• TABLE OF CONTENTS
HIDE
 Title Page
 Citation
 Introduction
 Results
 Discussion
 Population monitoring
 Applied research
 Population dynamics
 Acknowledgement
 Literature cited
 Table 1: Location of captures of...
 Table 2: Density estimates of Key...
 Table 3: A comparison of different...
 Figure 1: Location of the study...
 Figure 2: North Key Largo trapping...
 Figure 3: Summary of the tota number...
 Figure 4: Seasonal abundance of...
 Figure 5: Comparison of abundance...
 Figure 6: Sex ration fo small mammals...
 Figure 7: Age structure of Key...
 Figure 8: Age structure of Key...
 Figure 9: Recruitment as measured...






Group Title: Florida Cooperative Fish and Wildlife Research Unit technical reports
Title: Status survey for the Key Largo woodrat
CITATION PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00067422/00001
 Material Information
Title: Status survey for the Key Largo woodrat
Series Title: Florida Cooperative Fish and Wildlife Research Unit Technical Report 55
Physical Description: Book
Language: English
Creator: Frank, Philip A.
H., Franklin Percival
Keith, Britt W.
Sasso, Chris R.
Gaines, Michael S.
Duquesnel, James
Affiliation: University of Florida -- Florida Cooperative Fish and Wildlife Reserch Unit -- Institute of Food and Agricultural Sciences
Publisher: Florida Cooperative Fish and Wildlife Research Unit, U.S. Geological Survey, Biological Resources Division
Publication Date: August 1997
 Subjects
Spatial Coverage: North America -- United States of America -- Florida
 Record Information
Bibliographic ID: UF00067422
Volume ID: VID00001
Source Institution: Marston Science Library, George A. Smathers Libraries, University of Florida
Holding Location: Florida Agricultural Experiment Station, Florida Cooperative Extension Service, Florida Department of Agriculture and Consumer Services, and the Engineering and Industrial Experiment Station; Institute for Food and Agricultural Services (IFAS), University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Table of Contents
    Title Page
        Title Page
    Citation
        Section
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
    Results
        Page 6
        Page 7
        Page 8
    Discussion
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
    Population monitoring
        Page 14
    Applied research
        Page 15
    Population dynamics
        Page 16
        Page 17
    Acknowledgement
        Page 18
    Literature cited
        Page 18
        Page 19
        Page 20
    Table 1: Location of captures of small mammals on trapping transects on trapped between January-May, 1995
        Page 21
    Table 2: Density estimates of Key Largo woodrats and Key Largo cotton mice based on live trapping on four grids North Key Largo from July 1995 - June 1996
        Page 22
    Table 3: A comparison of different density estimates of Key Largo woodrats and Key Largo cotton mice based on grid trapping conducted on North Key Largo
        Page 23
    Figure 1: Location of the study area at the extreme southern tip of the Florida peninsula
        Page 24
    Figure 2: North Key Largo trapping locations
        Page 25
    Figure 3: Summary of the tota number of individuals captures on four 3.8 ha grids on North Key Largo trapped July 1995 through June 1996
        Page 26
    Figure 4: Seasonal abundance of small mammals captured on four 3.8 ha grids on North Key Largo trapped July 1995 through June 1996
        Page 27
    Figure 5: Comparison of abundance of Key Largo woodrats and black rats captured on four 3.8 ha grids on North Key Largo trapped July 1995 through June 1996
        Page 28
    Figure 6: Sex ration fo small mammals captured on four 3.8 ha grids on North Key Largo trapped July 1995 throug June 1996
        Page 29
    Figure 7: Age structure of Key Largo woodrats captured on four 3.8 ha grids on North Key Largo trapped July 1995 through June 1996
        Page 30
    Figure 8: Age structure of Key Largo cotton mice captured on four 3.8 ha grids on North Key Largo trapped July 1995 through June 1996
        Page 31
    Figure 9: Recruitment as measured by the proportion of immature individuals on four 3.8 ha grids on North Key Largo trapped July 1995 through June 1996
        Page 32
Full Text






A Status Survey for the Key Largo Woodrat (Neotomafloridana small) and Key Largo Cotton
Mouse (Peromyscus gossypinus allapaticola) on North Key Largo, Monroe County, Florida.



Technical Report #55
Florida Cooperative Fish and Wildlife Research Unit
U.S. Geological Survey
Biological Resources Division
Gainesville, Florida 32611-0450



Philip A. Frank, H. Franklin Percival, Britt W. Keith, Chris R. Sasso, Michael S. Gaines, and
James Duquesnel


August 1997









Citation: Frank, P.A.,, H.F. Percival2, B.W. Keith3, C.R. Sasso, M.S. Gaines3, and J.Duquesnel,.
1997. Status Survey for the Key Largo Woodrat (Neotomafloridana small) and Key Largo
Cotton Mouse (Peromyscus gossypinus allapaticola) on North Key Largo, Monroe County,
Florida. Florida Cooperative Fish and Wildlife Research Unit, U.S. Geological Survey, Biological
Resources Division, Technical Report #55, 32 pp.



































