A peer-reviewed open-access journal

NeoBiota 84: 23 |—246 (2023)

doi: 10.3897/neobiota.84.9098 | toy) NeoB 10ta

https:/ / neobi ota. pen soft. net Advancing research on alien species and biological invasions

Vertical spread of Hymenoscyphus fraxineus propagules

Milot Dvofak', Petr Stoidl?, Michal Rost?

| Department of Forest Protection and Wildlife Management, Faculty of Forestry and Wood Technology, Men-
del University in Brno, Zemédelskd 3, 613 00 Brno, Czech Republic 2. Department of Genetics and Agricul-
tural Biotechnologies, Faculty of Agriculture and Technology, University of South Bohemia, Studentska 1668,
370 05 Ceské Budéjovice, Czech Republic

Corresponding author: Milon Dvorak (milon.dvorak@mendelu.cz)

Academic editor: Andrea Battisti | Received 28 July 2022 | Accepted 25 October 2022 | Published 18 May 2023

Citation: Dvot4k M, Stoidl B Rost M (2023) Vertical spread of Hymenoscyphus fraxineus propagules. In: Jactel H,
Orazio C, Robinet C, Douma JC, Santini A, Battisti A, Branco M, Seehausen L, Kenis M (Eds) Conceptual and
technical innovations to better manage invasions of alien pests and pathogens in forests. NeoBiota 84: 231-246. https://
doi.org/10.3897/neobiota.84.90981

Abstract

Currently, the ash dieback causal agent Hymenoscyphus fraxineus is an established invasive pathogen in
most European countries. Its potential to spread quickly among invaded forests is based on its propagules:
airborne inoculum composed mainly of ascospores originated in apothecia growing on leaf litter infected
during the previous vegetation season. The spread of the inoculum by air masses to distant areas is prob-
able and depends on the availability of the ascospores in higher levels of air. Our study aimed to detect the
inoculum in an infected area at heights of more than 20 meters. Our study was conducted in a municipal
locality (Borgov nad Vltavou) with tens of infected ash trees (Fraxinus excelsior) in South Bohemia (SW
Czechia). The infected trees surround an agricultural silo where five rotating arm spore traps (rotorods)
were mounted for ten consequent 48h samplings during the peak of the sporulating season (17° July to 6"
August 2020). The spore traps were mounted 48, 37, 25, 14 and 0,3 meters above ground. Samples were
quantified by qPCR. Results clearly proved the ability of the spores to reach a height of 48 meters. Further-
more, H. fraxineus DNA was detected from all five spore traps during all ten samplings. Mostly, the amount
of detected spores showed a decreasing trend with height, and varied a lot. During some of the samplings,
higher spore concetrations were achieved at the top than at the lower traps, which can be explained by
horizontal air transfer of the inoculum from other infected areas. Based on GLM analyses, higher spore
concentrations were achieved during days without rain, lower air temperatures, after cloudy, humid and
rainy weather without strong winds. A combination of rotorod ROTTRAP 52 with qPCR quantification
proved to be an efficient technology for a study focused on the vertical spread of H. fraxineus propagules.

Keywords
airborne inoculum, ash dieback, riseability, rotorod, ROTTRAP, spore trap

Copyright Milori Dvorak et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC
BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

2352 Milon Dvoiak et al. / NeoBiota 84: 231-246 (2023)

Introduction

Hymenoscyphus fraxineus (T. Kowalski) Baral, Queloz, and Hosoya is the causal agent of
the ash dieback (Kowalski 2006). It is an invasive pathogen introduced to Europe from
Eastern Asia (Zhao et al. 2012) probably during the 1990’s (Przybyt 2002), first reported
in Poland (Kowalski 2006). Its current distribution apart from Eastern Asia covers the
whole of Europe apart from Portugal, Greece, Albania, Macedonia, Bulgaria and Mol-
dova (EPPO 2022). H. fraxineus severely attacks native European ash species Fraxinus
excelsior and FE angustifolia (Kowalski and Holdenrieder 2009) causing a serious threat
to mixed deciduous forest habitats, especially riparian forests where other deciduous
tree species are endangered by invasive pathogens (Jankovsky and Holdenrieder 2009).

The symptoms of the ash dieback and mechanism of infection of the host trees
have been well described in numerous publications (Kowalski 2006; Kowalski and
Holdenrieder 2009; Gross et al. 2012, 2014; Husson et al. 2012; Cleary et al. 2013;
Krautler et al. 2015; Chandelier et al. 2016; Fones et al. 2016; Hanackova et al. 2017)

The predominant way to spread this pathogen is via the airborne ascospores (Gross
et al. 2012), although the conidia of the imperfect stage Chalara fraxinea play an im-
portant role (Fones et al. 2016). The ascospores are typically released from apothecia
growing on shredded infected leaf rachises during the following growing season, main-
ly from June to September (Timmermann et al. 2011; Gross et al. 2012; Hietala et al.
2013; Chandelier et al. 2014; Dvorak et al. 2016). The ascospores have been shown to
be able to span long distances and in France they have been observed to even reach 50
to 100 km (Grosdidier et al. 2018b).

