Both Citizen Science Data and Field Surveys Detect Negative Impacts of Urbanization on Bird Communities
Caryn Ross1 and Sujan Henkanaththegedara1,*
1Department of Biological and Environmental Sciences, Longwood University, Farmville, VA 23909.
*Corresponding author
Urban Naturalist, No. 30 (2019)
Abstract
Birds have become a model species for observing the effects of urbanization on wildlife in urban ecology research. Although challenging to perform, broad scale analysis of urban impacts on bird populations may be necessary for conservation management strategies and urban planning. We assessed the feasibility of using citizen science data to determine the effects of urbanization on bird communities focusing on species richness, abundance, biomass, and feeding guild composition. We conducted field surveys to document narrow-scale differences of bird communities in non-urban (i.e., state parks) and urban (i.e., cities) sites in Virginia, USA. Subsequently, we collected comparable data using an online citizen science database (eBird) for the same sites to assess the correlation between field and eBird data. Additionally, we performed a broad-scale analysis using eBird data, covering 300 sites in five states in the southeastern United Sates. Our narrow-scale analysis showed that the average species richness in non-urban sites (36.80 ± 2.20) was significantly higher than that of urban sites (16.40 ± 1.50; W = 25, p = 0.0058). Additionally, Shannon-Wiener diversity of birds in non-urban sites were significantly higher than that of urban sites (W = 25, p = 0.0040). The feeding guild composition in urban areas was dominated by omnivores (48%) while insectivores (44%) dominated non-urban areas. Data gathered from eBird generated comparable outcomes for bird species richness and feeding guild composition for both non-urban and urban sites compared to our field survey results. The overall analysis showed a strong positive linear correlation between field survey data and eBird data for species richness (R2 = 0.8579, F1,8 = 55.33, p < 0.001). This indicates the importance of using untapped potential of eBird data to assess species richness, especially when traditional monitoring is not an option.
Download Full-text pdf
Site by Bennett Web & Design Co.
2019 Urban Naturalist 30:1–12
Both Citizen Science Data and Field Surveys Detect
Negative Impacts of Urbanization on Bird Communities
Caryn Ross1 and Sujan Henkanaththegedara1,*
Abstract - Birds have become a model species for observing the effects of urbanization on wildlife in
urban ecology research. Although challenging to perform, broad scale analysis of urban impacts on bird
populations may be necessary for conservation management strategies and urban planning. We assessed
the feasibility of using citizen science data to determine the effects of urbanization on bird communities
focusing on species richness, abundance, biomass, and feeding guild composition. We conducted field
surveys to document narrow-scale differences of bird communities in non-urban (i.e., state parks) and
urban (i.e., cities) sites in Virginia, USA. Subsequently, we collected comparable data using an online
citizen science database (eBird) for the same sites to assess the correlation between field and eBird data.
Additionally, we performed a broad-scale analysis using eBird data, covering 300 sites in five states
in the southeastern United Sates. Our narrow-scale analysis showed that the average species richness
in non-urban sites (36.80 ± 2.20) was significantly higher than that of urban sites (16.40 ± 1.50; W =
25, p = 0.0058). Additionally, Shannon-Wiener diversity of birds in non-urban sites were significantly
higher than that of urban sites (W = 25, p = 0.0040). The feeding guild composition in urban areas was
dominated by omnivores (48%) while insectivores (44%) dominated non-urban areas. Data gathered
from eBird generated comparable outcomes for bird species richness and feeding guild composition
for both non-urban and urban sites compared to our field survey results. The overall analysis showed
a strong positive linear correlation between field survey data and eBird data for species richness (R2 =
0.8579, F1,8 = 55.33, p < 0.001). This indicates the importance of using untapped potential of eBird data
to assess species richness, especially when traditional monitoring is not an option.