'Florida Game and Fresh Water Fish Commission Florida Keys Field Office, 2796 Overseas
Highway, Suite 213, Marathon, Florida 330502

2Florida Cooperative Fish and Wildlife Research Unit, 117 Newins-Ziegler Hall, University of
Florida, Gainesville, Florida 32611

3Department of Biology, University of Miami, Coral Gables, Florida 33146

"Florida Department of Environmental Protection, Key Largo Hammocks State Botanical Site,
P.O. Box 487, Key Largo, Florida 33037


_1~1~ ~ 111









Introduction

Key Largo woodrats (Neotomafloridana small) and Key Largo cotton mice (Peromyscus

gossypinus allapaticola) are small mammals endemic to the tropical hardwood hammocks of Key

Largo, Monroe County, Florida (Schwartz 1952, Sherman 1955). Historically, Key Largo

woodrats occurred over the length of Key Largo; Key Largo cotton mice occurred throughout

Key Largo and south to near Tavernier (Layne 1974, Brown 1878b). Both species are mainly

associated with mature tropical hardwood hammock, although a range of successional habitat

conditions are also used (Goodyear 1985, Humphrey 1992). Both species are nocturnal and may

feed on a variety of fruits and seeds of tropical hardwood trees. Key Largo woodrats nest in stick

nests which they construct on the forest floor although woudiats may ablo n2 below-ground in

crevices in the limestone (Hersh 1978, Barbour and Humphrey 1982). Key Largo cotton mice

may construct nests in hollow logs and crevices in the limestone (Barbour and Humphrey 1982).

Much of the original tropical hardwood hammock on Key Largo has been cleared for

development, and the present range of these rodents includes only the northern portion of Key

Largo (Barbour and Humphrey 1982). Because of their limited distribution on Key Largo, he

habitat specificity,h vulnerability to hurricanes, and the development pressure on the

remaining hardwood hammock, both rodents are listed as endangered by the Florida Game and

Fresh Water Fish Commission and the U.S. Fish and Wildlife Service.

A great deal of effort has gone into acquiring land for the protection of these and other

endangered species occurring on North Key Largo. The S. Fish and W ldl Service has

established Crocodile Lakes National Wildlife Refuge (CLNWR), and the State of Florid has

established North Key Largo State Botanical Preserve (KLSBP). Acquisition for these preserves


1









is ongoing, with much of the land targeted for acquisition already purchased. Current

management efforts focus on acquisition of remaining parcels and restoration of the badly

damaged ecosystem.

Hurricane Andrew struck North Key Largo on August 24, 1992, causing damage to large

areas of tropical hardwood hammock. The damage was not dibute P~ vly over North Key

Largo; d-wikmany areas were severely affected, other Peal' were spared ~'significant
A
damage. The most obvious damage resulting from the hurricane consisted of defoliation, limb

damage, and downed canopy-height trees. The effect of this damage on herodent habitat was to

create a more open canopy than existed previously. In the months immediately following the

storm, the hammocks responded with dense growth of vines and urt understory resulting from

the increased light levels from the canopy Mg. However, within three years following the
4
storm the hammock canopy has largely recovered to the point where canopy coverage has been

restored. The lasting results of the storm ontlrodent habitat has been to increase the amount of

fallen logs on the forest floor.

Because the Key Largo rodents are nocturnal and quite secretive, little is known regarding

their present status on North Key Largo. The most recent study evaluating the population status

of the Key Largo rodents was conducted in 1986 (Humphrey 1988). Given the endangered status

of the Key Largo woodrat and cotton mouse, the sustained efforts to acquire land for their

preservation, the recent hurricane, and the general paucity of current information, the present

status survey is justified.

The purpose of this study was to document the present population status and distribution

of Key Largo woodrats and cotton mice on North Key Largo, estimate population densities at


2


~ ___~









selected locations over a one-year period, and assess the effects of Hurricane Andrew on their

populations.

Methods

The Florida Keys comprise a series of islands extending southwest off the southern tip of

the Florida peninsula (Figure 1). Key Largo is the largest and northernmost island in the Keys and

is connected to peninsular Florida by a two-lane highway. North Key Largo is that portion of

Key Largo north of the point where this highway intersects Key Largo, and is sparsely developed
Cr-NA- A -ff ~ Aff f ^
4elaive to.Key Largo.

The Upper Florida Keys, including North Key Largo, are composed of limestone rock that

was once part of the Pleistocene emergent coral reef(Hoffmeister 1974). Plant communities in

the Keys are largely determined by elevation, with hardwood hammock the predominant upland

vegetation type on North Key Largo. Transitional buttonwood habitat occurs at lower elevations

and mangroves grow at elevations near sea level. Vegetation of hardwood hammocks is

predominantly West Indian, resembling hardwood hammocks found on many Caribbean islands.

Keys hammocks are speciose compared to mainland forest types, with 48 tree species being
C
recorded for Key Largo (Hilsenbgek 1976). Common tree species in the canopy include

poisonwood (Metopium toxiferum), wild tamarind (Lysiloma bahamensis), pigeon plum

(Cocoloba diversifolia), soldierwood (Colubrina eliptica), crabwood (Ateramnus lucidus),

mastic (MastichodendronfoAidissimum), willow bustic (Bumelia salicifolia), and Jamacia

dogwood (Piscidiapiscipula). Common shrubs in the understory include wild coffee (Psychotria

nervosa), stoppers (Eugenia spp.), and marlberry (Ardisia escallonioides).


3


1~-s~-1a_--11~_1_lll-.









The forests on North Key Largo have a long history of modification, and much of these

forests were cleared for agriculture prior to the turn of the 18th century. More recently, clearing

for development has occurred, reducing and fragmenting the hammocks. It is estimated that there

are approximately 850 ha of hammock remaining on North Key Largo (Humphrey 1988).

The structure of North Key Largo hammocks is variable depending on hammock age and

disturbance regime. Old, relatively undisturbed mature hammocks typically exhibit a well-

developed canopy of large diameter trees with a relatively open understory and few scattered

shrubs. Little light and wind penetrate the forest interior. The forest floor of the undisturbed

mature hammocks is covered with a deep layer of fallen leaves which comprise the soil overlying

the limestone. Numerous solution holes and crevices in the limestone are present, and fallen,

decaying logs are relatively common.