The probability of the long distance spread of the ascospores depends on the height
they can reach to be blown with the air masses (Chandelier et al. 2014; Oteros et al.
2015; Aguayo et al. 2021). Existing studies proved that the inoculum of H. fraxineus
rises from the ground where it is actively ejected from the apothecia (Timmermann et
al. 2011) and it is reliably detectable in aerobiological samples collected by the RNSA
(French Network of Aerobiology) at heights of between 10 to 20 meters (Aguayo et al.
2021). According to these authors, ascospores present at this height prove the presence
of the pathogen in the landscape scale in the range of tens of kilometres.

The aim of our study was i) to test a methodological approach for describing the ver-
tical spore dispersal pattern of H. fraxineus, and ii) to prove the presence of the inoculum
as high as possible, at such a height where the spread of the aerosols is more likely affected
by horizontal movement of air masses rather than by convection from the ground surface.

Methods
Sampling in ADB infected locality

The sampling point was located in an agricultural silo in Borsov nad Vltavou (South

Bohemia; SW Czech Republic); GPS: 48.9244°N, 14.4414°E; 412 ma. s. |. It is an

Vertical spread of Hymenoscyphus fraxineus propagules 233

industrial area of a village adjacent to Moldava River. It is located on the SE edge of a
large plain called Ceské Budéjovice Basin (Ceskobudéjovicka panev), where foothills
of Sumava mountains called Blansky les start to rise. Due to the surrounding landscape
being characterised by plains (especially in NW, N and E direction), the 52 m high silo
of Borsov is probably exposed to wind currents blowing from areas that are at least tens
of kilometres away. Trees in this area are mostly represented by mixtures of deciduous
middle-European species formed at the riverbanks, gardens and parks, or alleys along
railways and roads. Most of the groups of trees comprise a significant portion of ashes
(Fraxinus excelsior) with typical symptoms of ash dieback (Fig. 1A); in a few cases
they are even completely dead trees. The closest source of inoculum = rachises with
apothecia was found 60 m from the silo, where a lowest spore trap (R1) was installed.
This closest point is an edge of mostly untreated vegetation with many infected ashes
following adjacent railway and roads, partly visible in Fig. 1B at the right side of the
horizon. This forest-like suburban vegetation is in the northern direction altered with
recently recovered park with heavily infested trees; one of them, 170 m distanced, is
well visible in the foreground of the Fig. 1A. The sampling was carried out during the
peak of the ascospore production season in Czechia (Dvorak et al. 2016) continuously
from 17° July to 6" August, 2020. The presence of apothecia was checked in the litter
of ash trees adjacent to the sampling point.

Air samplers

To sample the air inoculum rotating arm spore traps (rotorods) ROTTRAP 52 (Milon
Dvorak, Borsov nad Vltavou, Czech Republic) were employed (Fig. 1C). Our air-
sampler was a further developed rotorod of Chandelier et al. (2014), based on the
description of Perkins and Leighton (1957) and improved by McCartney et al. (1997).
A 10 cm-long aluminium arm whirled 2067 rotations per minute with vertical squared
brass rods mounted on both ends. The impaction side of the rod (0.8 x 50 mm) was
covered by double-sided non-woven tape (Tesa SE, Norderstadt, Germany), which was
renewed every 48 hours; exposed stored in sterile 2-ml microtubes at -20 °C before fur-
ther processing. The rotorods were powered from an electric network via adaptor 220V
AC/9V DC. Such an arrangement of rotorod samples 52 litres of air per minute, with a
theoretical sampling efficiency close to 100% for particles with a diameter bigger than
10.94 um (Noll 1970; Dhingra and Sinclair 2017). Further methodological details are
described in Dvorak (2022).

Sampling points

The rotorods were installed at five different heights (Fig. 1B): 0.3 m (further men-
tioned as ,,R/“), 13.84 m (,,R2"), 25.09 m (,,R3"%), 36.57 m (,,R4") and 48.06 m (,,R5“)
above ground. ‘The R/ was positioned close to the ground of the nearest group (60 m
apart from the silo) of infected ash trees to monitor the source of inoculum. The R2
— R5 were mounted on windows in regular heights on the NNW side of the silo. This

234 Milon Dvorak et al. / NeoBiota 84: 231-246 (2023)

Figure |. Sampling locality and sampling tools A the sampling point in the focus of ash dieback.
Agricultural silo is surrounded by ashes infected with H. fraxineus B sampling spots in the windows of the
silo. ROTTRAPs 52 are installed in different heights R2 - R5 in the windows (red circles) of the silo on its
NNW side C rotating arm spore trap ROTTRAP 52 installed at R5.

Vertical spread of Hymenoscyphus fraxineus propagules 235

side was chosen for its non-exposure to sun (to avoid additional thermic air currents),
and due to the prevailing wind direction. It is exposed to the most common wind
(NW), which is supposedly bringing the airborne inoculum from the infection sources
present in that direction, including the closest inoculum source where R/ was installed.

Meteorological data

Meteorological data were partly measured by automatic meteorological station Sig-
nalizator (AMET, Velké Bilovice, Czech Republic) and partly received from the ar-
chive of the Czech Hydrometeorological Institute (CHM), station Ceské Budéjovice
— Roznov. From the sampling point the CHMI meteorological station is positioned
3.6 km to NE. Data taken from Signalizator were daily means of relative air humidity
(further in the text only air humidity). Data received from CHMI were: i) daily mean
of air temperature (air temperature); ii) daily duration of sunshine (sunshine); iii) daily
mean of air pressure (air pressure); iv) daily amount of precipitation (precipitation) and
v) daily mean of wind speed (wind speed).