Introduction
Environmental conditions are constantly changing due to human activities with
urbanization being one of the most significant human-induced issues in the contemporary
world (Partecke et al. 2006, Sushinsky et al. 2013, Silva et al. 2016). It is well known that
the global human population is projected to grow rapidly, increasing urbanization, while
consequently affecting many wildlife species and their habitats (Gagne and Fahrig 2011, Ikin
et al. 2013, Miller and Hobbs 2002). Changes associated with urbanization such as habitat
loss and fragmentation, changes in vegetation structure and food supply, and introduction of
exotic, predatory or competitor species cause an array of negative impacts on native species
and their habitats (McDonnell and Pickett 1993, Chase and Walsh 2006, Zhou and Chu 2012).
Recent work in the United States has shown urbanization to be a leading cause of decline in
more than 50% of threatened or endangered species declared under the Endangered Species
Act (Czech et al. 2000, McKinney 2002, Miller and Hobbs 2002, Smith and Chow-Fraser
2010). On the other hand, urban areas also support populations of urban adapted bird species
contributing to the regional biodiversity (Blair 1996; Fuller et al. 2009), and in some cases,
even healthy refuge populations of conservation concern species (see Fuller et al. 2009).
Birds have become a frequently used model species for studying the effects of urbanization
on wildlife populations (Blair 1996, Gagne and Fahrig 2011, Mills et al. 1989, Ormond et al.
2014, Zuckerberg et al. 2011). A plethora of studies have shown strong negative correlations
1 Department of Biological and Environmental Sciences, Longwood University, Farmville, VA 23909.
* Corresponding author: henkanaththegedarasm@longwood.edu
Manuscript Editor: Anett Richter
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
2
between degree of urbanization and bird species richness with relatively low richness of
native bird species in urbanized areas (Chase and Walsh 2006, MacGregor-Fors et al. 2012,
Mills et al. 1989, Zhou and Chu 2012). However, the abundance and biomass of bird species
are higher in urban settings than nearby natural areas due to large populations of exotic or
generalist species (Blair 1996, Chase and Walsh 2006, McKinney 2002, Mills et al. 1989,
Warren and Lepczyk 2012). Furthermore, Blair (1996) showed that the peak diversity is at
the moderately disturbed areas compared to both undisturbed and highly developed areas
in an urban gradient supporting intermediate disturbance hypothesis (Connell 1978). Bird
community composition also changed from mostly native species in the undisturbed areas to
mostly invasive and exotic species in highly developed areas (Blair 1996, Crooks et al. 2003,
Gagne and Fahrig 2011, MacGregor-Fors et al. 2012, McKinney and Lockwood 2001).
The majority of previous studies have a narrow focus and confined to localized areas
(Blair 1996, Cam et al. 2000, Crooks et al. 2003, Mills et al. 1989). However, broad
scale patterns of urban impacts on bird populations may be necessary to achieve sound
conservation management strategies (Miller and Hobbs 2002), and urban planning (Ikin
et al. 2013). Obtaining large-scale data sets covering wider geographic areas and long
spans of time is very challenging due to limited labor, time, and funding (Cohn 2008).
However, development of citizen science and publicly available citizen science data may
bridge this gap allowing large-scale exploration of ecological and conservation oriented
research questions (Bonney et al. 2009, Cohn 2008, Dickinson et al. 2012). These citizen
science databases provide large quantities of previously unavailable free data on species
occurrences and distribution at global scale, allowing conservation practitioners to assess
the status, distribution and ecological interactions of targeted species (Bonney et al. 2009,
Devictor et al. 2010, Dickinson et al. 2012).
The purpose of this study is to assess the feasibility of using citizen science data
(eBird, Sullivan et al. 2009) to determine the effects of urbanization on bird communities
focusing on species richness, abundance, biomass, and feeding guild composition. The
use of citizen science data gives access to a rich data set, allowing us to analyze and
compare vast geographic areas in state and/or national levels. Understanding broad scale
ecological patterns in bird communities across an urban gradient may help us implement
better management strategies within urban and non-urban areas (Dickinson and Bonney
2012, Ikin et al. 2013, Lepczyk and Warren 2012, Zhou and Chu 2012). First, we conducted
field surveys to document narrow-scale differences of bird diversity in selected non-urban
(i.e., state parks) and urban (i.e., cities) sites in Virginia. Second, we collected comparable
data using an online citizen science database (eBird.org) for the same sites to assess the
feasibility of using citizen-collected data for large-scale analyses. Third, we performed a
broad-scale analysis of the impacts of urbanization on bird diversity using citizen-collected
data, covering five states in the southeastern United Sates.