The more recently disturbed, youngeraagorests are typified by densely-spaced trees and

shrubs with eemparatively smaller diameter than the mature hammocks. An incomplete canopy

allows.apte light penetration, and shrubs and vines may grow densely in the understory. Leaf

litter is typically shallow and humus soils may be relatively poorly developed. Due to the young

age of many trees, large fallen logs are relatively rare in the less mature forests. Topographic

alteration of the limestone substrate associated with agriculture and development has reduced the

structural complexity of the limestone substrate in many areas, reducing the number of sinkhole

depressions and exposed limestone crevices.

Distribution ofwoodrats and cotton mice was studied by live trapping on a series of

transects placed over much of the length of North Key Largo (Figure 2). A total of 41 transects

were established along C.R. 905 using the numbered concrete power poles along the roadway as


4









reference points. These numbered power poles are located along the entire length of S.R. 905,

providing convenient permanent landmarks in the study area. Two transects were established at

each of 18 locations along C.R. 905 at approximately 1.0 km intervals, beginning just north of the

intersection of C.R. 905 and U.S. 1 and following C.R. 905 to the Ocean Reef Club. Five

transects were placed in hammock located off C.R. 905 at the intersection of the old (abandoned)

S.R. 905 and abandoned section of Card Sound Road.

Each transect consisted of 25 stations set at 10-meter intervals with two traps per station.
A
Transects were oriented approximately perpendicular to the road with one transect directed into

hammock on the KLSBP and one transect directed onto CLNWR. Trapping was conducted on

only privately-owned properties with the landowner's permission.

Each transect was trapped for three consecutive nights between January through May

1995. Data collected included location, rodent species, sex, age, weight, and reproductive

condition. Each individual rodent was identified with a uniquely numbered monel ear tag. All

rodents were released immediately following data collection.

The population ecology of Key Largo rodents was studied on trapping grids established at

four locations on North Key Largo (Figure 2). Grids 1 and 2 were located on CLNWR; grid 1

was located at power pole #214,and grid 2 was located at power pole #204. Grids 3 and 4 were

located on KLSBP; grid 3 was at power pole #144 and grid 4 was at power pole # 134. All grids

were placed in relatively mature closed-canopy hammock, although there was considerable

variation in microhabitat structure both between grids and also within areas on the grids. Grids

were located in hammocks known to be occupied by both woodrats and cotton mice as indicated

in the distribution transect trapping previously described.


5


l___________~_ssll__I_ ~_~~I









Each grid consisted of 144 trap stations placed at 15-meter intervals and arranged in a 12

x 12 array. This arrangement resulted in a grid area of 32,400 square meters (3.24 hectares). A

single trap was set at each trapping station and trapping was conducted for four consecutive

nights on each grid bi-monthly over the 12-month period beginning July 1995 through June 1996.
/'
Woodrat and cotton mouse density were calculated by dividing the grid area by the

minimum number of individuals alive on each grid and converting this value to number per

hectare. This simplistic enumeration estimate was selected over more complex density estimation

techniques such as the nested grid approach in CAPTURE (Humphrey 1988) because the data

requirements to meet model assumptions were not met in this study. For example, the

CAPTURE nested grid density estimation procedure requires a minimum of 75-100 individual

captures and the number of trapping occasions from 7-10 days (Otis et al., 1978); requirements

simply not realistic for this study. The data obtained in this study are simply too few to justify the

use of the CAPTURE density estimation procedures. While we recognize that the enumeration

method of density estimation may produce biased results (Hallet et al. 1991, Humphrey 1988),

this approach may be appropriate for this study given the obvious logistical constraints of meeting

the CAPTURE model assumptions.

Results

A total of 20,124 trap nights were accumulated on North Key Largo; 6,300 trap nights

were involved in the transect trapping and 13,824 were involved in the grid study. Although trap

disturbance by raccoons has been reported to be problematic on North Key Largo by others, we

did not experience this problem. Because the occurrence of sprung traps was negligible relative

to total trap nights, we report total trap nights as unadjusted for disturbance. We used specially


6


_~llg~ _I~ ~l~lls ~









designed raccoon-proof traps (H.B. Sherman, Inc., Tallahassee, Florida) and recorded no

raccoon-caused mortality to animals restrained in traps.

A total of 290 individual small mammals were captured on transects in 6,300 unadjusted

trap nights including 169 cotton mice, 42 woodrats, 61 black rats, and 17 cotton rats (Sigmodon

hispidus) (Table 1). Captures of both cotton mice and woodrats extended along the length of

North Key Largo, from Ocean Reef Club south along C.R. 905 to power pole 234. Trapping at

power poles 234, 244, 254, and 264 near the western end of North Key Largo was discontinued

after the initial night as large numbers of fire ants invaded the traps, causing the death of two

cotton mice and one black rat. Woodrats and cotton mice were found at both KLSBP and

CLNWR. Cotton mouse captures were approximately equal at both sites, but twice as many

woodrats were captured at KLSBP than at CLNWR (Table 1).

A total of 718 individual small mammals were captured on trapping grids in 13,824

unadjusted trap nightsincluding 568 cotton mice, 84 woodrats, and 66 black rats (Figure 3).
A
Summing er all trapping sessions, there was a significant difference in captures of the three

rodent species on the grids (X2 =43, P < 0.0005), with cotton mice the most common small

mammal captured on all four grids. Woodrats were the second most abundant species captured

on grids II, III, and IV, but more black rats than woodrats were captured on grid I (Figure 3).

Seasonal variation in individual species abundance on the grids was apparent, especially

for cotton mice where data was adequate. Cotton mice populations were at their lowest levels

during the summer (July-September 1995), increased to peak levels during the fall (September-

November 1995), remained high through the winter (Novemberl995-March 1996), and showed

signs of decreasing through spring and early summer (March-May 1996, Figure 4).


7


I~I









The sample size for woodrats is small relative to cotton mice, although a similar pattern of

seasonal abundance is apparent. Woodrat populations were at their lowest levels during the

summer (July-September 1995), increased to peak levels through the fall (September-

November1995), and remained at high levels through the winter and spring, 1996 (Figure 5).

