A Survey of Flora and Fauna on Bracken Cave Property Techniques in Wildlife Management Project Spring 2013

Texas State University, San Marcos‐ Department of Biology

Ring‐tailed cat photo by Kendall AuBuchon

Edited by Thomas R. Simpson Jacqueline Ferrato

INTRODUCTION

Bat Conservation International (BCI) was founded in 1982 with a mission to conserve the world’s bats and their ecosystems in order to ensure a healthy planet. In 1991 BCI pursued that commitment by purchasing five acres surrounding the world’s largest bat colony, Bracken Bat Cave (BCI 1999). Bat

Conservation International stewards the entire property by protecting endangered birds and providing efforts to restore the land by removing invasive species and revitalizing the native plants and animals on the property (Moore 2005). The Bracken Cave property is noted to be an area with a high concentration of invertebrates and vertebrates. In order to accomplish the desired restoration efforts a baseline inventory must be conducted to acquire a basal knowledge of the plants and animals residing in the area.

The Bracken Cave property is primarily composed of oak‐juniper and mixed brush woodland with juniper encroachment. The comfort‐rock outcrop complex and rumple‐comfort association characterizes the land (Moore 2005). The parent material is residuum weathered from limestone. The landform is comprised of plains with mixed grassland and midgrass oak savannah including a landscape of plateaus (USDA 2012). The hydrologic soil group consist of gravelly clay loam 0‐10 inches, very gravelly clay 10‐28 inches and bedrock from 28‐36 inches (More 2005). Historically the property is believed to be an area that was once predominantly grassland studded with occasional oak trees

(Moore 2005).

Our objectives involved determining the species richness and diversity of avifauna, herpetofauna, mammals, and vegetation on the Bracken Bat Cave and Natural Reserve area through a baseline inventory of various survey techniques. The purpose of conducting a baseline inventory involves recording the important community conservation values such as the species composition and richness of a variety of different biota and the current conditions of the property. Baseline information establishes the foundation from which future management plans will be generated. By documenting

3 current conditions a foundation is established to provide future stewardship guidelines and management activities (Land Trust Alliance 2011).

STUDY SITE

The Bracken Bat Cave property consists of 282 ha (697 ac) of grassland and woodland habitat located northeast of San Antonio in Comal County. This area can potentially support unique flora and fauna as it lies at the juncture between two eco‐regions, the Edwards Plateau and the Blackland Prairie.

This property is of ecological importance because it is home to the largest bat colony in the world and provides habitat for endangered birds; however, it is surrounded by housing development, and urban expansion has increased over the past 20 years since BCI acquired ownership of the cave.

The survey was conducted on three separate weekends with the first being Feb. 15‐17, second

March 8‐10, and the final trip March 28‐30. Ten survey points were selected with a minimum distance between each point of 250 m to prevent sample overlap. Ten points were selected based on habitat type, either forested or open grassland, and used as sampling sites. The sites were selected to provide an equal mixture of grass prairie and dominate tree cover (Figure 1).

Figure 1. Bracken Bat Cave property boundary with ten sampling sites, including the bat cave located at point 8 (29.687057°N 98.352532°W). Coordinates for all points found in Appendix B.

AVIFAUNA

Sara Miller, Mindy Murray, Michelle Rivera, Katelyn Turner

METHODS

Point counts were chosen for sampling the bird community because this technique is widely used and requires less time and effort than other survey methods (Forcey et al. 2006). Points were located in two habitat types, grassland and woodlands, to adequately represent the habitat found on the BCI property. A group of 4 researchers undertook this study, two as scribes and two as observers, working in pairs. Periodically other researchers had to stand in for individual surveyors due to schedule conflicts; however, efforts were made to curtail personnel substitutions to maximize consistency.

Morning observations began at first light, or as soon as possible thereafter, in order to survey during the

“dawn chorus” when birds are more active; and, on average, evening observations began 3‐4 hours before dusk and ended no later than sundown (Carlton and Mooy 2005). Each team of researchers

6 surveyed the same 5 points each iteration (points 1‐5 or 6‐10), and the order of site visits was reversed on alternating days to reduce the possibility of skewed data among sites due to time of day.

After arriving at each point, surveyors observed a 1‐min rest period followed by a 5‐min observation period (Hipperson 2010). During the observation period, all birds seen or heard within a 50‐m radius were recorded (Dettmers 1999). Aside from birds of prey, flyovers and flushed birds were excluded from the counts. An audio recorder was kept on hand, and field experts were consulted to confirm identification of unfamiliar bird calls. Researchers recorded GPS coordinates of points observed (Table

4), time of day, species, and number of individuals. (All tables are located in Appendix A).

All data analysis was based on direct counts from the 10 survey points. We used the Shannon‐Wiener

Index, H’ = ‐∑[(pi) × ln(pi)] to quantify diversity, and Pielou’s Index,

J’ = H’/ H’max., to measure evenness for each site, for the two habitat types, and for the total study area.

RESULTS AND DISCUSSION

From 80 point count surveys, we identified 310 individuals of 37 different species (Appendix A).

We also detected 4 more species incidentally: Cedar Waxwing (Bombycilla cedrorum), Common Ground

Dove (Columbina passerine), Great Blue Heron (Ardea herodias), and Western Scrub‐jay (Aphelocoma californica). Over 63% of all individuals identified at point counts consisted of six species: Northern

Cardinal (Cardinalis cardinalis), Baeolophus spp., Bewick’s Wren (Thryomanes bewickii), Mourning Dove

(Zenaida macroura), Carolina Chickadee (Poecile carolinensis), and Ruby‐crowned Kinglet (Regulus calendula) (Appendix A). Cumulative analysis consisted of species richness, abundance, relative abundance, diversity, and evenness. Figure 3 shows the distribution of cumulative data.

For each point, data analysis included species richness, abundance, relative abundance, diversity, and evenness (Fig. 4, 5). Species richness values varied from 6‐19 at points 6 and 8, respectively. The Golden‐cheeked Warbler (Setophaga chrysoparia), an endangered species in central

Texas, was found at points 1, 3, and 10, all of which were forested areas. Evenness was lowest at point

10, with a value of 0.8469, and highest at point 6, with a value of 0.9435, closely followed by point 5, at

0.9415. Raw abundance values varied from 19 at point 6, to 87 at point eight.

Species richness, abundance, relative abundance, diversity and evenness were calculated for each habitat type. In the grassland habitat 23 species were identified while 28 were identified in the woodland habitat. Abundance was also much higher in woodland habitat (188) than grassland habitat

(122) as well as evenness (0.8638 and 0.6466, respectively) (Fig. 6).

