Mosses (Hypnaceae) and lichens (Parmeliaceae) were collected from three distinct ecological environments in Korea between May and July 2019: (1) a high-altitude mountainous region (Sam-Bul Peak, Gyeryong Mountain, 775 m above sea level); (2) a high-humidity coastal area (Bong-Rae Waterfall, Ulleung Island); and (3) urban community parks (Daejeon City). Mosses of carpet-forming growth types were collected from rock surfaces. The moss carpet covering bulky rocks was first separated from rock surfaces using a garden trowel and stored in plastic bags. Special care was taken to prevent cross-contamination between samples after transporting them to the laboratory.

A total of 15 samples were gathered (Table 1). Ten moss samples (200–300 g fresh weight each) were collected from tree trunks, such as oriental oak, acasia, and cedar, at heights of 1–2 m above the ground, as well as from the surface of rocks and soil (Fig. 1A). Three lichen samples (50–100 g fresh weight each) were obtained from the rock surfaces and trunks of cherry and zelkova trees (Fig. 1B). Two topsoil samples (1–1.5 kg fresh weight each) directly below the mosses were collected from the same locations as the moss samples.

Small pieces of bark, dead insects, and other debris attached to the samples were removed using tweezers. The bottom part of the moss samples separated from rocks or soil was rinsed with distilled water while gently shaking the moss. All samples were naturally dried for a few days and then oven-dried at approximately 100 °C for 48 hours. Subsequently, the samples were transferred and ground to a powder in a glovebox using a blender. The ground powder was homogenized by passing through a 500 μm mesh sieve and then ashed in a furnace at approximately 400 °C for 24 hours. After weighing, each sample was enclosed in a polyethylene cylindrical container of the same geometry as the calibration source container for gamma spectrometry [15], i.e., 60 mm in diameter and 40 mm in height.

The radioactivity concentrations of radiocesium in the moss and lichen samples were determined via gamma spectrometry conducted using a high-purity Ge coaxial detector (HPGe; EG&G Ortec) and a multichannel analyzer. The peak areas at 661.7 keV (Pγ, 0.850) for ¹³⁷Cs, along with 604.7 keV (Pγ, 0.976) and 795.9 keV (Pγ, 0.854) for ¹³⁴Cs were calculated using spectrum analysis software (Gennie 2000; CANBERRA). Additionally, the radioactivity concentrations of ⁴⁰K and ⁷Be were determined from their respective gamma-ray energy lines at 1,461 keV (Pγ, 0.107) and 477.6 keV (Pγ, 0.103). Peak efficiencies of the high-purity Ge spectrometry system were calibrated using a mixed radionuclides solution prepared by diluting multiple gamma-ray-emitting standard sources (Product No. MX-8600; Eckert & Ziegler). To cover a wide photon energy range, the standard source contained a known mixture of 10 individual isotopes (²⁴¹Am, ¹⁰⁹Cd, ⁵⁷Co, ¹³⁹Ce, ⁵¹Cr, ¹¹³Sn, ⁸⁵Sr, ¹³⁷Cs, ⁸⁸Y, and ⁶⁰Co). The 11 diluted calibration sources were manufactured at 5 mm intervals in height of the sample geometry of polyethylene cylindrical container (Φ60 mm× 40 mm) as samples to correct geometric height differences for each sample via interpolation. The measuring time for the samples was 80,000 seconds. The MDAs for ¹³⁷Cs and ¹³⁴Cs in the samples were 1–3 Bq· kg⁻¹, depending on the sample amount. The MDA was insufficient for some samples because of small sample amounts. The reference date for decay correction was based on the sampling dates, and the radioactivity concentration in the sample is given as Bq· kg⁻¹ on a dry weight basis.

The measured activity concentrations of ¹³⁷Cs, ⁴⁰K, and ⁷Be are presented in Table 1 and Fig. 2 with ¹³⁷Cs in moss samples ranging from 1.89 Bq· kg⁻¹ to 47.8 Bq· kg⁻¹, and this range depends on the type of substrate and ecological environment. Additionally, the ¹³⁴Cs activity concentrations in all samples were less than their respective MDAs, likely because of insufficient sample amounts and counting time. The ¹³⁷Cs values in the moss covering rock substrate measured in this study agree with the results of Park et al. [16], which had values between 15 Bq· kg⁻¹ and 41 Bq· kg⁻¹ for moss samples collected from Jeju Island in 2011, shortly after the Fukushima nuclear accident.