Black rats were only slightly less abundant on the grids than woodrats, and exhibited a similar

pattern of seasonal abundance (Figure 5).

Woodrat density on the grids ranged from 0-1.8 woodrats per hectare, with a mean of 0.9

summing over all grids and seasons (Table 2). Cotton mouse density was greater than woodrat

density, ranging from 1.3-9.5 per hectare, with an overall mean of 6.2 (Table 2). These density

estimates are contrasted with estimates from previous studies (Table 3).

The sex ratios of adult woodrats, cotton mice, and black rats were approximately equal,

with the proportion of males 55%, 60%, and 54% respectively (Figure 6). The age structure of

the woodrat population was biased towards adults during all times of the year, although immature

(juvenile and subadult) individuals were present throughout the year (Figure 7). / r.erecruitment

of new individual woodrats onto the grids was highest during the November, January, and March

trapping sessions, indicating peak reproduction was occurring from late fall through winter

(Figure 9).

Adult cotton mice also were always more abundant on the grids than immatures, and

immature individuals also were recorded throughout the year (Figure 8). Cotton mouse

recruitment was highest during the November and January, suggesting peak breeding during late

fall and winter. Recruitment peaked again during May, illustrating the variability in the breeding

season of this species (Figure 9).


8


~ ~~ 1~_1_









Black rat adults were present on the grids during all trapping sessions. The proportion of

immature individuals was highest during January and March, and no immature individuals were

recorded during May or July, suggesting reproduction in this species was highest during winter

(Figure 9).

Discussion

Both the Key Largo woodrat and Key Largo cotton mouse continue to inhabit hardwood

hammocks over most of the length of North Key Largo, from approximately power pole #234 at

the southern end of North Key Largo to the hammocks near the Ocean Reef Club. This(en

generally confirms previous survey work done on these species (Humphrey 1978) but illustatefs a

troubling decline of both species in the southern end of the range.

Humphrey (1988) recorded both Key Largo woodrats and cotton mice in the Port

Bougainville hammocks near power pole # 240. Our survey failed to find either species south of

power pole #234; black rats were the only rodent species captured south of this area. In addition,

exotic fire ants (Solenopsis invicta) were abundant at trapping locations towards southern North

Key Largo, from power pole 224 south to power pole 265 where the hammocks are adjacent to

-tre developed isnrrd of Key Largo. Animals caught in traps where fire ants were abundant were

either dead or badly injured by the ants and trapping was discontinued in these areas following the

initial night. In all, one woodrat, three cotton mice, and two black rats were killed by ants in one

night of trapping on transects at power poles 224, 234, and 244.

The area between power poles 224 and 265 is generally disturbed. There are three

residential subdivisions partly developed with houses and one small mobile home park. Much of

the undeveloped area in the vicinity of power poles 240-265 is highly disturbed from past land


9


_1~ ~11_1~1~ ___ _









clearing, unfinished developments, and rock mining. The hammock habitat in this area is

.e--eemmtely bisected by old roads and several large abandoned buildings associated with the

unfinished Port Bougainville development are present. Free-ranging house cats, black rats, and
A
fire ants are abundant is ar, and combined with the disturbed and fragmented hammock

habitat, may have caused the decline of the native rodents in this area.

---T-past alteration of the hammock on southern North Key Largo has reduced the amount

and quality of remaining aRmoehabitat. Te three residential subdivisions in the souther cnd--

---of North Key Large have caused a permanent loss of/t hammock reTand the resulting

disturbance has badly fragmented the southern portion of hammock from the much larger areas to

the north. In addition to the direct loss and fragmentation of habitat, th4disturbances tisre -

has caused an secondary impacts, including domestic cats, fire ants, and invasive exotic

vegetation, irn1ti am~hat may affect the endangered rodent populations.

The domestic cat is the most widespread terrestrial carnivore on earth, and the fact that

cats negatively affect a vast array of wildlife species, especially birds and small mammals, is well

documented (McMurry and Sperry 1941, Pearson 1964, Iverson 1978). Cats are extremely

flexible in food habits and social organization (Jones and Coleman 1981, Liberg 1984) and hunt

even when fed daily by humans (Warner 1985, Churcher and Lawton 1989). Studies of food

habits of feral cats have shown that mice often compose a large proportion of the diet (Jones and

Coleman 1981, Churcher and Lawton 1989). The free-ranging domestic cats on North Key Largo
A
undoubtedly impact woodrat and cotton mouse populations, and in areas of relatively dense

human population such as exists between power pole 224-265 could be a principle factor in these

species absence
A


10









The role of fire ants in the ecology of the North Key Largo hammocks is unknown.

However, the fire ant has been documented to seriously impact wildlife populations in other areas

(Killion and Grant 1993, Killion et al. 1990). Because both the woodrat and cotton mouse are A

ground nesting and nocturnal, these species likely would be vulnerable to fire ant predation while

taking refuge in their nests during the daytime. In addition, both woodrats and cotton mice bear

helpless young in the nest. The newborn rodents would be extremely vulnerable to fire ant

predation. In contrast, the black rat is known to nest arboreally reducing their vulnerability to fire

ant predation.

While this study did not gather direct evidence that the above factors caused a decline in

the woodrat and cotton mouse populations in the southern North Key Largo hammock,-the- cw.

c.cmBriative evidence suggests these factors may be influencing thepatteir of distribution. 4the

5.2faeae-t Humphrey (1988) documented a comparatively robust population of both species at the

southern end of North Key Largo, suggest that there has been a decline in the last decade.
A
Twice as many individual woodrats were captured at KLSBP~,hnt at CLNWR. KLSBP
A
has larger areas of contiguous hammock than CLNWR, especially north towards Ocean Reef

Club, possibly accounting for the greater numbers of woodrats captured there. This difference

may reflect the apparent requirement woodrats have for large areas of mature hammock. In

contrast, cotton mice were equally represented on both sides ofC.R. 905, possibly reflecting the

wider habitat tolerances of this species. Black rats were more abundant on CLNWR than KLSBP

(Table 1), possibly resulting from the presence of the Monroe County solid waste transfer station

(dump) on this location. The presence of black rats on both sides of the road is indicative of the

high degree of disturbance on North Key Largo, and documents that this species is well-

established in hammocks on North Key Largo.