Figure 3. Cumulative abundance data of birds from all identifications and point counts. Abbreviations accepted by American Ornithological Union are listed in Appendix 1.

Figure 4. Species richness estimates at each point.

Figure 5. Evenness estimates at each point

Figure 6. Evenness, abundance, and species richness values in grassland and woodland habitats. Evenness values were multiplied by 100 to incorporate these variables into one figure. Bars reflect 95% confidence intervals.

The most abundant bird species found on this property was the Northern Cardinal, making up

20.32% of all sightings, followed by Baeolophus spp. which made up 12.58% of all sightings (Table 3).

The surprising part of this data however was the appearance of the endangered Golden‐cheeked

Warbler making up 1.29% of all sightings (Table 3). It was unique to see several raptor species within 50 meters of our points; these included Turkey Vulture, Sharp‐shinned Hawk, Great Horned Owl, Red‐tailed

Hawk, and Red‐shouldered Hawk, each making up less than 1% of total sightings (Table 3). When we examined the abundance, richness, and evenness of species found in woodlands versus grassland habitats in Figure 5, all 3 indices were higher in woodland habitats. This could indicate that this is the preferred habitat for a large number of the bird species found on the BCI property. Point 8 was unique among the points in that it was the location of the Bracken Bat Cave entrance. This point can be considered, at least in part, edge habitat; it is shown also to have significantly higher species richness and abundance than the other points

HERPETOFAUNA

Kristin Covert, Michelle Durfee, Jose Martinez, Ann McMaster, Sarah Reverman

METHODS

Sampling techniques employed consisted of using drift‐fence funnel traps, cover boards, and time constraint active searches with the use of stump rippers. Identification of herpetofauna was determined with the use of A Field Guide to Reptiles and Amphibians: Eastern and Central North

America (Conant and Collins 1998).

Drift‐Fence Funnel Trap. ‐ Drift‐fence funnel traps have been shown to be superior in overall trap success with lowest incidents of capture fatalities (Todd et al. 2007, Farallo et al. 2010). Combining drift‐fence funnel traps with large drift‐fence pit fall traps is the best practice for maximizing trap efficiency and yields (Todd et al. 2009; Farallo et al. 2010). Using these sampling techniques increases the potential of collecting fossorial, semifossorial and cryptic species (Todd et al. 2009, Farallo et al.

2010). However, the exceedingly rocky soil did not lend itself to the employment of pitfall traps and as such, only drift fence funnel traps were employed.

Due to limited resources, three drift fence funnel traps were deployed, two at contrasting habitats of covered and uncovered (sites 8 and 10 respectively) and one at site 9 to take advantage of the pond. Refer to Figure 2 for location of survey points. Each drift fence funnel trap consisted of metal flashing roughly 25‐30 feet in length that was anchored in place by metal stakes. The amount of stakes needed to successfully anchor each piece of flashing varied depending on the number and size of rocks encountered in relation to the length of flashing. The ends of each flashing were bisected by a double‐ ended entrance only funnel trap. Placing the funnel traps with the ends of the flashing bisecting the trap entrance facilitated the trapping of herpetofauna by one of the traps that traversed either side of the metal flashing. The flashing was left in place for the duration of the three surveying trips, while the funnel traps were removed during non‐sampling days. During sampling days, funnel traps were left in place and checked three times a day, once upon arrival to BBC, once in the afternoon and once upon departure from BBC.

Cover Boards. ‐ Cover boards were implemented at canopied, non‐canopied sites, and at the pond as an alternative to using pitfall traps. Previous research indicates that those cover boards placed on non‐canopied sites will typically yield more incidence of visitations from herpetofauna than those placed on canopied sites (Hampton 2007). It was also noted in Hampton’s study that the use of corrugated tin provided slightly higher incidence of trapping reptiles and amphibians (2007). The cover boards were roughly 30x30 inches and made of compressed plywood with identifiable markings spray painted on the edges and upper surface of the board for easy detection. Cover boards were placed at site 8 for covered habitat, site 10 for uncovered habitat and at site 9 for access to the pond water. Each cover board was left at their respective site for the duration of the survey time as incidence of use by herpetofauna increased with time as the cover boards begin to create their own microhabitat (Hampton

2007). Cover boards were checked twice a day in the mornings and evenings or upon arrival to BBC and upon departure from BBC. Each cover board was checked by lifting the far side of the cover board by hooking it with the curved end of a stump ripper, lifting the board toward one’s body. This method ensured the safety of the biologist in the event that there was a venomous snake encountered under the cover board.

Time Constrained Active Searches. ‐ Time‐constrained active searches were conducted at each site in order to collect data on the presence or absence of herpetofauna. Typically, fewer species are found during active searches than trapping and cover board sampling but aid in detecting herpetofauna that are sedentary or difficult to trap (Crosswhite et al. 1999). Three sites were surveyed by five people searching for one hour at each of three sites on two different days, with one hour each at 4 sites on one day giving a total of 50 person‐hours between 10 sites. The searches were conducted from the center of the point, expanding outward, with one person in charge of the time. Searches included flipping of logs, rocks and brush piles as well as scans of shrub and tree canopies. Specimens were identified upon collection to limit as much stress on the individual as possible. Specimens that were small enough were placed in a glass mason jar with a cotton scarf locked in place by the screw on lid to help limit handling time and ease the identification process. Larger herpetofauna were held in pillow cases fitted with a

550 cord to cinch the pillow case shut. This helped to reduce incidences of specimens biting the handlers and served to limit the handing time during the identification process.

RESULTS AND DISCUSSION

Due to not having sighted much herpetofauna, our original plan to evaluate richness would not have been appropriate. In order to evaluate species richness, there needs to be a significant number of individuals for each species. Instead, we evaluated the presence or absence of every species reported. A total of 10 species of herpetofauna were observed with a total of 22 individual specimens observed.

Site 8 had the highest incidence of herpetofauna presence while site 1 yielded no observed herpetofauna (Table 2). The time constraint search method accounted for 77% of the observed herpetofauna with incidentals providing the other 33% (Table 1). No specimens were observed in or around the cover boards or the drift fence funnel traps. Though the detection of the Coral snake

(Micrurus fulvius) was during a time‐constraint search, the snake escaped before exact identification could be ascertained. The common ground skink was observed across more sites and habitat types.