Methods
Narrow-scale Analysis
Study area. We conducted narrow-scale field surveys to document the bird species
richness and abundance within the Piedmont region of Virginia in five state parks (nonurbanized)
and in five cities (urbanized). All state parks selected (see Table 1) are frequently
visited for recreational activities, such as, multi-use hiking trails, biking, kayaking, and
swimming. Each park contains at least one source of water, all of those being man-made
reservoirs with the exception of the James River State Park, which has three miles of
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
3
shoreline along the James River. Holliday Lake State Park and Bear Creek Lake State Park
are both located within State Forests, and James River State Park is located at the foothills
of the Blue Ridge Mountains, making these three parks furthest from urbanized areas. The
two remaining state parks are in closer proximity to an urban area. Twin Lakes State Park is
located outside of Farmville, VA and Pocahontas State Park is located outside of Richmond,
VA (Table 1). The five cities selected (see Table 1) as field sites have a wide range of
populations; however, all exceeded the minimum requirement of 2,500 people to be deemed
urban by the US Census Bureau (uscensus.gov). Midlothian contains a large reservoir and
multiple creeks that ultimately feed into the James River below the fall line located within
Richmond. The James River also runs through the cities of Charlottesville and Lynchburg.
Farmville is the only urban field-site to be traversed by the Appomattox River, the major
southern tributary of the James River. Each urban field survey was conducted in the center
of the cities, where the landscape is primarily man-made structures, roads, and heavily
trafficked by humans (Table 1).
Field surveys. Field surveys were conducted in order to generate baseline data on bird
diversity and to ground-truth citizen-collected data. In May–June 2014, we visited five
selected state parks and five selected cities in Virginia and conducted bird surveys. Each site
was surveyed once during the study period using a series of random point counts along a
pre-determined transect documenting all bird species and number of individuals recognized
by sight or calls. Each point count lasted 10–15 minutes and each location was surveyed
approximately 2 hours between the hours of 7.00 and 11.00 am. We included only species
identified with certainty for the analysis.
Citizen science data mining. We utilized eBird citizen science database (www.ebird.org;
Sullivan et al. 2009) to gather citizen-collected data on bird occurrences. Information on bird
species richness for the same state parks and cities covered by field surveys were gathered
using Explore Hotspots tool. We restricted our search to include only the past 10 years for
all months and collected data only on bird species richness (i.e., the total number of species
Table 1. Description of the field survey sites. Number of visitors for each state park was obtained from
Virginia Department of Conservation and Recreation (dcr.virginia.gov/state-parks/) and residential population
for each city was obtained from US Census Bureau (uscensus.gov).
Field site Area (km2) Human population size
State Park (Non-urban) Visitor Population (2015)
Holliday Lake State Park 2.27 46,076
Bear Creek Lake State Park 1.33 90,398
Twin Lakes State Park 2.22 112,480
James River State Park 6.32 116,335
Pocahontas State Park 32.84 1,142,601
City (Urban) Residential Population (2015)
Farmville 19.00 8,169
Midlothian 83.92 48,386
Richmond 161.90 220,289
Lynchburg 128.49 79,812
Charlottesville 26.70 46,597
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
4
recorded for each site). We utilized this citizen-collected data to compare with our field-survey
data and to assess the feasibility of using citizen-collected data for broad-scale analysis.
Data Analysis. The mean differences of bird species richness between urban (cities)
and non-urban (state parks) areas for field-collected data were compared using nonparametric
Wilcoxon rank sum test due to small sample size (N = 5). The number of
bird species was treated as the response variable and location (urban vs. non-urban) was
treated as the predictor variable. We also compared Shannon-Wiener species diversity
index (H'), overall abundance, overall biomass and Simpson’s dominance index (D) of
birds between non-urban and urban field sites in a similar fashion using non-parametric
Wilcoxon rank sum test. Shannon-Wiener diversity index (H') and Simpson’s dominance
index (D) were computed using the formulae (Krebs 1999):
H' = -Σ (Pi * ln Pi)
D = Σ(Pi)2
Where: Pi = fraction of the entire population made up of species i.