11


~ ~II_









Density estimates ofwoodrats and cotton mice found in this study were generally lower

than comparable studies, although there is considerable variation in historic woodrat and cotton

mouse population density estimates (Table 3). Humphrey (1988) found a mean population

density of 7.6 woodrats and 21.2 cotton mice per hectare trapping on six grids during the spring,

1986. An earlier status survey by Humphrey and Barbour (1982) estimated woodrat density at

4.5 per hectare and cotton mouse density at 11.5 per hectare. A detailed study ofwoodrats on

North Key Largo by Hersh (1978) found a mean density of 2.2 woodrats per hectare trapping in a

5.6 hectare study site judged to be excellent habitat. Brown (1978) estimated a density of 1.2

woodrats per hectare on North Key Largo using unspecified methods.

Density estimates of 0.9 woodrats and 6.2 cotton mice per hectare in the present study are

the lowest densities reported for these species at this location. The cause for low densities are

unknown, but it is unlikely that impacts from hurricane Andrew are a contributing factor. The

damage to the North Key Largo hammocks from Andrew was generally minimal, consisting

mainly of broken limbs. The majority of the severe damage on North Key Largo was to the north

at the Ocean Reef Club.

Small mammal populations are known to fluctuate widely. The observed densities in this

study may represent natural fluctuations within the range of normal densities for these species at

this location. Interpreting patterns of abundance from small trap samples over only a short time

period, such as in this study, is inadequate to characterize normal population levels. This

problem highlights the need for long-term systematic monitoring on populations of small

mammals that typically vary greatly in their population densities.


12


~--PII^-ll~lil_~__~__I__.. -X~-_









The woodrat stick nests characteristic ofNeotpma on North Key Largo were uncommon

at our study sites. On the four trapping grids)oe-rig .t..tal of 15.2 hectareswe encountered

only a single woodrat stick nest, a lod compared ton

clbcrv--cd in past ycar;. iAn ditiuonal ,bsewt the hammocks of North Key Largw

Str'cIPrt trapping, rnfirms the ohrv"ation that tha large, well-developed stick nests are generally

uncommon (personal observation). In contrast, Barbour and Humphrey (1982) counted 3,666

actively used woodrat stick nests on an estimated 475 ha of hammock on North Key Largo and

estimated an average of 7.7 active stick nests per ha of hammock.

Woodrat stick nest loss was observed on Lignuniitae Key during a decline and

subsequent extirpation of an introduced population ofwoodrats (Jim Duquesnelle, personal

communication, 1997). In 1970, 19 Key Largo woodrats and 14 Key Largo cotton mice were

introduced onto the 90-ha island (Brown and Williams 1971). This population apparently

survived, and Hersh (1978) counted a total of six active stick nests on the island in 1978. In

1979, Barbour and Humphrey (1982) counted 90 active stick nests on the island and estimated an

average of 5.3 active stick nests per ha. Using a conversion factor of one woodrat per 5.6 active

stick nests, the total population on Lignurvitae Key was estimated at 85 individuals, representing

an annual growth rate of 18% (Barbour and Humphrey 1982). However, by 1992 all stick nests

on the island were decomposing and actively maintained stick nests were not present. Trapping

surveys conducted in 1993 and again in 1997 confirmed that both woodrats and cotton mice were

extirpated from the island (Jim Duquesnelle, personal communication, 1997; Chris Sasso, personal

communication, 1997).


13









Stick nests are not reliable indicators ofwoodrat populations on Key Largo. Goodyear

(1985) found woodrats on North Key Largo nesting in root mounds, holes beneath tree roots, and

wind-throws in addition to stick nests. Goodyear (1985) suggested that the presence of stick

nests in hammocks may be an indicator of habitat conditions; old, mature hammock with abundant

natural cavities may preclude the need to construct stick nests. In contrast, younger hammock or

hammock that has been topographically altered to decrease natural cavities may force woodrats to

construct the large stick nests considered characteristic of the species.

Similarly, population densities may also influence the presence of stick nests in the

hammock. At low population densities, woodrats may be able to locate suitable homesites in

crevices without the need to construct above-ground stick nests. As population densities increase

and natural homesites are used,surplu~sndividuals may be forced to construct above-ground stick

nests. Rather than a rquisitejaspect of this species' ecology, the stick nest may represent an

alternative homesite when natural limestone crevice sites are not available for occupancy.

Management Recommendations '

Population Monitoring

Prior to the initiation of this study, Key Largo woodrats and cotton mice had not been

observed on North Key Largo since Humphrey (1988) sampled the island in 1986, an interval of

approximately ten years. These animals are secretive and nocturnal, and periodic monitoring of

their populations is needed it g i aditi small

mammals are vulnerable to extinction, and several insular subspecies have perished despite the

presence of apparently suitable habitat. o simply assume that these animals will persist given a

protected area in which to live is naive. The Allegheny woodrat (Neotoma magister) has declined


.r;414

U L~-4LeaSAA -f
^~~GI ^S^wf c<^JL,^M ^-


________~__11______1__~ ~1~









drastically throughout New York, Pennsylvania, and New Jersey,yapareniy without an obvious]

cause (Hicks 1989, Beans 1992).

