Content uploaded by Gretchen LeBuhn
Author content
All content in this area was uploaded by Gretchen LeBuhn
Content may be subject to copyright.
The influence of urban park characteristics on ant
(Hymenoptera, Formicidae) communities
Kevin M. Clarke &Brian L. Fisher &Gretchen LeBuhn
Published online: 29 July 2008
#Springer Science + Business Media, LLC 2008
Abstract As rapid urbanization continues, existing natural areas within urban zones will
play a critical role in safeguarding remaining biodiversity. Ants are an integral part of
almost every terrestrial ecosystem, including urban environments, and understanding which
environmental characteristics influence their persistence is critical. In this study, 24
protected natural areas within urban parks including mosaic, scrub, herbaceous and forest
habitats were surveyed for ants with 563 pitfall traps. The data provide insights into the
distribution and abundance of ant fauna in San Francisco natural areas, as well as which
characteristics of parks have the most influence on ant community composition. A total of
2,068 ant individuals representing 15 species were collected. A regression analysis revealed
that urban forests reduced ant richness and abundance and that there was little or no impact
of the Argentine ant on native ants. Natural area size and shape were not important in
explaining variations in overall ant species richness and abundance, with many smaller
natural areas harboring ant populations that are just as diverse and robust as larger areas.
Keywords Urbanization .Natural areas .Argentine ant .San Francisco .Forest .
Conservation
Introduction
The process of urbanization has led to a pervasive transformation of natural ecosystems.
This process not only consumes natural habitats, but also fragments, degrades, and isolates
remaining natural areas, which are then entrenched in a matrix of anthropocentric land uses.
This transformation has profound and often detrimental effects on the remaining native
Urban Ecosyst (2008) 11:317–334
DOI 10.1007/s11252-008-0065-8
K. M. Clarke :G. LeBuhn (*)
Department of Biology, San Francisco State University, 1600 Holloway Avenue,
San Francisco, CA 94132, USA
e-mail: lebuhn@sfsu.edu
B. L. Fisher
Department of Entomology, California Academy of Sciences, San Francisco, CA 94118, USA
biota (Suarez et al. 1998; Gibb and Hochuli 2002) and the process of urbanization shows no
signs of slowing globally. By 2008, for the first time ever, more than half of the world’s
human population will live in urban areas. This trend is projected to swell to almost two-
thirds of the world’s population by the year 2025 (United Nations 2004). With urbanization
increasing in both size and scope, in addition to growing concerns about the magnitude of
global biodiversity loss (Gaston 1996), natural areas within the urban envelope will play a
growing and critical role in species preservation (Adams 2005; McFrederick and LeBuhn
2006). Previous studies have recognized that urban landscapes may present considerable
opportunities to conserve existing biodiversity that might otherwise be lost (Frankie and
Ehler 1978; Dickman 1987; Suarez et al. 1998; Bolger et al. 2000; Connor et al. 2003;
McFrederick and LeBuhn 2006). Within cities, urban parks provide a significant
opportunity to retain some of this biodiversity. Understanding the characteristics of areas
that enhance biodiversity is critical in order to preserve and manage them effectively
(Niemela 1999; McIntyre et al. 2001).
In order to survive in urban parks, species must adapt to a wide array of biotic and
abiotic factors related to urbanization. The most evident impact of urbanization is habitat
loss due to fragmentation. A decrease in habitat area can reduce population sizes (Bolger et
al. 2000) and cause local extinctions (Soulé 1987; Drayton and Primack 1996; Mangel and
Tier 1994). Furthermore, reduced gene flow attributed to small, isolated populations can
lead to inbreeding and reduced fitness in urban populations (Hitchings and Beebee 1998).
Fragmentation also increases edge effects, which change the microclimate, physical
properties and livable area available to some species (Saunders et al. 1991). Edge effects
also facilitate the invasion of exotic species (Suarez et al. 1998) and can increase predation
by domestic animals (Crooks and Soulé 1999). Urban species must also adapt to higher
urban temperatures compared to rural areas (Baur and Baur 1993; Angilletta et al. 2007)
and urban parks often have increased levels of pollution and soil compaction (Pyle et al.
1981; Jim 1998). Yet even these degraded and challenged ecosystems can yield sizable
numbers of important species, especially in areas of high native diversity and endemism.
San Francisco, California is the second most densely populated city (with populations
over 200,000) in the USA (US Census Bureau 2000) and is located within a biodiversity
hotspot harboring a number of rare and endemic species (Connor et al. 2003; EIP
Associates 2005). Urbanization has already taken a heavy toll on the biodiversity of San
Francisco. Three butterfly extinctions, including the first recorded in the USA, have been
recorded within city limits and are attributed to habitat loss due to development (Tilden
1956; Connor et al. 2003). Hafernik and Reinhard (1995) estimated that 43% of the city’s
butterfly species have been lost due to habitat loss. Furthermore, several bumblebees were
found to be locally extinct in urban parks (McFrederick and LeBuhn, 2006) and the
endangered forktail damselfly (Ishnura gemina) has gone extinct in a large protected park
due to overgrowth of its habitat (Connor et al. 2003). These species disappearances, many
of which occurred in and around protected areas, demonstrates the importance of
identifying habitat characteristics that encourage the persistence of biodiversity. Unfortu-
nately, information for land managers in urban areas is lacking for most taxa (Connor et al.
2003). This is particularly true for invertebrates such as ants (Suarez et al. 1998; Yamaguchi
2005; Thompson and McLachlan 2007).
Ants play an integral role in almost every ecosystem they inhabit (Hölldobler and
Wilson 1990), including urban environments (Thompson and McLachlan 2007). Ants move
and enrich soil, distribute plant seeds, are an important food for many animals, and can act
as chief predators of small invertebrates (Beattie 1985; Montanucci 1989; Hölldobler and
Wilson 1990; Folgarait 1998; Thompson and McLachlan 2007). Ants may also be excellent
318 Urban Ecosyst (2008) 11:317–334
indicators of land management practices and restoration efforts because they are sensitive to
habitat variation, respond quickly to changes in habitat quality and are easy to collect and
identify (Andersen et al. 2002; Underwood and Fisher 2006). California has a rich diversity
of ants (~281 species) and, due to its climatic and geographic features, has a high level of
endemism (13.9%) compared to adjacent regions (Ward 2005).
The effects of urbanization on ants and the characteristics of urban natural areas that
enhance their biodiversity are still unclear. However, a few studies are starting to expose
urbanization’s impact on ants and reveal characteristics that are key to ant persistence
(Suarez et al. 1998; Lopez-Moreno et al. 2003; Lessard and Buddle 2005; Yamaguchi 2005;
Pacheco and Vasconcelos 2007; Thompson and McLachlan 2007). Among these is the size
of remaining natural fragments. In general, larger natural fragments support a greater
number of species while smaller fragments reduce population sizes and species richness
(Suarez et al. 1998; Bolger et al. 2000; Yamaguchi 2005). However, a few studies have
shown no difference in species richness between large and small fragments and, in some
cases, smaller fragments may exhibit greater species richness (Gibb and Hochuli 2002;
Dauber et al. 2006). Another feature of fragmentation is that it increases edge effects, which
can promote the establishment of invasive species and reduce native ant populations
(Suarez et al. 1998).