Relative abundance and percentage occurrence of bird species were estimated using following
formulae (Krebs 1999):
Relative abundance =
Abundance of a given species
The total abundance of all species combined
Percentage occurrence =
Number of occurrences of a given species in sampling units
Total number of sampling units
In order to analyze functional diversity of bird communities, we utilized feeding guild
composition (Mills et al. 1989, Blair 1996). Each bird species was assigned to one of
seven feeding guilds (insectivore, omnivore, carnivore, herbivore, granivore, nectarivore
and frugivore) based on their major food source. The information of the diet of birds was
extracted from The Birds of North America Online (Poole 2005). Differences of feeding
guild composition between non-urban and urban sites were analyzed using a two-way
ANOVA test considering number of species as the response variable and both location
(Non-urban vs. urban) and feeding guild as predictor variables.
The mean differences of bird species richness between urban (cities) and non-urban
(state parks) areas for eBird data (N = 30) were compared using parametric two sample t-test
using number of bird species as the response variable and the degree of urbanization as the
predictor variable. Differences of feeding guild composition between non-urban and urban
sites were also analyzed for eBird data using a two-way ANOVA test considering number
of species as the response variable, and both sites (non-urban vs. urban) and feeding guild
as predictor variables. Finally, any correlations of species richness of birds between field
survey data and eBird data were assessed using a simple linear regression model on pooled
data (N = 10) to determine the feasibility of using eBird data to assess broad-scale analysis.
Broader-scale analysis
Study area. We selected 30 state parks and 30 cities for this analysis from Virginia, North
Carolina, South Carolina, Georgia, and Florida covering the southeastern United States.
When selecting state parks, we attempted to cover all physiogeographic zones of the state to
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
5
get a wider coverage of habitat types. When selecting cities from each state, we used 2012
census data from US Census Bureau (uscensus.gov) and chose the top 30 most populated
cities/ towns in each state to represent our urban areas.
Citizen science data mining. Each state park and city were then searched for bird species
richness using the eBird Explore Hotspots tool with a time range set to the past 10 years for
all months. In some rare occurrences, a new randomly selected state park was added to the
list when there was no eBird data for the selected state park. Similarly, if a city/town did
not have eBird data, the next highest populated location in that state was added to the list.
Overall, we sampled 150 cities and 150 state parks in the southeastern United States for bird
species richness.
Data analysis. Average differences of species richness of birds between non-urban
and urban sites for each state and for pooled data for all five states were individually
analyzed using two sample t-test. Subsequently, a two-way ANOVA test was conducted
on pooled data to partition the variation across states considering number of species
as the response variable and both site (non-urban vs. urban) and state as predictor
variables including interaction terms between site and state. All statistical analyses were
conducted using the R statistical software program (version 3.2.2; R Development Core
Team 2016).
Results
Narrow-scale analysis
Our field surveys yielded an overall of 77 bird species belonging to 64 genera and 33
families (see Supplemental File 1, available online at https://eaglehill.us/urnaonline2/supplfiles/
urna-152-Ross-s1.pdf). A total of 72 bird species was reported from non-urban sites
(i.e., state parks) and 28 species from urban sites (i.e., cities). The average species richness
in non-urban sites (36.80 ± 2.20) was significantly higher than that of urban sites (16.40 ±
1.50; W = 25.0, p = 0.0058). Additionally, Shannon-Wiener diversity of birds in non-urban
sites were significantly higher than that of urban sites (W = 25.0, p = 0.0040). Furthermore,
Simpson’s dominance index was significantly higher in urban sites compared to non-urban
sites (W = 0.0, p = 0.0079) due to large populations of a few dominant species, indicating
higher species diversity of birds in non-urban areas. However, overall bird abundance (W
= 5.0, p = 0.9524) and overall biomass (W = 10.0, p = 0.7262) did not vary significantly
between non-urban and urban sites (Table 2).
