Roost Characteristics of a Tricolored Bat Perimyotis subflavus in the Missouri Ozarks

Abstract

The tricolored bat (Perimyotis subflavus) once commonly occurred throughout the eastern and central United States, but is now experiencing range-wide population declines primarily due to white-nose syndrome and forest fragmentation. Conservation efforts for this species require more information regarding summer maternity roost characteristics, especially in the Ozark Highlands of Missouri, where the species has declined precipitously. Herein we report the capture and tracking of one female tricolored bat to two summer roosts in southeastern Missouri. Tree roosts differed by their substrates, as one was within a cluster of dead leaves on a snag and the other was located under exfoliating bark; otherwise, the roosts and surrounding vegetation were similar. Both roosts were located in large (>30 cm diameter breast height) oak (Quercus spp.) trees under a dense canopy of trees with similar basal areas. While our capture success was limited over two years, details regarding tricolored bat roost ecology are imperative for the species’ conservation and management.

Full text

Roost characteristics of a tricolored bat Perimyotis subavus in the
Missouri Ozarks
1,2
1
1. Department of Biology, Pittsburg State University, Pittsburg, Kansas
cbrodsky@pittstate.edu
2. U.S. Forest Service, Spearsh, South Dakota
3. U.S. Fish and Wildlife Service, Missouri Ecological Services Field Oce, Columbia,
Missouri
The tricolored bat (Perimyotis subavus) once commonly occurred throughout the
eastern and central United States, but is now experiencing range-wide population declines
primarily due to white-nose syndrome and forest fragmentation. Conservation eorts for
this species require more information regarding summer maternity roost characteristics,
especially in the Ozark Highlands of Missouri, where the species has declined
precipitously. Herein we report the capture and tracking of one female tricolored bat to
two summer roosts in southeastern Missouri. Tree roosts diered by their substrates,
as one was within a cluster of dead leaves on a snag and the other was located under
exfoliating bark; otherwise, the roosts and surrounding vegetation were similar. Both
roosts were located in large (>30 cm diameter breast height) oak (Quercus spp.) trees
under a dense canopy of trees with similar basal areas. While our capture success was
limited over two years, details regarding tricolored bat roost ecology are imperative for
the species’ conservation and management.
Keywords: conservation, forest, Quercus, roost ecology, tricolored bat


Vol. 125, no. 3-4
p.159-164 (2022)

The tricolored bat (Perimyotis subavus)
is a tree-roosting species that ranges
throughout the eastern and central United
States (Barbour and Davis 1969). Impacts of
white-nose syndrome (WNS) and forest loss
have signicantly aected many bat species,
especially the tricolored bat (Frick, Kingston,
and Flanders 2020). Range-wide declines
observed throughout North America resulted in
the U.S. Fish and Wildlife Service (USFWS)
to propose to list the tricolored bat as an
endangered species under the Endangered
Species Act of 1973 (USFWS 2022). Due to
the population status of the tricolored bat, it is
imperative to understand the species’ habitat
needs, especially maternity roost features, in
order to develop conservation strategies and
maintain critical habitats necessary for their
reproductive eorts.
Summer maternity roosts used by tricolored
bats include solitary or small colonies of bats
roosting in vegetation clusters in live and
dead hardwood species (e.g., oak Quercus
spp.), and occasionally tree cavities (Menzel
et al. 1999; Veilleux, Whitaker, and Veilleux
2003; Perry and Thill 2007; Shute et al.
2021). Vegetation characteristics identied
in areas occupied by tricolored bats include
those with taller and larger trees, high basal
areas, and within riparian areas (Ford et al.
2005; O’Keefe et al. 2009). While roost
descriptions throughout this species’ range are
informative, the literature lacks a description
of tricolored bat roost characteristics in the
Ozark Highlands of Missouri (Chapman et al.
2002). Additionally, maternity colonies have
shown high site delity (Veilleux, Whitaker,
and Veilleux 2003); therefore, knowledge of
roost characteristics of local tricolored bats will
inform future natural resource management and
the conservation of the species in Missouri and
the Ozark Highlands region.
This study evaluated post-WNS maternity
ecology of tricolored bats in southeastern

160 Hammesfahr et al.
Missouri. Since the introduction of WNS
to Missouri in 2012, tricolored bats have
declined by 53.2% according to statewide
winter hibernacula surveys (Colatskie
2017). Additionally, forests in southeastern
Missouri are managed for a variety of
purposes, including logging and wind energy
development, two habitat disturbances known
to negatively impact bat species (Yates and
Muzika 2006; Frick, Kingston, and Flanders
2020). Our goal was to track and describe
tricolored bat roosts utilized during the
maternity season, and surrounding forest
features. Results of this study will help provide
a better understanding of roosting requirements
of this species of conservation concern, with
the goal of informing state agencies and land
managers about critical forest habitats needed
for this species.

Our study area consisted of seven Missouri
Department of Conservation (MDC)
conservation areas in southeast Missouri (i.e.
Shannon, Carter, and Reynolds counties, Fig.
1). We collected data during the bat maternity
season of May – August in 2018 and 2019, for
47 mist-net nights (23 in 2018 and 24 in 2019).
Mist-netting for tricolored bats occurred at 12
locations, located at least one mile apart. Sites
were visited up to three times per year, as we
revisited sites that had successful captures
of tricolored bats with the goal of increasing
capture rates of the target species. Mist-net
surveys began at sunset and lasted for ve
hours (USFWS 2019). We positioned mist-nets
(38 mm polyester mesh nets; Avinet Research
Supplies, Maine, United States) across and
within forested waterbodies (i.e. ponds,
ephemeral, and perennial streams), resulting
in the optimal mist-net coverage for each site.
We separated mist-net attempts at a study site
by at least one week (Kunz and Parsons 2009).
We followed procedures and recommendations
for wildlife handling under the most current
version of the USFWS Range-wide Indiana
Bat Survey Protocol (USFWS 2019) and
Figure 1. Map of Missouri Department of Conservation (MDC) Conservation Areas and
three Missouri counties included in the study. The male tricolored bat was captured in
Shannon County, while the female was captured in Carter County. Both roosts were
located in Carter County, near the capture location.

 125(3-4), 2022 161
through the MDC’s Institutional Animal Care
and Use Committee. Additionally, we adhered
to the most recent WNS decontamination
protocol to prevent the spread of WNS (WNS
Decontamination Team, 2018).
When a tricolored bat female was captured,
we glued a 0.27 g radio-transmitter (Holohil
Systems Ltd., Ontario, Canada) between the
individual’s scapula. In adherence to USFWS
guidelines, we insured that the transmitter and
surgical application glue weighed less than
5% of the captured bats’ weight (USFWS
2019). We attempted to track the radio-tagged
individual to diurnal roosts for up to fourteen
days (i.e. the maximum lifespan of transmitter
batteries) or until the transmitter fell o. We
listened to the bat’s frequency during the day
with a three-element Yagi antenna to locate
their diurnal roost.
Once we located a roost, we collected
vegetation data within an 11.3 m radius around
the roost tree (James and Shugart 1970). At
the roost tree, we described its decay score
(i.e. value of one [live tree] to nine [stump],
Bartels et al. 1985) and substrate type (i.e.
cavity, crevice, exfoliating bark, coniferous
leaf cluster, squirrel nest, deciduous leaf
cluster, or roost suspended in hanging branch).
We measured canopy cover with a spherical
convex densiometer at the roost and 5 m from
the roost (Forestry Suppliers, Inc., Jackson,
MS) and roost tree height with a digital
rangender (Forestry Pro Laser Rangender,
Nikon, Melville, NY). For live trees and snags
surrounding the roost, we grouped each tree’s
measured values (i.e. total counts, height) into
size classes based on their diameter at breast
height (DBH), including: saplings (DBH 5
– 14.5 cm), poles (DBH 14.6 – 27 cm), and
sawtimber (DBH > 27.1 cm). We measured tree
DBH with a diameter tape (Forestry Suppliers,
Inc., Jackson, MS). The same size classes
were used for deciduous and conifer trees. We
estimated basal area of live trees and snags
with a 10-factor prism (Forestry Suppliers,
Inc., Jackson, MS). For each roost tree, we
attempted to monitor emergence to estimate the
number of bats utilizing the roost. We counted
the number of bats that emerged from the roost
from twenty minutes before sunset until ten
minutes after the last bat emerged or when
the contrast of the sky and forest dissipated
(USFWS 2019).

We captured two tricolored bat individuals: one
male (June 23, 2019; Birch Creek Conservation
Area) and one non-reproductive female (May
17, 2019; Peck Ranch Conservation Area; Fig.
1). No tricolored bats were captured in 2018.
We radio-tagged and tracked the female to two
roosts within large (i.e. > 30 cm DBH) oak trees.
The roost substrate diered between the two
roosts: Roost 1 was in a cluster of dead leaves
in a snag, while Roost 2 was under exfoliating
bark. Both roosts were under dense canopy
cover (i.e. > 95%) and surrounded by deciduous
trees with several sapling trees and no observed
conifers (Table 1). The heights of both roosts
fell within the average height for pole timber
within each plot, and the average height of the
trees around the roost tree were taller than the
roost’s height. Vegetation around the roosts
were similar, as they both resided within the
mid-canopy of the forest and had a basal area
between 11-14 m²/ha (Table 1).
We observed at least four individuals emerge
from Roost 1 (May 19, 2019), in addition to
the tagged female, from the same dead leaf
cluster near the top of the tree. During the
second observation period (May 20, 2019),
we observed at least seven individuals emerge
from Roost 1. We tracked the tagged female
to Roost 2 on May 22, 2019 after the previous
evening’s heavy rain event (7.6 cm). We were
unable to perform an emergence count at this
location due to other research priorities. The
next day, the female left Roost 2 and returned
to Roost 1 until we could not pick up the
battery’s transmission, which occurred on
June 1, 2019. The female ew 2.4 km from the
capture location to Roost 1, and then 1.5 km
to Roost 2. Roost 2 was closer to the capture
location (0.4 km).

We described the rst documented summer
roosts of a tricolored bat within the Ozarks
Highlands of southeastern Missouri, providing

162 Hammesfahr et al.
important details concerning roosting habitats
of this species’ population. Both roosts used
by the tracked tricolored bat were oaks, a
tree commonly preferred for roosts by this
species throughout their range (Menzel et
al. 1999; Veilleux, Whitaker, and Veilleux
2003). Tricolored bats were also reported to
use maples (Acer spp. in Indiana, Veilleux,
Whitaker, and Veilleux 2003), sweetgum
(Liquidambar stryaciua in South Carolina,
Shute et al. 2021), and conifers for roosts,
albeit less commonly (e.g., three roosts located
in Pinus echinata vs. 37 roosts in oak species,
reported Perry and Thill 2007). Roost 1’s
substrate material (i.e. leaf clusters, Menzel
et al. 1999; Veilleux, Whitaker, and Veilleux
2003; Perry and Thill 2007) and roost tree
diameter (Veilleux, Whitaker, and Veilleux
2003; Shute et al. 2021) were similar to other
reported studies, although Roost 2’s roost
substrate did not t the published descriptions
for maternity roosts, as it was a snag without
leaves. We were only able to nd one study
that described the use of a cavity as a possible
maternity roost (i.e. Menzel 1996), while
non-reproductive tricolored bats have been
reported to use hollow snags, dead tree limbs,
and bark as winter roosts (Newman, Loeb,
and Jachowski 2021). We recommend that
additional details of the roost tree, such as
decay status, be reported when describing the
roosts of this species.
Vegetation and forest characteristics
surrounding each roost were similar to one
another and typical of the vegetation in
the Ozark Highlands of Missouri. Forest
characteristics surrounding our two tricolored
bat roosts resembled those described
throughout the species range (e.g., Menzel
et al. 1999; Veilleux, Whitaker, and Veilleux
2003; Perry and Thill 2007; Shute et al.
2021), as this species typically roosts in dense
hardwood forests. Even so, Roosts 1 and 2
were located in forest stands with greater
canopy coverage (> 95%, Table 1) compared
to studies in South Carolina (i.e. 85% canopy
closure, Shute et al. 2021), Arkansas (i.e.
70.3%, Perry and Thill 2007), and Indiana (i.e.
41.1%, Veilleux, Whitaker, and Veilleux 2003).
Additionally, stand densities surrounding our
two roosts were greater than other reports
(Table 1; Shute et al. 2021). These stand
characteristics are important to recognize,
as forest fragmentation and thinning makes
tricolored bats less likely to utilize forested
habitats (Farrow and Broders 2011), potentially
due to the lack of mature trees for roosts.
Female tricolored bats typically use multiple
Table 1. Roost structure and vegetation
characteristics surrounding roost trees (11.3
m radius). Roost tree decay scores can
range from a value of 1 (live) to 9 (stump)
(Bartels et al. 1985). Standard deviations
follow averaged values.
Forest Characteristic Roost 1 Roost 2
Tree species Quercus alba Quercus spp.
Roost substrate Deciduous leaf
cluster
Exfoliating
bark
DBH (cm) 47.5 30.2
Roost tree decay
score
3 (Dead) 6 (Broken)
Roost tree height (m) 32.7 10.6
Roost height (m) 21.9 6.6
Canopy coverage at
roost
95.83% ± 3.70 97.65% ± 0.99
Canopy coverage 5
m from roost
95.57% ± 3.84 98.44% ± 0.60
Total deciduous
saplings
23 9
Total deciduous
poles
6 8
Total deciduous
sawtimber
5 3
Total conifer saplings 0 0
Total conifer poles 0 0
Total conifer
sawtimber
0 0
Total snag saplings 2 2
Total snag poles 1 0
Total snag sawtimber 0 0
Mean sapling height
(m)
13.53 ± 11.27 4.21 ± 2.97
Mean pole height (m) 26.88 ± 11.47 14.68 ± 6.79
Average sawtimber
height (m)
41.1 ± 15.92 22.2 ± 4.33
Snag basal area
m²/ha
0 0
Live tree basal area
m²/ha
11.47 13.76

 125(3-4), 2022 163
roost trees during maternity season, yet the
species has high site delity to their roost
habitat during the summer season across
multiple years (Veilleux, Whitaker, and
Veilleux 2003; Veilleux and Veilleux 2004). We
recorded the tagged female using two roosts
located 1.5 km apart, switching roosts after
approximately four days. We never recaptured
our tagged female to determine if she used
more roost trees for the remainder of the
season. Veilleux and Veilleux (2004) reported
roost tree delity of four female tricolored
bats to range between one to six roost trees
per summer season, spending between 1.2 to
seven days at each roost before switching. Of
the females that switched roosts, distances
between each roost ranged from 13 – 857 m in
one season (Veilleux and Veilleux 2004). We
hypothesized that heavy precipitation overnight
(7.6 cm) caused our radio-tagged female to
search for a less exposed roost when switching
to Roost 2. What remains unclear is why the
female used Roost 2 when it was located so far
from Roost 1, as the travel distance exceeded
reported summer roost-to-roost distances
in Veilleux and Veilleux (2004). However,
other studies of non-reproductive tricolored
bats describe large travel distances between
roosts and foraging locations (Krishon et al.
1997) and distances between winter roost
trees (Newman, Loeb, and Jachowski 2021),
indicating that seasonality and reproductive
status may be important factors in how far
a tricolored bat may travel in a short period.
Distances between roosts and the types of trees
selected for roosting habitats are important to
understand, especially in landscapes with high
rates of forest management, development, and
resulting fragmentation.
We still lack information regarding the roosting
ecology for tricolored bats within southeastern
Missouri. The unfortunate reality is that
individuals are so rare in the Missouri Ozarks
that gathering large datasets to determine
average roost characteristics is challenging.
The limited published information on tricolored
bat roost ecology post-WNS provokes inquiries
whether the studies were not published due
to small sample sizes, or if researchers failed
to capture these species and thus were unable
to study their roosting ecology. While our
capture success was limited, details regarding
tricolored bat roost ecology are imperative for
the species’ conservation in times of range-
wide population declines. This discovery of
the rst summer roosts of tricolored bats in the
Missouri Ozarks provides insight into potential
roost habitat use in southeastern Missouri and
warrants additional research eorts.

The MDC provided the funding for this
research. A generous Pittsburg State University
anonymous donor supported three intern salaries
during the project. Mist-netting was permitted
under the federal permit TE61451C-1 and
Missouri Wildlife Collector Permits (17893,
18119, and 18685 for 2018-2020, respectively).
Research assistants who helped with this study’s
eldwork included Jena Staggs, Monae Taylor,
Brandi Christiano, Maggie Murray, Jack Varallo,
Ryan McGinty, Michael Barnes, Zack Telford,
and Kennedy Johnson.

Barbour, T.W. and Davis, W.H. 1969. Bats of
America. University Press of Kentucky,
Lexington, Kentucky, 312 pp.
Bartels, R., Dell, J.D., Knight, R.L., and
Schaefer, G. 1985. Dead and down woody
material. In: Brown, E.R., technical editor,
Management of wildlife and sh habitats in
forests of western Oregon and Washington.
Part I - Chapter narratives. PNW R6-
F&WL-192-1985. USDA Forest Service,
Portland, Oregon, pp. 171-186.
Chapman, S.S., Omernik, J.M., Grith, G.E.,
Schroeder, W.A., Nigh, T.A., and Wilton,
T.F. 2002. Ecoregions of Iowa and Missouri,
Reston, Virginia, U.S. Geological Survey
(map scale 1:1,800,000). [https://clicktime.
symantec.com/15uBY2Lf9buyxNh6dhoW
m?h=EMGyQorDw1EtLoa4bKbayVIyVb
P74ne6BdqJ7-WsI-I=&u=https://clicktime.
symantec.com/15uBY2LiDbGTUZmuV
W7X9?h=PW2rRKlqQtm2mhplwgIEaM
z7qyBICK5JlbPBXAGK7pg=&u=http://
ecologicalregions.info/data/mo/moia_front.
pdf]. Date accessed [Aug. 2022].

164 Hammesfahr et al.
Colatskie, S. 2017. Missouri bat hibernacula
survey results from 2011-2017, following
white-nose syndrome arrival. Missouri
Department of Conservation, Technical Brief,
14 pp.
Farrow, L.J. and Broders, H.G. 2011. Loss of
forest cover impacts the distribution of the
forest-dwelling tri-colored bat (Perimyotis
subavus). Mammalian Biology 76(2):172-
179.
Ford, W.M., Menzel, M.A., Rodrigue, J.L.,
Menzel, J.M., and Johnson, J.B. 2005.
Relating bat species presence to simple
habitat measures in a Central Appalachian
Forest. Biological Conservation 126:528–539.
Frick, W.F., Kingston, T., and Flanders, J.
2020. A review of the major threats and
challenges to global bat conservation.
Annals of the New York Academy of
Sciences 1469(1):5-25.
James, F.C. and Shugart, H.H. 1970. A
quantitative method of habitat description.
Audubon Field Notes 24:727–736.
Krishon, D.M., Menzel, M.A., Carter, T.C.,
and Laerm, J. 1997. Notes on the home
range of four species of Vespertilionid bats
(Chiroptera) on Sapelo Island, Georgia.
Georgia Journal of Science 55:215-223.
Kunz, T.H. and Parsons, S. 2009. Ecological
and behavioral methods for the study
of bats. The John Hopkins University,
Baltimore, Maryland, 920 pp.
Menzel, M.A. 1996. Possible use of a tree
cavity as a maternity roost by the eastern
pipistrelle bat Pipistrellus subavus. Bat
Research News 37:115–116.
Menzel, M.A., Krishon, D.M., Carter, T.C.,
and Laerm, J. 1999. Notes on tree roost
characteristics of the northern yellow bat
(Lasiurus intermedius), the Seminole bat
(L. seminolus), the evening bat (Nycticeius
humeralis), and the eastern pipistrelle
(Pipistrellus subavus). Florida Scientist
62(3/4):185-193.
Newman, B.A., Loeb, S.C., and Jachowski,
D.S. 2021. Winter roosting ecology of
tricolored bats (Perimyotis subavus) in
trees and bridges. Journal of Mammalogy
102:1331-1341.
O’Keefe, J.M. Loeb, S.C., Lanham, J.D.,
and Hill, H.S. 2009. Macrohabitat factors
aect day roost selection by eastern red
bats and eastern pipistrelles in the Southern
Appalachian Mountains, USA. Forest
Ecology and Management 257:1757–1763.
Perry, R.W. and Thill, R.E. 2007. Tree roosting
by male and female eastern pipistrelles in a
forested landscape. Journal of Mammalogy
88:974–981.
Shute, K.E., Loeb, S.C., and Jachowski, D.S.
2021. Summer roosting ecology of the
northern yellow bat and tri-colored bat
in coastal South Carolina. Southeastern
Naturalist 20(3):459-476.
United States Fish and Wildlife Service
(USFWS). 2019. Range-wide Indiana Bat
Survey Guidelines, 62 pp.
United States Fish and Wildlife Service
(USFWS). 2022. Service proposes to list
the tricolored bat as Endangered under the
Endangered Species Act. Press release, 13
September 2022.
Veilleux, J.P., Whitaker, J.O., and Veilleux,
S.L. 2003. Tree-roosting ecology of
reproductive female eastern pipistrelles,
Pipistrellus subavus, in Indiana. Journal of
Mammalogy 84:1068–1075.
Veilleux, J.P. and Veilleux, S.L. 2004. Intra-
annual and interannual delity to summer
roost areas by female eastern pipistrelles,
Pipistrellus subavus. The American
Midland Naturalist 152(1):196-200.
WNS Decontamination Team. 2018. National
white-nose syndrome decontamination
protocol. White Nose Syndrome Response
Team, 7 pp.
Yates, M.D. and Muzika, R.M. 2006. Eects
of forest structure and fragmentation on site
occupancy of bat species in Missouri Ozark
forests. Journal of Wildlife Management
70(5):1238-248.