One of the most successful invasive species in urban environments, the Argentine ant
(Linepithema humile; Knight and Rust 1990), has expanded its range to San Francisco.
Since the Argentine ant was first recorded in California in 1907 (Newell and Barber 1913),
urbanization has hastened its proliferation (Suarez et al. 2001). This invasive pest causes a
number of negative impacts on native species. Many studies have found that the Argentine
ant reduces native ant diversity (Ward 1987; Human and Gordon 1996; Human and Gordon
1997; Suarez et al. 1998) and causes a number of deleterious secondary effects. For
example, Bond and Slingsby (1984) found a disruption of ant-plant mutualisms when
Argentine ants displace native ants that are seed dispersers. Sanders et al. (2001) found that
Argentine ants displace native ants that tend lycaenid butterfly larvae, which could
potentially impact the endangered mission blue butterfly (Icaricia icarioides missionensis)
that inhabits many San Francisco parks.
This study aims to provide baseline data on ant species diversity and community
composition and identify natural area characteristics important for San Francisco’s ant
communities. Specifically, it seeks to determine (1) the distribution and abundance of ant
fauna in San Francisco natural areas; (2) which characteristics of natural areas influence the
observed patterns of distribution and abundance; (3) whether species richness and
community composition differences exist in the city’s four dominant natural habitats; and
(4) whether the Argentine Ant has an impact on native ants.
Methods
Study sites
San Francisco has a coastal Mediterranean climate with dry moderate summers and wet
mild winters (Ahrens 1994). The dry season typically extends from May to October and the
wet season from November to April. On average, San Francisco receives 95% of its annual
rainfall from late October through March (Null 1992). The combined effects of regional
climate trends, San Francisco’s hilly topography and maritime surroundings produce a
number of extremely varied microclimates within city limits.
Urban Ecosyst (2008) 11:317–334 319
We sampled 24 urban parks in San Francisco, California (37°46′0″N 122°26′0″W) that
contain sites designated as Significant Natural Areas (hereafter called “natural areas”;
Fig. 1). Natural areas are defined as having remnant fragments of the historic Franciscan
Landscape (Forman 1995) that have persisted largely unchanged by human activity (EIP
Associates 2005). These areas contain a myriad of plant and animal communities that differ
dramatically from one another and support many sensitive plant and animal species
(Connor et al. 2003; EIP Associates 2005). Habitats include coastal scrub, dunes, annual
and perennial grasses, chaparral, riparian wetlands, lakes, rock outcrops, and native patches
of coastal live oak and laurel trees. These natural areas, however, should not be considered
pristine, as many are dominated by non-native plant species. For example, non-native
eucalyptus covers large portions of many of these areas. San Francisco residents and
visitors access these natural areas for passive recreational purposes such as hiking, nature
watching, and dog walking. There are 30 urban parks with natural areas in the city, which
range in size from 0.12 ha to over 66 ha. We selected 24 parks for sampling. Six parks were
eliminated from the study due to the potential for significant pitfall disturbance or concern
for pet safety. Also, The Presidio, a large area in the city with areas of natural area, was not
sampled because it is not managed by the San Francisco Park and Recreation Department.
Selected parks ranged in size from 0.12 acres to 24.28 ha (Appendix 1).
Van Ness
19th Avenue
01
The Presidio
Golden Gate Park
Glen Canyon
Mt Davidson
Buena Vista
Pine Lake
Bernal Hill
Twin Peaks
Bayview
Corona Heights
Interior Green Belt
Hawk Hill
Tank Hill
Brooks
Grandview
Golden Gate Heights
Rock Outcropping
Edgehill Mountain
Palou Phelps
Duncan/Castro
15th Avenue Steps
San Francisco Bay
Pacifi c Ocean
O’Shaughnessy Hollow
Billy Goat Hill
Fairmount
Lakeview
Kite Hill
Kilometers
Bay Bridge
2
Fig. 1 Map of sampled parks in San Francisco, CA
320 Urban Ecosyst (2008) 11:317–334
Study organisms
According to a survey of museum records, 21 species of ants have been collected in the city
of San Francisco: Aphaenogaster occidentalis,Camponotus laevigatus,Camponotus
vicinus,Crematogaster coarctata,Formica francoeuri,Formica fusca,Formica lasioides,
Formica moki,Formica subpolita,Formica transmontanis,Hypoponera opacior,Lasius
pallitarsis,Linepithema humile,Monomorium ergatogyna,Pogonomyrmex subdentatus,
Prenolepis imparis,Stenamma diecki,Tapinoma sessile,Technomyrmex difficilis,Temno-
thorax andrei, and Tetramorium caespitum (Fisher et al. 2007). All but Linepithema humile,
Technomyrmex difficilis, and Tetramorium caespitum are considered native (Ward 2005).
Sampling method—ant abundance and species richness
To estimate ant abundance and species richness, we sampled parks using pitfall traps and
general collecting methods. Pitfall traps are considered an effective way to obtain a
representative sample of ant communities (Andersen 1995). Each pitfall trap consisted of a
28 mm-wide (internal diameter of mouth) 50 ml Fisher-brand test tube filled halfway with
a 50:50 mixture of water and non-toxic antifreeze. Because of the large number of dogs and
other mammals that use city parks, Sierra-brand antifreeze (Safe Brands, Omaha, Nebraska,
USA) was used because it is non-toxic and doubles as an excellent insect preservative.
Traps were buried with minimal disturbance with the lip of the pitfall trap flush with the
ground surface. In each park, five pitfall traps were placed in each hectare of natural area,
with two pitfall traps considered the minimum for fragments smaller than a hectare. Each
point was sampled once from August to September 2006, with each of the 545 pitfall traps
set in the 24 natural areas left open for a 48-h sampling period.
To assign pitfall points in each park, we used a Geographical Information System (GIS)
random sampling program (Beyer 2004), which placed between 2 to 117 pitfall traps in
each park (Appendix 1). The order in which parks were sampled was randomized and
points in parks were located with a GPS unit and printed GIS maps.
In addition to pitfall sampling, extensive visual surveys were conducted in each park to
ensure that all ground-foraging ants were detected within the park. Surveys consisted of
walking between each pitfall point during pitfall trap collection, overturning objects, and
aspirating samples of all ants observed. Specimens collected in visual surveys were only
included in species richness results and not abundance data. Solenopsis and Stennama
species were excluded from analysis because they are not effectively sampled with pitfall
traps (Suarez et al. 1998). These species, however, were included in species distribution
results (Tables 1and 2).