The relative abundance of individual species in non-urban sites was much lower
than that of urban sites. The most abundant bird species in non-urban sites were Branta
Canadensis (Canada Goose, relative abundance [RA] = 0.072) and Polioptila caerulea
(Blue-gray Gnatcatcher, RA = 0.071) followed by Baeolophus bicolor (Tufted Titmouse,
RA = 0.046), Bombycilla cedrorum (Cedar Waxwing, RA = 0.043), and Spinus tristis
(American Goldfinch, RA = 0.043). In urban sites, the most abundant bird species were
non-native Sturnus vulgaris (European Starling, RA = 0.300), Chaetura pelagic (Chimney
Swift, RA = 0.162), and non-native Passer domesticus (House Sparrow, RA = 0.133).
The highest biomass in non-urban sites was represented by relatively large bodied birds
such as, Canada Goose (70.51%), Cathartes aura (Turkey Vulture, 5.73%), Meleagris
gallopavo (Wild Turkey, 5.45%), and Corvus brachyrhynchus (American Crow, 3.95%).
Similarly in urban sites, relatively large-bodied birds such as Turkey Vulture (39.22%)
and American Crow (21.40%) represented the highest biomass. Although relatively
small, it is surprising that non-native European Starlings represented 18.23% of the
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
6
total biomass in urban sites, due to large population size. Only 18% of bird species were
reported in each of five survey sites for both urban (5 species) and non-urban (13 species)
sites (see Supplemental File 1).
Field survey data also revealed significant differences of feeding guild composition
of birds between non-urban and urban sites (F1,56 = 80.97, p < 0.001), between feeding
guilds (F6,56 = 113.92, p < 0.001) and significant interactions between sites and feeding
guilds (F6,56 = 36.77, p < 0.001; Fig. 1). Feeding guild composition from non-urban sites
was dominated by insectivores (44%) followed by omnivores (39%). By contrast, the
feeding guild composition in urban sites was dominated by omnivores (48%) followed
Table 2. A comparison of ecological parameters of bird species diversity between non-urban (i.e., state
parks) and urban (i.e., cities) field sites (N = 5). Values in parentheses show standard error of the mean.
Ecological parameter Non-urban site (State parks) Urban site (City) W value p value
Species richness 36.8 (± SE 2.20) 16.4 (± SE 1.50) 25.0 0.0058
Shannon-Weiner 3.252 (± SE 0.024) 2.236 (± SE 0.155) 25.0 0.004
Diversity Index (H)
Overall abundance 138 (± SE 18.46) 193 (± SE 38.19) 5.0 0.9524
Overall biomass (kg) 63.82 (± SE 45.39) 29.86 (± SE 7.89) 10.0 0.7262
Simpson’s Dominance 0.053 (± SE 0.009) 0.266 (± SE 0.046) 0.0 0.0079
Index (D)
Figure 1. Bird species
richness in each feeding
guild based on field surveys
(A) and eBird data (B) for
five non-urban (i.e. state
parks) and urban (i.e. cities)
sites in Central Virginia.
Both field surveys and eBird
data showed insectivoredominant
non-urban sites
and omnivore-dominant
urban sites.
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
7
by granivores (19%) and insectivores (18%, see Supplemental File 2, available online at
https://eaglehill.us/urnaonline2/suppl-files/urna-152-Ross-s2.pdf).
Citizen-collected data mined from eBird database for the same non-urban sites (i.e.,
state parks) and urban sites (i.e., cities) also showed high diversity of birds in non-urban
sites compared to urban sites. The average species richness of birds in non-urban sites
(120.4 ± SE 10.84) was significantly higher (t = 7.223, p < 0.001) than that of urban
sites (40.0 ± SE 5.56), indicating higher bird diversity in non-urban areas compared to
urban sites. Furthermore, this also indicates that eBird data yields much higher species
richness values probably due to data saturation. Similarly, eBird data revealed significant
differences of feeding guild composition of birds between non-urban and urban sites
(F1,56 = 47.64, p < 0.001), between feeding guilds (F6,56 = 20.95, p < 0.001) and significant
interactions between sites and feeding guilds (F6,56 = 19.35, p < 0.001; Fig. 1). Feeding
guild composition generated with eBird data showed a very similar pattern compared to
our field survey data. Feeding guild composition in non-urban sites was dominated by
insectivores (40%) followed by omnivores (32%). Additionally, eBird data picked up a
considerable fraction of carnivores (19%) for non-urban sites. By contrast, the feeding
guilds in urban sites were dominated by omnivores (45%) followed by insectivores
(28%). Surprisingly, the contribution from granivores was very limited (7%) and a
considerable fraction of carnivores (15%) was represented compared to the field survey
data (see Supplemental File 2).
