Landscape and Urban Planning 167 (2017) 348–355

Contents lists available at ScienceDirect

Landscape and Urban Planning

journal homepage: www.elsevier.com/locate/landurbplan

Research Paper

Bryophytes as bioindicators of the atmospheric environment in urban-forest MARK

landscapes
⁎
Yoshitaka Oishia, , Tsutom Hiurab
a
Center for Arts and Sciences, Fukui Prefectural University, 4-1-1 Kenjojima, Matsuoka, Eiheiji-cho, Yoshida-gun, Fukui, 910-1195, Japan
b
Tomakomai Research Station, Field Science Center for Northern Biosphere, Hokkaido University, Tomakomai, Hokkaido, 053-0035, Japan

A R T I C L E I N F O A B S T R A C T

Keywords: Bryophytes have been used as indicators to evaluate atmospheric conditions in urban areas. However, further
Bioindicator research is needed for their effective application. In this study, we focused on four metrics related to atmospheric
Hygrophilous life-forms problems in urban areas (nitrogen concentration [% N], isotope ratio of nitrogen [δ15N], index of atmospheric
Index of atmospheric purity purity [IAP], and richness of hygrophilic life-forms [RHL]). Then, using linear and generalized linear models, we
Land use
examined the influence of land use on these four metrics in urban atmospheric environments, and evaluated the
Nitrogen pollution
usefulness and limitation of them. The % N and δ15N models were well explained by the influence of nitrogen
Urban heat island
sources and/or sinks. The RHL models were significantly affected by urban and forest areas, reflecting urban
heat islands. Surprisingly, IAP was higher in urban areas, but comparisons of IAP are not informative in areas
with narrow pollution gradients. Land use strongly affected % N and δ15N models in smaller areas because of a
point-source influence of nitrogen pollution, whereas RHL was strongly influenced by land use at larger scales
owing to drought stress in urban settings. Correlations among the metrics revealed that severe drought stress
tended to occur in areas with high nitrogen pollution. The nitrogen pollution sources were diverse, with no
significant correlation of δ15N values with % N.

1. Introduction vehicles). This increase in N input affects both species composition

In urban areas, changes in atmospheric environments have caused
serious problems for both human health and the ecosystem as a whole
(Breitner et al., 2009; Gurjar et al., 2010; Kopáček & Posch, 2011;
Organization for Economic Cooperation and Development, 2012;
Shibata et al., 2015; The World Meteorological Organization and the
International Global Atmospheric Chemistry, 2012). Generally, urban
development itself is accompanied by the loss of or decrease in green
areas, which often causes forest fragmentation. At the edges of these
fragmented forests, pronounced drought stress occurs because light
intensity and wind exposure increase (Murcia, 1995). These environ-
mental changes (edge effects) have a negative effect on biodiversity, as
drought-sensitive species disappear (Aragón, Abuja,
Belinchón, & Martínez, 2015; Gignac & Dale, 2005; Laurance et al.,
2006, 2007; Zartman, 2003) and alien species increase (Honnay,
Verheyen, & Hermy, 2002; Pauchard & Alaback, 2006). Bettez and
Groffman (2013) investigated the effects of urbanization on nitrogen
(N) deposition, and revealed that the increase in N deposition was a
result of increased amounts of dry deposition associated with stationary
sources (e.g., power plants) and mobile sources (e.g., highway
through changes in competitive biological interactions (Bobbink et al.,
2010) and ecological processes (Groffman & Pouyat, 2009; Thomas,
Canham, Weathers, & Goodale, 2010).
While environmental evaluation using physicochemical techniques
is more accurate and stable than evaluation methods using bioindica-
tors, the use of physical equipment is often costlier and time consuming
(Holt & Miller, 2010). In addition, bioindicators also have the following
advantages: first, they can convey the cumulative effects of both che-
mical pollutants and habitat alteration during the life span or residence
time of a given organism; second, they can reflect the indirect biotic
effects of pollutants; and finally, biotic indices are a more effective
means of ecosystem monitoring, as they reflect changes at the eco-
system level (Holt & Miller, 2010).
Bryophytes are one of the most popular taxonomic groups used as
bioindicators of atmospheric environments. They are nonvascular
plants that can grow on the surfaces of tree trunks or rocks, and gen-
erally absorb water and nutrients directly through leaf surfaces from the
immediate environment (Dymytrova, 2009; Harmens et al., 2011;
Onianwa, 2001; Pearson et al., 2000; Schröder et al., 2010;
Zechmeister, Dirnböck, Hülber, & Mirtl, 2007; Zechmeister et al.,

⁎
Corresponding author.
E-mail addresses: oishiy@fpu.ac.jp (Y. Oishi), hiura@fsc.hokudai.ac.jp (T. Hiura).