Following trapping, pitfall traps were capped and removed to the laboratory for
processing. Ant specimens collected were washed and stored in 95% ethanol until
identification. Worker ants in each trap were counted and identified to species. Brian Fisher
of the California Academy of Sciences and Phil Ward of UC Davis verified most species
identifications. All specimens for this study are housed in the Department of Entomology at
the California Academy of Sciences, San Francisco, CA except for a voucher set at San
Francisco State University.
Sampling method—park characteristics
We measured park characteristics that are known to influence the distribution and
abundance of ants (Appendix 1). Variables included park size, park shape (edge effect),
Urban Ecosyst (2008) 11:317–334 321
Table 1 All species found in San Francisco parks (parks 1–12)
Species Endemis
Score
Brooks
Park
Twin
Peaks
Bayview Bernal
Hill
Lakeview
Mini Park
Golden
Gate Park
Palou
Phelps
Golden Gate
Heights
Hawk
Hill
Rock
Outcrop
Grandview
Park
Billy
Goat Hill
Linepithema humile 0.04 1 1 1 1 1 1 1 1 1 1 1 1
Temnothorax andrei 0.29 1 1 1 1 1 1 1 1 1
Monomorium ergatogyna 0.42 1 1 1 1 1 1 1 1
Tapinoma sessile 0.50 1 1 1 1 1 1 1 1
Aphaenogaster occidentalis 0.71 1 1 1 1 1
Formica lasioides 0.79 1 1 1 1 1
Prenolepis impairs 0.83 1 1 1
Formica fusca 0.92 1 1
Pheidole californica 0.92 1
Camponotus vicinus 0.96 1
Formica subpolita 0.96 1
Temnothorax nevadensis 0.96 1
Tetramorium caespitum 0.96 1
Solenopsis molesta
a
11 1
Stennema diecki
a
1
Park species total 7 9 5 6 6 1 3 2 5 4 4 5
Park endemism score 0.55 0.50 0.50 0.49 0.48 0.48 0.43 0.42 0.41 0.41 0.41 0.39
Top table lists parks 1–12. Table includes endemism score (high scores indicate more unique species) for each species as well as presence/absence per park. Species are sorted
from most common to least common and parks are sorted by endemism scores from high to low
a
Detected but not included in analysis (see methods)
322 Urban Ecosyst (2008) 11:317–334
Table 2 All species found in San Francisco parks (parks 12–24)
Species Mt. Davidson O'shaughnessy Kite
Hill
Glen
Canyon
Tank
Hill
Pine
Lake
Corona
Heights
Fairmount
Park
Edgehill
Mountain
Buena
Vista Park
15th Avenue
Steps
Interior
Green Belt
Ubiquity
Linepithema humile 11111111111 23
Temnothorax andrei 111111 11 17
Monomorium ergatogyna 111111 14
Tapinoma sessile 1111 12
Aphaenogaster occidentalis 11 7
Formica lasioides 5
Prenolepis imparis 1 4
Formica fusca 2
Pheidole californica 1 2
Camponotus vicinus 1
Formica subpolita 1
Temnothorax nevadensis 1
Tetramorium caespitum 1
Solenopsis molesta
a
11
Stennema diecki
a
Park species total 5 3 4 5 4 4 2 2 2 1 1 0
Park endemism score 0.39 0.39 0.36 0.33 0.31 0.25 0.23 0.17 0.17 0.04 0.04 0.00
The bottom table lists parks 13–24. Table includes endemism score (high scores indicate more unique species) for each species as well as presence/absence per park. Species are
sorted from most common to least common and parks are sorted by endemism scores from high to low
a
Detected but not included in analysis (see methods)
Urban Ecosyst (2008) 11:317–334 323
soil moisture, the abundance of the invasive Argentine ant, percentage of four dominant
vegetation types and habitat heterogeneity for each park. Park size was derived from
existing data (EIP Associates 2005). Park shape was quantified using a complex perimeter/
area ratio (Patton 1975) called an irregularity index (Suarez et al. 1998). It was calculated
for each fragment using GIS maps and dividing the amount of edge around the park by the
circumference of a circle (the geometric shape with the smallest amount of edge per area) of
the same area as the fragment. As fragment shape becomes more irregular, the index
increases without limit above a value of 1 and represents the amount of edge specific to a
selected fragment relative to the fragment size. Soil moisture for each park was determined
by taking a soil core for each point at the completion of the 48-h sampling period. A bulb
planter was used to take a soil core within a one meter radius of each pitfall, which was then
placed in a paper lunch bag and brought to the laboratory to have the wet weight measured.
The samples were then dried at 90°C for 48 h and re-weighed. Soil moisture was calculated
by taking the difference between dry and wet weight and dividing by the wet weight. Soil
samples for each park were pooled together and divided by the total number of samples to
determine an average for the park. The percentage of the four dominant vegetation types
(forest, grassland, scrub and mosaic—areas of mixed vegetation, e.g. ‘grassland-scrub’)
were estimated using available data from the San Francisco Recreation and Parks
Department (EIP Associates 2005). Habitat heterogeneity, which was based on the number
of vegetation types in each natural area, was derived data provided by the same agency (EIP
Associates 2005).
Data analysis
Species richness for each park was determined by total number of species pooled from pitfall
sampling and visual surveys. The total number of workers divided by number of pitfall traps
quantified species abundance for each park. Because pitfall sampling may bias ant abundance
due to the chance proximity of a pitfall trap to a nest, the effect of a substrate on forager
activity, and variable species response to pitfall traps (Bestelmeyer and Wiens 2001), all ant
abundances were log-transformed [log(n+1)] at the trap level to normalize data.
To model the park characteristics which best explain differences in ant species richness,
native species richness, ant abundance, Argentine ant abundance and endemism between
parks, we analyzed the data using ordinary least squares regression. Response variables
were transformed when necessary to meet the assumptions of regression (Zar 1999). Prior
to regression analysis, we tested for collinearity and constructed a correlation matrix to
examine possible relationships among park characteristics. Two sets of models were built
using forward regression (pto enter the model <0.05) with SPSS (2001, v. 11.01), one with
number of pitfall traps for each park as a covariate and one without a covariate. Since the
results were consistent, we are only presenting results with a covariate. To quantify
endemism for each park, we created an endemism index for each ant species based on the
number of parks in which each species was found. We divided the number of parks a
species occurred in by the total number of parks and subtracted this number from one
(Tables 1and 2). The index ranged from 0.93 (unique) to 0 (ubiquitous; note that the
highest index does not equal 1, because there must be at least one park per species and a
score of one would indicate zero parks per species). We calculated an average endemism for
each park (summed all scores averaged over number of ants) and used this metric in
building the models.