While not detected at more than three sites, the Rose Belly Lizard was tie with the common ground skink for the most detected herpetofauna . More individuals and species were observed at the canopied sites than at the open sites. There was an abundance of leaf litter in the mixed woodland canopied sites that provided a duff layer utilized as cover (Table 3).

Common Ground skink Rose‐bellied lizard

Western Coachwhip

Table 1. List of species caught, site, and method of capture.

Common Name Species Name Method of Observation Incidental Time‐ Cover Drift Constrained Board Fence searches Common Ground Skink Scincella lateralis 3 2 0 0 Rough Earth Snake Virginia striatula 1 1 0 0 Flathead Snake Tantilla gracilis 0 1 0 0 Short‐lined Skink Eumeces tetragrammus 0 2 0 0 brevilineatus Rose Belly Lizard Sceloporus variabilis 0 5 0 0 marmoratus Western Coach Whip Masticophis flagellum 1 0 0 0 testaceus Coral Snake Micrurus fulvius 0 1 0 0 Texas Rat Snake Elaphe obsoleta 0 1 0 0 lindheimerii Six Lined Racerunner Cnemidophorus 1 0 0 0 sexlineatus Texas Spiny Lizard Sceloporus olivaceus 1 2 0 0 Total 7 15 0 0

Table 2. Species presence and occurrence by site.

Species Sites Count Common Ground Skink 4,6,8,9 5 Rough Earth Snake 5,9 2 Flathead snake 8 1 Short‐lined Skink 3,10 2 Rose Belly Lizard 2,3,8 5 Western Coach Whip 2 1 Coral Snake 4 1 Texas Rat Snake 3 1 Six Lined Racerunner 8 1 Texas Spiny Lizard 7,8 3 Total : 22

Table 3. Species occurrence by habitat type .

Species Forested Grassland Count Common Ground Skink 3 2 5 Rough Earth Snake 1 1 2 Flathead Snake 1 Short‐lined Skink 1 1 2 Rose Belly Lizard 4 1 5 Western Coach Whip 0 1 1 Coral Snake 1 0 1 Texas Rat Snake 1 0 1 Six Lined Racerunner 1 0 1 Texas Spiny Lizard 1 2 3 Total 14 8 22

Herpetofaunal abundance was low on this survey, with only ten species detected on the property, five of which detected at site 8. We believe this is due to the terrain and resources provided by that site, which not only is the location of the bat cave opening, but a mosaic of covered, uncovered, grassy, and rocky habitat. Other reasons for the low detection we believe to be attributed to variations within the high canopied plant compositions, seasonal timing of the survey, insufficient available time to spend trapping and searching, and the terrain did not lend itself to adequate employment of trapping techniques. Though the survey was attempting to compare high canopied to open sites, there was a noticeable difference in the plant composition between the high canopied sites that may have had an effect on our results. The two high canopied types could be distinctly separated into an Oak (Quercus spp.) mix series and an Ashe Juniper thicket. Each of these two series then formed observable transition zones with open sites that were distinctly unique from each other. In the cedar‐grassland sites had no transition zone to speak of, but rather an abrupt edge joining cedar canopy and mixed grasses with limited to no woody shrub along the series perimeter. The Ashe Juniper thicket understory was made up almost exclusively of cedar pine duff that was tightly compacted roughly 2 inches thick. In the Oak series, however, there was a relative transitional zone noticeable by the presence of sparsely growing

16 woody shrubs. The understory of the Oak series was composed of a mix of various woody shrub and vine with scattered dead wood throughout. The duff layer was roughly 3 to 4 inches thick, loosely packed, with a noticeable under layer of decomposition. Having this much variation among our canopied sites alone, perhaps affected our detection of herpetofauna. There was a noticeable difference in species and specimen detection between the two canopy cover sites, with more species and specimen detection occurring in the Oak species series. A reason for this may be that the Oak species series provides enough variation of habitat and forage potential that it lends itself to greater species presence, especially when adjacent to an open site. Possible reasons for low trap success rate include season (temperatures too low), time restrictions, and the rocky terrain when building drift fences.

Flathead snake

LARGE MAMMALS

Virginia Brown, Meredith Hominick, Malcolm Kiddle, Joshua Robbins

METHODS

Each if the 10 sites had at least one camera trap and one Bio‐Foam scent station with some oats.

Bio‐Foam is an easily manipulated piece of foam that is used as an impression kit for podiatry in the medical field. In the setting of this survey Bio‐Foam is being used to obtain tracks of the biota at each site. During the first sampling event cat food in conjunction with predator lure was used at every site without a scat survey. For the other two sampling events predator lure was used by itself with a completed scat survey at each site. Scat surveys consisted of 3 line transects at each site. Each member of the team would walk an estimated 50m in different directions searching for scat of any large mammals. Three types of cameras were used: Bushnell, Cuddebacks, and BirdCam cameras. Each was equipped with a memory card that consisted of a storage capacity of 2 to 8 gigabyte. Each camera was initially set up within 50m of the site. Due to the low number of captures the protocol changed to accommodate the game trails, open areas, and water sources. Incidental sightings were also included in

18 the survey that could have involved seeing or hearing a large mammal outside the site but still within the Bracken Bat Cave property.

RESULTS AND DISCUSSION

Out of the 5 nights sampled a total of 9 species of large mammal recorded. There were 3 species recorded with the biofoam, 7 species on the cameras, and 2 species seen or heard incidentally (Table 4).

The highest recorded species, Sus scrofa, was recorded 6 times between sites 8 and 9. The 4 lowest recorded species on camera were Urocyon cinereoargentus, Odocoileus virginianus, and Cervus axis with one sighting each. Mephitis mephitis, Procyon lotor, and Dedelphis virginiana were recorded twice at site 8 and 10B.

Table 4. Total number of species captured, location, and method of each capture.