The measured activity concentrations of ⁴⁰K and ⁷Be in mosses are 156–506 Bq/kg and 60.6–1,227 Bq/kg dry, respectively. Potassium-40 (⁴⁰K) and ⁷Be have different origins, with ⁴⁰K originating from the Earth’s crust, and ⁷Be produced as an airborne radionuclide via the spallation reactions of cosmic rays with ¹⁴N, ¹⁶O, and ¹²C [17]. Therefore, it is highly likely that a majority of ⁷Be in the moss and lichen of mountain samples result from atmospheric deposition, which varies with meteorological factors, geographic location, etc. Additionally, ⁴⁰K adsorption on moss and lichen could occur primarily through airborne dust soil particles carried by wind and spattering of raindrops containing soil particles.

Fig. 3 illustrates the comparison of the measured ¹³⁷Cs radioactivity concentrations in moss and lichen growing on rock substrates in the high-altitude mountainous region. Both samples were collected from the same place and substrate. It has been previously shown [18] that mosses tend to accumulate higher contents of resuspended soil than lichens because of their higher surface-to-volume ratio. The data obtained in this study also indicates that mosses accumulate greater amounts of ¹³⁷Cs. Lichens are symbiotic associations of a fungus and either an alga or a cyanobacterium [4]. Unlike in moss, radionuclides in lichens are actively transported to the lower parts of the thallus [19]. Consequently, as thallus’ decompose with age, radionuclides are removed, and therefore, it is likely that ¹³⁷Cs in lichen samples originates from deposition during at least a few years prior to sample collection; however, this depends on the biological residence time. It has been previously shown [18] that the degree of radionuclide accumulation in mosses and lichens is largely determined by their species and age, as well as the ecological conditions of their habitat.

Fig. 4 displays the comparison of measured ¹³⁷Cs radioactivity concentrations in moss with three different substrates within the same ecological environment (mountainous region). The ¹³⁷Cs radioactivity concentration in moss growing on bark is substantially lower than that in moss covering rock or soil substrates. The relatively low ¹³⁷Cs activity concentrations in bark were attributed to differences in the ¹³⁷Cs resuspension effect, which may occur via the spattering of raindrops and the effects of wind. A similar implication was suggested by Sawidis et al. [18], which stated that epilithic mosses reflect soil conditions more directly than epiphytic ones because of their proximity to the substrate. This was due to a majority of moss samples in this study being collected from tree trunks at a height of 1–2 m above the ground, whereas some samples were collected from rock and soil surfaces.

Cesium-137 (¹³⁷Cs) and ⁴⁰K activity concentrations in moss samples were compared with those in the soil covering the moss. To intuitively assess the mosses capability of adsorbing ¹³⁷Cs and ⁴⁰K, radionuclide transfer from soil to moss was quantified in terms of the transfer factor (TF), which were calculated as follows in Equation (2) [20]:

where A_moss and A_soil are the radioactivity concentrations of radionuclides in moss and soil, respectively.

The TF values for ¹³⁷Cs, ⁴⁰K, and ⁷Be are 0.61, 0.65, and 0.46, respectively with similar TF values for ¹³⁷Cs and ⁴⁰K implying that these radionuclides behave similarly in moss. However, the TF value for ⁷Be is smaller than that for ¹³⁷Cs, suggesting that the atmospheric deposition effect was a more dominant factor for ⁷Be.

We have previously conducted measurements for the national monitoring program for radioactive fallout over the past 20 years. These measurements were carried out using annual mugwort and perennial pine needles as bioindicators. According to the annual reports [21], the ¹³⁷Cs radioactivity concentrations in almost all mugwort samples were less than the MDA (0.023–0.103 Bq/kg fresh), as displayed in Fig. 5. The results indicate that mugworts and pine needles are not suitable bioindicators for estimating temporal variations in radioactive fallout contamination. The present study suggests that mosses show potential as bioindicators of long-term radiocesium fallout monitoring due to the following features: (1) perennial plants are widely distributed in Korea; (2) they exhibit relatively high capacity for radionuclide accumulation; (3) a close relationship exists between activity concentrations in moss and radioactive fallout [22]; (4) mosses are convenient for measurements because they accumulate a greater amount of airborne substances than vascular plants, e.g., mugwort, growing in the same habitat [23]; and (5) their sampling, pretreatment, and measurement are relatively simple and inexpensive.