Five different regression models were built. The dependent variables were species
richness, native species richness, ant abundance, Argentine ant abundance, and park
324 Urban Ecosyst (2008) 11:317–334
endemism. Potential explanatory variables were park size, park shape, soil moisture,
Argentine ant abundance (not used in model with abundance of Argentine Ant as
dependent), percentage of the four dominant vegetation types and habitat heterogeneity. We
examined the pattern of the residuals for each regression model and found no evidence to
suggest that linear regression was not the appropriate model for these data.
To compare the four dominant vegetation types, pitfall traps were pooled together in
each vegetation type. Dominant vegetation types were forest (dominated by Eucalyptus
globulus), herbaceous (dominated by annual grassland, mainly Avena sp.), scrub
(dominated by Baccharis pilularis and Genista monspessulana) and mosaic (dominated
by Avena sp./Baccharis pilularis mosaic). To assess community composition similarity
among vegetation types, a pairwise species composition similarity was calculated using the
Marczewski–Steinhaus (M–S) distance index (Colwell and Coddington, 1994):
MS ¼w
aþbw
where a=number of ant species of one vegetation type, b=number of ant species of the
second vegetation type, and w=number of ant species common to both vegetation types.
Results
Community structure
A total of 2,068 ants were captured between August and September 2006 belonging to 11
genera and 15 species. Four of these species had not been previously recorded in San
Francisco County (Pheidole californica,Solenopsis molesta,Stenamma sp. 1 cf.
punctatoventre and Temnothorax nevadensis), though are known from surrounding
counties. The species detected in this study comprise 60% of the 25 species (including
four new species detected in this study) that have historically been recorded in San
Francisco. Species richness in each park ranged from zero in a park dominated by
eucalyptus (Interior Green Belt, Tables 1and 2) to nine species found in a park dominated
by patches of grassland and scrub (Twin Peaks, Tables 1and 2). Average ant abundance
between parks ranged from 0 (Interior Green Belt) to 33.5 (Brooks Park) individuals per
trap. The most abundant species, accounting for 29% of individuals collected, was
Monomorium ergatogyna, a small (about 2 mm long), common, ground-nesting species.
This species was found in 14 parks. The second most abundant species, accounting for 26%
of individuals collected, was the Argentine ant. This species was the most commonly
encountered species, appearing in every park but one. There were also two uncommon
species collected: one Stenamma sp. 1 cf. punctatoventre in Golden Gate Park and two
Temnothorax nevadensis in Bernal Hill park. Endemism, which is a measure of park
uniqueness (a higher value represents a more unique fauna), ranged from 0 in the Interior
Green Belt to 0.55 in Brooks Park (Tables 1and 2), a smaller park dominated by grassland.
Influence of park characteristics
The model examining species richness revealed that soil moisture and percent of natural
area that is forest explained 58% of the variation in species richness between parks (F
3, 20
=
11.525, p<0.0005, adjusted R
2
=0.579, Table 3). Soil moisture was positively related to
species richness and percent of natural area that is forest was negatively related.
Urban Ecosyst (2008) 11:317–334 325
The model testing ant abundance as the dependent revealed the percent of park that is
forest was significantly negatively correlated and accounted for 47% of the variation in ant
abundance among parks (F
2, 19
=10.263, p=0.001, adjusted R
2
=0.469, Table 3).
The model investigating native species richness as the dependent revealed soil moisture
and percent of park that is forest explained 62% of the variation in native species richness
between parks (F
3, 20
=13.418, p<0.0005, adjusted R
2
=0.618, Table 3). Soil moisture was
positively correlated with native species richness and percent of natural area that is forest
was negatively correlated.
The model testing mean park endemism as the dependent revealed that the percent of
forested area was significantly negatively correlated and accounted for 52% of the variation
in mean endemism among parks (F
2, 21
=13.469, p<0.0005, R
2
=0.520, Table 3).
Impact of the Argentine ant on native ant species
No model explained the variation in Argentine ant abundance. Argentine ants were detected
in 23 of the 24 parks, indicating it is ubiquitous in San Francisco parks. However, in 12 of
the parks, Argentine ants were not detected in pitfall traps. Instead, their presence was
confirmed in visual surveys. In these parks, Argentine ants were observed only around park
perimeters (personal observation). This suggests Argentine ants are limited to park
perimeters and are not abundant in park interiors. Of the pitfall traps that contained
Argentine ants, over 75% contained fewer than five Argentine ants, suggesting these parks
also had relatively low abundances in their interiors. Overall, Argentine ants were found in
only 18% (n=100 traps) of total traps with a mean abundance of 5.5 individuals per trap. Of
the traps with Argentine ants, 65% did not contain any other ant species. The traps that did
contain other species included only Monomorium ergatongyna,Prenolepis imparis,
Temnothorax andrei and Tapinoma sessile.
Ant community in four dominant vegetation types
Ant species richness was noticeably lower in forest habitats (Table 4). The five species
detected within forests were also found in the three other habitats. Scrub and mosaic have
the most similar ant communities, sharing eight species. The Marczewski–Steinhaus (M–S)
Complementarity Index, which is a measure of beta diversity (Magurran 2004), shows that
species turnover is greatest between forest and herbaceous vegetation habitats (Table 5).
The difference in ant communities between forest and the three other habitats examined
may be explained, in part, by soil moisture. The forest vegetation had an average soil
moisture of 10.3%, compared to just 2.8% in herbaceous, 3.4% in mosaic, and 4.3% in
scrub (Fig. 2). Within forest vegetation types, traps (n=19) that contained ants had a mean
Table 3 Regression model variables and coefficients with adjusted R
2
values in bold and positive or
negative correlation indicated
Dependent variable Independent variables Correlations
(respectively)
Adjusted R
2
for
whole model
Species richness Percent forest and soil moisture Negative and positive 0.58
Native species Percent forest and soil moisture Negative and positive 0.62
Ant abundance Percent forest Negative 0.47
Argentine ant abundance No variable fit model
Endemism index Percent forest Negative 0.52
326 Urban Ecosyst (2008) 11:317–334
soil moisture of 4.5%. This was very different than traps in forests without ants, which had
a mean soil moisture of 11.5% (n=171) suggesting soil moisture plays a role the
distribution of ants in the urban forest. Herbaceous habitat had the richest ant community
and the most unique species.
