor from Heidelberg University (HRM; Levin et al., 2002), and an Atmospheric Radon MONitor (ARMON, Grossi et al., 2012), an electrostatic deposition monitor from the Universitat Politecnica de Catalunya (UPC

The
wo comparison campaigns were carried out at a coastal and a mountain site, with sampling in both cases from 10 m a.g.l.These comparisons revealed that the responses of both monitors were in agreement except for w

er s
turated atmospheric conditions or periods of rainfall.Again, the quantity of comparison data was not suffi

ent
o confirm any statistical correlation.

Loss of aerosols in the a

int
ke systems can also complicate the derivation of 222 Rn activity concentrations from one-filter

stem
such as the HRM.Levin et al. (2017) carried out an assessment of 222 Rn progeny loss in long tubing by laboratory and field experiments.Results of these experiments, for 8.2 mm inner diameter (ID) Decabon tubing, gave an empirical correction function for 222 Rn progeny measurements

whic
enables the correction of measurements for this specific experimental setup (e.g.tubing type and diameter, flow rate, aerosol size distribution).

Finally, Schmithüsen et al. (2017) conducted an extensive European-wide 222 Rn/ 222 Rn progeny comparison study in order to evaluate the comparative performance of one-filter and two-filter measurement systems, determining potential systematic biases between them, and estimating correction factors that could be applied to harmonize 222 Rn activity concentration estimates for

eir
se as a tracer in various atmospheric applications.In this case, the a

hors
employed a HRM monitor as the reference device.It was taken to nine European measurement stations to run for at least one month at each o

them
This monitor was run in parallel to other one-filter and two-filter radon monitors operating at each station of interest.

Although several inter-comparison campaigns have been carried out in the past, none of them has included simultaneous observations from one-filter, two-filter and electrostatic deposit

n me
hods.Here, we present the results of a three-month inter-comparison campaign carried out in the fall and winter of 2016-2017 in Gif Sur Yvette (France) where, for the first time, co-located measurements from monitors based on the three measurement principles were included.Two two-filter 222 Rn monitors, two single-filter 222 Rn progeny monitors and an electrodeposition monitor were run simultaneously under differen

mete
rological and aerosol conditions sampling from heights of 2 and 100 m a.g.l.

The main objectives of the present study were to: i)

ompa
e the calibration and correction factors between all monitors required to derive harmonized atmospheric radon activity concentrations; and ii) analyze the influence that meteorological and environmental parameters, as well as sampling height, can have on the finally determined 222 Rn activity concentration.

In the present manuscript the applied methodology is reported, including a short presentation of the 222 Rn / 222 Rn progeny monitors participating in the campaigns, the sampling sites and the statistical analysis carried out.Finally, the outcomes of the present study are discussed and compared with the ones from Schmithüsen et al. (2017).


Methods

In section 2.1 a short de

ript
on is given of the monitors compared in the experiment, mainly focusing on measurement techniques, in