http://dx.doi.org/10.1016/j.landurbplan.2017.07.010
Received 8 November 2016; Received in revised form 13 July 2017; Accepted 16 July 2017
0169-2046/ © 2017 Elsevier B.V. All rights reserved.
Y. Oishi, T. Hiura
2008). Owing to these properties, bryophytes have been used to eval-
uate atmospheric pollution in many parts of the world (e.g., Agnan,
Séjalon-Delmas, Claustres, & Probst, 2015; Boquete, Fernández, Aboal,
Real, & Carballeira, 2009; Fernández, Aboal, Real, & Carballeira, 2007;
Oishi, 2013; Schröder et al., 2010; Vuković et al., 2015).
Among various types of atmospheric pollution, N pollution has re-
ceived a lot of attention in recent years because of its increasing threat
to environments at a global scale (e.g., Kopáček & Posch, 2011; Shibata
et al., 2015). The evaluation of N pollution by bryophyte indicators is
based on the strong correlation of N content (% N) in bryophytes with
atmospheric N concentration (Harmens et al., 2011; Schröder et al.,
2010). Furthermore, the stable isotope ratio of N in bryophytes (δ15N)
has been used to diagnose sources of nitrogen, as the signature of N in
nitrogen oxides (NOx) is higher than that of reduced nitrogen (NHy)
(Liu, Xiao, Liu, & Xiao, 2008; Pearson et al., 2000; Zechmeister et al.,
2008). The response of δ15N to nitrogen sources is more complicated
than that of % N; while this value was lower near NHy sources (e.g.,
farms; Liu et al., 2008), higher values were observed near NOx sources
(e.g., motor way; Pearson et al., 2000).
Not only the chemical properties of bryophytes but also the di-
versity (species richness, cover, and life-forms) has been used to eval-
uate atmospheric environments. The index of atmospheric purity (IAP)
is one of the most popular indicators of bryophyte diversity. This in-
dicator was first proposed in the 1970s and is calculated from the
species richness and cover of epiphytic bryophytes and lichens
(LeBlanc & De Sloover, 1970). Previous studies reported that lower IAP
values were recorded in urban areas because epiphytes are generally
vulnerable to atmospheric pollution (e.g., Krommer, Zechmeister,
Roder, Scharf, & Hanus-Illnar, 2007; Taoda, 1972). Recently, bryophyte
life-forms have also been utilized as indicators for the evaluation of
environments (Oishi, 2009; Oishi & Morimoto, 2016; Pardow, Gehrig-
Downie, Gradstein, & Lakatos, 2012; Vieira, Séneca, Sérgio, & Ferreira,
2012). The use of bryophyte life-forms as indicators is based on the
close relationships between life-forms and the surrounding environ-
ments. For example, in fragmented forests with severe drought stress,
the richness of hygrophilous life-forms (RHL) could reflect overall hu-
midity, providing an estimate for total species richness
(Oishi & Morimoto, 2016).
As shown above, bryophytes have been reported to be useful in-
dicators for the evaluation of atmospheric environments in urban areas.
However, further study is required to fully understand their utility as
indicators. Researchers do not yet know how effective the combina-
tional use of bryophyte features is for understanding atmospheric en-
vironments in urban areas, or what limitations these features have in
349
Landscape and Urban Planning 167 (2017) 348–355
Fig. 1. The study site is located in Hachioji-shi, Tokyo, Japan, and sam-
pling plots were established in each 3 × 3 km grid cell.
practical applications. Until now, few studies have adopted a combi-
national use of bryophyte features for the evaluation of urban en-
vironments (e.g., Giordano et al., 2004; Krommer et al., 2007). These
studies have examined urban environments from diverse viewpoints;
thus, the effectiveness of such combinational use warrants further
study. When examining the usefulness of such parameters, however, we
have to keep in mind that the sensitivity of bryophyte features to tar-
geted environments is expected to decrease with increasing distance
from sampling points. This limitation has been implied by several stu-
dies (e.g., Bignal, Ashmore, & Headley, 2008; Pitcairn et al., 2003;
Skinner et al., 2006). Therefore, an examination of the influenced area
(zone of influence) is essential to correctly interpret the environment
evaluated by bioindicators.
In this study, focusing on the possible strong correlation of bryo-
phyte features with land use types (e.g., urban, agricultural, and in-
dustrial areas), we first evaluated the atmospheric environments using
bryophyte features, and then analyzed the influence of land use on
them. Based on these results, we examined the usefulness of their
combinational application and determined the zone of influence in
urban-forest landscapes.
2. Materials and methods
2.1. Study site and plots
The study site was located in Hachioji City in the northwestern part
of Tokyo, Japan (Fig. 1). This city is part of the Tokyo capital region
and has a population of nearly 0.6 million people. The altitude ranges
from 63.0 to 862.7 m. The climate is characterized by large differences
in temperature between summer and winter owing to its inland basin
location. The annual mean temperature is ca. 14.4 °C, with the highest
temperature in August (26.1 °C) and the lowest in January (3.2 °C)
(averages from 1981 to 2010) (Japan Meteorological Agency, 2016).
Hachioji is roughly divided into two parts. The eastern part has been
developed as a city matrix, while the western part is a mountainous
areas. A highway crosses the middle of the city from east to west. This
city is ideal for examination of the practical use of bryophyte metrics, as
it contains variable land-use types from urbanized to mountainous
areas.
We established a 3 × 3 km grid across the city and sampled one or
two sites in each grid cell, except in inaccessible areas. Previous studies
proposed that grid sampling (sampling sites are established at the
vertices of the grid) and random sampling (sampling points are estab-
lished randomly in a particular size of grid) are the most effective
Y. Oishi, T. Hiura Landscape and Urban Planning 167 (2017) 348–355

40 Fig. 2. The comparison of the species richness and cover of epiphytic

16 bryophytes among sampling plots.
The bars in the graph show standard deviation. No significant differences
were found among the three categories for species richness and cover.
14 30
Mountain = sampling plots in mountainous areas, Suburb = sampling

% cover
plots in suburban areas, Urban = sampling plots in urban areas.

12
20

10
10

Mountain Suburb Urban Mountain Suburb Urban

Table 1 2.2. Tree and bryophyte surveys

Pearson product-moment correlation coefficients among bryophyte metrics and the spe-
cies richness and cover. Our study plots had a high diversity of tree species, so we targeted
%N δ N
15
IAP RHL SR Cover
the epiphytes growing on the following tree species: Acer palmatum
Thunb., Carpinus laxiflora (Siebold & Zucc.) Blume, Celtis sinensis Pers.
%N 1 −0.087 0.146 −0.534 −0.358 0.234 var. japonica Nakai, Ginkgo biloba L., Cerasus spp. (C. jamasakura
δ15N 1 0.142 0.203 0.213 0.086 (Siebold ex Koidz.) H.Ohba, C. leveilleana (Koehne) H.Ohba,
IAP 1 0.040 0.359
C. × yedoensis (Matsum.) A.V.Vassil.), Quercus acutissima Carruth., and
RHL 1 −0.100
SR 1 0.169 Zelkova serrata (Thunb.) Makino. Taoda (1972) used these and other
Cover 1 tree species to calculate IAP, demonstrating their utility for the calcu-
lation. The surveyed trees were selected based on the following rules: 1)
1
Underlined items indicate significance at p < 0.05; double underlining indicates sig- trees with epiphytic bryophytes were chosen at the approximate center
nificance at p < 0.01 % N = nitrogen concentration, δ15N = stable isotope ratio of ni- of each plot, at least 50–100 m away from main roads or buildings, to
trogen, IAP = index of atmospheric purity, RHL = richness of hygrophilous life-form
avoid the influence of adjacent urban environments; 2) two to three
species, SR = species richness, Cover = total cover of bryophytes.
deciduous species were surveyed in each plot to reduce the influence of
differences between host species; and 3) upright trees without lower
branches were selected for the samples. We sampled the tree trunks
methods for biomonitoring of atmospheric pollution (Fernández, Real, 0.5–1.5 m above the ground and recorded the occurrence and percent
Couto, Aboal, & Carballeira, 2005). Boquete et al. (2009) also suggested cover of epiphytic bryophytes. All collected bryophytes were preserved
that different grid sizes could be used simultaneously to evaluate at- and returned to the laboratory for identification. The nomenclature
mospheric pollution at a variety of spatial scales. However, we adapted followed that proposed by Iwatsuki (2000). The bryophyte cover of
the random sampling design in this study because the grid sampling each species was recorded in accordance with the Braun-Blanquet scale:
technique was difficult to apply with the limited amounts of green 0.1 (< 1% cover; median 0.5), 1 (1–10% cover; median 5), 2 (11–25%
space in our study site. cover; median 17.5), 3 (26–50% cover; median 37.5), 4 (51–75% cover;
According to the random sampling design, the established study median 62.5), and 5 (76–100% cover; median 87.5). In addition, we
plots were classified into three categories: urban (11 plots), suburban (7 sampled Pylaisiadelpha tenuirostris (Bruch & Schimp.) W.R.Buck (ca.
plots), and mountainous areas (4 plots). The urban category is defined 0.1 g in dry weight) for the N analysis. This ectohydric moss species was
as sampling sites located close to the city center and surrounded by city abundant in all study plots. The samples for N analysis were tightly
matrix. The suburban areas are characterized by rich green space sealed in plastic bags and refrigerated until analysis.
around the sampling plots (more than 50% green area within 1-km
buffer around the sampling plots). The mountainous areas are within 2.3. Analyses of physicochemical features and diversity metrics
mountainous regions (more than 90% forest cover within a 1-km buffer
around the sampling plots). We examined the following four bryophyte metrics: % N content,
The vegetation types varied greatly between urban/suburban and δ15N, IAP, and RHL. Bryophyte samples (P. tenuirostris) were air-dried
mountainous areas. Ornamental trees (e.g., Cerasus jamasakura (Siebold for 48 h and preserved at room temperature until the analysis.
ex Koidz.) H.Ohba) or secondary forest trees (e.g., Quercus serrata Subsequently, % N and δ15N in these samples were analyzed using an
Murray) dominated in urban and suburban areas, while the mountai- elemental analyzer with an isotope ratio mass spectrometer (EA-IRMS,
nous areas included large patches of natural forests (e.g., Fagus crenata Flash EA 1112-DELTA V PLUS ConFlo III System; Thermo Fisher
Blume) and conifer plantations (e.g., Cryptomeria japonica (Thunb. ex Scientific, Inc., Waltham, MA, USA). The IAP was calculated based on
L.f.) D.Don). the methods of Asta and Rolley (1999):
In each plot, we calculated the various types of land-use area and n Qi n aij × bij
the total length of the roads within seven zones of influence (50, 100, IAP = Σ Σ
200, 500, 1000, 1500, and 2000 m) from the sampling points. Land use
i 10 i m
was classified into five types: forest, grassland, agriculture, urban (e.g., where n = number of species, and Qi = ecological index of each spe-
residential/commercial areas), and industrial areas. The road types cies. This index shows the number of accompanying species for species
were grouped into three categories: highway, multiple-lane road, and i, aij is the coverage class of species i, bij is the occurrence of species i
single-lane road. with the coverage of class j, and m indicates the number of coverage
classes of species i. The coverage classes in this study were based on

350
Y. Oishi, T. Hiura Landscape and Urban Planning 167 (2017) 348–355

Table 2
Results of linear model (LM) and generalized linear model (GLM) analyses for the correlation between bryophyte metrics and environmental variables.
Coefficients of LMs and GLMs and the coefficients of determination are presented.

% Na Environmental Variables Statistics

Z.I. Forest Agri. Urban Ind. Multiple AIC R2 p

−4
50 m 2.19 × 10 −19.6 0.149 0.043
100 m 8.92 × 10−5 −24.7 0.257 0.011
−5 −5 −6
200 m −1.74 × 10 −1.35 × 10 7.59 × 10 −22.7 0.297 0.025
500 m 8.88 × 10−5 −21.9 0.207 0.022
1000 m 3.05 × 10−7 −21.7 0.197 0.025
1500 m 3.17 × 10−5 −21.6 0.195 0.026
2000 m 1.92 × 10−5 −21.3 0.182 0.031

δ15N

I.Z. Grass Urban Ind. H.W. AIC R2 p

50 m −4.16 × 10−4 17.9 0.088 0.103

100 m 4.74 × 10−2 14.8 0.214 0.021
200 m 4.28 × 10−5 1.12 × 10−2 13.1 0.303 0.015
500 m 2.33 × 10−3 15.4 0.112 0.071
1000 m 7.74 × 10−4 16.9 0.130 0.060
1500 m 1.915 × 10−4 20.0 0.00 0.377
2000 m −5.95 × 10−7 20.2 0.00 0.438

IAP

I.Z. Forest Grass Agri. Single AIC R2 p

−4 −4
50 m −1.40 × 10 −3.23 × 10 −13.5 0.290 0.015
100 m −3.09 × 10−5 −8.47 × 10−5 2.28 × 10−3 −18.2 0.447 < 0.010
200 m −20.0 0.471 < 0.010
500 m −13.1 0.249 0.011
1000 m −15.0 0.309 < 0.010
1500 m 1.60 × 10−6 −12.7 0.233 0.013
2000 m 1.04 × 10−6 −11.2 0.179 0.028

RHL

I.Z. Forest Urban Single AIC pseudo R2 p

−4
50 m 2.43 × 10 38.9 0.008 0.129
100 m 9.70 × 10−5 37.3 0.050 0.049
−3
200 m −3.75 × 10 34.3 0.131 < 0.01
500 m 1.77 × 10−5 2.57 × 10−5 −1.12 × 10−3 33.7 0.146 < 0.01
1000 m 1.05 × 10−6 32.2 0.186 < 0.01
1500 m −6.53 × 10−7 31.7 0.200 < 0.01
2000 m 31.3 0.212 < 0.01

a
Underlined items indicate significance at p < 0.05; double underlining indicates significance at p < 0.01 % N = nitrogen concentration, δ15N = stable isotope ratio of nitrogen,
IAP = index of atmospheric purity, RHL = richness of hygrophilous life-forms, Z.I. = zone of influence, Agri. = agricultural area, Ind. = industrial area, H.W. = Highway,
Multiple = multiple-lane road, Single = single-lane road, R2 = adjusted R2.

those recorded with the Braun-Blanquet scale. RHL was calculated as
the total number of bryophyte species with hygrophilous life-forms
according to Oishi and Morimoto (2016). The criteria for the life-form
classification and those of hygrophilous forms were based on those
proposed by Bates (1998).
2.4. Statistical analyses
We first calculated the Pearson product-moment correlation coeffi-
cients among the metrics and the correlation coefficients of these me-
trics with species richness and cover. Then, to understand the influence
of land–use types on the metrics, we analyzed the relationships between
% N, δ15 N, and IAP and land-use types using linear models (LMs). The
best-fit models were identified using the step Akaike information cri-
terion (AIC) function. Generalized linear models (GLMs) were adopted
to analyze the RHL (count data). Modeling was performed using ex-
planatory variables (land-use types) that correlated with each indicator
value (r > 0.3). However, in cases where all the environmental vari-
ables were weakly correlated (r < 0.3), the environmental variable
with the highest correlation was used for modeling. Multicollinearity
was examined based on the variance inflation factor (< 10). All data
were analyzed using the statistical program R (R Development Core
Team, 2015).
The zone of influence was determined based on changes in the
coefficients of determination (R2) of LMs with increases in distance
from the sampling point. For the GLMs, we calculated pseudo R2

351
Y. Oishi, T. Hiura
(McFadden’s adjusted pseudo R squared) instead of R2. The levels of
significance of GLM model fits were tested using analysis of deviance
with χ2 distribution.
Finally, we used an ordinary kriging method to extrapolate the
values of bryophyte metrics to map atmospheric environments across
the city. This analysis was performed using SuperMap Deskpro 6 GIS
software (SuperMap Japan Co., Ltd., Tokyo, Japan).
3. Results
3.1. Bryophyte diversity and metrics
We observed a total of 48 bryophyte taxa, including two genera that
could not be identified to species on the surveyed trees (Appendix A).
The mean species richness in each plot was 13.5 ± 2.9 (mean ±
standard deviation). The most frequently occurring species were P. te-
nuirostris and Frullania muscicola Steph., followed by Lejeunea ulicina
(Tayl.) Gottsche, Lindenb. & Nees, Glyphomitrium humillimum (Mitt.)
Card., Orthotrichum consobrinum Card., and Entodon challengeri (Paris)
Card. Regarding dominant species, P. tenuirostris was the most domi-
nant in 16 sites, including urban, suburban, and mountainous areas.
Other dominant species in urban areas were G. humillimum, Fabronia
matsumurae Besch. Venturiella sinensis (Vent.) Müll. Hal., Cololejeunea
japonica (Schiffn.) Mizut., and O. consobrinum, whereas Macvicaria
ulophylla (Steph.) S. Hatt was dominant in mountainous areas. The
comparison of total species richness and cover among the forest patch
types (urban, suburban, and mountainous areas) is shown in Fig. 2.
Both species richness and cover showed no significant differences
among the patch types in multiple comparisons.
The N analysis was conducted for all sampling plots except one that
did not contain trees meeting the tree selection criteria. The values of %
N, δ15N, IAP, and RHL ranged from 1.60 to 4.15% (2.90 ± 0.63),
−9.91 to −3.41‰ (−6.81 ± 1.53), 0.83 to 4.36 (2.32 ± 0.82), and
0.0 to 3.0 (0.36 ± 0.79), respectively (mean ± SD). The plots in
352
Landscape and Urban Planning 167 (2017) 348–355
Fig. 3. Maps of atmospheric variables evaluated with
bryophyte metrics in this study (For interpretation of
the references to color in this figure legend, the
reader is referred to the web version of this article.).
The color gradient from green to red indicates a scale
for each variable. The circles with solid lines indicate
optimal zone of influence sizes, and circles with
dotted lines indicate the largest valid zone. %
N = nitrogen concentration, δ15N = stable isotope
ratio of nitrogen, IAP = index of atmospheric purity,
RHL = richness of hygrophilous life-forms.
urban areas tended to have higher values of both % N and IAP and
lower values of RHL. The δ15N did not show clear trends along an
urban-rural gradient.
The Pearson product-moment correlation coefficients showed sig-
nificant strong positive correlations between RHL and species richness
and between IAP and total cover (Table 1). A negative correlation was
found between % N and RHL.
3.2. LM/GLM models
Using these data, we calculated the values of the bryophyte metrics
and constructed their respective LMs and GLMs (Table 2). The % N was
positively affected by agricultural areas (50–100-m zones), industrial
areas (500-m zone), and multiple-lane roads (1500–2000-m zones), and
was negatively affected by forest areas (200-m zone). The δ15N was
positively affected by total highway length (100–200-m zone). Contrary
to our expectations, forest and grassland areas had significantly nega-
tive influences on IAP values. RHL was significantly and positively
correlated with forest areas (1000- and 2000-m zones) and negatively
correlated with urban areas (1500-m zone) (Table 1). Notably, % N,
δ15N, and IAP had significant and higher values in 100–200-m zones; in
contrast, the pseudo R2 of RHL increased with increases in zone size
(Table 2).
3.3. Mapping of atmospheric environments
We mapped bryophyte features across the city (Fig. 3). In these
maps, we indicated the zone sizes with the highest (pseudo) R2 value as
the “optimal zone size” and those with significant values as “valid zone
sizes.” These maps revealed varying patterns of high and low values
among the metrics. While RHL showed higher values in the western
part of the city, % N had higher values in the eastern parts. The value of
IAP tended to be higher in urban areas than the surroundings. The
optimal zone size for % N, δ15N and IAP was 200 m, and that for RHL
Y. Oishi, T. Hiura
was 2000 m. The valid zone sizes for the metrics (% N, δ15N, IAP, and
RHL) were 50–2000 m, 100–200 m, 50–2000 m, and 100–2000 m, re-
spectively (Fig. 3).
4. Discussion
4.1. Influence of N pollution on N/δ15N models
The % N models had positive correlations with agricultural areas,
urban areas, industrial areas, and multiple-lane roads and a negative
correlation with forest areas (Table 2). These correlation can be ex-
plained by the fact that agriculture, industrial combustion, and traffic
are known to be major sources of atmospheric N input (Shibata et al.,
2015), while forests can function as N sinks (Yin et al., 2011). In con-
trast, only one land-use type (highway) was a significant variable in the
δ15N models. This result suggests the strong influence of NOx emissions
from motor vehicles on nitrogen pollution in areas close to the highway.
The R2 values of N/δ15N models had higher values in 100–200-m
zones. These results agree with the strong influence of N pollution
within ca. 200 m of its sources (e.g., Bignal et al., 2008; Pitcairn et al.,
2003; Skinner et al., 2006).
Our results also showed no significant correlation between %N and
δ15N (Table 1). This lack of correlation indicates the existence of var-
ious N sources in our study site. This is because although %N is de-
termined by total atmospheric N deposition, δ15N is influenced posi-
tively or negatively according to N emission sources. Hence, a
significant correlation between %N and δ15N is observed only in cases
where a particular N source strongly affects N deposition (e.g., Foan
et al., 2014). By utilizing the correlations among these metrics, we can
characterize N pollution in relation to the variety of N sources.
4.2. Environmental factors associated with IAP and RHL
Our results showed that higher IAP values were recorded in urban
areas and that NOx sources such as industrial areas had no negative
influence on IAP models (Table 2). These results contrast with those of
previous studies, which found that IAP values were lower in urban
areas because of the serious atmospheric pollution (e.g., Taoda, 1972)
and that NO2 pollution decreased IAP values (Krommer et al., 2007).
The correlation of IAP with other metrics may explain these con-
trasting results. The IAP values were significantly and strongly corre-
lated with the bryophyte cover (Table 1), which had higher values in
urban areas (Fig. 2), indicating that higher IAP values in urban areas
can be explained by the higher bryophyte cover. It is likely that no
serious atmospheric pollution occurred in the surveyed urban areas;
pollution decreases IAP values as more sensitive epiphytic bryophytes
disappear. Recently, epiphyte diversity in some urban areas has re-
covered from past decreases in response to improvements in air quality
(Sérgio et al., 2016; Taoda, 1992). Similar trends have also been re-
ported in epiphytic lichens (Davies, Bates, Bell, James, & Purvis, 2007).
RHL was significantly and positively correlated with species rich-
ness (Table 1). This correlation suggests that the epiphytic bryophyte
species richness of a site can be strongly affected by humidity, because
RHL can be an indicator for humidity, an important factor for species
richness (Oishi, 2009).
Appendix A
See Table A1.
353
Landscape and Urban Planning 167 (2017) 348–355
Notably, the RHL showed the largest zone of influence. Focusing on
the high sensitivity of RHL to drought stress (Oishi & Morimoto, 2016),
we can explain the higher pseudo R2 values for RHL models in larger
zones of influence. Drought stress occurs not only at the forest edges but
also in areas with large, populated cities, owing to the influence of
urban heat islands (UHI) (Gago, Roldan, Pacheco-Torres, & Ordonez,
2013; Rizwan, Dennis, & Liu, 2008). UHIs cause a decrease in the var-
iations of daily temperature because heat held by man-made materials
(e.g., concrete) is released at night (Fujibe, 2009; Gago et al., 2013).
This reduction in temperature variation negatively affects drought-
sensitive bryophytes because the narrower the daily temperature range,
the less frequently fog and dew occur; both are important water sup-
plies for bryophytes (Csintalan, Takács, Proctor, Nagy, & Tuba, 2000;
Proctor et al., 2007). Considering that UHIs are widely occurring phe-
nomena in urban areas (e.g., Gago et al., 2013), this influence can be
reflected at a city-wide scale, resulting in higher values of pseudo R2 of
RHL in larger zones.
These results are remarkable in that they indicate the importance of
analysis based on a city-wide scale for the conservation of drought-
sensitive species. In previous studies, the conservation of these species
has often been discussed in terms of environmental factors within
fragmented forests or the surrounding environments, such as the size
and shape of forests (Murakami, Maenaka, & Morimoto, 2005;
Oishi & Morimoto, 2016). However, evaluation of the environment at a
forest-wide scale may not be sufficient to assess the influence of RHI on
drought-sensitive species.
5. Conclusions
Our results will have implications for the effective application of
bryophyte metrics and can be a model for the evaluation of atmospheric
environments in other urban areas. First, the combinational use of
bryophyte metrics can contribute to comprehensive understanding of
atmospheric environments The RHL had a significant negative corre-
lation with % N, indicating that severe drought stress can occur in areas
highly polluted with nitrogen. These areas are likely to be a particular
concern for both biodiversity conservation and human health.
Furthermore, the sources of nitrogen pollution can be diverse among
sampling points, as suggested by the lack of significant correlation of
δ15N values with % N. Second, the usefulness of IAP can be limited in
areas where pollution gradients are narrow or nonexistent. Finally, our
results showed the importance of careful attention to the zones of in-
fluence of these metrics. The extrapolated maps in Fig. 3 are useful for
visualizing atmospheric environments at landscape scales, although
they might include areas that were poorly evaluated owing to their
greater distances from the sampling points. To overcome this potential
issue, we have denoted both valid and optimal zones of influence in the
maps.
Acknowledgments
This research was supported by Japan Society for the Promotion of
Science (Green Network of Excellence) and by the Ministry of the
Environment, Japan (S9-3) to T. Hiura.
Y. Oishi, T. Hiura Landscape and Urban Planning 167 (2017) 348–355

Table A1
Observed bryophyte species arranged in descending order of frequency.

Species Qa Freq.b

Pylaisiadelpha tenuirostris (Bruch & Schimp.) W.R.Buck 5.04 22

Frullania muscicola Steph. 5.86 22
Lejeunea ulicina (Tayl.) Gottsche, Lindenb. & Nees 5.27 21
Glyphomitrium humillimum (Mitt.) Card. 5.48 19
Orthotrichum consobrinum Card. 6.30 19
Entodon challengeri (Paris) Card. 6.38 19
Cololejeunea japonica (Schiffn.) S.Hatt. ex Mizut. 5.92 15
Haplohymenium pseudo-triste (Müll. Hal.) Broth. 6.16 14
Sematophyllum subhumile (Müll. Hal.) M.Fleisch. 5.33 14
Venturiella sinensis (Vent.) Müll. Hal. 6.00 13
Acrolejeunea pusilla (Steph.) Grolle & Gradst. 6.29 13
Macvicaria ulophylla (Steph.) S.Hatt. 5.83 12
Fabronia matsumurae Besch. 6.95 10
Brachymenium nepalense Hook. 7.63 7
Brothera leana (Sull.) Müll. Hal. 4.78 6
Frullania parvistipula Steph. 6.13 6
Trocholejeunea sandvicensis (Gottsche) Mizut. 6.50 6
Metzgeria lindbergii Schiffn. 5.09 5
Haplocladium angustifolium (Hampe & Müll.Hal.) Broth. 7.75 3
Chiloscyphus minor (Nees) J.J.Engel & R.M.Schust. 4.50 3
Frullania tamarisci (L.) Dumort. subsp. obscura (Verd.) S.Hatt. 6.33 3
Lejeunea japonica Mitt. 8.33 3
Dicranum flagellare Hedw. 4.67 2
Leucobryum juniperoideum (Brid.) Müll. Hal. 4.50 2
Ulota crispa (Hedw.) Brid. 6.50 2
Aulacopilum japonicum Broth. ex Card. 10.50 2
Haplohymenium sieboldii (Dozy & Molk.) Dozy & Molk. 6.00 2
Haplocladium microphyllum (Hedw.) Broth. 8.00 2
Herpetineuron toccoae (Sull. & Lesq.) Card. 5.00 2
Frullania ericoides (Nees) Mont. 9.00 2
Dicranum viride (Sull. & Lesq.) Lindb. var. hakkodense (Card.) 4.00 1
Takaki
Bryum argenteum Hedw. 6.00 1
Zygodon sp. 3.00 1
Neckera humilis Mitt. 8.00 1
Forsstroemia cryphaeoides Card. 5.00 1
Haplohymenium triste (Ces.) Kindb. 7.00 1
Rauiella fujisana (Paris) Reimers 6.00 1
Brachythecium sp. 7.00 1
Rhynchostegium pallidifolium (Mitt.) A.Jaeger 5.00 1
Hypnum plumaeforme Wilson 5.00 1
Hypnum plumaeforme var. minus Broth. ex Ando 8.00 1
Hypnum tristo-viride (Broth.) Paris 4.00 1
Plagiochila sciophila Nees ex Lindenb. 8.00 1
Radula constricta Steph. 5.00 1
Cololejeunea minutissima (Sm.) Schiffn. 12.00 1
Cololejeunea raduliloba Steph. 8.00 1
Leptolejeunea elliptica (Lehm. & Lindenb.) Schiffn. 8.00 1
Metzgeria temperata Kuwah. 5.00 1

a
Q = ecological index of each species.
b
Freq. = frequency in our study plots.

References
Agnan, Y., Séjalon-Delmas, N., Claustres, A., & Probst, A. (2015). Investigation of spatial
and temporal metal atmospheric deposition in France through lichen and moss
bioaccumulation over one century. Science of the Total Environment, 529, 285–296.
http://dx.doi.org/10.1016/j.scitotenv.2015.05.083.
Aragón, G., Abuja, L., Belinchón, R., & Martinez, I. (2015). Edge type determines the
intensity of forest edge effect on epiphytic communities. European Journal of Forest
Research, 134(3), 443–451. http://dx.doi.org/10.1007/s10342-015-0863-5.
Asta, J., & Rolley, F. (1999). Biodiversité et bioindication lichénique: Qualité de l’air dans
l’agglomération Grenobloise. Bulletin de l’Association Internationale Lichénologique de
France, 3, 121–126.
Bates, J. W. (1998). Is life-form a useful concept in bryophyte ecology? Oikos, 82,
223–237.
Bettez, N. D., & Groffman, P. M. (2013). Nitrogen deposition in and near an urban eco-
system. Environmental Science & Technology, 47(11), 6047–6051. http://dx.doi.org/
10.1021/es400664b.
Bignal, K. L., Ashmore, M. R., & Headley, A. D. (2008). Effects of air pollution from road
transport on growth and physiology of six transplanted bryophyte species.
Environmental Pollution, 156(2), 332–340. http://dx.doi.org/10.1016/j.envpol.2008.
02.011.
Bobbink, R., Hicks, K., Galloway, J., Spranger, T., Alkemade, R., Ashmore, M., et al.
(2010). Global assessment of nitrogen deposition effects on terrestrial plant diversity:
A synthesis. Ecological Applications, 20(1), 30–59. http://dx.doi.org/10.1890/08-
1140.1.
Boquete, M. T., Fernández, J. A., Aboal, J. R., Real, C., & Carballeira, A. (2009). Spatial
structure of trace elements in extensive biomonitoring surveys with terrestrial
mosses. Science of the Total Environment, 408(1), 153–162. http://dx.doi.org/10.
1016/j.scitotenv.2009.09.017.
Breitner, S., Stölzel, M., Cyrys, J., Pitz, M., Wolke, G., Kreyling, W., et al. (2009). Short-
term mortality rates during a decade of improved air quality in erfurt, Germany.
Environmental Health Perspectives, 117(3), 448–454. http://dx.doi.org/10.1289/ehp.
11711.
Csintalan, Z., Takács, Z., Proctor, M. C. F., Nagy, Z., & Tuba, Z. (2000). Early morning
photosynthesis of the moss Tortula ruralis following summer dew-fall in a Hungarian
temperate dry sandy grassland. Plant Ecology, 151(1), 51–54. http://dx.doi.org/10.
1023/A:1026590506740.
Davies, L., Bates, J. W., Bell, J. N. B., James, P. W., & Purvis, O. W. (2007). Diversity and
sensitivity of epiphytes to oxides of nitrogen in London. Environmental Pollution,
146(2), 299–310. http://dx.doi.org/10.1016/j.envpol.2006.03.023.
Dymytrova, L. (2009). Epiphytic lichens and bryophytes as indicators of air pollution in
Kyiv city (Ukraine). Folia Cryptogamica Estonica, 46, 33–44.
Fernández, J. A., Real, C., Couto, J. A., Aboal, J. R., & Carballeira, A. (2005). The effect of

354
Y. Oishi, T. Hiura
sampling design on extensive bryomonitoring surveys of air pollution. Science of the
Total Environment, 337(1–3), 11–21. http://dx.doi.org/10.1016/j.scitotenv.2004.07.
011.
Fernández, J. A., Aboal, J. R., Real, C., & Carballeira, A. (2007). A new biomonitoring
method for detecting sources of small scale pollution. Atmospheric Environment,
41(10), 2098–2110. http://dx.doi.org/10.1016/j.atmosenv.2006.10.072.
Foan, L., Leblond, S., Thöni, L., Raynaud, C., Santamaría, J. M., & Sebilo, M. (2014).
Spatial distribution of PAH concentrations and stable isotope signatures (δ13C, δ15N)
in mosses from three European areas—Characterization by multivariate analysis.
Environmental Pollution, 184, 113–122. http://dx.doi.org/10.1016/j.envpol.2013.08.
006.
Fujibe, F. (2009). Detection of urban warming in recent temperature trends in Japan.
International Journal of Climatology, 29(12), 1811–1822. http://dx.doi.org/10.1002/
joc.1822.
Gago, E. J., Roldan, J., Pacheco-Torres, R., & Ordonez, J. (2013). The city and urban heat
islands: A review of strategies to mitigate adverse effects. Renewable and Sustainable
Energy Reviews, 25, 749–758. http://dx.doi.org/10.1016/j.rser.2013.05.057.
Gignac, L. D., & Dale, M. R. T. (2005). Effects of fragment size and habitat heterogeneity
on cryptogam diversity in the low–boreal forest of western Canada. The Bryologist,
108(1), 50–66. http://dx.doi.org/10.1639/0007-2745(2005)108[50:EOFSAH]2.0.
CO;2.
Giordano, S., Sorbo, S., Adamo, P., Basile, A., Spagnuolo, V., & Cobianchi, R. C. (2004).
Biodiversity and trace element content of epiphytic bryophytes in urban and extra-
urban sites of southern Italy. Plant Ecolgy, 170(1), 1–14. http://dx.doi.org/10.1023/
B:VEGE.0000019025.36121.5d.
Groffman, P. M., & Pouyat, R. V. (2009). Methane uptake in urban forests and lawns.
Environmental Science and Technology, 43(14), 5229–5235. http://dx.doi.org/10.
1021/es803720h.
Gurjar, B. R., Jain, A., Sharma, A., Agarwal, A., Gupta, P., Nagpure, A. S., et al. (2010).
Human health risks in megacities due to air pollution. Atmospheric Environment,
44(36), 4606–4613. http://dx.doi.org/10.1016/j.atmosenv.2010.08.011.
Harmens, H., Norris, D. A., Cooper, D. M., Mills, G., Steinnes, E., Kubin, E., et al. (2011).
Nitrogen concentrations in mosses indicate the spatial distribution of atmospheric
nitrogen deposition in Europe. Environmental Pollution, 59(10), 2852–2860. http://
dx.doi.org/10.1016/j.envpol.2011.04.041.
Holt, E. A., & Miller, S. W. (2010). Bioindicators: Using organisms to measure environ-
mental impacts. Nature Education Knowledge, 3(10), 8.
Honnay, O., Verheyen, K., & Hermy, M. (2002). Permeability of ancient forest edges for
weedy plant invasion. Forest Ecology and Management, 16(1-3), 109–122. http://dx.
doi.org/10.1016/S0378-1127(01)00490-X.
Iwatsuki, Z. (Ed.), (2000). Mosses and liverworts of Japan. Tokyo: Heibonsha.
Japan Meteorological Agency (2016). Past meteorological data. Japan Meteorological
Agency. [Accessed 11 April 2016] http://www.jma.go.jp/jma/index.html.
Kopáček, J., & Posch, M. (2011). Anthropogenic nitrogen emissions during the Holocene
and their possible effects on remote ecosystems. Global Biogeochemical Cycles, 25,
1–16. http://dx.doi.org/10.1029/2010GB003779 [GB2017].
Krommer, V., Zechmeister, H. G., Roder, I., Scharf, S., & Hanus-Illnar, A. (2007).
Monitoring atmospheric pollutants in the biosphere reserve Wienerwald by a com-
bined approach of biomonitoring methods and technical measurements. Chemosphere,
67(10), 1956–1966. http://dx.doi.org/10.1016/j.chemosphere.2006.11.060.
Laurance, W. F., Nascimento, H. E. M., Laurance, S. G., Andrade, A., Ribeiro, J. E. L. S.,
et al. (2006). Rapid decay of tree-community composition in Amazonian forest
fragments. Proceedings of the National Academy of Sciences of the United States of
America, 103(50), 19010–19014. http://dx.doi.org/10.1073/pnas.0609048103.
Laurance, W. F., Nascimento, H. E. M., Laurance, S. G., Andrade, A., Ewers, R. M., Harms,
K. E., et al. (2007). Habitat fragmentation, variable edge effects, and the landscape-
divergence hypothesis. PLoS One, 2(10), e1017. http://dx.doi.org/10.1371/journal.
pone.0001017.
LeBlanc, F., & De Sloover, J. (1970). Relation between industrialization and the dis-
tribution and growth of epiphytic lichens and mosses in Montreal. Canadian Journal
of Botany, 48(8), 1485–1496. http://dx.doi.org/10.1139/b70-224.
Liu, X. Y., Xiao, H. Y., Liu, C. Q., & Xiao, H. W. (2008). Atmospheric transport of urban-
derived NHx: Evidences from nitrogen concentration and δ15N in epilithic mosses at
Guiyang, SW China. Environmental Pollution, 156(3), 715–722. http://dx.doi.org/10.
1016/j.envpol.2008.06.011.
Murakami, K., Maenaka, H., & Morimoto, Y. (2005). Factors influencing species diversity
of ferns and fern allies in fragmented forest patches in the Kyoto city area. Landscape
and Urban Planning, 70(3–4), 221–229. http://dx.doi.org/10.1016/j.landurbplan.
2003.10.019.
Murcia, C. (1995). Edge effects in fragmented forests: Implications for conservation.
Trends in Ecology & Evolution, 10(2), 58–62. http://dx.doi.org/10.1016/S0169-
5347(00)88977-6.
Oishi, Y., & Morimoto, Y. (2016). Identifying indicator species for bryophyte conservation
in fragmented forests. Landscape and Ecological Engineering, 12(1), 107–114. http://
dx.doi.org/10.1007/s11355-013-0220-0.
Oishi, Y. (2009). A survey method for evaluating drought-sensitive bryophytes in frag-
mented forests: A bryophyte life-form based approach. Biological Conservation,
142(12), 2854–2861. http://dx.doi.org/10.1016/j.biocon.2009.04.011.
Oishi, Y. (2013). Comparison of pine needles and mosses as bio-indicators for polycyclic
aromatic hydrocarbons. Journal of Environmental Protection, 4(8A), 106–113. http://
355
Landscape and Urban Planning 167 (2017) 348–355
dx.doi.org/10.4236/jep.2013.48A1013.