Discussion
Influence of park characteristics
This study indicates that both the amount of forest vegetation and soil moisture are
important characteristics in explaining the distribution of ants in San Francisco parks. The
percent of park that is forest was a significant variable in four of the five models tested,
suggesting this vegetation type has a strong negative impact on the ant community. Urban
forests embedded in parks supported very few ants and may limit the habitable areas to
remaining non-forested habitat. The most dramatic example of this was on Mount
Davidson, which contains 75% closed canopy eucalyptus forest and 25% scrub and mosaic
vegetation. The larger forested area did not support a single ant while the smaller scrub and
mosaic areas supported five species and 4.04 ants per trap. Thompson and McLachlan
(2007) reported a similar effect in Canada and found a simplified ant community in urban
compared to suburban and rural forests. In California, previous investigations of urban ants
Table 4 List of ant species found in four park dominant vegetation types
Forest Mosaic Scrub Herbaceous
Linepithema humile ** **
Monomorium ergatogyna ** **
Temnothorax andrei ** **
Tapinoma sessile ** **
Prenolepis impairs ** **
Aphaenogaster occidentalis ***
Temnothorax nevadensis *
Camponotus vicinus ***
Pheidole californica **
Formica lasioides ***
Formica fusca *
Formica subpolita **
Tetramorium caespitum *
Total species 5 9 10 12
Table 5 The Marczewski–Steinhaus (M–S) complementarity measure between four vegetation types in San
Francisco parks
Forest Herbaceous Mosaic Scrub
Forest –0.58 (5) 0.44 (5) 0.50 (5)
Herbaceous –– 0.36 (7) 0.33 (8)
Mosaic –– – 0.11 (8)
Higher values represent greater distinctness with the number of species shared between vegetation types
presented in parentheses
Urban Ecosyst (2008) 11:317–334 327
have focused on scrub, grassland and riparian habitats (Human and Gordon 1996; Suarez et
al. 1998; Bolger et al. 2000; Holway and Suarez 2006). This is the first study to document
the negative effect of urban forests in California.
Urban forests are structurally different than natural forests. Besides being smaller,
fragmented and more isolated than non-urban forests, urban forests also show increased
canopy cover (Dreistadt et al. 1990), greater disturbances due to human traffic and
pollution, and differences in leaf litter accumulation (Thompson and McLachlan 2007).
These changes can lead to lower plant and animal species richness (McIntyre et al. 2001;
Moffatt et al. 2004) and a greater proportion of exotic plant and animal species (Dreistadt et
al. 1990; Poland and McCullough 2006) compared to non-urban forests. In San Francisco,
trees were not a conspicuous part of the historic landscape, and a majority of the trees in
San Francisco today are not native to the area. The ants currently found in San Francisco
(and based on historic collections) suggest that the majority of ants were adapted to open
coastal prairie, grassland, or sandy (dune) habitats (Brian Fisher, pers. comm.). The
alteration of biotic and abiotic characteristics of the city’s natural areas by forests may help
explain these impacts on ant communities.
Microclimatic conditions are also important in determining ant distributions (Perfecto
and Vandermeer 1996). Studies have shown that increases in soil moisture, canopy cover,
and habitat complexity are all associated with decreases in ant diversity and composition
(Wang et al. 2001; Gotelli and Ellison 2002; Lassau and Hochuli 2004). A combination of
high soil moisture, dense canopy cover, and dense understory (habitat complexity) may
help explain the lack of ground-foraging ants in urban forests.
Urban forests in this study had increased average soil moisture compared to the three
other major vegetation types (see Fig. 2). Even though soil moisture is positively correlated
to overall species richness in parks (Table 3), very high levels of soil moisture may be
unsuitable for ants in San Francisco because ant broods cannot grow in cold and wet
environments. Prior studies have found increased soil moisture to be associated with lower
nest densities in urban forests (Thompson and McLachlan 2007) and decreased ant
abundance and diversity in North American forests (Wang et al. 2001). All ant species in
San Francisco (besides the Argentine ant, which typically nests in leaf litter), are soil
nesters, suggesting that increases in soil moisture have a direct impact on nest architecture,
brood rearing and/or the incidence of harmful fungal infections at nest sites (Clark and
Prusso 1986; Hölldobler and Wilson 1990). Within forest types examined, eucalyptus
forests contained significantly more soil moisture than other forest types and also had lower
ant richness and abundance. The increased moisture in eucalyptus is due to the fact that
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
Mosaic Herbaceous Scrub Forest
Habitat Type
0.0
2.0
4.0
6.0
8.0
10.0
12.0
% Soil Moisture
ant abundance % soil moisture
Fig. 2 Relationship between ant
abundance and soil moisture in
four habitat types
328 Urban Ecosyst (2008) 11:317–334
summer fog tends to condense on eucalyptus leaves and branches and drip down to the soil
below. Such fog drip can add as much as 42 cm of water to eucalyptus forest soil during a
single summer (Azevedo and Morgan 1974). Eucalyptus leaves also produce high
concentrations of aromatic oils with allelopathic effects. As ants can be sensitive to soil
properties (Johnson 1992; Fisher 1997), these effects may further inhibit the nesting habits
of local ants.
Argentine ants were also absent or found in low abundances in urban forests. This
finding is surprising, especially because the species is a ubiquitous urban pest and several
studies in similar Mediterranean-type climates indicate this species prefers mesic over xeric
sites (Suarez et al. 1998; Holway and Suarez 2006). In this study, Argentine ant abundance
was not correlated with increased soil moisture (r= 0.075) and no Argentine ants were
found in traps (n=63) with soil moisture of 10% or greater. Wet environments with high
levels of rain may not permit these ants to forage adequately (Vega and Rust 2001).
Climatic differences may help explain discrepancies between this study and others showing
that Argentine ants prefer moist conditions (Holway et al. 2002; Holway and Suarez 2006).
For example, San Diego (Holway et al. 2002; Holway and Suarez 2006) has an average
rainfall of ~250 mm, approximately half the average rainfall of San Francisco. Additionally,
in San Francisco the dry season experiences a summer fog that may help maintain
Argentine ant presence in drier areas.
The dense understory common in many of San Francisco’s urban forests may also
contribute to the scarcity of ants in this habitat. Urban forests in San Francisco are often
carpeted with a dense, homogeneous, invasive understory of either Cape ivy (Delairea
odorata), English ivy (Hedra helix) or Himalayan blackberry (Rubus discolor) (EIP
Associates 2005). In San Francisco forests with a dense understory, no ants were detected,
compared with a few ants found in urban forests with little or no understory (personal
observation). Nor were ants detected outside of urban forests with similarly dense
vegetation (i.e: iceplant—Carpobrotus edulis). A previous study found ant species richness
to be negatively correlated with ground herb cover in a eucalypt-dominated habitat in
Australia (Lassau and Hochuli 2004). Way et al. (1997) reports a similar finding with the
Argentine ant, which dislikes dense vegetation. However, this study’s result may be biased
by the sampling method because pitfall traps are known to catch fewer insects in dense
vegetation (Melbourne 1999). Nevertheless, ants were neither found in dense vegetation nor
the forest understory during visual surveys supporting pitfall data.
