JIMENEZ-THESIS-2016.Pdf (685.2Kb)

IDENTIFYING AND CHARACTERIZING ROOSTS OF SOUTHERN AND NORTHERN

YELLOW BATS (LASIURUS EGA AND LASIURUS INTERMEDIUS)

A Thesis

Presented to the

Faculty of the College of Graduate Studies of

Angelo State University

In Partial Fulfillment of the

Requirements for the Degree

MASTER OF SCIENCE

PATRICIA CITLALLY JIMENEZ

May 2016

Major: Biology

IDENTIFYING AND CHARACTERIZING ROOSTS OF SOUTHERN AND NORTHERN

YELLOW BATS (LASIURUS EGA AND LASIURUS INTERMEDIUS)

PATRICIA CITLALLY JIMENEZ

APPROVED:

Dr. Loren K. Ammerman

Dr. Robert C. Dowler

Dr. Ben R. Skipper

Dr. Biqing Huang

April 5, 2016

APPROVED:

Dr. Susan E. Keith Date Dean, College of Graduate

DEDICATION

This thesis is dedicated to my family, my future husband James Kiser, and my forever adorable yellow bats; “I can do all things through Him who gives me strength.”

Using palm fronds as roosts,

Yellow bats await.

Hide-and-seek on the loose,

Is the game that they play.

iii

ACKNOWLEDGMENTS

I wish to start by thanking my thesis committee. I thank Dr. Ammerman for her never

ending patience with my naivety and kookiness throughout this project, for her determination

and knowledge to mold my skills to become a good researcher, and for teaching me how a

strong work ethic, perseverance and a little creativity can lead to success. I thank Dr. Dowler

for his reassurances and for always ensuring I produced quality work. I thank Dr. Skipper for

being the best committee cheerleader a graduate student could ever hope for; without his

guidance, understanding, and positive encouragement, I would still be stumbling through this

project. And lastly, I’d like to thank Dr. Huang, who assured that I was never beat too badly

by the rest of my committee.

I also want to express my appreciation to those who graciously shared their wisdom

and resources when I asked for help: Roger Rodriguez for providing contacts for sites where

I could conduct my research, to Pablo Deyturbe and Stephanie Galla from Resaca de la

Palma State Park, for granting the time and space to search for these yellow bats; John

Karges and Max Pons from The Nature Conservancy for his knowledge and help finding the bats and identifying some key characteristics of yellow bat roosts; to the managers and staff of Sabal Palm Sanctuary, especially Seth Patterson and Guillermo Aguilar, for their invaluable willingness to open their doors to us and offering their time to facilitate this project; to Dr. Negovetich for advising in the initial understanding of the confusing world of statistics; and to Dr. Dixon, for his guidance and ‘outside-the-box’ thinking that spurred different aspects of this project.

I am indebted to my funding sources that permitted every facet of this project: Head of the River Ranch Grant, Angelo State University’s Graduate Research Fellowships, and the

Bob Berry Holohil Grant from the Western Bat Working Group.

But I also would not have been able to complete any data-gathering without the priceless time given, (and sometimes suffered), by my volunteers. Words cannot describe how thankful I feel towards you, my sweet friends, who were willing to withstand long hours in the hot and humid, mosquito and hornet’s nest ridden, and sometimes bat-less nights during my field season. Thank you Max Abrahamson, Loren Ammerman, Shelby Bessette,

Kylie Vincent Briggs, Gwyn Carmean, Linette Castañeda, Krysta Demere, Catheline

Froehlich, Stephanie Galla, Malorri Hughes, Becca Thomas-Kuzilik, Stephanie Martinez,

Isidro Montemayor, Seth Patterson, Max Pons, Greta Schmidt, Alison Shepherd, Ben

Skipper, Christrina Straway, Craig Tipton, Juliet Vallejo, Ullisa Uribe-Zepeda, and Jaime

Zepeda.

To all those aforementioned, I owe a debt of gratitude for your encouragement and support throughout the journey to the completion of this project.

ABSTRACT

Previous research has demonstrated the southern yellow bat (Lasiurus ega) to roost in

the dead fronds of native palms (Sabal mexicana) and non-native palms (Washingtonia

robusta). Roost use by the northern yellow bat (L. intermedius) is similar, with the addition

of Spanish moss (Tillandsia spp.). Quantitative assessments of these roosting substrates,

however, are lacking. My objective was to identify and quantitatively characterize the diurnal

roosts of L. ega and L. intermedius in the Lower Rio Grande Valley of Texas. Through radiotelemetry from May – November of 2015, I located a total of 20 roosts in S. mexicana palms used by 8 yellow bats. Comparison of characteristics between roosts and randomly selected palms showed that yellow bats selected sabal palms with significantly taller, thicker frond skirts and smaller trunk diameters. A predictability model was subsequently constructed to aid in the management of roosting habitat for these species of yellow bats.

TABLE OF CONTENTS Page DEDICATION ...... iii

ACKNOWLEDGMENTS ...... iv

ABSTRACT ...... vi

TABLE OF CONTENTS ...... vii

LIST OF FIGURES ...... viii

LIST OF TABLES ...... ix

INTRODUCTION ...... 1

MATERIALS AND METHODS ...... 6

Study sites ...... 6

Sampling strategies ...... 6

Bat capture and radiotracking ...... 8

Roost site characterization ...... 9

Data analysis ...... 10

RESULTS ...... 13

DISCUSSION ...... 23

LITERATURE CITED ...... 28

APPENDICES ...... 36

BIOGRAPHY ...... 42

vii

LIST OF FIGURES

FIGURE 1. Proportion of bats submitted to the Texas Department of State Health Services annually that were a yellow bat species encountered in Texas. In total, 780 out of 18,342 submissions were yellow bats across all years (2005 – 2014)...... 3

FIGURE 2. Map of Cameron County (black outline) in Texas indicating the three study sites used to research the southern and northern yellow bats: Green – Resaca de la Palma, Orange – Southmost Preserve, Blue – Sabal Palm Sanctuary ...... 7

FIGURE 3. Diagram of measurements taken for each roost and random palm using a clinometer at ground level from 5 m away, or from a distance where palm was in visible. Total palm height was taken from the tallest live frond of the palm; the height at the top of the skirt was taken from the topmost dried frond of the skirt; the height of the bottom of the skirt was taken from the bottommost dried frond of the skirt, whose stem attachment to the trunk was measured. Thickness of the dried frond skirt was calculated by subtracting the bottommost dried frond height from the topmost dried frond ...... 11

FIGURE 4. Radio-tagged bats, 6 Lasiurus ega (LAEG) and 2 L. intermedius (LAIN) were tracked to 20 roosts in Sabal mexicana palms within Sabal Palm Sanctuary in Brownsville, Texas from May 2015 – November 2015 ...... 15

viii

LIST OF TABLES

TABLE 1. Summary of bat captures for three sites in Brownsville, Texas (Cameron County) over a total of 23 trapping nights, shown in parentheses for each site, from November 2014 – November 2015...... 14

TABLE 2. The sexes, number of roosts used, number of days tracked, mean number of days per roost of radio-tagged Lasiurus ega (LAEG) and L. intermedius (LAIN) at Sabal Palm Sanctuary, Brownsville, Texas...... 17

TABLE 3. Measurements of characteristics of roost and random Sabal mexicana palms used by both Lasiurus ega and L. intermedius. Numbers are means ± SD...... 20

TABLE 4. Comparison of candidate models for Lasiurus ega or L. intermedius roost selection in Sabal mexicana based on AIC (Akaike information criterion scores), AICc (second order scores), delta AICc (Δi), and model probabilities (ωi) ...... 21

TABLE 5. Top model (75% accuracy) including variables that best influenced Lasiurus ega or L. intermedius roost selection. The model was constructed using a logistic regression function ...... 22

INTRODUCTION

Texas has the greatest bat diversity in the United States (Ammerman et al. 2012).