Common Name Species Total # Point Method Reorded Gray Fox Urocyon cinereoargentineus 1 8 Camera Striped Skunk Mephitis mephitis 2 8, 10B Camera, Bio foam Raccoon Procyon lotor 2 8, 10B Camera, Bio foam Feral Hog Sus scrofa 6 8, 9 Camera Virginia Opossum Didelphis virginianus 2 8 Camera, Bio foam White‐tailed Deer Odocoileus virginianus 1 8 Camera Axis Deer Cervus axis 1 8 Camera Coyote Canis latrans 2 10B Biofoam, Incidental Ring‐tailed Cat Bassariscus astuts 1 Incidental

We provide evidence of the presence of several large mammal species and witnessed incidental sightings of a few more. This kind of presence/absence survey was performed for BCI in order to further their goal of returning the Bracken Bat Cave and Nature Reserve to historic conditions. It is important to know what species of large mammal would historically exist in this area and what species are present today that are relatively new. This will aid management strategies in the future not just with the

Raccoon Striped skunk

Axis deer Feral hogs

Bio‐foam prints: coyote and skunk

20 change in distribution of native fauna but with invasive species, such as the Axis Deer, as well.

Whitetail deer, axis deer, and feral hogs were present in the area and were recorded with camera traps, on scat surveys, and actually sighted during site visits. Feral hogs, an invasive and problematic species, were recorded with high frequency, more so than whitetail deer. Feral hogs are well known for their hardiness and their ability to survive in a myriad of environments. Whitetail deer, however, are browsers and depend on woodland vegetation such as mesquite (Prosobis glandulosa) and Mexican persimmon (Diospyros texana) for mast, twigs, fruits, and new shoots. Whitetail deer also depend heavily on forbs, but will not readily eat grasses if mid‐level woody plants and low‐level forbs are available (Teer 1965). Axis deer are an introduced species that primarily occupy woodland habitat and as a relatively recent introduction to Texas has not historically occurred in the surveyed area.

Mesopredators such as raccoons, striped skunks, gray fox, and the Virginia opossum were recorded on camera, biofoam, or sighted within the area of Bracken Cave. Raccoons and the Virginia opossum are often known to occupy the same area and are putative competitors in habitats that contain a vegetative mosaic of a wooded area with clear openings. The two species thrive in areas with habitat fragmentation, allowing for an increased area of edge habitat in which a vegetative mosaic might occur. The expanse of the Bracken Bat Cave and Nature Reserve was patchy in terms of vegetation, with copses or thickets of Juniper often surrounding patches of open grass. Raccoons and

Virginia opossums select for woodland and forest habitats and do not readily select for grassland areas.

This means that, historically, these mesopredators are unlikely to have occurred in the surveyed area in large numbers (Ginger 2003). The striped skunk is a habitat generalist, and does just as well in woodland surroundings as in grassland areas studded with stands of trees (Neiswenter 2007). The gray fox, similarly, does well in any area with favorable climate and a good supply of small mammal prey

(Chamberlain 200). Both the gray fox and striped skunk were recorded as present in the surveyed area, and are likely to have been present in the area historically.

SMALL MAMMALS

Trevor Baum, Lauren Cody, Thomas Lynn, Nicole Pearsall, Salem Reyna (Wednesday)

Kendall AuBuchon, Casey Burch, Samuel Frantzen, Jordon Moore, Madison Torres (Friday)

METHODS

Sherman live traps were used (Sherman folding small animal trap; 7.5 X 9 X 23 cm) at each site.

At half of the sites traps were arranged along four parallel lines with 10 traps per line while the other half of the sites had traps arranged along five parallel lines with 8 traps per line. Line transects were situated 20 meters apart. We baited traps with a rolled oats and birdseed mixture. On trap nights where temperatures were expected to fall below 10°C, we also provided cotton bedding material to minimize accidental take (Sikes et al. 2011). We set traps before dusk and checked them at dawn; on one occasion, we left traps open during the day and checked them before dusk, as temperatures did not exceed 27°C (Jones et al. 1996, Manley et al. 2006).

Trapped animals were transferred to a plastic bag for handling and species level identification, and then released (Manley et al. 2006). We then disinfected those traps that contained an animal with ethanol before resetting or removal from the site (Mills et al. 1995, Yunger and Rana 1999). We used

22 protective gloves when handling traps and animals in order to minimize exposure to fecal material and disease and for protection from animal bites and scratches. All animals were handled in accordance with guidelines set forth by the American Society of Mammalogists (Sikes et al. 2011).

RESULTS AND DISCUSSION

We caught 74 small mammals with 1,395 trap‐nights total, giving us a capture rate of 5.3%. We captured 5 species: the Texas mouse (Peromyscus attwateri), the white‐ankled mouse (Peromyscus pectoralis), the white‐footed mouse (Peromyscus luecopus), the deer mouse (Peromyscus maniculatus), northern pygmy mice (Baiomys taylori), and the hispid cotton rat (Sigmodon hispidus) (Table 5). The species that was most abundant was Peromyscus attwateri with 35 individuals (47.3%). Peromyscus pectoralis was the next most abundant with 19 individuals (25.7%). Peromyscus leucopis and

Peromuscus maniculatus were least abundant with one individual each.

There were 42 small mammals from 4 species caught in the grassland sites: B. taylori, P. attwateri, P. pectoralis, and S. hispidus (Table 6). P. attwateri was the most abundant in both forest and grassland habitats (68.8% and 30.9%, respectively), while P. luecopus, P. maniculatus, and S. hispidus were least abundant at forest sites (3.1% each), and B. taylori was least abundant in grassland sites

(14.3%). The S‐W index is lower for forest sites at 0.91, and higher for grassland sites at 1.35. There were a grand total of 74 small mammals from 6 species, with an overall diversity index of 1.32 (Table 5).

Table 5. Species list and sites of occurrences

Common name Species Total Number Texas mouse Peromyscus attwateri 35 White‐ankled mouse Peromyscus pectoralis 19 Deer mouse Peromyscus maniculatus 1 White‐footed mouse Peromyscus luecopus 1 Hispid cotton rat Sigmodon hispidus 12 Northern Pygmy mouse Baiomys taylori 6

Table 6. Species list and diversity by habitat type.

Habitat Type Species Number Total Number S‐W Index Peromyscus attwateri 22 Peromyscus pectoralis 7 Peromyscus luecopus 1 32 0.91 Forest Peromyscus maniculatus 1 Sigmodon hispidus 1

Baiomys taylori 6 Grassland Peromyscus attwateri 13 42 1.35 Sigmodon hispidus 11 Peromyscus pectoralis 12

Peromyscus sp.

Hispid cotton rat

Polydactyly in Peromyscus sp.