We recommend that the endangered small mammal populations on North Key Largo be

monitored through live-trapping at least semi-annually to monitor their status. An arrangement of

trapping transects such as was used in this study would be adequate to document general patterns

of distribution and abundance on the island. Trapping should be of short duration to minimize

stress on the animals; we recommend three nights of trapping to document presence/absence.

Traps used should be the extra-long, raccoon-proof, ventilated type used in this study. These

traps are effective and minimize risk to animals-retrained. Animals should not be permanently

marked unless long-term research requiring permanent identification is being conducted. There is

simply no sense in permanently disfiguring individuals if simple presence/absence is the objective.

India ink can be used to identify individuals over short durations and allows count statistics to be

generated. Monitoring at KLSBP and CLNWR should be coordinated so that trapping results are

comparable.

Applied Research

Despite several studies conducted on woodrats and cotton mice on North Key Largo,

many important aspects of their ecology are still poorly understood. The past studies were of

short duration and not intended to answer detailed questions about these species ecology. Several

questions tat narneedef to prndl ad&qatp management for these species need to be

addressed.


15


~I~









Population Dynamics:

Because there have been no long-term studies ofwoodrat and cotton mouse population

dynamics, managers have no reference point to gauge the health of the populations. There is a

need for long-term data on population densities so that future monitoring efforts can put results in

a proper perspective. For example, we are unable to conclude if the population densities observed

in this study are high, low, or well within some range of acceptable limits.

Habitat Structure and Stick Nests:

The structural requirements ofwoodrat habitat-isunknown, but it appears that older, more

structurally complex habitats are better for woodrats. The construction of the stick nest may be a

response to structurally simple habitats rather than a general ecological requirement. Given the

large-scale restoration efforts underway on North Key Largo, are managers addressing the needs

of these species? Increasing the structural complexity of restored habitats through the placement

of rubble, fallen timber, and topographic alteration may be a valuable management tool for

woodrats and cotton mice. However, the data to justify these actions do not exist while the

opportunity to manage for this aspect of the habitat is missed.

Exotic Species and Secondary Effects of Development:

isignifiat anecdotal evidence suggest that ieotic species,

specifically the house cat, black rat, and fire ant, are negatively impacting native small mammal

z community on North Key Largo through competition and predation. Rgads oi

sin ificane ofe evidence in the jusltri omnfimie of immdite management action, the prnohlem

- l-shuld t, tuied. Specific research on the impacts of exotic species on the endangered North

Key Largo small mammal populations should be initiated. .


16


__ __~ _I_









In addition to hebeotrdirect loss of hammock abiton North Key Largo, there is

significant fragmentation of what habitat remains. Obvious influences on the continuity of the

habitat include the Monroe County Solid Waste Transfer Station (CLNWR), abandoned buildings

(CLNWR, KLSBP), a trailer park (CLNWR), housing developments (KLSBP), missile sites

(CLNWR, KLSBP), borrow pits and canals (CLNWR, KLSBP), and numerous roads (CLNWR,

KLSBP). The impact of these habitat discontinuities on woodrat and cotton mouse populations

should be studied so thaithe level of management activity focused on these features of the habitat

can be determine~. 0 bv' -A f

Habitat Restoration and Acquisition:

The impacts to Key Largo woodrat and cotton mouse habitat described above are

extensive atd-theresearch required to understand the specific impacts these disturbances have on

the ecology of the forest is aemplicated and will take time. In the absence of data, management

action addressing the fragmented nature of the North Key Largo forest is justified and indeed has

already been initiated at several sites on North Key Largo (Harrison Tract, Carysfort Yacht Club,

and Port Bougainville). A plan detailing the impacts to the North Key Largo forest that identifies

the specific projects needed to restore forest continuity should be developed. The feasibility of

permanently removing/relocating large impacts such as the solid waste transfer station and the

two missile sites should be examined ieta. A comprehensive list of remaining land acquisition

needs on North Key Largo should also be developed.

The development of a comprehensive habitat restoration plan as-ahiwill greatly ai4


iffermatieft-byplanners and decision makers. Several large-scale restoration efforts are underway


17


~ _1_1~









in south Florida at the present time, and we anticipate funding will become increasingly available
A
in the future. ti i th fl detailed habitat restoration plan will

-alow land managers on North Key Largo tvrecessge funding opportunities as they become

available. In addition, there is great potential to use mitigation resources generated by existing

regulatory programs such as the Monroe County Comprehensive Plan, Florida Department of

Environmental Protection, U.S. Army Corps of Engineers, and South Florida Water Management

District to implement habitat restoration, but these resources can only be accessed through the

identification of specific projects.

Acknowledgments

Many people assisted in the completion of this study. We thank M. Bentzien of the U.S. Fish and

Wildlife Service; J. Duquesnelle, R. Wilhelm, and R. Skinner of the Florida Department of

Environmental Protection; C. Sasso, L. Lund, and M Gaines of the University of Miami. This

project was conducted with funding provided by the U.S. Fish and Wildlife Service, Coop

Agreement #14-16-0009-1544, RWO #123 through the Florida Cooperative Fish and Wildlife

Research Unit (University of Florida, Game and Fish Commission, U.S. Geological Survey-

Biological Resources Division, and Wildlife Management Institute cooperating).

Literature Cited --- P /

Barbour, D. B. and S. R. Humphrey. 1982. Status and habitat of the Key Largo woodrat and
cotton mouse (Neotomafloridana small and Peromyscus gossypinus allapaticola). J.
Mamm. 63:144-148.

Beans, B.E. 1992. Without a trace: The puzzling demise of the Allegheny woodrat. Audubon
(January) 1992:32-34.


18









in south Florida at the present time, and we anticipate funding will become increasingly available
A
in the future. ti i th fl detailed habitat restoration plan will

-alow land managers on North Key Largo tvrecessge funding opportunities as they become

available. In addition, there is great potential to use mitigation resources generated by existing

regulatory programs such as the Monroe County Comprehensive Plan, Florida Department of

Environmental Protection, U.S. Army Corps of Engineers, and South Florida Water Management

District to implement habitat restoration, but these resources can only be accessed through the

identification of specific projects.