Although temperature was not explicitly examined in this study, it may also play a role
in determining the observed distributions. Ants are primarily thermophilic and Andersen
(1995) has suggested that low temperature is the principal stress affecting ant community
structure, to the point of limiting taxa diversity. Urban forests are distinctly cooler than
other urban areas (Yilmaz et al. 2007), mainly due to the shade provided by the forest
canopy. In San Francisco, shade from the tree canopy compounded by coastal fog during
the summer can reduce forest temperatures by as much as 15°(F) in some areas of the city
during the day (Gilliam 1962). The lack of solar radiation due to the thick forest canopy and
extreme temperature changes may create an environment too stressful to support ants in
urban forests. Add to this the collective effects of increased soil moisture, and a dense
understory, and the result may be a microenvironment unsuitable for urban ants.
Temperature may also play a role in explaining the distribution of ant communities in the
other habitats examined in this study. In San Francisco’s cool Mediterranean climate,
vegetation types that allow more solar radiation to penetrate may be more ideal for
thermophilic ants. Across all parks, the herbaceous vegetation, which is dominated by
perennial and annual grasslands and characterized by low complexity, supported the most
Urban Ecosyst (2008) 11:317–334 329
species (see Table 5). However, there were no clear differences in species composition
between the mosaic, scrub, and herbaceous habitats (see Table 5).
Area is often a predictor of the number of species supported on an island or fragment
(MacArthur and Wilson 1967), and is therefore a factor often considered in reserve design.
However, park size did not correspond to any of the models in this study, suggesting size
does not explain differences in richness and abundances of ants in San Francisco. This
analysis holds true whether or not the number of pitfall traps was included as a covariate.
We found that even the smallest park in San Francisco, Lakeview Mini park, which is just
0.16 ha, supports a rich ant fauna (6 species). Suarez et al.( 1998) and Yamaguchi (2005),
however, found different results in their urban ant studies, and report that larger parks
support more species. Lastly, the number of habitats within a park fragment did not
correspond to any regression models either. This may be because a majority of ants in San
Francisco are categorized as generalist species and are not limited to a specific habitat.
Impact of Argentine ant on native species
Numerous studies have shown that Argentine ants displace native ants in California (Ward
1987;Majer1994; Human and Gordon 1996,1997; Suarez et al. 1998; Holway 1999;
Laakkonen et al. 2001). Surprisingly, this study found no correlation between Argentine ant
abundance and native species richness in parks, suggesting the impact of Argentine ants may
be subdued here. This result may be explained by the low abundance of Argentine ants in
many sampled parks. Pitfall traps failed to detect Argentine ants in fully half of parks
sampled. In these parks, the Argentine ant’spresencewasconfirmedbyvisualsurveys,but
individuals were only spotted around the perimeter of parks. These results suggest Argentine
ants are largely confined to the edges of these parks. The low abundance of this species in
San Francisco parks stands in stark contrast to a recent study conducted in urban parks in
Australia. Walters (2006) found that 95% of ants in pitfall traps in invaded sites were
Argentine ants, compared to 26% in this study. The differences between the two studies may
be due to habitat type. Walters (2006) sampled in highly disturbed (mowed and watered with
sprinklers) parks, whereas this study was conducted in more “natural”urban parks.
In traps that did contain Argentine ants, only four other species (M. ergatongyna, P.
imparis, T. andrei, and T. sessile) were found, suggesting they can coexist with the
Argentine ant. This has been supported by previous studies showing T. andrei and P.
imparis to persist in areas with Argentine ants (Ward 1987; Human and Gordon 1996;
Suarez et al. 1998). Suarez et al. (1998) suggested T. andrei’ssmall size and cryptobiotic
nature reduces competition for resources such as nesting space and food with the larger
Argentine ant, which forages above ground. P. imparis tends to be more active in winter
months (Tschinkel 1987), when Argentine ant activity is lower. Additionally, Holway
(1999) documented that M. ergatogyna will use a highly repellent chemical defensive
compound to fend off Argentine ants.
Implications for conservation
Urbanization is a serious threat to many of the rare and declining plants and animals in the
San Francisco Bay area. Protecting and maintaining remaining urban natural areas not only
helps conserve the city’s biological diversity, but also provides valuable opportunities for
human enjoyment. Conserving species in urban landscapes will depend on effective
management planning, which includes both ecologically sensitive park design and the
creation of a system of urban ecological reserves (Adams and Dove 1989).
330 Urban Ecosyst (2008) 11:317–334
The results of this study indicate that natural areas within urban parks play a critical role
in supporting ant biodiversity. Many habitats in the natural areas of San Francisco’s parks
support healthy, diverse ant communities. Areas of non-native forest, however, reduce this
diversity. Maintaining open grasslands, reducing tracts of non-native forest, removing the
invasive understory, and thinning forest canopy may all help support a healthier ant
community and ecologically valuable parks.
Acknowledgements We thank The San Francisco Recreation and Parks Natural Areas Program for data and
access to study sites, Edward F. Connor for advice on statistical analysis and manuscript comments and Barry
Nickel at San Francisco State for help with GIS. We are grateful for help in the field from Maureen Harrison
and the constructive comments from members of the LeBuhn Lab. Kevin M. Clarke was supported by the
California Academy of Sciences Graduate Assistantship Program and SFSU SEPAL GK-12 Partnership
Program funded by the National Science Foundation.