Table 7. Species, individual captures, and Shannon‐Wiener diversity indices by site

Site Species Number S‐W Index 1 Peromyscus attwateri 3 0.69 P. pectoralis 3

2 Baiomys taylori 4 0.69 P. attwateri 5

3 P. attwateri 6 Sigmodon hispidus 1 0.74 P. pectoralis 1

4 S. hispidus 7 0.38 P. pectoralis 1

5 P. attwateri 10 0.30 P. maniculatus 1

6 P. attwateri 2 0.45 P. pectoralis 10

7 B. taylori 2 0.67 S. hispidus 3

8 P. attwateri 7 0.53 P. pectoralis 2

9 P. attwateri 1 0.69 S. hispidus 1

10 P. attwateri 1 P. luecopus 1 1.04 P. pectoralis 2

The majority of small mammal species captured were in the genus Peromyscus with one

Sigmodon and one Baiomys. The largest number of captures was the species P. attwateri with a total of

35 individuals captured. The second largest number of captures was P. Pectoralis with 19 individuals captured. Both S. hispidus and P. leucopus only had fewer individual captures. The forest habitats had a

26 total of 32 captures, and the grassland having a total of 42 captures and we could conclude that the grassland is more productive due to the higher capture rate. Some future research to look into would be whether or not grasslands are better for food quality or abundance, or protection from predators and nest locations. When compared to other surveys of small mammals in the Texas area, we have similar species richness and diversity (Baccus et al. 2000). The study done by Baccus et al. shows highest richness and diversity in both of their grassland habitats (2000). Our data show similar results where grassland sites overall had a higher capture rate as well as higher diversity.

VEGETATION

Marcus Patterson, Dominic DeSantis, Calvin Kettler, Eric Torkildsen, Christopher Balcom (Wednesday)

Jim Ma, Mark Miranda, Brittni Molinar, Brian Podgurski, Ashley Seagroves (Friday)

METHODS

We used a line transect technique at each survey site. The lines were run from a central point at each site, and spanned 100m in each direction (N, E, S, and W). We stopped every 10m along each line to conduct the following analyses: Line intercept (woody cover), Daubenmire frame analysis

(herbaceous ground cover), and the vegetation profile board (vertical cover). The three previously listed analyses were run a total of 40 times at each site on the property (every 10m along each of the four

100m line transects). We used a handheld compass to determine the direction of our transect lines.

Additionally, lines were run in a straight‐line, these details helped us eliminate any transect navigation bias.

Line Intercept. – At each survey site a center point was randomly selected and four 100‐m lines were run at 90° intervals in the four cardinal directions (N, S, E, W). The 100‐m transect lines were run

28 using a Keson 100‐m graduated metric tape measure. All hard and soft wood plant species that crossed the line (above or below) were recorded to species level. Additionally, we measured total length coverage of each intercepting woody plant along and across the line (Canfield 1941; Kaiser 1983). Direct percent cover was calculated per species by dividing the total length covered by the total length of the line (100‐m). Sample size, species richness, and abundance were calculated for each site. Diversity for each site was calculated using the Shannon‐Wiener Diversity Index.

Daubenmire Technique. – The daubenmire frame technique involves the placement of small quadrats or frames (20x50cm) on the ground and visually estimating ground cover using several percent coverage classes (1=0‐5%, 2=5‐25%, 3=25‐50%, 4=50‐75%, 5=75‐95%, 6=95‐100% coverage)

(Daubenmire 1959; Abrahamson et al. 2011). We chose four categories of cover to record that consisted of grasses, forbs, rocks, litter (dead material and mast), and bare ground. Daubenmire readings were taken along each line at every 10m interval. We took the mean of the midpoints of each category to estimate percent ground cover for each category type.

Vegetation Profile Board (VPB). – Visual obscurity was estimated at the center point (0‐m), the midpoint (50‐m), and the endpoint (100‐m) of each line. A group member would move 15m in one perpendicular direction with the board and stop, while another group member would estimate the class of coverage for each block on the VPB. Following this, the group member with the board move 15m in the exact opposite perpendicular direction and the same procedure was followed (Nudds 1977). Vertical coverage was estimated based upon coverage classes. Classes for the VPB included: 1=0‐20%, 2=21‐

40%, 3=41‐60%, 4=61‐80%, and 5=81‐100% (Nudds 1977). Percent vertical cover was estimated by averaging the means of the midpoints of the blocks.

RESULTS AND DISCUSSION

Daubenmire Technique. – Daubenmire frame technique displayed little difference in percent ground cover across sites for the forb, grass, and bare ground categories for the forested sites (table 8).

The category entitled, “litter”, made up the largest proportion of ground cover, this comprises both dead material and mast. Of the ten total survey sites of the property, four were characterized as predominately grassland habitat, and the remaining six sites were characterized as predominately forest habitat. The four grassland habitats (4, 6, 7, and 9) had more grass and forb percent ground cover (by proportion), and less litter.

Table 8. Results from the Daubenmire analysis and percent ground cover for 4 cover categories: grass, forbs, litter and bare ground.

Grassland Habitat Site Grass (%) Forbs (%) Litter (%) Bare Ground (%) 2 21.26 7.53 56.96 14.25 4 47.7 5.25 31.91 15.14 6 43.75 7.44 29.95 16.75 7 25.17 16.79 40.62 17.42 9 25.38 17.5 36.44 38.56 Mean 32.65 10.9 39.18 20.42

Forested habitat Site Grass (%) Forbs (%) Litter (%) Bare Ground (%) 1 13.25 12.88 54.2 9.95 3 19.63 9.86 58.45 12.05 5 19.58 11.21 48.48 20.72 8 25.38 39.81 23.5 21.31 10 6.38 5.88 42.19 22.21 Mean 16.84 15.93 45.36 17.25

Vegetation Profile Board. – Vertical cover was fairly constant across site 6, 8, 9, and 10, however site 1 displayed moderately higher percent vertical cover (71.98%) than all other sites (Table 9).

The grassland habitat provides slightly less vertical cover than the forested habitat.

Table 9. Vegetation Profile Board ‐ % cover by site

Grassland Forested Site Avg. % Vertical Site Avg. % Vertical Cover Cover 2 17.67 1 71.98 4 46.76 3 58.67 6 51.47 5 83.67 7 39.67 8 55.21 9 50.48 10 43.61 All sites 41.21 All sites 62.63

Line Intercept. – A total of 15 woody plant species were encountered with a total of 751 individuals intercepting the tape measure. Ashe Juniper (Juniperus ashei) accounted for over half of the observed woody species (N=328), while Texas Persimmon (Diospyros texana, N=98) and Plateau Live

Oak (Quercus fusiformis, N=77) were the next two most commonly encountered species. The Shannon‐

Wiener diversity index yielded an H=1.3791 rating among all sites, although there was some variation between sites (Table 11).