Acknowledgments

Many people assisted in the completion of this study. We thank M. Bentzien of the U.S. Fish and

Wildlife Service; J. Duquesnelle, R. Wilhelm, and R. Skinner of the Florida Department of

Environmental Protection; C. Sasso, L. Lund, and M Gaines of the University of Miami. This

project was conducted with funding provided by the U.S. Fish and Wildlife Service, Coop

Agreement #14-16-0009-1544, RWO #123 through the Florida Cooperative Fish and Wildlife

Research Unit (University of Florida, Game and Fish Commission, U.S. Geological Survey-

Biological Resources Division, and Wildlife Management Institute cooperating).

Literature Cited --- P /

Barbour, D. B. and S. R. Humphrey. 1982. Status and habitat of the Key Largo woodrat and
cotton mouse (Neotomafloridana small and Peromyscus gossypinus allapaticola). J.
Mamm. 63:144-148.

Beans, B.E. 1992. Without a trace: The puzzling demise of the Allegheny woodrat. Audubon
(January) 1992:32-34.


18









Brown, L.N. 1978. Key Largo cotton mouse. Pp. 10-11 in J.N. Layne, ed., Inventory of rare
and endangered biota of Florida. Vol. 1. Mammals. University Presses of Florida,
Gainesville. 52 pp.

Brown, L.N. and R.L. Williams. 1971. The Key Largo woodrat (Neotomafloridana small) and
Key Largo cotton mouse (Peromyscus gossypinus allapaticola) on Lignum Vitae Key,
Florida. Florida Nat. 44:95-96.

Churcher, P.B. and Lawton, J.H. 1989. Beware of well-fed felines. Natural History 7:40-46.

Goodyear, N.C. 1985. Results of a study of Key Largo woodrats and cotton mice: Phase I,
spring and summer 1985. Unpublished report, 76 p.

Hallet, J.G., M.A. O'Connell, G.D. Sanders, and J. Seidensticker. 1991. Comparison of
population estimators for medium-sized mammals. J. Wildl. Manage. 55(1):81-93.

Hersh, S. L. 1978. The ecology of the Key Largo woodrat (Neotomafloridana small) J.
Mamm. 62:201-206.

Hicks, A. 1989. Whatever happened to the Allegheny woodrat? Pages 34-38 in The
Conservationist (March-April). New York State Department of Environmental
Conservation, Albany.

Hilsenbeck, C.E. 1976. A comparison of forest sampling methods in hammock vegetation.
Unpubl. M.S. thesis, University of Miami, Coral Gables. 106 pp.

Hoffmeister, J.E. 1974. Land from the sea: the geologic story of South Florida. Univ. Miami
Press, Coral Gables, 143 pp.

Humphrey, S. R. 1988. Density estimates of the endangered Key Largo woodrat and cotton
mouse (Neotomafloridana small and Peromyscus gossypinus allapaticola) using the
nested grid approach. J. Mamm., 69:524-531.

Iverson, J.B. 1978. The impact of feral cats and dogs on populations of the West Indian rock
iguana, Cyclura carinata. Biological Conservation. 14:63-72.

Jones, T. C. and Coleman, B.J. 1981. Ecology of the feral cat in south- eastern Australia. I. Diet.
Australian Wildlife Research. 8:537-547.

Killion, M. J. and W. E. Grant. 1993. Scale effects in assessing the impact of fire ants on small
mammals. The Southwest Naturalist 38:393-396.


19


~1 11111_.._ .._









Killion, M. J., Grant, W. E. and S. B. Vinson. 1990. The influence of red imported fire ants
(Solenopsis invicta) on small mammal habitat utilization. Pages 43-44 in proceedings of
the 1990 Imported Fire Ant Conference (M. E. Mispagel, ed.), University of Geogia,
Athens. r

Layne, J.N. 1974. The land mammals of south Florida. Pages 386-413 in Environments of
South Florida: Present and Past (P.J. Gleason, ed.). Memoir 2 Miami Geological Society.
452 pp.

Liberg, O. 1984. Home range and territoriality in free ranging house cats. Acta. Zool. Fennica
171:283-285.

McMurry, F.B. and Sperry, C.C. 1941. Food of feral house cats in Oklahoma, a progress report.
Journal of Mammalogy 22:185-190.

Menkens, G.E. and S.H. Anderson. 1988. Estimation of small-mammal population size.
Ecology. 69:1952-1959.

Otis, D.L., K.P. Burnham, G.C. White, and D.R. Anderson. 1979. Statistical inference from
capture data on closed animal populations. Wildlife Monographs, 62:1-135.

Pearson, O.P. 1964. Carnivore-mouse predation: an example of its intensity and bioenergetics.
Journal of Mammalogy. 45:177-188.

Schwartz, A. 1952. Three new mammals from southern Florida. J. Mamm., 33:381-385.

Sherman, H. B. 1955. Description of a new race of woodrat from Key Largo, Florida. J.
Mamm., 36:113-120.

Warner, R. 1985. Demography and movements of free-ranging domestic cats in rural Illinois.
Journal of Wildlife Management 49(2):340-346.


20










Table 1. Location of captures of small mammals on trapping transects on trapped between
January-May, 1995 on North Key Largo, Monroe County, Florida. KLSBP refers to public lands
on the south side of C.R. 905, and CLNWR refers to public lands on the north side of C.R. 905.




Location Woodrats Cotton Mice Black Rats Cotton Rats

KLSBP 28 89 20 8

CLNWR 14 80 41 9

TOTAL 42 169 61 17


21


1~ 111111^_1__ ..