Appendix 1
Table 6 List of parks surveyed, including summary of park characteristics
Park Total
Area (ha)
#of
Habitats
Soil
Moisture (%)
Irregularity
index
Forest
(%)
Grassland
(%)
Scrub
(%)
Mosaic
(%)
#of
pitfalls
Glen Canyon 24.28 7 3.36 2.13 29 23 39 3 117
Bayview 17.72 6 2.15 2.05 39 25 23 2 87
Mt. Davidson 16.15 6 15.65 1.52 75 12 8 4 76
Twin Peaks 11.86 7 3.68 5.04 3 18.7 36 35 57
Bernal Hill 9.75 6 2.70 2.95 6 82 2 2 44
Interior Green
Belt
6.63 2 15.85 1.62 100 0 0 0 27
Golden Gate Park 5.22 3 8.92 2.89 98 2 0 0 25
Corona Heights 3.84 7 2.98 1.53 27 50 10 3 19
Pine Lake 2.75 3 6.85 1.72 48 0 29 0 14
Buena Vista Park 2.47 4 8.24 2.03 100 0 0 0 18
Hawk Hill 1.78 5 2.20 1.64 4.2 0 84.1 0 9
O'shaughnessy
Hollow
1.74 4 2.13 2.02 0.4 22 67 0 7
Grandview Park 1.58 5 1.90 1.17 15 0 0 77.8 8
Billy Goat Hill 1.38 5 4.07 1.62 15 72 8 0 7
Tank Hill 1.13 5 4.77 1.21 19 55 17 0 6
Kite Hill 0.97 4 3.68 1.41 16 76 4 0 3
Edgehill
Mountain
0.93 3 15.97 1.33 80.9 5.1 0 0 3
Palou Phelps 0.81 5 3.15 1.88 24 51 0 22 3
Brooks Park 0.77 6 4.65 1.75 15.1 59.6 0 10.9 4
Rock Outcrop 0.65 5 1.57 1.72 4.2 8.8 4.4 0 3
Golden Gate
Heights
0.32 3 4.85 2.42 46.3 0 0 51.2 2
Fairmount Park 0.28 4 8.55 1.35 81 13.5 0 0 2
Lakeview Mini
Park
0.16 3 2.60 1.48 0 75.6 0 0 2
15th Avenue
Steps
0.12 3 2.65 1.71 48 0 48 4 2
Urban Ecosyst (2008) 11:317–334 331
References
Adams LW (2005) Urban wildlife ecology and conservation: a brief history of the discipline. Urban Ecosyst
8:139–156
Adams LW, Dove LE (1989) Wildlife reserves and corridors in the urban environment. National Institute for
Urban Wild, Columbia, MD
Ahrens CD (1994) Meteorology today. West Publishing, St. Paul, MN
Andersen AN (1995) A classification of Australian ant communities, based on functional-groups which
parallel plant life-forms in relation to stress and disturbance. J Biogeogr 22:15–29
Andersen AN, Hoffmann BD, Muller WJ, Griffiths AD (2002) Using ants as bioindicators in land
management: simplifying assessment of ant community responses. J Appl Ecol 39:8–17
Angilletta MJ, Wilson RS, Niehaus AC, Sears MW, Navas CA, Ribeiro PL (2007) Urban physiology: city
ants possess high heat tolerance. PLoS ONE 2:e258
Azevedo J, Morgan DL (1974) Fog precipitation in coastal California forests. Ecology 55:1135–1141
Baur B, Baur A (1993) Climatic warming due to thermal-radiation from an urban area as possible cause for
the local extinction of a land snail. J Appl Ecol 30:333–340
Beattie AJ (1985) The evolutionary ecology of ant–plant mutualisms. Cambridge Univ. Press, Cambridge,
UK
Bestelmeyer BT, Wiens JA (2001) Ant biodiversity in semiarid landscape mosaics: the consequences of
grazing vs. natural heterogeneity. Ecol Appl 11:1123–1140
Beyer HL (2004) Hawth's analysis tools for ArcGIS. Available at http://www.spatialecology.com/htools
Bolger DT, Suarez AV, Crooks KR, Morrison SA, Case TJ (2000) Arthropods in urban habitat fragments in
southern California: area, age, and edge effects. Ecol Appl 10:1230–1248
Bond W, Slingsby P (1984) Collapse of an ant–plant mutualism—the Argentine ant (Iridomyrmex humilis)
and myrmecochorous Proteaceae. Ecology 65:1031–1037
Clark WH, Prusso DC (1986) Desmidiospora myrmecophila found infesting the ant Camponotus
semitestaceus. Mycologia 78:865–866
Colwell RK, Coddington JA (1994) Estimating terrestrial biodiversity through extrapolation. Philos Trans R
Soc Lond B Biol Sci 345:101–118
Connor EF, Hafernik J, Levy J, Moore VL, Rickman J (2003) Insect conservation in an urban biodiversity
hotspot: the San Francisco Bay area. J Insect Conserv 6:247–259
Crooks KR, Soulé ME (1999) Mesopredator release and avifaunal extinctions in a fragmented system. Nature
400:563–566
Dauber J, Bengtsson J, Lenoir L (2006) Evaluating effects of habitat loss and land-use continuity on ant
species richness in seminatural grassland remnants. Conserv Biol 20:1150–1160
Dickman CR (1987) Habitat fragmentation and vertebrate species richness in an urban-environment. J Appl
Ecol 24:337–351
Drayton B, Primack RB (1996) Plant species lost in an isolated conservation area in Metropolitan Boston
from 1894 to 1993. Conserv Biol 10:30–39
Dreistadt SH, Dahlsten DL, Frankie GW (1990) Urban forests and insect ecology. Bioscience 40:192–198
EIP Associates (2005) Significant natural resources management plan. Prepared for San Francisco Recreation
and Parks Department, Natural Areas Program. June 2005
Fisher BL (1997) A comparison of ant assemblages (Hymenoptera, Formicidae) on serpentine and non-
serpentine soils in northern California. Insectes Soc 44:23–33
Fisher BL, Davis L, Ward PS (2007) Antweb: ants of the bay area, Available via http://www.antweb.org/
bayarea.jsp. Cited August 23, 2007
Folgarait PJ (1998) Ant biodiversity and its relationship to ecosystem functioning: a review. Biodivers
Conserv 7:1221–1244
Forman RT (1995) Land mosaics: the ecology of landscapes and regions. Cambridge Univ. Press, NY
Frankie GW, Ehler LE (1978) Ecology of insects in urban environments. Ann Rev Entomol 23:367–387
Gaston KJ (1996) Biodiversity. A biology of numbers and differences. Blackwell, London, UK
Gibb H, Hochuli DF (2002) Habitat fragmentation in an urban environment: large and small fragments
support different arthropod assemblages. Biol Conserv 106:91–100
Gilliam H (1962) Weather of the San Francisco Bay region. Univ. of California Press, Berkeley, CA
Gotelli NJ, Ellison AM (2002) Biogeography at a regional scale: determinants of ant species density in New
England bogs and forests. Ecology 83:1604–1609
Hafernik JE, Reinhard H (1995) Butterflies by the bay: winners and losers in San Francisco’s urban jungle.
Am Butterflies 3:4–11
Hitchings SP, Beebee TJ (1998) Loss of genetic diversity and fitness in common toad (Bufo bufo)
populations isolated by inimical habitat. J Evol Biol 11:269–283
332 Urban Ecosyst (2008) 11:317–334
Hölldobler B, Wilson E (1990) The ants. Harvard Univ. Press, Cambridge, MA
Holway DA (1999) Competitive mechanisms underlying the displacement of native ants by the invasive
Argentine ant. Ecology 80:238–251
Holway DA, Suarez AV (2006) Homogenization of ant communities in Mediterranean California: the effects
of urbanization and invasion. Biol Conserv 127:319–326
Holway DA, Suarez AV, Case TJ (2002) The role of abiotic factors in governing susceptibility to invasion: a
test with a widespread invasive social insect. Ecology 83:1610–1619
Human KG, Gordon DM (1996) Exploitation and interference competition between the invasive Argentine
ant, Linepithema humile, and native ant species. Oecologia 105:405–412
Human KG, Gordon DM (1997) Effects of Argentine ants on invertebrate biodiversity in northern California.
Conserv Biol 11:1242–1248
Jim CY (1998) Soil characteristics and management in an urban park in Hong Kong. Environ Manag
22:683–695
Johnson RA (1992) Soil texture as an influence on the distribution of the desert seed-harvester ants
Pogonomyrmex rugosus and Messor pergandei. Oecologia 89:118–124
Knight RL, Rust MK (1990) The urban ants of California with distribution notes of imported species.
Southwest Entomol 15:167–178
Laakkonen J, Fisher RN, Case TJ (2001) Effect of land cover, habitat fragmentation and ant colonies on the
distribution and abundance of shrews in Southern California. J Anim Ecol 70:776–788
Lassau SA, Hochuli DF (2004) Effects of habitat complexity on ant assemblages. Ecography 27:157–164
Lessard JP, Buddle CM (2005) The effects of urbanization on ant assemblages (Hymenoptera: Formicidae)
associated with the Molson Nature Reserve, Quebec. Can Entomol 137:215–225
Lopez-Moreno IR, Diaz-Betancourt ME, Landa TS (2003) Social insects in human environments—ants in
the city of Coatepec, Veracruz, Mexico. Sociobiology 42:605–621
MacArthur RH, Wilson EO (1967) The theory of island biogeography. Princeton Univ. Press, Princeton, NJ
Magurran AE (2004) Measuring biological diversity. Blackwell, Maldan, MA
Majer JD (1994) Spread of Argentine ants (Linepthema humile), with special reference to Western Australia.
In: Williams DF (ed) Exotic ants: biology, impact, and control of introduced species. Westview Press,
Boulder, CO, pp 163–173
Mangel M, Tier C (1994) 4 facts every conservation biologist should know about persistence. Ecology
75:607–614
McFrederick QS, LeBuhn G (2006) Are urban parks refuges for bumble bees Bombus spp. (Hymenoptera:
Apidae)? Biol Conserv 129:372–382
McIntyre NE, Rango J, Fagan WF, Faeth SH (2001) Ground arthropod community structure in a
heterogeneous urban environment. Landsc Urban Plan 52:257–274
Melbourne BA (1999) Bias in the effect of habitat structure on pitfall traps: an experimental evaluation. Aust
J Ecol 24:228–239
Moffatt SF, McLachlan SM, Kenkel NC (2004) Impacts of land use on riparian forest along an urban–rural
gradient in southern Manitoba. Plant Ecol 174:119–135
Montanucci RR (1989) The relationship of morphology to diet in the horned lizard genus phrynosoma.
Herpetologica 45:208–216
Newell W, Barber TC (1913) The Argentine ant. Bureau of Entomology Bulletin 122. USDA, Washington,
DC
Niemela J (1999) Ecology and urban planning. Biodivers Conserv 8:119–131
Null J (1992) The climatology of San Francisco rainfall 1849–1991. MA Thesis, San Jose State University
Pacheco R, Vasconcelos HL (2007) Invertebrate conservation in urban areas: ants in the Brazilian Cerrado.
Landsc Urban Plan 81:193–199
Patton DR (1975) A diversity index for quantifying habitat ‘edge’. Wildl Soc Bull 3:171–173
Perfecto I, Vandermeer J (1996) Microclimatic changes and the indirect loss of ant diversity in a tropical
agroecosystem. Oecologia 108:577–582
Poland TM, McCullough DG (2006) Emerald ash borer: invasion of the urban forest and the threat to North
America's ash resource. J For 104:118–124
Pyle R, Bentzien M, Opler P (1981) Insect conservation. Ann Rev Entomol 26:233–258
Sanders NJ, Barton KE, Gordon DM (2001) Long-term dynamics of the distribution of the invasive
Argentine ant, Linepithema humile, and native ant taxa in northern California. Oecologia 127:123–130
Saunders DA, Hobbs RJ, Margules CR (1991) Biological consequences of ecosystem fragmentation—a
review. Conserv Biol 5:18–32
Soulé ME (1987) Viable populations for conservation. Cambridge Univ. Press, NY
SPSS (2001) SPSS for Windows, Rel. 11.0.1. SPSS, Chicago
Suarez AV, Bolger DT, Case TJ (1998) Effects of fragmentation and invasion on native ant communities in
coastal southern California. Ecology 79:2041–2056
Urban Ecosyst (2008) 11:317–334 333
Suarez AV, Holway DA, Case TJ (2001) Patterns of spread in biological invasions dominated by long-
distance jump dispersal: insights from Argentine ants. Proc Natl Acad Sci U S A 98:1095–1100
Thompson B, McLachlan S (2007) The effects of urbanization on ant communities and myrmecochory in
Manitoba, Canada. Urban Ecosyst 10:43–52
Tilden JW (1956) San Francisco's vanishing butterflies. Lep News 10:3–4
Tschinkel WR (1987) Seasonal life-history and nest architecture of a winter-active ant, Prenolepis imparis.
Insectes Soc 34:143–164
Underwood EC, Fisher BL (2006) The role of ants in conservation monitoring: if, when, and how. Biol
Conserv 132:166–182
United Nations (2004) World urbanization prospects: the 2003 revision. Population Division, Department of
Economic and Social Affairs, United Nations, New York
U.S. Census Bureau (2000) County and city data book: 2000. US Department of Commerce, Bureau of the
Census, Washington, DC Table C-1
Vega SY, Rust MK (2001) Developing marking techniques to study movement and foraging of Argentine
ants (Hymenoptera: Formicidae). Sociobiology 37:27–39
Walters AC (2006) Invasion of Argentine ants (Hymenoptera: Formicidae) in South Australia: impacts on
community composition and abundance of invertebrates in urban parklands. Austral Ecol 31:567–576
Wang CL, Strazanac JS, Butler L (2001) Association between ants (Hymenoptera: Formicidae) and habitat
characteristics in oak-dominated mixed forests. Environ Entomol 30:842–848
Ward PS (1987) Distribution of the introduced Argentine ant (Iridomyrmex mumilis) in natural habitats of the
lower Sacramento Valley and its effects on the indigenous ant fauna. Hilgardia 55:1–16
Ward PS (2005) A synoptic review of the ants of California (Hymenoptera: Formicidae). Zootaxa 936:3–68
Way MJ, Cammell ME, Paiva MR, Collingwood CA (1997) Distribution and dynamics of the Argentine ant
Linepithema (iridomyrmex)humile (Mayr) in relation to vegetation, soil conditions, topography and
native competitor ants in Portugal. Insectes Soc 44:415–433
Yamaguchi T (2005) Influence of urbanization on ant distribution in parks of Tokyo and Chiba City, Japan II.
Analysis of species. Entomol Sci 8:17–25
Yilmaz S, Toy S, Irmak MA, Yilmaz H (2007) Determination of climatic differences in three different land
uses in the city of Erzurum, Turkey. Build Environ 42:1604–1612
Zar JH (1999) Biostatistical analysis, 4th edn. Prentice-Hall, Upper Saddle River, NJ
334 Urban Ecosyst (2008) 11:317–334