Onianwa, P. C. (2001). Monitoring atmospheric metal pollution: A review of the use of
mosses as indicators. Environmental Monitoring and Assessment, 71(1), 13–50. http://
dx.doi.org/10.1023/A:1011660727479.
Organization for Economic Cooperation and Development (2012). OECD environmental
outlook to 2050: The consequences of inaction. Organization for Economic Cooperation
and Development1–8. http://dx.doi.org/10.1787/9789264122246-en.
Pardow, A., Gehrig-Downie, C., Gradstein, S. R., & Lakatos, M. (2012). Functional di-
versity of epiphytes in two tropical lowland rainforests, French Guiana: Using
bryophyte life-forms to detect areas of high biodiversity. Biodiversity and Conservation,
21(14), 3637–3655. http://dx.doi.org/10.3389/fpls.2011.00117.
Pauchard, A., & Alaback, P. B. (2006). Edge type defines alien plant species invasions
along Pinus contorta burned, highway and clearcut forest edges. Forest Ecology and
Management, 223(1–3), 327–335. http://dx.doi.org/10.1016/j.foreco.2005.11.020.
Pearson, J., Wells, D. M., Seller, K. J., Bennett, A., Soares, A., Woodall, J., et al. (2000).
Traffic exposure increase natural 15N and heavy metal concentrations in mosses. New
Phytologist, 147(2), 317–326. http://dx.doi.org/10.1046/j.1469-8137.2000.00702.x.
Pitcairn, C. E. R., Fowler, D., Leith, I. D., Sheppard, L. J., Suttone, M. A., Kennedy, V.,
et al. (2003). Bioindicators of enhanced nitrogen deposition. Environmental Pollution,
126(3), 353–361. http://dx.doi.org/10.1016/S0269-7491(03)00248-3.
Proctor, M. C. F., Oliver, M. J., Wood, A. J., Alpert, P., Stark, L. R., Cleavitt, N., et al.
(2007). Desiccation tolerance in bryophytes: A review. The Bryologist, 110(4),
595–621. http://dx.doi.org/10.1639/0007-2745(2007)110[595:DIBAR]2.0. CO;2.
R Development Core Team (2015). R: A language and environment for statistical computing.
Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org/.
Rizwan, A. M., Dennis, L. Y. C., & Liu, C. (2008). A review on the generation, determi-
nation and mitigation of urban heat island. Journal of Environmental Sciences, 20(1),
120–128. http://dx.doi.org/10.1016/S1001-0742(08)60019-4.
Sérgio, C., Carvalho, P., Garcia, C. A., Almeida, E., Novais, V., Sim-Sim, et al. (2016).
Floristic changes of epiphytic flora in the metropolitan Lisbon area between 1980 and
1981 and 2010–2011 related to urban air quality. Ecological Indicators, 67, 839–852.
http://dx.doi.org/10.1016/j.ecolind.2016.03.022.
Schröder, W., Holy, M., Pesch, R., Harmens, H., Fagerli, H., Alber, R., et al. (2010). First
europe-wide correlation analysis identifying factors best explaining the total nitrogen
concentration in mosses. Atmospheric Environment, 44(29), 3485–3491. http://dx.doi.
org/10.1016/j.atmosenv.2010.06.024.
Shibata, H., Branquinho, C., McDowell, W. H., Mitchell, M. J., Monteith, D. T., Tang, J.,
et al. (2015). Consequences of altered nitrogen cycles in the coupled human and
ecological system under changing climate: The need for long-term and site-based
research. Ambio, 44(3), 178–193. http://dx.doi.org/10.1007/s13280-014-0545-4.
Skinner, R. A., Ineson, P., Jones, H., Sleep, D., Leith, I. D., & Sheppard, L. J. (2006).
Heathland vegetation as a bio-monitor for nitrogen deposition and source attribution
using δ15N values. Atmospheric Environment, 40(3), 498–507. http://dx.doi.org/10.
1016/j.atmosenv.2005.09.054.
Taoda, H. (1972). Mapping of atmospheric pollution in Tokyo based upon epiphytic
bryophytes. Japanese Journal of Ecology, 22(7), 125–133.
Taoda, H. (1992). Long term change of epiphytic vegetation in Tokyo and the neighboring
area, with special reference to surveying methods. Proceedings of the Bryological
Society of Japan, 5(12), 193–196.
The World Meteorological Organization and the International Global Atmospheric
Chemistry (2012). Impacts of megacities on air pollution and climate. The World
Meteorological Organization and the International Global Atmospheric Chemistry.
[Accessed 11 April 2016] http://www.wmo.int/pages/prog/arep/gaw/documents/
Final_GAW_205.pdf.
Thomas, R. Q., Canham, C. D., Weathers, K. C., & Goodale, C. L. (2010). Increased tree
carbon storage in response to nitrogen deposition in the US. Nature Geoscience, 3,
13–17. http://dx.doi.org/10.1038/ngeo721.
Vieira, C., Séneca, A., Sérgio, C., & Ferreira, M. T. (2012). Bryophyte taxonomic and
functional groups as indicators of fine scale ecological gradients in mountain streams.
Ecological Indicators, 18, 98–107. http://dx.doi.org/10.1016/j.ecolind.2011.10.012.
Vuković, G., Urošević, M. A., Goryainova, Z., Pergal, M., Škrivanj, S., Samson, R., et al.
(2015). Active moss biomonitoring for extensive screening of urban air pollution:
Magnetic and chemical analyses. Science of the Total Environment, 521–522, 200–210.
http://dx.doi.org/10.1016/j.scitotenv.2015.03.085.
Yin, S., Shen, Z., Zhou, P., Zou, X., Che, S., & Wang, W. (2011). Quantifying air pollution
attenuation within urban parks: An experimental approach in Shanghai, China.
Environmental Pollution, 159(8–9), 2155–2163. http://dx.doi.org/10.1016/j.envpol.
2011.03.009.
Zartman, C. E. (2003). Habitat fragmentation impacts on epiphyllous bryophyte com-
munities in Central Amazonia. Ecology, 84(4), 949–954. http://dx.doi.org/10.1890/
0012-9658(2003)084[0948:HFIOEB]2.0.CO;2.
Zechmeister, H. G., Dirnböck, T., Hülber, K., & Mirtl, M. (2007). Assessing airborne
pollution effects on bryophytes—Lessons learned through long-term integrated
monitoring in Austria. Environmental Pollution, 147(3), 696–705. http://dx.doi.org/
10.1016/j.envpol.2006.09.008.
Zechmeister, H. G., Richter, A., Smidt, S., Hohenwallner, D., Roder, I., Maringer, S., &
Wanek, W. (2008). Total nitrogen content and δ15N signatures in moss tissue:
Indicative value for nitrogen deposition patterns and source allocation on a nation-
wide scale. Environmental Science and Technology, 42(23), 8661–8667. http://dx.doi.
org/10.1021/es801865d.