As this survey represents one of the first comprehensive vegetation analyses of the Bracken Bat

Cave property, the data should provide useable baseline information that can be applied in future studies to access any fluctuation in the properties plant communities over time. Further encroachment by J. ashei would lead to increased homogenization of the available habitat to wildlife, and thus potentially decrease not only diversity within the plant communities, but also the diversity of native wildlife. As characterized by the data collected from our Daubenmire analysis, J. ashei often significantly decreases herbaceous growth in a system while increasing the proportion of bare ground and dead material (Davenport et al. 1998). Litter

(dead material and mast) constituted the largest portion of estimated “ground cover” across all

31 surveyed sites. The five grassland habitats (2, 4, 6, 7, and 9) were seen to have far more grass and forb ground cover, less litter (Table 4, 5), and lower percent vertical cover (Table 7, 8) when compared to the

Table 10. Percent Canopy Cover

Percent Cover (%) Common name Species Grassland Forested Sweet acacia Acacia farnesiana 0.75 0 Agarita Berberis trifoliolata 1.36 5.63 Sugarberry elm Celtis laevigata 0.04 0.37 Brasil Condalia hookeri 0.132 0.27 Texas Persimmon Diospyros texana 2.41 6.41 Kidneywood Eysenhardtia texana 0 0.15 Deciduous holly Ilex decidua 0.49 0 Ashe juniper Juniperus ashei 22.15 38.1 Prickly pear Opuntia engelmanii 1.21 1.56 Christmas cactus Opuntia leptocaulis 0.028 0.16 Mistletoe Phoradendron tomentosum 0 0.05 Plateau Live oak Quercus fusiformis 14.72 19.74 Chinkapin oak Quercus mehlenbergii 0 0.13 Green Briar Smilax bona‐nox 0.07 0 Cedar elm Ulmus crassifolia 0.1 1.14

Table 11. Shanon‐Wiener Diversity index per site

Point S‐W index

1 0.89 2 1.77 3 1.29 4 0.9 5 0.81 6 1.65 7 1.38 8 1.95 9 1.54 10 0.87 Overall 1.38

five forest habitats (1, 3, 5, 8, and 10). Juniper (J. ashei) had slightly higher percent cover in grassland habitat, but its cover was fairly constant throughout all sites and habitats (Table 1, 2). Plateau Live Oak

(Quercus fusiformis) displayed the most significant variation in percent cover between habitat types, as its percent cover nearly doubled from grassland habitat to forest habitat (Table 1, 2). This is expected as

J. ashei and Q. fusiformis are the two most common tree species (along with D. texana) on the property, making the Oak‐Juniper woodland community the dominant forest type present. When analyzing the data by habitat type, it is important to note that classification of sites into grassland and forest habitat was not based on any strict definition of the two, only that they were either predominately grassland or forested. Additionally, the grassland habitat present did not ever completely cover any survey site, thus the four “grassland habitat” sites were more a mosaic of grassland and woodland rather than the thick oak‐juniper woodlands that dominated the six “forest habitat” sites.

Overall the forested sites had higher productivity for birds, herpetofauna, and large mammals.

Grassland sites were more productive for small mammals, however the difference was very little. This report provides BCI with an inventory of flora and fauna species for which to manage for. Both grassland and forested habitats are important for supporting native wildlife and both habitats should be managed for using various methods as BCI sees fit.

LITERATURE CITED

Abrahamson, I. L., C. R. Nelson, and D. L. R. Appleck. 2011. Assessing the performance of sampling designs for measuring the abundance of understory plants. Ecological Application 21:452-464.

Ansley, R. J., W. E. Pinchak, and D. N. Ueckert. 1995. Changes in redberry juniper distribution in northwest Texas. Rangelands 17:49–53.

Ansley, R.J. and G.A. Rasmussen. 2005. Managing Native Invasive Juniper Species Using Fire. Weed Technology 19:3 517-522.

Baccus, J. T., H. M. Becker, T. R. Simpson, and R. W. Manning. Mammals of the Freeman Ranch, Hays County, Texas. Freeman Ranch Publication Series 1-2000:1-31.

Bat Conservation International [BCI]. 1999. Annual Report 1998 – 1999. . Accessed 31 Jan 2013.

Bat Conservation International [BCI]. 2011. Annual Report 2010-2011. . Accessed 31 Jan 2013.

Bat Conservation International [BCI]. 2013. Bracken bat cave. . Accessed 30 January 2013.

Bat Conservation International [BCI]. 2013. Bracken Bat Cave. < http://www.batcon.org/ index.php/get- involved/visit-a-bat-location/bracken-bat-cave/subcategory.html?layout =subcategory&utm_campaign=bracken&utm_source=external&utm_medium=redirect>. Accessed 27 Jan 2013.

Benson, D.E. 2001. Wildlife and recreation management on private lands in the United States. Wildlife Society Bulletin 29:359-371.

Braun, C.E. 2005. Techniques for Wildlife Investigations and Management. Baltimore: Port City Press.

Brooks, M. L. and D. A. Pyke. 2001. Invasive plants and fire in the deserts of North America. In K.E.M. Galley and T. P. Wilson, eds. Proceedings of the Invasive Species Workshop: The Role of Fire in the Control and Spread of Invasive Species. Fire Conference 2000: The First National Congress on Fire Ecology, Prevention and Management. Miscellaneous Publ. 11. Tallahassee, FL: Tall Timbers Research Station. 15–30.

Brown, D.J., J.R. Dixon and M.R.J. Forstner. 1994. Visual summary of herpetofaunal diversity in Texas. The Southwester Naturalist 57:465-467.

Burger, L. 2006. "Creating Wildlife Habitat through Federal Farm Programs: An Objective Approach." Wildlife Society Bulletin 34: 994-999. Print

Burkhardt, J. W. and E. W. Tisdale. 1976. Causes of juniper invasion in southwestern Idaho. Ecology 57:472–484.

Cameron, G. and S. Spencer. 1981. Sigmodon hispidus. Mammalian Species. 158:1-9 Canfield, R. H. 1941. Application of the line-intercept method in sampling range vegetation. Journal of Forestry 39:388-394.

Carey, A. B., and M. L. Johnson. 1995. Small mammals in managed, naturally young, and old-growth forests. Ecological Applications 5:336-352.

Carlton, C., and R. Mooy. 2005. Bird survey methods. Community Biodiversity Survey Manual 9:1-10.

Caro, T. M., and G. O’Doherty. 1999. On the use of surrogate species in conservation biology.Conservation Biology 13:4.

Clark, D. R., C. O. Martin, and D. M. Swineford. 1975. Organochlorine insecticide residues in the free-tailed bat (Tadarida brasiliensis) at Bracken Cave, Texas. Journal of Mammalogy 56:429- 443.

Collier, B. A., M. L. Morrison, S. L. Farrell, A. J. Campomizzi, J. A. Butcher, K. B. Hays, D. I. MacKenzie, and R. N. Wilkins. 2010. Monitoring golden-cheeked warblers on private lands in Texas. Journal of Wildlife Management 74:140-147.

Coulloudon, B., Eshelman, K., Gianola, J., Habich, N., Hughes, L., Johnson, C., Pellant, M., Podborny, P., Rasmussen, A., Robles, B., Shaver, P., Spehar, J., and John Willoughby. 1996. Sampling Vegetation Attributes. USDA Forest Service, BLM, NRCS. Denver, Colorado.

Crosswhite, D.L, F.F. Stanley, and R.E. Thill. 1999. Comparison of methods for monitoring reptiles and amphibians in upland forests of the Ouachita Mountains. Proceedings of the Oklahoma Academy of Sciences 79:45-50.

Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Science 33:43-64.

Davenport, D.W., D.D. Breshears, B. P. Wilcox, and C. D. Allen. 1998. Viewpoint: sustainability of pinyon-juniper ecosystems—a unifying perspective of soil erosion thresholds. Journal of Range Management 51:231-240.

Dettmers, R., D. A. Buehler, J. G. Bartlett, and N. A. Klaus. 1999. Influence of point count length and repeated visits on habitat model performance. Journal of Wildlife Management 63:815- 823.

Dixon, K. L. 1955. An ecological analysis of the inter-breeding of crested titmice in Texas. University of California Publications in Zoology 54:125-206.

Dixon, K. L. 1990. Constancy of margins of the hybrid zone in titmice of the parus bicolor complex in coastal Texas. Auk 107:184-188.

Easton, W.E. and Martin K. 1998. The Effect of Vegetation Management on Breeding Bird Communities in British Columbia. Ecological Applications 8:1092-1103.

England, A. 2002. Members in action. BATS: Bat Conservation International 20: 3.

Eshelman, B. & G. Cameron. 1987. Baiomys taylori. Mammalian Species. 285:1-7. Ferguson, A.W., F.W. Weckerly, J.T. Baccus, and M.R. Forstner. 2008. Evaluation of predator attendance at pitfall traps in Texas. The Southwestern Naturalist 53:450-457.

Farallo, V.R., D.J. Brown, and M.J. Forstner. 2010. An improved funnel trap for drift fence surveys. The Southwestern Naturalist 55:457-460.

Forcey, G. M., J. R. Anderson, F. K. Ammer, and R. C. Whitmore. 2006. Comparison of two double- observer point-count approaches for estimating breeding bird abundance. Journal of Wildlife Management 70:1674-1681.

Fowler, N. L., and M. Simmons 2009. Savanna dynamics in central Texas: just succession? Applied Vegetation Science 12:23-31.

Ginger, S.M., E.C. Hellgren, M.A. Kasparian, and D. Levesque 2003. Niche shift by Virginia opossum following reduction of a putative competitor, the raccoon. Journal of Mammalogy 84: 1279- 1291.

Gurnell, J., and J. R. Flowerdew. 1990. Live trapping small mammals: a practical guide. Occasional Publications of the Mammal Society of London 3:1-39.

Hampton, P. 2007. A comparison of the success of artificial cover types for capturingamphibians and reptiles. Amphibia-Reptilia 28:433-437.

Hellmann, J. J., and G. W. Fowler. 1999. Bias, precision, and accuracy of four measures of species richness. Ecological Applications 9(3): 824-834.

Hipperson, C. 2010. A comparison of three bird survey methods used to identify 12 target species within forest fragments in Northern Madagascar. Frontier. . Accessed 22 Jan 2013.

Horsley, S.B., S.L. Stout, and D.S. DeCalesta. 1994. White-tailed deer impact on the vegetation dynamics of the Northern Hardwood Forest. Ecological Applications 13: 98-118.

Horvitz, C.C., S. Tuljapurkar, and J.B. Pascarella. 2005. Plant-animal interactions in random environments: habitat-stave elasticity, seed predators, and hurricanes. Ecology 86: 3312- 3322

Johnson, T. N. 1962. One-seed juniper invasion of northern Arizona grasslands. Ecological Monographs 32:187–207.

Jones, C., W. J. McShae, M. J. Conroy, and T. H. Dunz. 1996. Capturing Mammals. Pages 115- 155 in

D. E. Wilson, F. R. Cole, J. D. Nichols, R. Rudran, and M. S. Foster, eds. Measuring and monitoring biological diversity: Standard methods for mammals. Smithsonian Institution Press, Washington, D. C.

Kaiser, L. 1983. Unbiased estimation in line-intercept sampling. Biometrics 39:965-976

Kutac, E.A., and S.C. Caran. 1994. Birds and other wildlife of South Central Texas. University of Texas Press, Austin. Land Trust Alliance. 2011. Practical Pointers Services. Baseline Documentation Reports.

Lee, Y., and G. F. McCracken. 2005. Dietary variation of Brazilian free-tailed bats links to migratory populations of pest insects. Journal of Wildlife Management 86:67-76.

Manley, P.N., B. Van Horn, J. K. Roth, W. J. Zielinski, M. M. McKenzie, T. J. Weller, F. W. Weckerly, and C. Vojta. 2006. Multiple species inventory and monitoring technical guide. General Technical Report WO-73. Washington, DC: U.S. Department of Agriculture, Forest Service, Washington Office. 204 p.