Table 2. Density estimates of Key Largo woodrats and Key Largo cotton mice based on live
trapping on four grids North Key Largo from July 1995 June 1996. Grid area was estimated at
3.8 ha and naive density was calculated by dividing the number of captures by grid area.


Trapping Session



July 1995







September 1995







November 1995







January 1995







March 1995







May 1995


Grid


Woodrat Woodrat density (per
captures ha)


0.3

1.3

0.3

0.8

0

0.5

0.5

0.5

1.0

1.3

1.0

1.3

1.0

1.0

1.8

1.6

0.5

0.8

0.8

1.8

1.0

0.8

1.6

1.0


Cotton mouse
captures


cotton mouse
density (per ha)


5

26

6

23

19

20

14

16

34

36

22

19

31

25

31

25

29

30

25

34

24

25

20

29


1.3

6.8

1.6

6.0

5.0

5.3

3.7

4.2

8.9

9.5

5.8

5.0

8.1

6.6

8.1

6.6

7.6

7.9

6.6

8.9

6.3

6.6

5.3

7.6


22


_I ~LI1I~






















Lf)


0

0
o









o
0
0


'C




-o
a



0t


-o



Ma












Cd
a
' U




&o











ei
C;





-o .4
a U











Fo
.^


00

CI

cl


Ca


NC A

S a.
C4r- 0
,_*tu( (


oN

N

0i


10

00


C,


.2







-e
0
u
Cd
4-
-o!
Cd
-ou


m


Cd


C.- 0
sC,'






cl

eCC
1 Cd
B c

0).S


Cd

U


c
00
oo
<3


00




00











00


00
rci'


I
Cd


0










Cd
f~


110
c\
oN
c^

1I1
oN
oN


0 EL
m~ t% -r


C-I l
- C"l


t:
0





Cd


CO'
-0





<0
0
r=
0<




Cd


I



0,
VI


0
0
e-I


\0



In
H o




Cd S

I Id


O .
Cl~


0
*-

4.X
Q






'i:
Cd

'
Cd
-3
*-
cd


0










0
o
Cd


a-
4-.
i3
s,


0


u
at


__IC_ __ g


_~_~_~~I_ _I_ __I~















Florida


North Key Largo


Florida Keys


Y- C


*t9s


Marathon


Key West






Figure 1. Location of the study area at the extreme southern tip of the Florida peninsula.


24


I


o, e






North Key Largo Trapping Locations

a :Ocean
Reef Club


/


I.


b j N


1 0 1 2 3 Miles


Habitat types
g Hammocks
Buttonwoods
I-I Salt Marsh
B Scrub Mangroves
Lii Mangroves
I Developed
Exotics


Figure 2. Land cover map of North Key Largo indicating major habitat types. Trapping grids are
indicated by yellow squares and trapping transects by red lines.


25


_s~~


(t


63




























o
0
L-

E
Z


180


160


140


120


100


80


60


40


20


0


Grid I


Grid II


Grid III


Grid IV


Cotton Mouse Woodrat Black Rat





Figure 3. Summary of total number of individuals captured on four 3.8 ha grids on North Key
Largo trapped July 1995 through June 1996.


26






















/
/
/
/
/


U_- _-n


COTTON MICE


WOODRATS



AIr BLACK RATS


JULY SEPT NOV
JULY SEPT NOV


I I I
JAN MAR MAY


TRAPPING SESSION (1995-96)


Figure 4. Seasonal abundance of small mammals captured on four 3.8 ha grids on North Key
Largo trapped July 1995 through June 1996.


27


120 -


100 -


-A,


Z
0
LU


O


80


60 -


40 -


20-


0-


~I~


1 I I I I I


I





















WOODRATS
BLACK RATS


/


/


/ \

A


I
JULY


I
SEPT


I
NOV


I I
JAN MAR


MAY
MAY


TRAPPING SESSION (1995-96)


Figure 5. Comparison of abundance of Key Largo woodrats and black rats captured on four 3.8
ha grids on North Key Largo trapped July 1995 through June 1996.


28


*0
A


20 -


Co




z
U-
0





w
Lt



z
U-
C0


/


/


0 -


~_~~__~ __~























100 -


50 -


0


Woodrat


Cotton Mouse Black Rat


Figure 6. Sex ratio of small mammals captured on four 3.8 ha grids on North Key Largo trapped
July 1995 through June 1996. Adult individuals only were included in the analysis.


29


C/
a)

E
a)
LL.




o
0
n---

x


X


III
















WOODRATS


ADULTS


16 -


14 -


12 -


10 -


8


6-


4-


,A /


/


JUVENILES


JULY SEPT NOV


JAN MAR MAY


TRAPPING SESSION (1995-96)






Figure 7. Age structure of Key Largo woodrats captured on four 3.8 ha grids on North Key
Largo trapped July 1995 through June 1996.


30


SUBADULTS
*


Co




-,
IU
Z
LL
0
LU
03


2


0-


_I~-P~~_-i(-~-~~_l__..~.. .II_.1~L



















COTTON MICE


ADULTS


SUBADULTS

.^ --,\


/


JUVENILES


SEPT NOV


JAN MAR MAY


1995-1996


Figure 8. Age structure of Key Largo cotton mice captured on four 3.8 ha grids on North Key
Largo trapped July 1995 through June 1996.


31


100


80 -



60 -



40 -



20 -


ILl
C
0
Z
LL
0


CD
z


/
--UJ.


0 -


JULY
















0.6


(C


E
E
0.3
O

0
0
0-








0.0
JULY SEPT NOV JAN MAR MAY

TI rapping Session -96)





Figure 9. Recruitment as measufk4b orovnf'o n d nature individuals captured on four
3.8 ha grids on North Key Largo trapped July 1995 1trou--g une 1996.


32


I I Woodrat
Cotton Mouse
Black Rat




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs