Atmospheric Pollution Particulate Matter Absorption Efficiency by Bryophytes in Laboratory Conditions

Abstract

The World Health Organization (WHO) has recognized Particulate Matter (PM) as the main threat to human health from air pollution. One of the solutions is Green Infrastructure (GI), which uses different plants to mitigate pollution. Among these plants are bryophytes (or more commonly used mosses), which have easier maintenance, lighter weight, and durability compared to vascular plants. However, currently, there is limited knowledge of its effectiveness in air pollution mitigation. By addressing this gap in current scientific knowledge, more effective deployment of GI could be introduced by municipalities for society’s health benefits. This study aimed to evaluate three species of mosses (Dicranum scoparium, Plagiomnium affine, and Hypnum cupressiforme) and one thuja (Thuja plicata) as a control species for a possible GI vertical barrier for local de-pollution. The objective was to assess different moss species’ effectiveness in air pollution PM_2.5 and PM₁₀ absorption in a laboratory setting. The practical experiment was conducted from June–July 2024 in the Laboratory of the Physics and Chemistry of Environment and Space in Orleans (LPC2E-CNRS), France. For the experiment, a unique air pollution chamber was engineered and built with a linear barrier of GI inside to measure pollution absorption before and after the barrier. With the obtained data from the sensors, the efficiency of the vegetation barrier was calculated. The total average efficiency of all 18 tests and tested moss species is 41% for PM_2.5 and 47% for PM₁₀ mass concentrations. Efficiency shows moss species’ maximum or optimal ability to absorb pollution PM_2.5 and PM₁₀ in laboratory environments, with the limitations indicated in this article. This research is an essential step towards further and more profound research on the effectiveness of GI barriers of mosses in urban environments. It significantly contributes to understanding GI effects on air pollution and presents the results for specific moss species and their capacity for PM_2.5 and PM₁₀ mitigation in the air. The novelty of the study lies in a particular application of the chosen moss species.

1. Introduction

Air pollution as a threat to human health [1] has been recognized since the time of Hippocrates around 400 BCE [2] and is currently a global issue [3,4], resulting in more than 8 million deaths per year [5]. A major pollutant [6,7] is Particulate Matter (PM), which can be divisible by its size, for example, PM₁₀ are inhalable particles with a diameter that is generally 10 μm and smaller [1], but the most hazardous is PM_2.5, fine dust particles typically 2.5 µm in diameter or smaller, according to ref. [8,9,10] that are not visible to the naked eye [11]. The chemical composition of PM is complex [12,13]—these are tiny particles consisting of both liquid droplets and solid particles composed of carbonaceous materials, metals, mineral dust, and salts [9,14,15]. PM sources are both natural and anthropogenic [16,17]. Man-made PM sources are combustion in mechanical and industrial processes, vehicle emissions, tobacco smoke [18,19,20,21], and wood burning for heating, which is a significant PM source [22,23,24,25,26,27] and is epidemiologically linked to various adverse effects on health and mortality [28,29,30,31]. In some countries, wood and coal stoves are the largest source of the pollutant PM_2.5. This type of pollution consists of tiny particles absorbed deep into human bodies, including the lungs and blood. It has been recognized by the World Health Organization (WHO) as the most severe air pollutant for human health [32].

The biggest issue globally occurs in cities where outdoor air quality in most areas does not meet the WHO’s guidelines for healthy living [33]. Although reducing pollutant emissions is always the most direct way to improve urban air quality [34,35,36], authorities worldwide have struggled to deliver adequate improvements in air quality through emission control strategies alone. Thus, policymakers increasingly turn to additional methods to reduce human exposure to air pollutants, but as cities expand, the number of motor vehicles increases, and the distance traveled enlarges [37,38]. The presence of diesel vehicles, many of which do not comply with emission regulations, becomes a significant additional disadvantage [39,40].

Although air pollution is a global problem, solutions can be found in an urban, local context. For instance, an efficient solution is based on nature—using different plants for air purification with phytoremediation [41,42,43], green infrastructure (GI), and urban greening methods [44,45]. GI is the network of natural environmental components and green spaces in cities that provide ecosystem services: parks, street trees, gardens, and urban woodland [46,47,48]. Implementing GI in cities can mitigate exposure to air pollution [49,50], as GI can act as a porous barrier [49,51,52], altering local air dispersion patterns and providing a relatively high surface area for pollutant deposition [49,52,53]. Particularly linear barriers—such as hedges or fences—between source and receptor zones [Figure 1] redirect the air pollutants upwards, effectively extending the length of the air pathway from source to receptor. They may also promote concentration reduction by increasing turbulence [46].

Thus, hedges and fences can decrease air pollution concentrations in sidewalks and other pedestrian areas adjacent to traffic [55,56,57,58].

However, selecting plants for GI is challenging due to the lack of information on species resilience to heat and water fluctuations and their ability to provide various benefits [59,60,61], like air pollution mitigation [59,62]. Bryophytes (or mosses, more commonly used) can endure challenging environments, and have easier maintenance, lighter weight, and durability compared to vascular plants [63]; therefore, they might be used for air mitigation purposes, but there is limited data on their efficiency yet. Research has shown that utilizing mosses [63,64,65] in the GI can enhance ecosystem benefits [66,67]. Some preliminary studies of moss utilization within GI have shown effective rainwater management, the ability to reduce surface temperature [63], and the ability to trap atmospheric pollutants [67,68,69]. This feature is facilitated by mosses having rhizoids instead of roots [70], which was addressed in a study by Haynes et al. 2019 [71]. The study was conducted by analyzing three highways with high pollution levels, three suburban roads with moderate pollution levels, and three quiet suburban roads with low pollution levels, which were tested to compare the effects of moss and tree leaves on PM levels. Researchers found that when the site was more polluted, the moss cleaned up exponentially higher traces of PM, while the amount in tree leaves did not change as much. Overall, in this experiment, moss removed 3.8 times more PM than a tree leaf of the same dry weight, with higher efficiency of PM size 10–100 µm and lower for PM 0.2–2.5 [71]. On the contrary, another study in Italy in 2021 also evaluated the capacity of three moss species—Barbula unguiculata, Grimmia pulvinata, and Homalothecium sericeum to mitigate urban air pollution. After a three-month exposure in a high-traffic urban setting, samples were examined using Field Emission Scanning Electron Microscopy (SEM) to quantify particle deposition, and results revealed that the mosses collected up to 45,580 particles per mm². Approximately 45–55% of captured particles were in the PM_0.5 μm range, underscoring the potential of these moss species to retain fine particles effectively within urban environments [67]. The study suggests that most of the found particles on the surface are in diameters between 2.5 and 10 μm, rarely more significant than 25 μm [67]. Mosses might be effective in pollution mitigation, but there is a gap in knowledge on the effectiveness, specifically of vertical barrier placement of mosses, and both laboratory and field studies must be conducted. This article represents the results of a laboratory experiment on the effectiveness of mosses on air pollution mitigation and is part of broader research on mosses for urban application of pollution removal.

Wind direction also must be taken into account for GI, as it significantly impacts the dispersion of PM_2.5 [72]. Prevailing winds can enhance and diminish air pollutants depending on wind strength and duration [73]. Thus, intense and sustained prevailing winds can cleanse the air of contaminants [74]. Still, conversely, they can also transport air pollutants from upwind regions, worsening air quality in the direction of the wind [75,76]. It is also important to determine other variables for the GI linear barrier placement location, such as average wind speed, humidity, atmospheric pressure, cardinal directions, etc. Therefore, measurements at the location must be taken before barrier placement, and later, PM_2.5 concentration differences between the traffic area and the area next to the barrier must be taken and evaluated. Valid air quality sensors must be used to obtain air quality improvement results [77].

This research rationale, with a practical experiment, is based on the literature analysis and results of existing data on the topic. Models may often and significantly underestimate the contribution of GI to air quality, thus limiting their reliability for application in urban management [78]. The objective of this study was to assess different moss species for their effectiveness of air pollution PM_2.5 and PM₁₀ absorption in a laboratory setting; therefore, a novel laboratory experiment design was created and tested. The study aimed to answer the research question: Can mosses absorb air pollution PM_2.5 and PM₁₀ in a laboratory setting, and what is the average efficiency of moss species Dicranum scoparium, Plagiomnium affine, and Hypnum cupressiforme in a vertical position? This research provides a clear contribution to further research on GI barriers of mosses in an urban environment and serves as preliminary knowledge for further field research.

2. Materials and Methods

The practical experiment was conducted from June to July 2024 in the Laboratory of the Physics and Chemistry of Environment and Space in Orleans (LPC2E-CNRS), France. There were 3 bryophyte species tested in a special experimental chamber.

2.1. Selection of Experimental Material

The following bryophyte species—Dicranum scoparium, Plagiomnium affine. and Hypnum cupressiforme (Figure 2)—were chosen based on these specifications:

• According to the literature review [79,80,81,82,83,84], selected moss species were identified as most effective in absorbing urban air pollution;
• Species are cosmopolitan, growing in different parts of the world, commonly widespread in Europe’s temperate climate zone and urban environments;
• Species can sustain different environmental conditions, including hot sun and shading areas, a wide range of temperature fluctuations, and have a high capacity to absorb water.

Selected moss species were purchased from a certified moss cultivation company in France (“The Bryophyta Nursery”) to ensure the project’s sustainability and the moss species’ precision. Each moss sample was 4680 cm² in size and was stored under appropriate conditions in the laboratory, where the environment was controlled to prevent the moss from drying out before the experiment. Moreover, Thuja plicata was collected in Orleans city from a hedge and chosen as a control species for mosses, as thujas are typical for urban hedges in Europe.

2.2. Laboratory Design, Equipment Properties, and Validation

For the experiment, a special air pollution chamber (Figure 3b) was engineered and built with a size of 340 × 65 × 72 cm. The chamber was built from different materials: metal for the main construction (3 × 3 cm); transparent polyethylene film was used for the walls and roof (width of 3 m and a thickness of 50 mkr), and it was rolled over the metal construction 5 times to ensure durability. Adhesive tape was used all over the chamber connection places to ensure it was hermetically sealed. Inside the chamber, a linear barrier of vegetation was displayed, with a size of 65 × 72 cm. Vegetation was placed between the metal mesh (Figure 3a), also referred to as a tile. One tile was prepared for each species and tested repeatedly for all tests (6 tests for each species, 18 in total).

The experimental chamber (Figure 4) had a fan to create traction with 2 levels of speed (lower speed created traction speed of 0.5 m s⁻¹, higher speed created traction speed of 1.2 m s⁻¹), and pollution was injected from the other side of the chamber using a manual nebulizer. During the experiment, different types of pollution were injected: Icelandic volcanic dust, granite, and concrete, for each test, 1 g of pollution. Pollution was injected directly towards the vegetation tile (distance between 200 cm). To clean the chamber after each test, an extractor was used, which cleaned the chamber with the airflow. Before the beginning of each test, the background concentration was measured using both Pollutrack sensors, and tests started only when the concentrations in both sensors for either PM_2.5 or PM₁₀ were less than 5 µm (mass concentration). To obtain the data, atmospheric pressure daily data were taken directly from Meteo France’s daily web, and for other measurements, several instruments were used: a wind speed measurement device (Anemometer (manufactured by XRCLIF, China) certified by CE, RoHS, and FC), humidity and temperature station (ThermPro TP358-GB, manufactured by ThermoPro, Toronto, Canada), two calibrated PM_2.5 and PM₁₀ measurement sensors Pollutrack (manufactured by company Pollutrack, Bretigny-sur-Orge, France) [85], one installed before vegetation, another after. Both fractions, PM_2.5 and PM₁₀ mass concentrations, were measured by Pollutrack sensors in parallel at the same session. The Pollutrack sensor is a compact optical particle counter that measures PM concentrations by analyzing light scattered from particles crossing a laser beam. It detects particle sizes from 0.5 to 10 µm and converts counts into PM_2.5 and PM₁₀ mass concentrations using internal calibrations [86]. Measurements from sensors were also taken once every 10 s. Each test was conducted for 20 min: (a) background concentration in the chamber was measured for the first 10 min; (b) pollution injection was performed; (c) to measure changes in background concentration, measurements were taken for 9–10 more minutes. The key target of these tests was to obtain data on different moss species absorption effectiveness of PM_2.5 and PM₁₀.

To calculate the efficiency of the vegetation barrier, the following equation was used:

$$E=100\times {\displaystyle \frac{{\mathrm{M}}_{\mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e}}-{\mathrm{M}}_{\mathrm{a}\mathrm{f}\mathrm{t}\mathrm{e}\mathrm{r}}}{{\mathrm{M}}_{\mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e}}}}$$

E = efficiency;

M_before—measurements before the injection;

M_after—measurements after the injection.

To determine total efficiency, it is valuable to analyze the efficiency during the injection and calculate the average injection efficiency for each species at different wind speeds. This experiment had limitations such as temperature, humidity, and atmospheric pressure in laboratory conditions. Each parameter was measured, and all data on the limitations are presented in

Table 1. Experiment limitations.

Wind Speed (WS)	Humidity Inside the Chamber	Temperature Inside the Chamber	Atmospheric Pressure
Two WS tested: 0.5 m s−1 and 1.2 m s−1	78–93% (mostly around 82–87%)	20.2–21.2 °C	1012 to 1015 Pa

. The limitation of vegetation samples was considered, as there was one sample for each of the vegetation types after each test chamber cleaning with the extractor fan was conducted.

After efficiency data were calculated, statistical analysis of variance (ANOVA) was conducted to assess the difference between the means of these three test groups (each test group represents each moss species). For each moss species, a separate group was created, and four variables to this group were added (average efficiency for wind speed 0.5 m s⁻¹ for PM_2.5, wind speed 1.2 m s⁻¹ for PM_2.5, wind speed 0.5 m s⁻¹ for PM₁₀, wind speed 1.2 m s⁻¹ for PM₁₀). Further, each of the numbers in each group (x) was squared (x²), and then the Data Summary table was created using the following formulas:

∑x = x1 + x2 + x3 + x4

(1)

$$\mathrm{M}\mathrm{e}\mathrm{a}\mathrm{n}={\displaystyle \frac{\mathsf{\Sigma }}{\mathrm{n}}}$$

(2)

∑ = the total of all values in the group;

N = number of observations (in this case, 4).

∑x² = x₁² + x₂² + x₃² + x₄²

(3)

$$\mathrm{Std}.\mathrm{Dev}. \mathrm{or} \mathrm{s}=\surd {\displaystyle \frac{{\mathsf{\Sigma }(x}_i- {\left.\overline{x}\right)}^2}{n-1}}$$

(4)

x_i = each value;

$\overline{x}$ = mean of the group;

n = number of values in the group (4).

As a last step, the ANOVA summary table was filled in by following calculations with a target to calculate F and p values. It is important to obtain these values to determine how much the group means differ from each other relative to how much the values vary within each group (F value) and to calculate the probability that the differences between group means happened by chance (p value).

$$\mathrm{F} \mathrm{value} \mathrm{F}={\displaystyle \frac{{\mathrm{M}\mathrm{S}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}}{{\mathrm{M}\mathrm{S}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}}}$$

$${\mathrm{M}\mathrm{S}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}={\displaystyle \frac{{\mathrm{S}\mathrm{S}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}}{{\mathrm{d}\mathrm{f}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}}}$$

$${\mathrm{M}\mathrm{S}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}={\displaystyle \frac{{\mathrm{S}\mathrm{S}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}}{{\mathrm{d}\mathrm{f}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}}}$$

SS = sum of squares;

Df = degrees of freedom;

MS = mean square (variance estimate).

To obtain all the needed variables for F value calculations, it is necessary to obtain DF (degrees of freedom) and SS (Sum of squares) using calculations, that are standard for ANOVA analysis.

A validation session was conducted using the Light Optical Aerosol Counter (LOAC) instrument to obtain validation of the measurements taken by the Pollutrack sensors during this experiment. LOAC is an instrument used for scientific purposes, and it is involved in several atmospheric research programs for the determination of the concentration and properties of tropospheric and stratospheric aerosols [87]. This aerosol counter uses a laser beam and two scattering angles, 12°, and 60°, respectively, which allow the determination of aerosol concentration (number of aerosols per cm³) in 19 size classes from 0.2 to 30 µm, as well as identify the typology of the aerosols according to five leading particle families (Carbons, Droplets, Ice, Salt, and Minerals).

Overall, the selected materials and methods proved effective in collecting data for analysis and interpretation of results.

3. Results

Results are depicted in graphs (Figure 5) where three different color lines show different PM_2.5 efficiency after injection: Efficiency AB shows the first 10 s of injection, followed by Efficiency CD with the next 10 s of injections (second 20), and Efficiency EF shows the last 10 s (second 30). PM₁₀ injection was performed in the same manner as Efficiency GH in the first 10 s of injection, Efficiency JK in the second 10 s, and Efficiency LM in the last 30 s after the start of injection. After the first 30 s, the pollution level drops significantly to a level close to the initial amount of pollution. Each of the six tests on each moss species shows slightly different efficiency, although there is a correlation between all tests and the timing after pollution injection. For example, the graph shows that in the first 10 s (Efficiency AB), the average efficiency of the Dicranum scoparium moss species linear barrier is 33% (Figure 5a).

In this round of tests (Figure 5), two wind speeds were tested: 0.5 m s⁻¹ and 1.2 m s⁻¹, which show different efficiency results.

Table 2. Total average efficiency for both sizes of PM, different wind speeds, and different vegetation.

		PM2.5			PM10
Moss Species	Number of Tests	Average Efficiency (WS 0.5 m s−1)	Average Efficiency (WS 1.2 m s−1)	Average Efficiency	Average Efficiency (WS 0.5 m s−1)	Average Efficiency (WS 1.2 m s−1)	Average Efficiency
D. scoparium	6	36.82	35.54	36.18	43.30	42.99	43.15
P. affine	6	50.06	33.19	41.63	55.73	47.16	51.45
H. cupressiforme	6	42.58	48.06	45.32	39.28	52.58	45.93
3 moss species	18	43.15	38.93	41.04	46.10	47.58	46.84
T. plicata	4	No data	−20.34	−20.34	No data	−34.57	−34.57

shows all the data on the limitations. The moss species D. scoparium showed higher PM₁₀ absorption efficiency compared to PM_2.5. The average absorption efficiency for PM₁₀ was 43% against the average efficiency for PM_2.5 of 36%, indicating that the moss species D. scoparium better absorbs larger particles. Similarly, another moss species showed better results for the efficiency of larger particles. For example, during the tests on the moss species Plagiomnium affine PM_2.5 absorption, correlations were found between the tests, showing an average overall efficiency of 46.5% (Figure 5c). At the same time, P. affine PM₁₀ absorption efficiency was found to be even higher, with similar results between all three tests with a wind speed of 0.5 m s⁻¹, showing a significant result of an average of 86% in efficiency AB (Figure 5d). Furthermore, the moss species Hypnum cupressiforme showed the highest efficiency (45%), among all three tested moss species during the seven tests with PM_2.5 pollution (Figure 5e). However, there were different patterns across the tests, which are explained in the Discussion paragraph. Data on all three moss species tests are available in the Supplementary Material. As a control species against mosses to assess differences (Figure 5g,h), the coniferous tree species Thuja plicata was tested. The same experimental protocol was applied, and each test with thuja showed the same result, with no absorption capacity and additional pollution detected in the second sensor. Thus, indicating that T. plicata has no efficiency in pollution absorption. T. plicata obtained a negative result (−20% for PM_2.5 and −35% for PM₁₀), which means that the existing pollution was displaced due to the traction force created by the fan (wind speed 1.2 m s⁻¹). This leads to the conclusion that T. plicata does not absorb any pollution in the laboratory environment, and the air is even more polluted on the other side of the barrier because it is closer to the fan that created the traction. Further research is needed to analyze the pollution absorption of T. plicata in the outdoor environment. Still, this study shows that T. plicata cannot remove PM_2.5 and/or PM₁₀ pollution from the air. The average efficiency per two wind speeds for separately PM_2.5 and PM₁₀ and all tested vegetation is shown in Table 2. The table depicts that the most effective is the moss species P. affine for PM₁₀ in wind speed 0.5 m s⁻¹, followed by H. cupressiforme in 1.2 m s⁻¹.

Analysis of variance (ANOVA) statistical test was used to assess the difference between the means of three tested moss species and their effectiveness in PM_2.5 and PM₁₀ absorption (

Table 3. Three species efficiency values for both PM_2.5 and PM₁₀ absorption effectiveness, ANOVA test.

Group 1 (D. scoparium)	Group 2 (P. affine)	Group 3 (H. cupressiforme)
37	50	43
36	33	48
43	56	39
43	47	53

).

Table 3 shows the average efficiency of each moss species for PM_2.5 and PM₁₀ absorption, and

Table 4. Groups (3 species) square summary, ANOVA test.

Group 12	Group 22	Group 32
1369	2500	1849
1296	1089	2304
1849	3136	1521
1849	2209	2809

depicts the groups (moss species) square summary based on the ANOVA test.

Table 5. Data summary, ANOVA test.

Groups	N	∑x	Mean	∑x2	Std.Dev.
Group 1	4	159	39.75	6363	3.7749
Group 2	4	186	46.5	8934	9.7468
Group 3	4	183	45.75	8483	6.0759
Total	12	528	44	23,780

shows a data summary based on an ANOVA test, which consists of such values as: N (sample Size) represents the number of observations (data points) in each group (species). Each group has 4 observations, and the total number of observations is 12; Σx (sum of values) shows the sum of all individual values in each group. The mean represents the mean (average) efficiency for each moss species. Σx² (sum of squared values) illustrates the sum of the squares of individual values in each group and Std.Dev. stands for standard deviation, this value measures the dispersion or spread of values in each group. Standard deviation depicts how dispersed the data are against the mean.

From the data summary (Table 5), it is visible that Group 2 (P. affine) has the highest mean (45.5) and the highest standard deviation (9.7), indicating more variation in values. Group 1 (D. scoparium) has the lowest mean (40) but the lowest standard deviation. The total mean (44) gives the average across all groups combined. ANOVA summary is shown in

Table 6. ANOVA test summary.

Source	DF	SS	MS	F-Stat	p-Value
Between Groups	2	109.5	54.75	1.12	0.36665
Within Groups	9	438.5	48.72
Total	11	548

and it depicts five values: DF (degrees of freedom) represents the number of independent values; SS (Sum of Squares) represents the total variation; MS (Mean Square) shows the average variance for each source, calculated between and within groups; F-Stat (F-Ratio) measures whether the between-group variability is significantly larger than the within-group variability and p-value shows the probability of obtaining the observed results assuming the null hypothesis (no difference between groups) is true.

In the ANOVA test (Table 6), a larger F-value indicates a more significant difference among the group means and suggests that the group variations are significant. Since the p-value (0.37) is much higher than 0.05, there is no significant difference between the group means. At the same time, if the p-value is less than the chosen significance level (set at 0.05), it indicates statistically significant differences among the groups, which gives evidence to reject the null hypothesis, which assumes no significant differences. Since p > 0.05, the result is not statistically significant, meaning there is no strong evidence that moss species differ in their effectiveness in absorbing PM_2.5 and PM_10. Therefore, the authors focus on all three moss species’ average efficiency.

4. Discussion

Based on the results, it is possible to elaborate on the efficiency or the optimal ability of mosses to absorb pollution PM_2.5 and PM₁₀ in laboratory environments with the limitations indicated above in the article (such as humidity, atmospheric pressure, temperature) and under conditions when the pollution is injected directly towards the moss tile.

Different effectiveness across tests could be explained by the different forces of injection for each test. As injection was performed manually (using a hand nebulizer), speed and force did differ, but traction from the fan equated to the force. To validate the results obtained from Pollutrack sensors, another session was organized using the LOAC tool.

LOAC Instrument Testing

The same experiment protocol was used to obtain results with the LOAC instrument, and LOAC sessions were carried out in parallel with a Pollutrack experiment. As the mean values of PM_2.5 mass concentrations may change from the first to the second session, the LOAC values had to be adjusted for direct comparison. LOAC values for the second session are multiplied by the total value of the Pollutrack sensor in the first session before the moss wall and divided by the total value of the Pollutrack sensor in the second session before the moss wall. Average de-pollution efficiency with Pollutrack sensors in the first session: 4% considering all values and 10% considering values above 20 µg/cm³ (thus during dust injections). Average de-pollution efficiency with Pollutrack sensors in the second session: 15% considering all values and 35% considering measurements above 20 µg/cm³ (thus during injections). At the same time, the average de-pollution efficiency with LOAC sensors was 13% considering all values and 31% considering measurements above 20 µg/cm³ (thus, during injections). These values are close to Pollutrack results, providing a validation of the experiment, as well as LOAC sessions depict a broader view of the pollution that vegetation barriers absorb. For example, the figure below (Figure 6) shows the size distribution of the LOAC measurement. Squares represent the error bars. The coarse code of the sample (“Blanc de Meudon”, composed of CaCO₃) is visible between 1 and 20 µm. This shows that the moss wall absorbs the largest and heaviest particles, increasing efficiency with increasing size (if the size distribution remains the same in the absence of the moss wall, which is a plausible hypothesis given the particle transport wind system used in the experiment).

During the experiment, it was found that the moss absorption of the pollution is prompt, and the efficiency increases with larger particles. This experiment shows that the selected moss species can de-pollute in closed conditions with stable traction under stable meteorological conditions.

The results from the three moss species tiles absorbing fine dust pollution PM_2.5 and PM₁₀ experiment are promising for scientific purposes and the practical application of mosses in indoor or outdoor air quality improvement. The lessons from this experiment could also validate existing technologies using these three moss species. Further research is needed on other moss species and a more in-depth analysis of how mosses absorb, store, and remove pollution. For the experiment, different wind speeds were used to test all three moss species, yet there is no clear correlation between different wind speeds and all three moss species for example, Dicranum scoparium showed similar results within different wind speeds: 37% average efficiency at 0.5 m s⁻¹ and 36% at 1.2 m s⁻¹ for PM_2.5 and similarly for PM₁₀ with 43% for either lower and higher wind speed. Therefore, the authors conclude that this moss species is equally efficient in both tested wind speeds (0.5 and 1.2 m s⁻¹). On the contrary, moss species Plagiomnium affine depict different patterns with higher efficiency (50%) at lower wind speeds for PM_2.5 and 56% for PM₁₀ versus significantly lower efficiency for PM_2.5 at higher wind speed (33%) and less significant for PM₁₀ (47%). For the moss species Hypnum cupressiforme, the opposite pattern versus Plagiomnium affine was detected with higher efficiency (48% versus 43%) for higher wind speed for PM_2.5 and 53% versus 39% for PM₁₀. The authors conclude that three different patterns across three different species were observed: D. scoparium showed similar results for both wind speeds, P. affine species were more effective at lower wind speeds, and H. cupressiforme was more efficient at lower wind speeds. The authors observed that there were differences between efficiency for different sizes of particles: all three moss species demonstrated higher efficiency for PM₁₀ absorption vs. PM_2.5, with the highest difference for P. affine, with almost a 10% difference (41.63% for PM_2.5 versus 51.45% for PM₁₀). D. scoparium showed a similar pattern with the highest efficiency for PM₁₀ (43.15%) vs. PM_2.5 (36.18%). On the other hand, moss species, H. cupressiforme, showed just a slight difference in efficiency, with 45.93% for PM₁₀ versus 45.32% for PM_2.5. From all tests that were conducted, the moss species Plagiomnium affine obtained the highest efficiency, at 56%, during lower wind and PM₁₀ absorption; this is the highest result in this experiment.

Overall, the authors observed slight differences in different moss species and their effectiveness, but as per ANOVA test results, these differences are insignificant to view each moss species separately; therefore, the authors conclude that these three tested moss species are effective on air pollution PM_2.5 and PM₁₀ absorption in laboratory settings under the described controlled conditions. As the objective of this study was to assess different moss species on their effectiveness of air pollution PM_2.5 and PM₁₀ absorption in a laboratory setting, a novel laboratory experiment design was created and tested. The study answered the proposed research question with the clear understanding that moss species Dicranum scoparium, Plagiomnium affine, and Hypnum cupressiforme in a vertical position have an average absorption efficiency for PM_2.5 pollution of 41%, and for PM₁₀, it is 47%. Further research on the same three moss species in experiments outside the controlled environment is needed to determine absorption efficiency under outdoor conditions.