Miller, R. F. and R. J. Tausch. 2001. The role of fire in juniper and pinyon woodlands: a descriptive analysis. In K.E.M. Galley and T. P. Wilson (eds.) Proceedings of the Invasive Species Workshop: The Role of Fire in the Control and Spread of Invasive Species. Fire Conference 2000: The First National Congress on Fire Ecology, Prevention and Management. Miscellaneous Publ. 11. Tallahassee, FL: Tall Timbers Research Station. 15–30.

Mills, J. N, T. L. Yates, J. E. Childs, R. R. Parmenter, T. G. Ksiazek, P. E. Rollin, and C. J. Peters. 1995. Guidelines for working with rodents potentially infected with hantavirus. Journal of Mammalogy 76:716-722.

Moore, A. 2005. Creating the Bracken bat cave and nature reserve. BATS: Bat Conservation International 23:7-9.

Nudds, T.D. 1977. Quantifying the vegetative structure of wildlife cover. Wildlife Society Bulletin 5:113-117.

Reed, M. J. 1995. Ecosystem management and an avian habitat dilemma. Wildlife Society Bulletin 23:453-457

Romero, H. and F. Ordenes. 2004. Emerging urbanization in the Southern Andes: environmental impacts of urban sprawl in Santiago de Chile on the Andean Pledmont. Mountain Research and Development 24(3): 195-199.

Sikes, R. S., W. L. Gannon, and 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.

Silvy, N. J. 2012. The wildlife techniques manual. Seventh Edition. Johns Hopkins University Press, Baltimore, Maryland, USA.

Slade, N. A., and S. M. Blair. 2000. An empirical test of using counts of individuals captured as indices of population size. Journal of Mammology 81(4): 1035-1045.

Strickler, G.S. 1959. Use of the densiometer to estimate density of forest canopy on permanent sample plots. Number 180. USDA Forest Service, Pacific Northwest Forest and Ranger Experiment Station. Portland, Oregon, USA.

Teer, J.G., J.W. Thomas, and E.A. Walker 1965. Ecology and management of white-tailed deer in the Llano basin of Texas. Wildlife Monographs 15:3-62.

Texas Parks and Wildlife Department. 2007. TPWD: Hill Country Vegetation. Retrieved April 21, 2013, from http://www.tpwd.state.tx.us/landwater/land/habitats/hillcountry/vegetation/

Todd, B.D., C.T. Winne, J.D. Willson, and J.W. Gibbons. 2007. Getting the drift: Examining the effects of timing, trap type, and taxon on herpetofaunal drift fence surveys. American Midland Naturalist, 158:292-305.

U.S. Fish and Wildlife Service. 1992. Golden-cheeked warbler recovery plan. U.S. Fish and Wildlife Service, Albuquerque, New Mexico, USA.

USDA. 2012. United States Department of Agriculture. Natural Resources Conservation Service. Web Soil Survey.

Yunger, John A., and Lynda A. Randa. 1999. Trap Decontamination using hypochlorite: effects on trappability of small mammals. Journal of Mammalogy 80:1663-1340.

Appendix A. Cumulative data for all bird species identified during point counts. Abundance, relative abundance, Shannon-Wiener diversity index (H’) and evenness were calculated. H’=2.843, H’max=3.611, J’=0.7873. Relative Species AOU1 Abundance Abundance Species Richness (37) 310 Northern Cardinal (Cardinalis cardinalis ) NOCA 63 20.3% Black-crested and Tufted Titmouse (Baeolophus spp. TUTI/BCTI 39 12.6% )B2e wick's Wren (Thryomanes bewickii ) BEWR 32 10.3% Mourning Dove (Zenaida macroura ) MODO 23 7.4% Carolina Chickadee (Poecili carolinensis ) CACH 21 6.8% Ruby-crowned Kinglet (Regulus calendula ) RCKI 18 5.8% Northern Mockingbird (Mimus polyglottos ) NOMO 13 4.2% Chipping Sparrow (Spizella passerina ) CHSP 10 3.2% Lark Sparrow (Chondestes grammacus ) LASP 9 2.9% White-eyed Vireo (Vireo griseus ) WEVI 9 2.9% Lincoln's Sparrow (Melospiza lincolnii ) LISP 8 2.6% Savannah Sparrow (Passerculus sandwichensis ) SAVS 7 2.3% Vesper Sparrow (Pooecetes gramineus ) VESP 7 2.3% Cactus Wren (Campylorhynchus brunneicapillus ) CACW 6 1.9% Rufous-crowned Sparrow (Aimophila ruficeps ) RCSP 6 1.9% American Kestrel (Falco sparverius ) MAKE 4 1.3% Golden-cheeked Warbler (Setophaga chrysoparia ) GCWA 4 1.3% White-winged Dove (Zenaida asiatica ) WWDO 4 1.3% Carolina Wren (Thryothorus ludovicianus ) CARW 3 1.0% Eastern Phoebe (Sayornis phoebe ) EAPH 3 1.0% Field Sparrow (Spizella pusilla ) FISP 3 1.0% Turkey Vulture (Cathartes aura ) TUVU 3 1.0% Sharp-shinned Hawk (Accipiter striatus ) SSHA 2 0.6% Yellow-rumped Warbler (Setophaga coronata ) YRWA 2 0.6% Brown-headed Cowbird (Molothrus ater ) BHCO 1 0.3% Crested Caracara (Caracara cheriway ) CRCA 1 0.3% Great Horned Owl (Bubo virginianus ) GHOW 1 0.3% Ladder-backed Woodpecker (Picoides scalaris ) LBWO 1 0.3% Red-tailed Hawk (Buteo jamaicensis ) RTHA 1 0.3% Red-shouldered Hawk (Buteo lineatus ) RSHA 1 0.3% Song Sparrow (Melospiza melodia ) SOSP 1 0.3% Spotted Towhee (Pipilo maculatus ) SPTO 1 0.3% Western Meadowlark (Sturnella neglecta ) WEME 1 0.3% Rio Grande Wild Turkey (Meleagris gallopavo ) WITU 1 0.3% Yellow-throated Warbler (Setophaga dominica ) YTWA 1 0.3%

Appendix B. Decimal degrees GPS coordinates for sampling points.

Point Lat Lon 1 29.68776° 98.34940° 2 29.68311° 98.34523° 3 29.67950° 98.34976° 4 29.68090° 98.35511° 5 29.68221° 98.35825° 6 29.68456° 98.34693° 7 29.68305° 98.35219° 8 29.68695° 98.35254° 9 29.68931° 98.35342° 10 29.69181° 98.35157°