DNA extraction

The genomic DNA from samples was extracted with a DNeasy plant minikit (Diis-
seldorf, Germany). Each microtube with exposed tape was supplemented with one
3-mm sterile tungsten bead and 20 pcs 2-mm glass beads, 400 pl of AP1 buffer and
4 ul of RNase. This mix was ground twice for 60 seconds using a high speed homog-
enizer Millmix 20 (Domel, d.o.0., Zelezniki, Slovenia) set at 30 Hz and incubated for
10 minutes at 65 °C. The microtubes were inverted three times during the incubation.
Further steps were following the manufacturer's instructions; however, the last step
(elution) was not repeated to obtain higher concentration of DNA. DNA samples were
eluted in 100 ul and stored at -20 °C before further processing.

Real-time quantitative PCR conditions

Direct specific qPCR was performed using a QuantStudio 6 Flex Real-Time PCR Sys-
tem (Life Technologies Holdings Pte. Ltd., Singapore), Light Cycler 480 Probes master
(Roche Diagnostics Nederland BV, Almere, the Netherlands) and primers and probes
specific to H. fraxineus (Chandelier et al. 2010). The cycling conditions followed the
master mix manufacturer's instructions and the setting optimised by (Chandelier et al.
2010): preincubation: 10 min, 95 °C followed by 45 cycles of denaturation: 10 sec, 95
°C; annealing: 30 s, 60 °C; extension: 1 s, 72 °C. Composition of the reaction mixture
was following: 0,2 yl of each primer (final concentration 400 nM), 0,2 pl of TaqMan
probe (200 nM), 5 ul of TaqMan Universal PCR Master Mix, 1,4 ul of sterile deion-
ized water and 3 ul of template DNA. Every reaction was performed in two technical
repetitions together with a negative control containing the master mix without tem-

plate DNA.

236 Milon Dvorak et al. / NeoBiota 84: 231-246 (2023)

Absolute quantification

The concentrations of H. fraxineus DNA in the samples were expressed as numbers of
copies of the target sequence in 1 ul of template DNA (further only C/N). These values
were obtained using a standard curve generated from reactions with different C/Ns
(2.5x10? to 2.5x10°) of plasmid pCR 2.1 TOPO TA vector (Invitrogen, Carlsbad,
California, USA) by QuantStudioTM Real-Time PCR System Version 1.3 (Thermo
Fisher Scientific). Plasmids contained species - specific insert (PCR products amplified
with Cf-F and Cf-S primers). DNA was extracted from pure cultures of H. fraxineus
(collection of Mendel University in Brno).

To express the absolute amount of ascospores in every analysed sample, an ab-
solute quantification of ascospore suspension was performed. For that purpose ten
apothecia were collected from ash leave rachises and immersed in 1 ml of distilled
water in a 2-ml microtube. The following day, the clear liquid upper part without
apothecia and other debris was transferred into a clean 2 ml microtube. Vortexed
ascospore suspension was quantified in Biirker chamber by microscope. Ten-fold se-
rial dilutions from 18750 to 1.875 ascospores in 100 pl of distilled water were trans-
ferred into clean 2 ml microtubes and DNA was extracted with the same protocol
as for the spore trap samples. Extracted DNA samples were used as standards for a
qPCR absolute quantification of the plasmids previously used for the quantification
of the environmental samples. The lowest detectable concentration which turned
positive in all three technical repetitions of the sample was 18.75 ascospores per

sample (Ct = 36.328, SD = 0.862).

Statistical analysis

To describe the influence of meteorological factors on the ability of the inoculum to
spread vertically, meteorological variables were averaged for three particular days of
every sampling. Furthermore, the factor of riseability (FR) has been defined. It is ex-
pressed as the ratio of the ascospore concentration recorded at the highest sampler (5)
to the concentration at the lowest sampler (RZ). It takes values lower than 1.0 in case
the R5 ascospore concentration is lower than RJ concentration.

In order to describe the relationship between CN and the character of the weath-
er described by explanatory variables (air temperature, precipitation, sunshine, air
humidity, wind speed and air pressure), generalised linear regression models were
constructed with explanatory variables measured on the same day as the dependent
variable CN, or with explanatory variables recorded during previous four samplings
(i.e. one sampling lag = period of preceding two days before the sampling started,
two samplings lag = two to four days before, three samplings lag = four to six days
before and four samplings lag = six to eight days before sampling) to simulate the
lag of the pathogen’s reaction on the changing weather. Due to the nature of the
dependent variable (a strictly positive variable showing a positive skew for RI, R5
and FR), we used the gamma distribution with a logarithmic link function when

Vertical spread of Hymenoscyphus fraxineus propagules 237

R5 ° R5 0,736 7
Sampling 1 Sampling 2
R4 R4 1,396
R3 R3 2,009
R2 R2 | 2,059
R1 | R1
0 2 4 6 8 10 12 0 50 100 150 200
Quantity (CN) Quantity (CN)
. R5 0,030 °
Pope Sampling 3 Sampling4
R4 7,897 R4 0,165
R3 ,847 R3 0,123
R2 9 R2 0,124
R1 | R1
0 100 200 300 400 500 0 2 4 6 8 10 12
Quantity (CN) Quantity (CN)
evn: Sampling5
s PIING> ks [pss Sampling 6

0 20 40 60 80 0 1 2 3 4 5 6
Quantity (CN) Quantity (CN)

ee La Sampling 7

0 5 10 15 20 0,00 0,02 0,04 0,06 0,08 0,10
Quantity (CN) Quantity (CN)

R5

Sampling9 * | °3° Sampling 10

R4 R4 | 0,153
R3 R3 | 0,282
R2 R2 | 0,209
R1 R1
0,00 0,05 0,10 0,15 0,20 0,25 0 100 200 300 400 500 600 700
Quantity (CN) Quantity (CN)

Figure 2. Vertical profiles of spore concentrations recorded during the ten samplings.

fitting the regression models. To select variables for individual models, we used the
procedure described in Morgan and Tatar (1972) implemented in the “bestglm”
library for the R software (R Core Team 2018). This approach allowed the selection
of the best subset of explanatory variables for particular GLMs. However, because
of the low number of observations, in some cases the models did not converge

238 Milon Dvoiak et al. / NeoBiota 84: 231-246 (2023)

numerically, and it was not possible to obtain the optimal regression model. ‘The
selection of suitable models was made with regard to the AIC values achieved (the
smallest the best).

At the same time, we tried to model the relationship between CN and the height
of the sampler. Due to convergence problems with an exponential model based on a
differential equation, we used a somewhat simpler empirical exponential model with
the following analytical form:

CN = 8, - exp(-B, : 4)

where f, and §, are the estimated regression coefficients, and h is the height in meters.
All numerical calculations were performed using the R 4.2.0 programming environ-
ment (R Core Team 2018).

Results

Fresh apothecia were observed on infected rachises at the sampling locality from the
beginning of July until the end of the sampling (6" August, 2020). Consequently, all
ten 48-h samplings showed presence of H. fraxineus in the air (Fig. 2).

Every sampler positively detected inoculum during every sampling. The lowest
positively detected spore concentration was detected in samples from the highest
sampler R5 from the sampling started on 4" August (CN = 0.018, Ct = 37.428).
This least concentrated sample contained 1.89 ascospores; however, this amount
was calculated extrapolating the ascospore suspension standard curve, hence the
exact amount cannot be taken in consideration. The highest concentration was re-
corded at R/ during the following sampling started on 6" August (CN = 659.063,
Ct = 27.181). Taking in account the sampling rate 52 1/min and sampling period
of 48 h, we detected average spore concentration in a range from 0.013 to 462.12
spores per m’.

Most of the detected spore concentrations showed a clear decreasing trend follow-
ing the height of the sampling point. The resulting nonlinear regression model (Fig. 3;
residual SE = 99.85, DF = 96), corresponding to the aforementioned parameterization,
describing the generally decreasing trend, was estimated as follows:

CN = 150.3243 « exp(-0.2155- /)

GLM analyses resulted in 10 significant models to estimate C’Ns from meteoro-
logical variables measured during the sampling and lagged of 2—4, 4-6 and 6-8 days.
Their parameters can be found in supplementary file “Parameters of GLM models”.
The calculated p and positive or negative meaning of each parameter are displayed

in Table 1.

Vertical spread of Hymenoscyphus fraxineus propagules 239

40 50

0 10 20 30
Height of the sampling points [in m]

Figure 3. Model of the vertical spread of the inoculum. The nonlinear regression model described the

decreasing trend of spore concentrations with increasing height of sampling.

Discussion

Our results prove for the first time, that propagules of 1. fraxineus are reliably detect-
able at almost 50 meters above ground, where they have their main source in infected
rachises (Gross et al. 2012). Repeated sampling revealed . fraxineus DNA’s presence
to be more than twice as higher as in an aerobiological study, where air was sampled by
7-day volumetric spore traps (Hirst 1952) of RNSA (Aguayo et al. 2021).

The results of this study also confirm statements of other authors (Chandelier et al.
2014) who described decreasing trend of ascospore concentrations up to three meters,
regardless of the site and time of the sampling. This was a general trend among our
samplings. The ground sampling spot RJ always showed higher C/N than the top spot
R5. In one case even more than 20,000 times.

However, not in every sampling did the CN values descend with increasing height.
The sampling spot R4 showed lower values, than 25 for samplings 5, 6 and 7. ‘This was
probably due to overloading the sampling rods with dust, which is a critical handi-

240 Milon Dvorak et al. / NeoBiota 84: 231-246 (2023)

Table |. Meteorological variables as parameters of GLM models and their p during sampling and 0-8
days before sampling (d.b.s.).

RI 6-8 d.b.s. 4-6 d.b.s. 2-4 d.b.s. 0-2 d.b.s. sampling
Air temperature 014 .026
Sunshine 034
Precipitation 032
Air humidity .032 045
Wind speed 01 031
Air pressure 012
RS 6-8 d.b.s. 4-6 d.b.s. 2-4 d.b.s. 0-2 d.b.s. sampling
Air temperature .043 .040 .018
Sunshine
Precipitation
Air humidity
Wind speed
Air pressure
FR 6-8 d.b.s. 4-6 d.b.s. 2-4 d.b.s. 0-2 d.b.s. sampling
Air temperature .016
Sunshine 029
Precipitation 034 .018 ae .048
Air humidity O11
Wind speed .016 015
Air pressure
Legend: Influence

positive negative
Significant parameter (p < .05) P fi

Highly significant parameter (p < .01) lo» BEE

cap of rotorods (Lacey and West 2006; Chandelier et al. 2014; Dhingra and Sinclair
2017). The dust apparently came from outlets of ventilators of the silo adjacent to R4
while depositing the harvested grain.

Furthermore, the sampling site R3 gave higher C’Ns than it could be expected to
follow the descendent trend during samplings 8 and 9. Similarly, CNs at R5 from
samplings 1, 5 and 6 were also not lower than CN of lower sampling points. An
explanation for this anomaly could be the effect of horizontal air currents which
could bring a higher amount of inoculum from more distant localities, influencing
the results of sampling only in these cases of low local concentrations. Generally,
abnormalities in the trend of decreasing CN with the height of sampling can be
interpreted as a consequence of the fact that inoculum detected in heights above 10
m might be representative for areas within a perimeter of at least tens of kilometres
(Aguayo et al. 2021). Study focused on long distance spread of pollen grains found
correlation of pollen data from the sampler (elevated 10-20 m) and land use even
200 km apart (Oteros et al. 2015, 2017). In the case of our experiment, samplers
R2, R3, R4 and R5 were mounted higher than 10 meters; R5 even in more than 48
m. Taking into account the previously mentioned studies (Oteros et al. 2015, 2017,

Vertical spread of Hymenoscyphus fraxineus propagules 241

Aguayo et al. 2021) there is a theoretical presumption that the detected inoculum
can be related to sources at a distance of up to hundreds of kilometres. Also fungal
spores long distance proofs are known. Based on spatial air sampling and modelling
it was revealed that tobacco blue mould outbreak was caused by Peronospora tabacina
spores blown from sources several hundred km away (Aylor et al. 1982). Another
example of long distance transport of spores was studied by Vasaitis and Enderle
(2017) considering a possible way of ash dieback introduction to Great Britain.
Agricultural silo used for this study is located very beneficially from this point. Its
height is almost three times bigger than the canopy layer of surrounding vegetation,
which acts as a natural barrier for the vertical spread of the inoculum (Aylor 1999).
Taking into account the geomorphology of surrounding landscape and predominant
wind direction (NW), it is highly likely exposed to air masses from an adjacent plane
area called “Ceskobudéjovicka panev” (Ceské Budéjovice Basin), which is a tectonic
ditch oriented NW — SE, elevated 380-410 maz. s. L., almost 70 km long and 10-12
km wide. From personal experiences we know that infected ashes are very frequently
present there, especially in alleys along roads, riverside plantations and municipal
green spaces. We assume that inoculum detected in R2 — R5 may originate in this
area, not only from the sources adjacent to the sampling point. Transfer of spores by
air masses up to 100 km has been already proven (Grosdidier et al. 2018b), which is
enabling this hypothesis.

Long distance transfers of H. fraxineus by air masses have always been an impor-
tant issue. It was considered that H. fraxineus had been introduced into Great Britain
between 2008 and 2011 via long distance transfer of the air inoculum from mainland
Europe. This statement was strongly supported with a model (Vasaitis and Enderle
2017). Recently, H. fraxineus was found for the first time in Spain (Asturias, NW
Spain) on matured trees and its surrounding regeneration (Stroheker et al. 2021). The
fact that it has not yet been observed in NE Spain, which is adjacent to the closest
distribution of the pathogen, implies the possibility of long-distance transport of the
inoculum from France via the Bay of Biscay. However, the introduction of the patho-
gen through plant material import cannot be overlooked.

Results of the GLM modelling of the determination of the detected spore concen-
trations by meteorological factors discovered numerous relations. Although the low
number of repetitions decreases the reliability of the results’ interpretation, we would
like to highlight some of them:

Air temperature proved to have significantly positive effect two days preceding the
sampling at both RJ and R35. This confirms previous observations of Chandelier et al.
(2014), who found a positive influence of higher temperatures on spore concentra-
tions. On the contrary, sooner than two days before and directly during the sampling,
the effect of air temperature was significantly negative. This pattern indicates, and
confirms, that for the ascospore development and release the high air temperature is
not favourable (Grosdidier et al. 2018a). Since the entire sampling period can be char-
acterized like very warm with maximal temperatures exceeding 30 °C, this biological
limitation of the pathogen was more apparent within our experiment.

242 Milon Dvofak et al. / NeoBiota 84: 231-246 (2023)

The daily amount of precipitation showed different effects at different sampling
heights. R/ was significantly positively affected by precipitations two to four days be-
fore sampling. Through the enhanced humidity of the ground surface, rain proved to
be essential for successful ascospore maturation and release (Timmermann et al. 2011;
Hietala et al. 2013; Dvorak et al. 2016) and the consequent progress of ash dieback
(Chumanova et al. 2019). On the other hand, concentrations at R5 and riseability
were significantly negatively affected by rain during the days before the sampling, and
even highly significantly during the sampling. Two explanations for this phenomenon
are probable: i) ascospores currently present in the air are washed out by rain from the
aerosol which is a known factor (Aylor 1999), and ii) trapped ascospores are partly
splashed from the sticky sampling surface. For the significantly positive effect of pre-
cipitation on riseability during sampling we do not have any reasonable explanation;
however, the p for this parameter is approaching the limit of significance (p = .048)
hence its importance is ambiguous.

Daily duration of sunshine showed highly significantly negative effect on spore
concentration at RJ and R5 two days before the sampling and on riseability during the
preceding four to six days. However, the sunshine was significantly positive during the
sampling at R/ and R5 and during the preceding two to four days it was significantly
positively affecting the riseability. This partly confirms and partly neglects results of
Burns et al. (2022) who found a clear determination of 1. fraxineus spore release after
five-day-average net radiation and leaf moisture.

Air humidity was found to be an important factor through our sampling. At R5
the humid air was significantly determining the inoculum load after two to six days;
similarly, at RJ there was a significantly positive effect on inoculum concentrations
lagged by two to four days. The promoting effect of air humidity on the disease spread
and establishment has already been emphasised many times (Timmermann et al. 2011;
Hietala et al. 2013; Dvorak et al. 2016; Cermdkové et al. 2017; Chumanova et al.
2019; Volke et al. 2019). However, up to two days before the sampling, the air humid-
ity had an opposite effect at RJ and negatively affected also the riseability. This ambigu-
ity of the air humidity effect for sporulation emphasizes the necessity for more detailed
study of the ascospore discharge mechanism such as in Ingold (1999).

Wind speed proved to have a significantly negative influence on ascospore con-
centrations at RJ after two to six days. This effect is probably due to the desiccation
of leaf rachises and growing apothecia which are not able to mature and sporulate un-
der such conditions (Timmermann et al. 2011; Hietala et al. 2013). The significantly
negative effect of wind speed was also found to affect the sampling directly. Since the
wind speed is measured in horizontal direction, it is logical that strong wind cannot
be beneficial for vertical spore transfer. Horizontally, moving air masses are supposedly
continuously removing the local inoculum, which would otherwise reach higher air
levels by convection (Garratt 1994; Lacey and West 2006).

Air pressure did not show much importance in our experiment. It was a significant
parameter with negative influence on spore concentrations at R/ after two to four days.
At the same period and sampling height two other factors had positive influence on the

Vertical spread of Hymenoscyphus fraxineus propagules 243

spore concentrations: precipitations and air humidity. Naturally, rainy and humid weath-
er is characterized by low air pressure, which we assume to be a reason for this result.

From a methodological point of view, rotating arm spore trap ROTTRAP 52
proved to be a reliable tool for the detection of H. fraxineus inoculum. All samplers suc-
cessfully completed all ten 48h samplings at five sampling spots without any blackout
even in hot or rainy weather. This reliability has been improved compared to previous
experiments (Dvoidk et al. 2016, 2017; Cermakova et al. 2017). Rotorods combined
with quantitative real-time PCR detection and quantification of H. fraxineus in the air
continuously produce robust and valuable data (Chandelier et al. 2014, Dvorak et al.
2016; Cermakova et al. 2017). Similarly, in our study all samplings yielded positive
results. On the other hand, the possibility of overloading by other, more concentrated
particles (in this study’s case it was dust) must be always taken into consideration in
experiment planning and interpretation of the results.

Conclusion

Our study revealed the permanent presence of the H. fraxineus inoculum up to 48
meters above the ground during the whole sampling period. Its concentration is con-
tinuously changing depending on previous and current weather, and decreases with
height. It poses a persistent threat to ash trees, either at local or landscape scale. ‘This
finding supports a sceptical outlook for the future of ashes in European forests, but
also confirms the important role of high height air sampling of the propagules of this
invasive alien pathogen to ensure its reliable monitoring.

Acknowledgements

This work was funded by a HORIZON 2020 project (agreement no. 771271): Ho-
listic Management of Emerging Forest Pests and Diseases (HOMED). We would like
to thank to EVUROPASTA SE company for cooperation and full access to their silo
in BorSov nad Vltavou, namely to the director Ing. Josef Maska, Jaroslav Mrkvicka,
Ladislav Hakl and Jaroslav Hnili¢ka. Furthermore, we thank Dr. Leticia Botella
Sanchez for preparing the plasmid for absolute quantification.

References

Aguayo J, Husson C, Chancerel E, Fabreguettes O, Chandelier A, Fourrier-Jeandel C, Dupuy
N, Dutech C, Ioos R, Robin C, Thibaudon M, Margais B, Desprez-Loustau ML (2021)
Combining permanent aerobiological networks and molecular analyses for large-scale sur-
veillance of forest fungal pathogens: A proof-of-concept. Plant Pathology 70(1): 181-194.
https://doi.org/10.1111/ppa.13265

244 Milon Dvoiak et al. / NeoBiota 84: 231-246 (2023)

Aylor DE (1999) Biophysical scaling and the passive dispersal of fungus spores: Relationship to
integrated pest management strategies. Agricultural and Forest Meteorology 97(4): 275—
292. https://doi.org/10.1016/S0168-1923(99)00072-6

Aylor DE, Taylor GS, Raynor GS (1982) Long-range transport of tobacco blue mold spores. Ag-
ricultural Meteorology 27 (3-4): 217-232. https://doi.org/10.1016/0002-1571(82)90007-3

Burns P, Timmermann V, Yearsley JM (2022) Meteorological factors associated with the timing
and abundance of Hymenoscyphus fraxineus spore release. International Journal of Biomete-
orology 66(3): 493-506. https://doi.org/10.1007/s00484-021-0221 1-z

Cermakova V, Kudlaéek T, Rotkova G, Rozsypalek J, Botella L (2017) Hymenoscyphus fraxineus
mitovirus | naturally disperses through the airborne inoculum of its host, Hymenoscyphus
fraxineus, in the Czech Republic. Biocontrol Science and Technology 27(8): 992-1008.
https://doi.org/10.1080/09583157.2017.1368455

Chandelier A, André F, Laurent F (2010) Detection of Chalara fraxinea in common ash (Fraxi-
nus excelsior) using real time PCR. Forest Pathology 40(2): 87-95. https://doi.org/10.1111/
j.1439-0329.2009.00610.x

Chandelier A, Helson M, Dvorak M, Gischer F (2014) Detection and quantification of air-
borne inoculum of Hymenoscyphus pseudoalbidus using real-time PCR assays. Plant Pathol-
ogy 63(6): 1296-1305. https://doi.org/10.1111/ppa.12218

Chandelier A, Gerarts K San Martin G, Herman M, Delahaye L (2016) Temporal evolution
of collar lesions associated with ash dieback and the occurrence of Armillaria in Belgian
forests. Forest Pathology 46(4): 289-297. https://doi.org/10.1111/efp.12258

Chumanova E, Romportl D, Havrdova L, Zahradnik D, Pe’kova V, Cerny K (2019) Predicting
ash dieback severity and environmental suitability for the disease in forest stands. Scandinavian
Journal of Forest Research 34(4): 254-266. https://doi.org/10.1080/02827581.2019.1584638

Cleary MR, Daniel G, Stenlid J (2013) Light and scanning electron microscopy studies of the
early infection stages of Hymenoscyphus pseudoalbidus on Fraxinus excelsior. Plant Pathology
62(6): 1294-1301. https://doi.org/10.1111/ppa.12048

Dhingra OD, Sinclair JB (2017) Basic plant pathology methods. CRC Press, 434 pp. https://
doi.org/10.1201/9781315138138

Dvoiak M (2022) Detection of Airborne Inoculum of Hymenoscyphus fraxineus: The Causal
Agent of Ash Dieback. In: Luchi N (Ed.) Plant Pathology. Methods in Molecular Biology,
vol 2536. Humana, New York, 119-137. https://doi.org/10.1007/978-1-0716-2517-0_8

Dvoiak M, Rotkova G, Botella L (2016) Detection of airborne inoculum of Hymenoscyphus
fraxineus and H. albidus during seasonal fluctuations associated with absence of apothecia.
Forests 7(12): 1-13. https://doi.org/10.3390/£7010001

Dvorak M, Janos P, Botella L, Rotkova G, Zas R (2017) Spore dispersal patterns of Fusarium
circinatum on an infested monterey pine forest in North-Western Spain. Forests 8(11):
432. https://doi.org/10.3390/£8110432

EPPO (2022) Hymenoscyphus fraxineus: Distribution. EPPO Global Database. https://gd.eppo.
int/taxon/CHAAFR/distribution [July 14, 2022]

Fones HN, Mardon C, Gurr SJ (2016) A role for the asexual spores in infection of Fraxinus
excelsior by the ash-dieback fungus Hymenoscyphus fraxineus. Scientific Reports 6(1): 1-10.
https://doi.org/10.1038/srep34638

Vertical spread of Hymenoscyphus fraxineus propagules 245

Garratt JR (1994) Review: The atmospheric boundary layer. Earth-Science Reviews 37(1-2):
89-134. https://doi.org/10.1016/0012-8252(94)90026-4

Grosdidier M, loos R, Margais B (2018a) Do higher summer temperatures restrict the dis-
semination of Hymenoscyphus fraxineus in France? Forest Pathology 48(4): 1-8. https://doi.
org/10.1111/efp.12426

Grosdidier M, loos R, Husson C, Cael O, Scordia T, Margais B (2018b) Tracking the invasion:
Dispersal of Hymenoscyphus fraxineus airborne inoculum at different scales. FEMS Micro-
biology Ecology 94(5): fiy049. https://doi.org/10.1093/femsec/fry049

Gross A, Zaffarano PL, Duo A, Griinig CR (2012) Reproductive mode and life cycle of the
ash dieback pathogen Hymenoscyphus pseudoalbidus. Fungal Genetics and Biology 49(12):
977-986. https://doi.org/10.1016/j.fgb.2012.08.008

Gross A, Holdenrieder O, Pautasso M, Queloz V, Sieber TN (2014) Hymenoscyphus pseudoal-
bidus, the causal agent of European ash dieback. Molecular Plant Pathology 15(1): 5-21.
https://doi.org/10.1111/mpp.12073

Hanadtkova Z, Koukol O, Cmokova A, Zahradnik D, Havrdovaé L (2017) Direct evidence of
Hymenoscyphus fraxineus infection pathway through the petiole-shoot junction. Forest Pa-
thology 47(6): 1-6. https://doi.org/10.1111/efp.12370

Hietala AM, Timmermann V, B@rja I, Solheim H (2013) The invasive ash dieback path-
ogen Hymenoscyphus pseudoalbidus exerts maximal infection pressure prior to the onset
of host leaf senescence. Fungal Ecology 6(4): 302-308. https://doi.org/10.1016/j.fune-
co.2013.03.008

Hirst JM (1952) An automatic volumetric spore trap. Annals of Applied Biology 39(2): 257—
265. https://doi.org/10.1111/j.1744-7348.1952.tb00904.x