Some species roost in caves, rock crevices, buildings, and old mines, while others roost in trees. Among the tree roosting bats are the northern (Lasiurus intermedius) and southern

(Lasiurus ega) yellow bats (Ammerman et al. 2012). Their name describes their appearance, as the yellow hue of their pelage matches the shade of one of their known roosts, dried (dead) palm fronds (Mirowsky 1997). Due to their similarity, these two species are difficult to distinguish from one another without anatomical measurements or genetic techniques, and this is especially true for juveniles of both species (Baker et al. 1988; Morales and Bickham

1995). On average, adult L. intermedius are larger in total body length (133 mm) than adult

L. ega (118mm; Ammerman et al. 2012).

Both species have broad geographic ranges, with the range of L. intermedius extending from the southeastern United States and overlapping with L. ega through eastern

Mexico and Central America; additionally, L. ega is found across South America (Webster et al. 1980; Kurta and Lehr 1995). Where their distributions overlap in southern Texas, these species are thought to cohabitate in similar roosts (Spencer et al. 1988; Chapman and

Chapman 1990). In recent years, research on roosting ecology of bats has highlighted the importance of roosts for the lifecycle of bats and the need to conserve habitat for the residing species (Findley 1993; Kunz and Fenton 2003; Lacki et al. 2007). Typically, urbanization is thought to destroy habitat, and in the Lower Rio Grande Valley of Texas, Mirowsky (1997) noted both L. ega and L. intermedius are regularly encountered in palms growing in urban areas when dried fronds are trimmed. Yet, as Spencer et al. (1988) points out, urbanization

Journal of Mammalogy 1

might also be contributing to their expansion as a large number of ornamental palms are used for residential and commercial landscaping.

Despite this rise in palms used for landscaping, records of L. ega in Texas are few; they are considered threatened by Texas Parks and Wildlife Department due to lack of information on their life history, increased frond trimming practices, and pesticide use (Baker et al. 1971; TPWD 2015). Although L. intermedius is not considered threatened in Texas, the species’ faces similar threats as L. ega. Based on submissions to the Texas Department of

State Health Services (DSHS) in Austin, Texas between 2005 – 2014, human encounters with yellow bats remain constant (Fig.1). As many as 66% of all yellow bats submitted

(n=789) throughout Texas were identified as L. intermedius, 13% as L. ega, 1% as L. xanthinus, and the remaining 18% as juvenile yellow bats. Approximately 50% of these submissions (387/789) were from Cameron and Hidalgo Counties in the Lower Rio Grande

Valley of Texas (Bonnie Mayes [Department of State Health Services, Austin, Texas], personal communication [March 2015]; Fig. 1). Because little is known about their life history, it cannot be determined whether these encounters are associated with the expanding human population within the yellow bats’ range or an increase in the bat population.

According to the U.S. Census Bureau’s annual population estimates (U.S. Census Bureau

2015), there was a higher population growth in both Cameron (0.77%) and Hidalgo (1.62%) counties compared to other Lower Rio Grande Valley counties (0%) from 2010 – 2014. As the human population within these counties continues to grow, there is reasonable cause for concern with regard to monitoring the resident species of yellow bats to understand how urban expansion might impact them.

Roosting habits of several North American lasiurines (i.e. Lasiurus cinereus,

0.07 7%

0.06 5%

0.05 4% 4% 4% 4% 4% 0.04 4% 3% 3% 0.03

0.02 Proportion of yellow submittedbats yellow Proportion of

0.01

0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Year of Submission

FIG. 1. – Proportion of bats submitted to the Texas Department of State Health Services annually that were a yellow bat species encountered in Texas. In total, 789 out of 18,342 submissions were yellow bats across all years (2005 – 2014).

L. xanthinus, L. borealis, and L. seminolus) have been well studied. Menzel et al. (1998)

radiotracked red bats (L. borealis) and Seminole bats (L. seminolus) to diurnal roosts in forests of Georgia and South Carolina. They found that both species had low roost fidelity

and each species was tracked to different tree communities (mixed-pine hardwood and pine trees [Pinus spp.]), respectively. Mager and Nelson (2001) studied red bats in urban areas of central Illinois and supported the low roost fidelity typical of these lasiurines among individual species of trees, but noted some fidelity to small roost areas, which consisted of roosts that were approximately within 100 m of one another. Perry and Thill (2007) radiotracked hoary bats (L. cinereus) to 12 roost trees that were taller with larger trunk diameters than random trees in Arkansas; roosts were all within mature tree stands of several species of Quercus oaks and Pinus echinata. Perry et al. (2007) continued the study in

Arkansas, noting differences in roost selection of L. borealis according to sex, and showed females roosting in taller mature trees (white oaks, Quercus alba, and hickories, Carya spp.) than males, with males intermittently roosting in younger trees. Few studies have addressed roosting habits of yellow bats, and these have been conducted on western yellow bats (L. xanthinus), but all highlight the importance of dried fronds as roosts. After noting the first

record in Texas of the western yellow bat (Higginbotham et al. 1999), Higginbotham et al.

(2000) went on to document this species roosting in a frond skirt (Yucca carnerosana). Ortiz

and Barrows (2014) observed diurnal roosts of L. xanthinus at palm oases with different-

sized frond skirts located within the Colorado Desert of southern California.

Besides knowledge of yellow bats roosting in foliage (Kunz and Fenton 2003),

specific roosting habits of L. ega and L. intermedius remain poorly known. Spencer et al.

(1988) captured L. ega from beneath the native palm groves of S. mexicana and noted the use

of the non-native palm, W. robusta, as well as other ornamental palm trees in Corpus Christi,

Texas. Lasiurus intermedius has also been found roosting in ornamental palms, as well as

Spanish moss (Chapman and Chapman 1990). Quantitative assessments of roosting substrates, however, are lacking. Further studies are needed to provide an overview of roost tree use, habitat usage, seasonal behavior, and colony structure of yellow bats to implement effective conservation strategies. Analysis of these types of quantitative data are needed to address questions about roost preferences, document specific roost locations, and evaluate roost characteristics within the range of yellow bats (L. ega or L. intermedius) in southern

Texas. My objective was to identify and quantitatively characterize roosts of L. ega and L. intermedius in the Lower Rio Grande Valley of Texas. Based on previous research on these species, as well as research conducted on other Lasiurus species, I hypothesized yellow bats would select mature palms with dried frond skirts.

MATERIALS AND METHODS

Study sites.--To locate roosts of both species, I identified three possible field sites in

Cameron County, Texas (Fig. 2); the Nature Conservancy’s Lennox Foundation Southmost

Preserve (25°51’14.51” N; 97°23’50.09” W), Resaca de la Palma State Park (25°59’48.48”

N; 97°34’17.85” W), and Sabal Palm Sanctuary (25°51’08.38” N; 97°25’01.56” W). Each site was selected for the presence of plant species (native and non-native palms and/or yuccas) likely to be used as roosts by yellow bats based on previous reports (Spencer et al.

1988; Higginbotham et al. 2000). The Southmost Preserve is 418 hectares and is part of the

Lower Rio Grande Valley Wildlife Corridor that is comprised of native sabal palm forests, resacas and Tamaulipan thornscrub. Resaca de La Palma State Park encompasses a semi- tropical habitat that includes non-native palms and yuccas within 485 hectares. As the smallest study site with 213 hectares, Sabal Palm Sanctuary remains one of a few sites to retain natural sabal palm populations in the United States (Lockett and Read 1990). There is no consensus on the taxonomic status of sabal palm (Cook 1901; Beccari 1907; Lockett

1991; Lonard et al. 1991). I follow Jones et al. (2011) who recognize sabal palm as S. mexicana.

Sampling strategies.--Trapping sessions spanned from November 2014 – November

2015. Southmost Preserve was the first property sampled in November 2014 for 1 night.

During January 2015, both Southmost Preserve and Resaca de la Palma State Park were sampled for 1 night each, and Sabal Palm Sanctuary staff was contacted for future collaboration. Subsequent visits to Southmost Preserve, Resaca de la Palma State Park, and

Sabal Palm Sanctuary were made in March 2015, and lasted 2 – 3 nights per visit. Sampling

FIG. 2. -- Map of Cameron County (black outline), Texas indicating the three study sites used to research the southern and northern yellow bats: Green – Resaca de la Palma, Orange – Southmost Preserve, Blue – Sabal Palm Sanctuary.

at all three sites in May and June 2015 extended to at least 10 nights per visit to allow time to

track bats to roosts. After the presence of both species of yellow bats (L. ega and L.

intermedius) was detected only at Sabal Palm Sanctuary in June 2015, subsequent visits from

July – November 2015 were focused exclusively at this site. Visits in September – November

2015 lasted for 2 – 3 nights.

Bat capture and radiotracking.-- During each sampling trip, I captured bats by

setting out 2 – 4 mistnets of various lengths (2 m, 6 m, 9 m, or 12 m), including triple-high

mistnets, and one harp trap (Bat Conservation and Management Inc., Carlisle, PA) in flyways or trails, near previously identified roosts, and over water sources approximately 30 minutes before sunset, using the methods suggested by Kunz et al. (2009). Bats were removed from the single-tiered and triple-high mistnets, as well as harp traps, and placed inside individual cloth bags. I identified each bat to species, and recorded the age, sex, reproductive condition, mass (g), and forearm length (mm). After measurements were recorded, I attached a radio- transmitter (BD-2N, 0.54g, Holohil Systems Ltd., Carp, Ontario, Canada) to the back of adult yellow bats, using similar methods as Kunz and Weise (2009). Bats were held in place by applying light pressure on the wings and head. The hair between the shoulder blades was shaved using an electric trimmer (Finishing Touch Lumina), and further cut using scissors.

The trimmed area was cleaned with one-time-use alcohol pads, and surgical cement (Perma-

Type Company Inc., Plainville, CT) was used to attach the transmitter to the back of the bat

(Amelon et al. 2009). No bat weighing less than 11 g was tagged so that the weight of the transmitter would not exceed 5% of the bat’s bodyweight and would minimize negative effects on behavior (Aldridge and Brigham 1988).

Fecal samples and one biopsy punch (3 mm) from the wing membrane were collected

from yellow bats. Each biopsy punch was labeled and placed in 95% ethanol or lysis buffer

(Longmire et al. 1997). Bats are known to heal quickly, 2 – 3 weeks, from these procedures with little-to-no ill effects noted (Davis and Doster 1972; Bonaccorso et al. 1976). Bats were handled following the American Society of Mammalogists guidelines (Sikes et al. 2011) and all methods were approved by the Angelo State University Institutional Animal Care and Use

Committee (Approval No: 15-03). Holding time did not exceed one hour and bats were released on site.

The morning following capture, I tracked radio-tagged bats via walking transects

along the trails, (and at times, off-trail areas), of the property using a radio receiver (R1000,

Communication Specialists, Inc., Orange, CA) and Yagi antenna (RA-150, Communication

Specialists, Inc.). According to Amelon et al. (2009), signals from a stationary site indicate

possible roosts, whose locality can be narrowed by noting several reference points along with

triangulation and homing. The location of a roost was identified only when strong signals

could be detected at the lowest gain setting on the radio receiver without the Yagi antenna

(Amelon et al. 2009). Once a roost was identified either visually or via radiotracking, I

recorded its coordinates (UTM, NAD83, zone 14) on a handheld GPS (Garmin Ltd., Kansas

City, KS), and noted the date.

Roost site characterization.--Similar to roost characteristics measured by Menzel et

al. (1998), I measured the trunk diameter (Diam) 1 m from the ground with a diameter tape and the total height of the tree was measured using a clinometer (Suunto, Vantaa, Finland)

for each roost. I also measured heights of the dried frond skirt at roost palms (Fig. 3). Heights were measured at 5 m away from the roost palm, or at a distance where features were visible.

In addition to total height of the palm, measurements included 1) height of the topmost dried

frond of the skirt, 2) height of the bottommost dried frond of the skirt, and 3) height of bottommost dried frond’s attachment to the trunk of the palm. Percent frond coverage around the trunk of the roost was measured by taking pictures of the canopy at the 4 cardinal directions using a tripod 1 m from the base of the palm; images were analyzed using

SamplePoint software (Booth et al. 2006). One hundred grid points were overlaid onto each image and each point was categorized as ‘open sky’, ‘not palm’, ‘frond’, ‘stem of palm’, or

‘trunk of palm’. Distance of the roost from a resaca and distance from a trail was also measured using ArcMap (ESRI 2015). I measured the same palm characteristics and habitat features for randomly selected palms generated using ArcMap to compare used and unused palms. Random points were generated within a 250 m buffer from a resaca; if a random point did not correspond to a palm, the closest palm within 50 m was chosen as a random palm.

Data analysis.--The difference between the palm characteristics of roosts used by L. ega and L. intermedius were evaluated using two-sample t-tests using R-software v.3.1.2 (R

Core Team 2014). Due to a low sample size for each species, roost characteristics of L. ega and L. intermedius were combined in subsequent analyses. Palm characteristics between roosts and random palms were assessed via two-sample t-tests using R. After completing correlated analysis among the 8 palm characteristics, I chose 5 non-correlated variables to build 7 a priori models to explain roost selection by yellow bats; variables included height of the topmost dried frond of the skirt (Top), thickness of the dried frond skirt (Thick), trunk diameter (Diam), distance to a resaca (DR), and distance to a trail (DT). These habitat variables were selected on the basis of perceived biological importance, field observations, and habitat characteristics used in other lasiurine studies, such as trunk diameter and height of the roost as well as distance to water and edge (Menzel et al. 1998; Mormann and Robbins

FIG. 3. -- Diagram of measurements taken for each roost and random palm using a clinometer at ground level from 5 m away, or from a distance where palm was in visible. Total palm height was taken from the tallest live frond of the palm; the height at the top of the skirt was taken from the topmost dried frond of the skirt; the height of the bottom of the skirt was taken from the bottommost dried frond of the skirt, whose stem attachment to the trunk was measured. Thickness of the dried frond skirt was calculated by subtracting the bottommost dried frond height from the topmost dried frond.

2007). Similar to Klug et al. (2012) and O’Keefe et al. (2009), I used logistic regression analysis with roost or non-roost as the dependent variable to evaluate my 7 models. Logan

(2010) reiterates the benefit of using this type of regression for its lack of assumptions on equal variance and normality, as well as preference on model fitting for small sample sizes.

Guisan and Zimmermann (2000) state such models, specifically predictability models, are a useful tool when assessing species’ distribution, and Jaberg and Guisan (2001) affirm the need for such methods when studying bats that are difficult to capture.

The fit of the global model was verified using the le Cessie-van Houwelingen lack of fit test (Logan 2010). As per Burnham and Anderson (2002), I analyzed and evaluated each model by their second order Akaike’s information criterion scores (AICc), differences between the AICc of all models relative to the highest scoring model (Δi), and Akaike weights (ωi). The null model (model with only the intercept) was included in the analysis to decrease the likelihood that the ‘top’, model was not just the highest scoring of a cluster of

‘poor’ models. Models with Δi = 0 were considered top models (Burnham and Anderson

2002), and a Wald test was conducted to assure the significance of the variables (Hosmer and

Lemeshow 2000).

Following modeling, I created a resource selection probability function (RSPF) using the estimated coefficients of each variable from the top model (Manly et al. 2002). I did not have an independent data set to evaluate the RSPF due to a low sample size. The RSPF was evaluated using the training data set, and I considered a palm to be used if it scored a 0.7 on the scale (0.0 – 1.0). Although this method restricted the reliability of the function, it was the only option given the limiting factors.

RESULTS

Throughout the field season (November 2014 – November 2015), Southmost

Preserve was sampled for a total of 7 nights, Resaca de la Palma State Park for 5 nights, and

Sabal Palm Sanctuary for 11 nights. Three species of bats were captured during this study

(Table 1). Evening bats (Nycticeius humeralis) were captured at all three study sites. I

confirmed the presence of southern yellow bats (L. ega) at Southmost Preserve and northern

yellow bats (L. intermedius) at Resaca de la Palma State Park. Both species were captured at

Sabal Palm Sanctuary, thus all radiotracking occurred at this property. Sampling at Sabal

Palm Sanctuary resulted in the capture of 20 L. ega, 5 L. intermedius, and 75 N. humeralis

(Table 1); 18 fecal samples and 20 wing biopsies were collected from L. ega and 4 fecal

samples and 5 wing biopsy were collected from L. intermedius (Appendix I). Most of the

yellow bats (76% of both L. ega and L. intermedius) were captured in a triple-high net placed

over a trail near a mature sabal palm grove approximately 50 m from a resaca, though a few

were captured in mistnets over water (16%) or mistnets (4%) and harp traps (4%) along

trails.

Radio-tagged yellow bats (n=8) were found roosting in palms. In March 2015, a male

northern yellow bat was tagged at Resaca de la Palma State Park and its signal never located.

In September 2015, a female northern yellow bat was tracked to private property, for which I

did not have access. I located a total of 20 roost palms (S. mexicana); 6 L. ega were tracked

to 16 roosts, and 2 L. intermedius tracked to 6 roosts (Fig. 4; Appendix II). Two roosts were

occupied by both species on different dates. Of the 20 roost palms, 18 roosts used by yellow

bats were located within a mature palm grove that holds the oldest and tallest sabal palms

(Fig. 4) and were within 50 m from a resaca, a trail, or a netting site (Appendix II).

TABLE 1.--Summary of bat captures for three sites in Brownsville, Texas (Cameron County) over a total of 23 trapping nights, shown in parentheses for each site, from November 2014 – November 2015.

Resaca de la Southmost Sabal Palm Total captures Species Palma (5) Preserve (7) Sanctuary (11) per species

Lasiurus ega 0 4 20 24

Lasiurus intermedius 1 0 5 6

Nycticeius humeralis 5 15 75 95

Total captures per site 6 19 100 125

Resaca

Mature Grove of Sabal mexicana

FIG. 4. -- Radio-tagged bats, 6 Lasiurus ega (LAEG) and 2 L. intermedius (LAIN) were tracked to 20 roosts in Sabal mexicana palms within Sabal Palm Sanctuary in Brownsville, Texas from May 2015 – November 2015.

Radio-tagged bats were tracked from 1 – 5 days (L. ega) or 1 – 4 days (L.

intermedius) during each visit (Table 2). On average, L. ega individuals (n= 6) were tracked

for 3.3 days, and L. intermedius (n=2) were tracked for an average of 3.5 days. The number

of different roosts used by both species of bats ranged from 1 – 5 (average of 2.75). While

walking transects along, one roost was identified visually with a non-tagged yellow bat spp.,

(height of the topmost frond of the dried skirt = 5.95 m, thickness of frond skirt = 2.68 m,

Diameter = 0.43 m). This visual roost was located within 10 m of two radiotracked roosts and was not included in subsequent analyses. At one particularly young, short palm without a frond skirt, a male L. ega was tracked and photographed in a low hanging dried frond.

Individual yellow bats switched roost locations often (Table 2). In June, I located 8

roost palms from tracking two L. ega (male and female) and a female L. intermedius. The

male L. ega relocated to different roosts daily while the female L. ega and L. intermedius relocated only once and were tracked to two roost palms each; they were tracked to the same roost for two consecutive days. In July, two female L. ega and one female L. intermedius were radiotracked to 9 sabal palm roosts. Each female selected the same palm roost no more than twice in a row. In September, two female yellow bats (L. ega and L. intermedius) were

tracked. The L. ega individual was tracked to the same roost for three consecutive days while

the L. intermedius was tracked outside the property, and triangulated to be within a private

residence that had palms. In November only one male L. ega was captured and tracked to two

roost palms.

Roost characteristics did not differ between L. ega and L. intermedius sample (all t-

test, P>0.01), so roost characteristics of both species were pooled in following analyses.

TABLE 2.--The sexes, number of roosts used, number of days tracked, and mean number of days per roost of radio-tagged Lasiurus ega (LAEG) and L. intermedius (LAIN) at Sabal Palm Sanctuary, Brownsville, Texas.

Species Sex No. Roost Used No. Days Tracked Mean No. Days per Roost LAEG M 4 4 1.0 LAEG F 2 3 1.5 LAEG F 5 5 1.0 LAEG F 2 3 1.5 LAEG F 1 3 3.0 LAEG M 2 2 1.0 LAIN F 2 3 1.5 LAIN F 4 4 1.0 Total 20* *Two roosts were occupied by both species on different dates; repeated roosts were subtracted from total.

Comparison of roost palm characteristics (n=20) and randomly selected palms showed that yellow bats used sabal palms that had taller (x̅ = 8.18 m, sd = 3.05 m) thicker (x̅ =3.03 m, sd

= 1.60 m) dried frond skirts and had smaller trunk diameters (x̅ = 0.39 m, sd = 0.10 m) than randomly selected sabal palms (Table 3). The total height of roost palms was not significantly different from random palms (Table 3). Roost palms were on average 12.34 m tall and had an average trunk diameter of 0.39 m, while random palms were 10.61 m tall and had a trunk diameter of 0.53 m (Table 3). Canopy coverage around roost and random palms was similar (Table 3). Roost palms had, on average, canopy coverage consisting of 19% open area, 37% non-palm vegetation, 33% fronds (dead or alive), 6 % palm stem, and 3% palm trunk. Random palms had, on average, canopy coverage consisting of 17% open area, 31% non-palm vegetation, 34% fronds (dead or alive), 12% palm stem, and 2% palm trunk.

The full model containing all of the measured variables was evaluated for

collinearity and resulted in correlation between the height variables. Only variables with

biological importance and statistical significance were retained. The top model included the

effects of the following variables: height of the topmost dried frond of the skirt, thickness of

the dried frond skirt, trunk diameter, and distance to a resaca (Table 4). The Wald test statistic showed the overall significant effect of these variables towards roost selection (Wald test, P<0.05). The coefficients from the top model were then used to construct a resource selection probability function (RSPF) using the following formula (Logan 2010):

1 = 1 + ( … ) 𝑝𝑝 − 𝛽𝛽0+𝛽𝛽1𝑋𝑋1 +𝛽𝛽𝑛𝑛𝑋𝑋𝑛𝑛 where β is the partial regression coefficient𝑒𝑒𝑒𝑒𝑒𝑒 relating to the nth predictor variable and p is the

probability that a palm may be used as a roost (Table 5). After the RSPF was evaluated using

training data set, and rated using the scale 0.0 – 1.0, the RSPF correctly identified roosts 75% of the time (n=15/20).

TABLE 3.--Measurements of characteristics of roost and random Sabal mexicana palms used by both Lasiurus ega and L. intermedius. Numbers are means ± SD.

Parameter Roost (n = 20) Random (n= 20) P-value Trunk Diameter (m) 0.39 ± 0.10 0.53 ± 0.16 0.003* Heights (m) Total 12.34 ± 3.43 10.61 ± 3.99 0.151 Top Frond Skirt 8.18 ± 3.05 6.11 ± 2.79 0.031* Bottom Frond Skirt 5.15 ± 2.01 4.48 ± 2.62 0.376 Frond Skirt Attachment 6.20 ± 2.15 5.07 ± 3.12 0.190 Frond Skirt Thickness 3.03 ± 1.60 1.62 ± 0.83 0.001*

Open vs. Cover Canopy (%) Cover 80.57 ± 9.71 81.10 ± 11.61 0.663 Open 19.08 ± 8.95 17.99 ± 9.95 0.307

Distance to a Trail (m) 16.34 ± 16.56 40.08 ± 60.74 0.090 Distance to a Resaca (m) 52.43 ± 18.83 70.03 ± 68.03 0.271 *Statistically significant (Student’s t-test, P<0.05).

TABLE 4.--Comparison of candidate models for Lasiurus ega or L. intermedius roost selection in Sabal mexicana based on AIC (Akaike information criterion scores), AICc (second order scores), delta AICc (Δi), and model probabilities (ωi).

Modela -2LL AICc Δi ωi Top+Thick+Diam+DR 46.97 48.70 0.00 0.342 Thick+Diam+DR 47.54 49.30 0.57 0.258 Top+Thick+Diam+DT 47.78 49.50 0.81 0.228 Top+Thick+Diam+DT+DR 48.48 51.00 2.29 0.109 Top+Thick+DT+DR 50.95 52.70 3.98 0.047 Top+Diam+DT+DR 53.60 55.40 6.62 0.012 Null 57.45 57.60 8.62 0.004 a Top: height of topmost dried frond of the skirt Thick: thickness of the dried frond skirt Diam: trunk diameter 1 m from the ground DT: distance to trail DR: distance to resaca Null: all variables removed.

TABLE 5.--Top model (75% accuracy) including variables that best influenced Lasiurus ega or L. intermedius roost selection in Sabal mexicana. The model was constructed using a logistic regression function.

Variable Estimate SE z value P- value Top -0.214 0.212 -1.068 0.312 Thick 1.055 0.449 2.351 0.018 Diam -8.168 3.986 -2.049 0.040 DR -0.009 0.011 -0.883 0.376 Intercept 3.415 2.812 1.215 0.224

DISCUSSION

I confirmed the observations of Mirowsky (1997) that L. ega and L. intermedius roost

in dried fronds of palms trees in the Lower Rio Grande Valley of Texas. In fact, both species

were found occupying the same general area at Sabal Palm Sanctuary. This study is the first

to quantitatively characterize dried frond skirts of S. mexicana as roosts for L. ega and L. intermedius. Dried palm fronds as roosts have been poorly studied as the majority of the studies that conduct palm roosting analysis pertain mostly to tent-making bats in the neotropics (Kunz and McCracken 1996; Kunz and Fenton 2003). The dried fronds where L.

ega and L. intermedius were recorded roosting can be considered temporary structures, much

like the live palm fronds used by tent-making bats. These provide protection for only a

relatively short period of time, affirming the low roost fidelity seen in these species of yellow

bats, as with other lasiurines (Kunz and Fenton 2003; Coleman et al. 2012).

High frequency roost switching could be attributed to a variety of environmental

disturbances on the dried frond skirt. The length of time dried fronds remain on a palm can

depend on their durability against harsh wind and rain. Predation might also play a strong

role in yellow bats’ low roost fidelity; pack rats (Neotoma spp.) were often observed in the

leaf stalks and skirts of palms along with raccoons (Procyon spp.). Increased disturbances can possibly reduce the likelihood of repeated usage of a particular roost. Menzel et al.

(1998) observed roost switching as a common occurrence for L. borealis and L. seminolus

(average of 1.2 and 1.7 nights per roost, respectively), a characteristic also suggested for L.

xanthinus by Ortiz and Barrows (2014). This strongly supports the frequent day-to-day turnover seen with L. ega and L. intermedius, with an average of 1.23 and 1.16 days per roost, respectively (Table 2).

Perry and Thill (2007) and Perry et al. (2007) indicate that forest bats tend to roost in older mature forests. This is a similar characteristic of roost palms used by L. ega and L. intermedius, as most (n=18/20) were found within the mature palm grove. Mature palms likely provide a more stable shelter than young palms, gaining characteristics that are

attributed to their growth such as taller and thicker dried frond skirts. This could potentially reduce predation, provide a suitable microclimate for thermoregulation, and temporarily protect yellow bats from hazardous weather (wind or rain).

Similar to Hutchinson’s (2006) study of L. intermedius roosting in Spanish moss within a small patch of oaks (8-20 m wide), my study shows L. ega and L. intermedius roosts are relatively close to each other. Both species of yellow bats were tracked, at least once, to the same or previously used roost, though not concurrently. Although this study cannot statistically support the small home range for L. intermedius (10.5 ha) found by Krishon et al.

(1997), I can qualitatively state that both species selected roosts within a small geographic area during the length of this study.

Despite the commonality of lasiurine bats roosting in mature trees that, in turn, have a large diameter (Menzel et al. 1998; Perry and Thill 2007), L. ega and L. intermedius deviates from this pattern when roosting in S. mexicana palms. As they grow, the trunk of these palms becomes narrower as leaf-stalks are shed (Wasowski and Wasowski 2002), resulting in the smaller diameter variable that was significantly different compared to random palms. The presence of a dried frond skirt on these palms is also attributed to their maturity, as older palms are more likely to have one compared to younger palms (excluding harsh weather conditions). Image analysis revealed a high percentage of canopy coverage among both roost

and random palms, which could also be attributed to the shade preference and growth of S.

mexicana (Wasowski and Wasowski 2002).

My data suggests that yellow bats roost in sabal palms with frond skirts that are, on

average, 3 m thick and 8 m high on the trunk and have trunk diameter of 0.4 m which correspond to older S. mexicana palms. Additionally, the top model showed that the height of the frond skirt on the palm, its thickness, the diameter of the trunk, and a palms’ distance from a resaca influenced L. ega and L. intermedius roost selection. Although a small trunk diameter had a significant effect on roost selection, this might be interpreted as selection for mature trees. With mature palms, a higher and thicker dried frond skirt could potentially reduce the risk of predation by restricting access to yellow bats, as they tend to roost in the part of the dried frond closest to the leaf stem that makes them difficult to reach. The thickness of the skirt could serve as protection during adverse weather conditions (wind, rain, hail, etc.), and possibly play a role in microclimate conditions required for thermoregulation of bats (Lacki et al. 2007). Furthermore, Lacki et al. (2007) stated proximity to a water source, such as a resaca, can be important in reducing the energy spent commuting as well as

reduce exposure to predators for these bats.

The resource selection probability function (RSPF) estimated for this study accurately

predicted known roosts 75% of the time. The function accounted for the height of the skirt on

the trunk of the palm, the thickness of the skirt, the trunk diameter, and the distance of that

palm to the resaca. Roosts accurately classified as used had frond skirts at least 2 m thick,

and 5.5 m tall, and a trunk diameter of at least 0.2 m. There were random palms classified

used, which could be due to at least measurement was similar to a known roost. The high

accuracy of the RSPF can be attributed to the data it was estimated from, which was also

used to evaluate the function. This RSPF was developed using only data from S. mexicana;

therefore, I urge caution in its application to other species of palms. Although this function could potentially be implemented in other parts of the range for L. ega and L. intermedius for assessing probability of use for other species of palms with similar growth conditions, it is not within the scope of this study to extrapolate probabilities to other species of palms.

This study was limited to one field season which resulted in a small sample size and

low number of trapping/tracking visits. Further studies with longer sampling periods and

larger number of bats should be conducted to confirm my results, and evaluate the RSPF.

Such studies could also be performed throughout the range of these yellow bats species (L.

ega and L. intermedius), both in the United States and in Latin America, to expand our

knowledge of their roosting substrates besides S. mexicana, Washingtonia spp., and

Tillandsia spp. It would be useful to note if dried palm fronds are the most common roosting

substrates for these species throughout their range in order adequately manage and conserve

their habitat. The 250 m buffer used to select random palms might have introduced a bias in

the characteristics of random palms because this was one of the most cluttered areas of the

property. Further, it limited my ability to test the significance of “distance to water” because

most random palms were in the same general area as roost palms near the resaca. The use of

the RSPF for palm landscaping and management must be considered conservatively for it

was built and evaluated from non-dependent data. Meaning, radio-tagged yellow bats were

assumed to be representative of the population and bats were assumed to be absent from

random palms. Additional technologies that could be explored and developed for tree

roosting bats, such as thermal-imaging cameras, light detection lasers (Azmy et al. 2012) and

unmanned aerial vehicles (UAVs; Evershed 2014), might be useful to visually confirm or

deny bats in used/unused roosts.

The sabal palm grove at Sabal Palm Sanctuary is one of the last remaining natural

groves of S. mexicana in the United States (Locket and Read 1990) whose fronds are purposely never trimmed unlike palms found in urban areas. Because yellow bats have been

tracked and seen roosting in dried fronds, my results suggest that trimming dried palm frond

skirts may have a negative impact on roosting habitat for yellow bats. The RSPF developed

and evaluated here could potentially aid in managing palm groves for these species by

inputting the measured variables to calculate the likelihood that a particular palm might be suitable as a roost. If a high probability of use (p ≥ 0.7) is calculated for palms, this could

guide decisions about trimming. Although the best approach would be to leave natural frond

skirts on the palms to provide roosting habitat, frond skirts could be shortened during the

reproductive season (Sep – Dec) and parturition season (April – July) instead of entirely

removed, thereby potentially reducing detrimental impacts to the habitat of yellow bats.

Habitat management efforts for the state threatened L. ega, as well as L. intermedius,

by wildlife agencies could be enhanced by involving the local communities. Environmental

education workshops emphasizing reduced tree trimming practices could be implemented in

local state parks and regions with a high occurrence of palms to educate local communities

about bats, publicize the presence and benefits of yellow bats in the area, and refer them to

landscaping companies that are conscientious of wildlife should they decide to trim their

palms.

LITERATURE CITED

ALDRIDGE, H. D., J. N. AND R. M. BRIGHAM. 1988. Load carrying and maneuverability in an

insectivorous bat: a test of the 5% “rule” of radiotelemetry. Journal of Mammalogy

69: 379-382.

AMELON, S. K., D. C. DALTON, J. J. MILLSPAUGH, AND S. A. WOLF. 2009. Radiotelemetry. Pp.

57-77 in Ecological and behavioral methods for the study of bats (T.H. Kunz and S.

Parsons, eds.). Johns Hopkins University Press, Baltimore, Maryland.

nd AMMERMAN, L. K., C. L. HICE, AND D. J. SCHMIDLY. 2012. Bats of Texas. 2 ed. Texas

A&M University Press, College Station, Texas.

AZMY, S. N., S. A. M. SAH, N. J. SHAFIE, A. ARIFFIN, Z. MAJID, M. N. A. ISMAIL, AND M. S.

SHAMSIR. 2012. Counting in the dark: non-intrusive laser scanning for population

counting and identifying roosting bats. Scientific Reports 2: 524.

BAKER, R. J., T. MOLLHAGEN, AND G. LOPEZ. 1971. Notes on Lasiurus ega. Journal of

Mammalogy 52: 849-852.

BAKER, R. J., J. C. PATTON, H. H. GENOWAYS, AND J. W. BICKHAM. 1988. Genic studies of

Lasiurus (Chiroptera: Vespertilionidae). Mammalogy Papers: University of

Nebraska State Museum 97: 1-15.

BECCARI, O. 1907. Le palme americane della tribu delle Corypheae. Webbia 2: 1-343.

BONACCORSO, F. J., N. SMYTHE, AND S. R. HUMPHREY. 1976. Improved techniques for

marking bats. Journal of Mammalogy 57: 181-182.

BOOTH, D. T., S. E. COX, AND R. D. BERRYMAN. 2006. Point sampling digital imagery

‘SamplePoint’. Environmental Monitoring Assessment 123: 97-108.

BURNHAM, K. P. AND D. R. ANDERSON. 2002. Model selection and multimodel inference: a

practical information-theoretic approach. 2nd ed. Springer-Verlag, New York, New

York.

CHAPMAN, S. S. AND B. R. CHAPMAN. 1990. Bats from the coastal region of southern Texas.

Texas Journal of Science 42: 13-22.

COLEMAN, L. S., K. M. MORRIS, AND S. B. CASTLEBERRY. 2012. Characteristics of Lasiurus

intermedius (northern yellow bat) roosts on Sapelo Island, Georgia. Southeastern

Naturalist 11: 534-536

COOK, O. F. 1901. A sypnosis of palms of Puerto Rico. Bulletin of the Torrey Botanical Club

28: 525-569.

DAVIS R. AND S. E. DOSTER. 1972. Wing repair in pallid bats. Journal of Mammalogy 53:

377-378.

ESRI. 2015. ArcMap GIS. Ver.10.1. Environmental Systems. Research Institute, Redlands,

California.

EVERSHED, N. 2014. Drones in Australia: fruit bats, forest fires and invasions of privacy. The

Guardian. http://www.theguardian.com/world/2014/mar/05/drones-in-australia-fruit-

batsforest-fires-and-invasions-of-privacy. Accessed 22 March 2016.

FINDLEY, J. S. 1993. Bats: a community perspective. Cambridge University Press,

Cambridge, United Kingdom.

GUISAN, A. AND N. E. ZIMMERMANN. 2000. Predictive habitat distribution models in ecology.

Ecological Modeling 135: 147-186.

HIGGINBOTHAM, J. L., L. K. AMMERMAN, AND M. T. DIXON. 1999. First record of Lasiurus

xanthinus (Chiroptera: Vespertilionidae) in Texas. Southwestern Naturalist 44: 343-

347.

HIGGINBOTHAM, J. L., M. T. DIXON, AND L. K. AMMERMAN. 2000. Yucca provides roost for

Lasiurus xanthinus (Chiroptera: Vespertilionidae) in Texas. Southwestern Naturalist

45: 338-440.

nd HOSMER, D. W. AND S. LEMESHOW. 2000. Applied logistic regression. 2 ed. John Wiley &

Sons Inc., New York, New York.

HUTCHINSON, J. T. 2006. Bats of Archbold Biological Station and notes on some roost sites.

Florida Field Naturalist 34: 48-51.

JABERG, C. AND A. GUISAN. 2001. Modeling the distribution of bats in relation to landscape

structure in a temperate mountain environment. Journal of Applied Ecology 38:

1169-1181.

JONES, S. D., J. K. WIPFF, AND P. M. MONTGOMERY. 2011. Vascular plants of Texas: a

comprehensive checklist including synonymy, bibliography, and index. University

of Texas Press, Austin, Texas.

KLUG, B. J., D. A. GOLDSMITH, AND R. M. R. BARCLAY. 2012. Roost selection by solitary,

foliage-roosting hoary bat (Lasiurus cinereus) during lactation. Canadian Journal of

Zoology 90: 329-336.

KRISHON, D. M., M. A. MENZEL, T. C. CARTER, AND J. LAERM. 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 B. FENTON. 2003. Bat ecology. The University of Chicago Press, Chicago,

Illinois.

KUNZ, T. H., AND G. F. MCCRACKEN. 1996. Tents and harems: apparent defense of foliage

roosting by tent-making bats. Journal of Tropical Ecology 12: 121-137.

KUNZ, T. H., R. HODGKINSON, AND C. D. WEISE. 2009. Methods of capturing and handling

bats. Pp. 3 - 35 in Ecological and behavioral methods for the study of bats (T. H.

Kunz and S. Parsons, eds.), Johns Hopkins University Press, Baltimore, Maryland.

KUNZ, T. H. AND C. D. WEISE. 2009. Methods and devices for marking bats. Pp. 36-56 in

Ecological and behavioral methods for the study of bats (T. H. Kunz and S. Parsons,

eds.). Johns Hopkins University Press, Baltimore, Maryland.

KURTA, A. AND G. C. LEHR. 1995. Lasiurus ega. Mammalian Species 515: 1-7.

LACKI, M. J., J. P. HAYES, AND A. KURTA. 2007. Bats in forests: conservation and

management. Johns Hopkins University Press, Baltimore, Maryland.

LOCKETT, L. 1991. Native Texas palms north of the Lower Rio Grande Valley: Recent

Discoveries. Principes 35: 64-71.

LOCKETT, L. AND R. W. READ. 1990. Extension of native range of Sabal mexicana (Palmae)

in Texas to include Central Coast. SIDA, Contributions to Botany 14: 79-85.

LOGAN, M. 2010. Biostatistical design and analysis using R: a practical guide. Wiley-

Blackwell, West Sussex, United Kingdom.

LONARD, R. I., J. H. EVERITT, AND F. W. JUDD. 1991. Woody plants of the Lower Rio Grande

Valley, Texas. Miscellaneous publication No. 7. Texas Memorial Museum,

University of Texas, Austin, Texas.

LONGMIRE, J. L., M. MALTBIE, AND R. J. BAKER. 1997. Use of "lysis buffer" in DNA isolation

and its implications for museum collections. Occasional Papers, The Museum of

Texas Tech University 163: 1-3.

MAGER, K. J AND T. A. NELSON. 2001. Roost-site selection by eastern red bats (Lasiurus

borealis). American Midland Naturalist 145: 120-126.

MANLY, B. F. J., L. L. MCDONALD, D. L. THOMAS, T. L. MCDONALD, AND W. P. ERICKSON.

2002. Resource selection by animals. Kluwer Academic Publishers, Dordrecht,

Netherlands.

MENZEL, M. A., T. C. CARTER, B. R. CHAPMAN, AND J. LAERM. 1998. Quantitative

comparison of tree roosts used by red bats (Lasiurus borealis) and Seminole bats (L.

seminolus). Canadian Journal of Zoology 76: 630-634.

MIROWSKY, K. M. 1997. Bats in palms: precarious habitat. Bats 15:3-6.

MORALES, J. C. AND J. W. BICKHAM. 1995. Molecular systematics of the genus Lasiurus

(Chiroptera: Vespertilionidae) based on restriction-site maps of the mitochondrial

ribosomal genes. Journal of Mammalogy 76: 730-749.

MORMANN, B. M. AND L. W. ROBBINS. 2007. Winter roosting ecology of eastern red bats in

southwest Missouri. Journal of Mammalogy 71: 213-217.

O’KEEFE, J. M., S. C. LOEB, J. D. LANHAM, AND H. S. HILL JR. 2009. Macrohabitat factors

affect day roost selection by eastern red bats and eastern pipistrelles in the southern

Appalachian Mountains, USA. Forest Ecology and Management 257: 1757-1763.

ORTIZ, D. D. AND C. W. BARROWS. 2014. Occupancy patterns of western yellow bats

(Lasiurus xanthinus) in palm oases in the lower Colorado Desert. Southwestern

Naturalist 59: 381-388.

PERRY, R. W. AND R. E. THILL. 2007. Roost characteristics of hoary bats in Arkansas.

American Midland Naturalist 158: 132-138.

PERRY, R. W., R. E. THILL, AND S. A. CARTER. 2007. Sex-specific roost selection by adult red

bats in a diverse forested landscape. Forest and Ecology Management 253: 48-55.

R CORE TEAM. 2014. R: a language and environment for statistical computing, version 3.1.2.

R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/.

SIKES, R. S., W. L. GANNON, AND THE ANIMAL CARE AND USE COMMITTEE OF THE AMERICAN

SOCIETY OF MAMMALOGISTS. 2011. Guidelines of the American Society of

Mammalogists for the use of wild mammals in research. Journal of Mammalogy 92:

235-253.

SPENCER, S. G., P. C. CHOUCAIR, AND B. R. CHAPMAN. 1988. Northward expansion of the

southern yellow bat, Lasiurus ega, in Texas. Southwestern Naturalist 33: 493.

TPWD [TEXAS PARKS AND WILDLIFE]. 2015. Nongame and Rare Species Program: Federal

and State Listed Mammal Species.

http://tpwd.texas.gov/huntwild/wild/wildlife_diversity/nongame/listed-

species/mammals.phtml. Accessed 31 March 2015.

UNITED STATES CENSUS BUREAU, POPULATION DIVISION. 2015. Annual Estimates of

Resident Population: April 1, 2010 to July 1, 2014.

http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=CF.

Accessed 7 March 2016.

WASOWSKI, S. AND A. WASOWSKI. 2002. Native Texas plants: landscaping region by region.

2nd ed. Lone Star Books, Lanham, Maryland.

WEBSTER, W. D., J. K. JONES, JR., AND R. J. BAKER. 1980. Lasiurus intermedius. Mammalian

Species 132: 1-3.

APPENDIX I. Tissue catalog number (ASK of biopsy), date of capture, species, sex, reproductive condition (condition), mass (wt.; g), hind foot (HF; mm), ear length (mm), feces collected, and radio-tag frequencies of captured/collected Lasiurus ega (LAEG) and L. intermedius (LAIN) in Sabal Palm Sanctuary. Single sample with an asterisk indicates specimen number of ASNHC 17703.

ASK of Date Species Sex Condition Wt. FA HF Ear Feces Radio-tag biopsy collected frequency 13010 17-Mar-15 LAEG F Lactating 18.0 45.75 9.3 11.5 Yes -

13011 3-Jun-15 LAEG F Lactating 17.0 44.30 6.9 13.8 Yes -

13012 3-Jun-15 LAEG M Scrotal 12.7 44.20 6.5 13.9 Yes 150.697

13013 3-Jun-15 LAEG M Scrotal 12.0 43.50 6.9 13.8 Yes - 36

13014 3-Jun-15 LAEG F Lactating 21.0 47.00 7.6 12.0 Yes 150.737

13015 3-Jun-15 LAEG M Scrotal 15.0 42.20 5.6 13.0 Yes -

13016 3-Jun-15 LAEG M Scrotal 12.5 44.10 6.8 10.6 Yes -

13017 3-Jun-15 LAIN F Lactating 39.0 55.80 10.0 14.7 Yes 150.715

13018 4-Jun-15 LAEG F Lactating 16.0 47.00 6.8 13.1 Yes -

13019 4-Jun-15 LAEG F Open 19.0 45.30 7.0 13.9 Yes - vagina

ASK of Date Species Sex Condition Wt. FA HF Ear Feces Radio-tag biopsy collected frequency 13020 4-Jun-15 LAEG M Scrotal 14.0 42.5 7.0 13.8 Yes -

13021 4-Jun-15 LAEG F Pregnant 18.0 42.1 6.0 13.6 Yes -

13022 4-Jun-15 LAEG M Scrotal 15.0 44.2 7.9 12.8 Yes -

13023 4-Jun-15 LAIN F Pregnant 37.0 52.5 9.2 11.3 Yes -

13024 5-Jun-15 LAEG M Scrotal 19.0 46.1 7.8 11.1 Yes -

13025 5-Jun-15 LAIN F Pregnant 36.0 54.2 9.3 11.2 Yes - 37

13026 25-Jul-15 LAEG F None 12.0 46.4 7.5 13.8 No 148.268

13027 25-Jul-15 LAEG M Scrotal 11.5 45.9 9.2 13.4 No -

13028 25-Jul-15 LAIN F - 30.0 54.2 12.1 14.2 Yes 150.754

10857* 27-Jul-15 LAEG F None 11.5 41.5 7.6 14.3 No -

13029 28-Jul-15 LAEG F None 17.5 48.1 8.9 13.4 No 148.247

13030 4-Sep-15 LAEG F none 16.0 45.5 5.8 13.0 No 148.310

ASK of Date Species Sex Condition Wt. FA HF Ear Feces Radio-tag biopsy collected frequency 13031 4-Sep-15 LAEG M None 15.5 44.5 10.0 12.2 Yes -

13032 4-Sep-15 LAIN F - 31.0 56.2 8.1 14.0 Yes 149.255

13033 20-Nov-15 LAEG M Scrotal 13.0 44.1 7.0 14.0 No 149.586

APPENDIX II Summary measurements of Sabal mexicana roosts used by Lasiurus ega and L. intermedius (Fig. 4): GPS Coordinates, heights in meters (total height of the palm, height of the topmost dried frond of the skirt, height of the bottommost dried frond of the skirt, and height of the bottommost dried frond attachment to the trunk of the palm), the trunk diameter in meters (Diam), distance to a resaca in meters (DR), distance to a trail in meters (DT), and distance to a net in meters (DN)

Coordinates Total Top Bottom Attached Thick Diam DR DT DN 658468.42 2858975.88 15.555 11.400 6.701 7.101 4.698 0.350 47.367 5.710 8.225

658432.44 2859241.89 5.217 1.799 1.624 1.887 0.175 0.764 16.256 33.317 43.309

658442.02 39

2859945.47 13.135 7.809 4.612 6.745 3.197 0.334 34.090 16.909 37.188

658435.72 2859927.23 12.389 7.564 4.250 4.432 3.313 0.382 30.878 18.220 24.040

658248.36 2860327.89 9.771 5.796 2.696 3.572 3.099 0.471 52.406 54.801 66.489

658246.40 2860324.21 13.229 8.085 6.112 5.796 1.972 0.453 48.409 50.635 62.980

658507.19 144.912 11.350 7.609 5.796 7.408 1.813 0.440 96.717 40.963 2860043.74

Coordinates Total Top Bottom Attached Thick Diam DR DT DN 658495.10 2860040.27 15.197 9.149 5.645 7.201 3.503 0.430 84.273 28.812 137.467

658456.30 2859890.80 7.849 7.003 2.512 4.108 4.490 0.405 59.020 2.179 25.009

658452.74 2859861.95 9.451 6.627 4.574 3.997 2.053 0.271 55.933 7.167 47.741

659762.61 2749132.60 13.765 5.992 5.082 7.408 0.909 0.350 55.435 0.974 16.447

658454.59 40

2859891.55 15.187 13.825 7.201 10.470 6.623 0.479 56.701 0.061 18.302

658443.55 2859892.31 17.934 11.842 6.484 7.450 5.357 0.382 45.744 11.018 19.403

658446.14 2859852.35 10.110 7.003 4.951 4.336 2.052 0.319 48.231 2.169 57.668

658458.76 2859853.27 8.043 7.390 6.160 7.104 1.230 0.382 60.883 5.671 56.809

Coordinates Total Top Bottom Attached Thick Diam DR DT DN 658466.49 2859893.58 13.926 8.220 5.220 6.632 3.000 0.334 67.938 11.824 21.563

658444.71 2859913.04 15.971 9.771 7.408 8.085 2.362 0.335 42.615 4.250 8.174

658437.26 2860065.28 10.984 6.068 3.005 5.180 3.063 0.342 34.642 10.969 7.230

658447.92 9.951 5.546 2.886 5.141 2.659 0.377 38.815 11.100 113.359 2859954.18 41

BIOGRAPHY

Patricia Citlally Jimenez was born in Houston, Texas but raised in Mexico City until

she was ten years old. Upon her return to Houston in 2001, she pursued her love for animals

through a career in veterinary medicine. While progressing in her career at Texas A&M

University, she became enamored by bats and environmental education after several research

trips to Mexico and the Caribbean. Citlally went on to complete an undergraduate thesis

comparing the effectiveness in retention of different multi-interactive education tools

(videos, hands-on activities, and role-drama) available for educators to utilize when engaging students in Monterey, Nuevo Leon, Mexico. She graduated from Texas A&M University in

2014 with her Bachelor’s degree in Wildlife and Fisheries and a Theater minor.

During her pursuit of a Master of Science in Biology at Angelo State University with

Dr. Loren Ammerman, Citlally researched roosts of southern and northern yellow bats in the

Lower Rio Grande Valley. She is optimistic that her study will provide foundational knowledge over the roosting ecology and aid in the habitat management of these yellow bats.

In addition to entering married life, Citlally plans to continue her career in science education, pursuing a PhD at the University of Nebraska – Lincoln, researching the use of multi-interactive tools to teach critical thinking in science to the general public.

Phone: 832 – 868 – 0931. Email: [email protected], [email protected]