Husson C, Caél O, Grandjean JP, Nageleisen LM, Marcais B (2012) Occurrence of Hymenos-
cyphus pseudoalbidus on infected ash logs. Plant Pathology 61(5): 889-895. https://doi.
org/10.1111/j.1365-3059.2011.02578.x

Ingold CT (1999) Active liberation of reproductive units in terrestrial fungi. The Mycologist
13(3): 113-116. https://doi.org/10.1016/S0269-9 15X(99)80040-8

Jankovsky L, Holdenrieder O (2009) Short communication Chalara fraxinea - Ash die-
back in the Czech Republic. Plant Protection Science 45(2): 74-78. https://doi.
org/10.17221/45/2008-PPS

Kowalski T (2006) Chalara fraxinea sp. nov. associated with dieback of ash (Fraxinus excel-
sior) in Poland. Forest Pathology 36(4): 264-270. https://doi.org/10.1111/j.1439-
0329.2006.00453.x

Kowalski T, Holdenrieder O (2009) Pathogenicity of Chalara fraxinea. Forest Pathology 39(1):
1-7. https://doi.org/10.1111/j.1439-0329.2008.00565.x

Krautler K, Treitler R, Kirisits T (2015) Hymenoscyphus fraxineus can directly infect intact cur-
rent-year shoots of Fraxinus excelsior and artificially exposed leaf scars. Forest Pathology
45(4): 274-280. https://doi.org/10.1111/efp.12168

Lacey ME, West JS (2006) The Air Spora — A Manual for Catching and Identifying Airborne
Biological Particles. Springer, 156 pp. https://doi.org/10.1007/978-0-387-30253-9

McCartney HA, Fitt BDL, Schmechel D (1997) Sampling bioaerosols in plant pathology. Jour-
nal of Aerosol Science 28(3): 349-364. https://doi.org/10.1016/S0021-8502(96)00438-7

246 Milon Dvoiak et al. / NeoBiota 84: 231-246 (2023)

Morgan JA, Tatar JF (1972) Calculation of the Residual Sum of Squares for all Possible Regres-
sions. Technometrics 14(2): 317-325. https://doi.org/10.1080/00401706.1972.10488918

Noll KE (1970) A rotary inertial impactor for sampling giant particles in the atmosphere.
Atmospheric Environment (1967) 4: 9-19.

Oteros J, Garcia-Mozo H, Alcazar P, Belmonte J, Bermejo D, Boi M, Carifianos P, Diaz de la
Guardia C, Fernandez-Gonzalez D, Gonzalez-Minero F, Gutiérrez-Bustillo AM, More-
no-Grau S, Pérez-Badia R, Rodriguez-Rajo FJ, Ruiz-Valenzuela L, Sudrez-Pérez J, Trigo
MM, Dominguez-Vilches E, Galan C (2015) A new method for determining the sources
of airborne particles. Journal of Environmental Management 155: 212-218. https://doi.
org/10.1016/j.jenvman.2015.03.037

Oteros J, Valencia RM, del Rio S, Vega AM, Garcia-Mozo H, Galan C, Gutiérrez P, Man-
drioli P, Fernandez-Gonzalez D (2017) Science of the Total Environment Concentric Ring
Method for generating pollen maps. Quercus as case study. The Science of the Total Envi-
ronment 576: 637-645. https://doi.org/10.1016/j.scitotenv.2016.10.121

Perkins WA, Leighton PA (1957) The rotorod sampler. Second Semi-Annual Report No. CML
186, Aerosol Laboratory.

Przybyt K (2002) Fungi associated with necrotic apical parts of Fraxinus excelsior shoots. Forest
Pathology 32(6): 387-394. https://doi.org/10.1046/j.1439-0329.2002.00301.x

R Core Team (2018) R: A language and environment for statistical computing. https://www.r-
project.org/

Stroheker S, Queloz V, Nemesio-Gorriz M (2021) First report of Hymenoscyphus fraxineus caus-
ing ash dieback in Spain. New Disease Reports 44(2): 10-12. https://doi.org/10.1002/
ndr2.12054

Timmermann V, Berja I, Hietala AM, Kirisits T, Solheim H (2011) Ash dieback: Pathogen
spread and diurnal patterns of ascospore dispersal, with special emphasis on Norway. Bul-
letin OEPP. EPPO Bulletin. European and Mediterranean Plant Protection Organisation
41(1): 14-20. https://doi.org/10.1111/j.1365-2338.2010.02429.x

Vasaitis R, Enderle R (2017) Dieback of European Ash (Fraxinus spp.) -Consequences and
Guidelines for Sustainable Management The Report on European Cooperation in Science
&amp; Technology (COST) Action FP1103 FRAXBACK, 301 pp.

Volke V, Knapp S, Roloff A (2019) Survey of Hymenoscyphus fraxineus in a central European
urban area and exploration of its possible environmental drivers. Urban Forestry & Urban
Greening 40: 165-173. https://doi.org/10.1016/j.ufug.2018.05.013

Zhao YJ, Hosoya T, Baral HO, Hosaka K, Kakishima M (2012) Hymenoscyphus pseudoalbidus,
the correct name for Lambertella albida reported from Japan. Mycotaxon 122(1): 25-41.
https://doi.org/10.5248/122.25