Although eBird data resulted in significantly higher values for species richness than the
field survey data (F1,16 = 59.65, p < 0.001, Fig. 2), a simple linear regression model showed a
strong positive linear correlation between field survey data and eBird data (N = 10, Adjusted
R2 = 0.8579, F1,8 = 55.33, p < 0.001, Fig. 3). This indicates that it is reasonably appropriate
to use eBird data to assess broader-scale patterns of bird species richness.
Broad-scale Analysis
The average species richness of birds across five states for non-urban sites (i.e., state
parks) ranged from 105.1 (Georgia) to 159.4 (Florida). The average bird species richness
for urban sites (i.e., cities) varied within a narrow range from 27.4 (North Carolina) to
37.53 (Georgia). The bird species richness was significantly higher for non-urban sites
Figure 2. Relative species
richness of birds generated
by field surveys and eBird
data. eBird data resulted in
significantly higher values
for species richness than
the field survey data (F1,16
= 59.65, p < 0.001).
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
8
(125.82 ± 3.83) compared to urban sites (33.35 ± 1.16) for pooled data across all states
(F1, 290 = 612.984, p < 0.0001), and for individual states (see Table 3). Additionally, there
were significant differences of species richness between states (F4,290 = 7.091, p < 0.0001)
including significant interactions between sites and states (F 4,290 = 5.873, p < 0.001).
Discussion
Our narrow-scale analysis showed that bird species diversity in urban areas was
consistently lower than that of non-urban areas, providing evidence for negative impacts
of urbanization on bird species diversity. A few species dominated the bird community
structure in urban areas reducing the overall diversity compared to non-urban sites.
Specifically, three bird species including two non-native species (i.e., European Starling
and House Sparrow) represented nearly 60% of the overall abundance of the birds in urban
areas. By contrast, in non-urban areas, the most abundant three species represented less than
20% of the overall abundance of birds. Feeding guild composition also showed significant
differences between urban and non-urban areas. Urban areas were dominated by omnivores
(48%), while non-urban areas were dominated by insectivores (44 %).
Table 3. A comparison of average bird species richness in non-urban (i.e. state parks) and urban (i.e.
cities) sites across five southeastern states. Values in parentheses show standard error of the mean.
State Non-urban Urban N t value p value
Virginia 130.37 (± 8.94) 30.40 (± 2.16) 30 10.87 < 0.001
North Carolina 116.43 (± 5.70) 27.4 (± 1.89) 30 14.81 < 0.001
South Carolina 117.80 (± 10.83) 34.07 (± 3.02) 30 7.44 < 0.001
Georgia 105.1 (± 5.73) 37.53 (± 3.02) 30 10.44 < 0.001
Florida 159.4 (± 7.35) 37.37 (± 2.36) 30 15.8 < 0.001
All states 125.82 (± 3.83) 33.35 (± 1.16) 150 23.16 < 0.001
Figure 3. A simple linear
regression model showed
a strong positive linear
correlation between field
survey data and eBird
data (N = 10, Adjusted R2
= 0.8579, F1,8 = 55.33,
p < 0.001) for non-urban
(i.e. state parks) and
urban (i.e. cities) sites (N
= 10).
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
9
Consistent with previous research, our results clearly show significantly higher species
diversity of birds in natural areas compared to urban areas (Blair 1996, Chase and Walsh
2006, MacGregor-Fors et al. 2012, McKinney 2002). This is most likely due to the sensitivity
of birds to different effects of urbanization such as habitat loss and fragmentation, changes
in vegetation structure, altered resource availability and nesting sites, new competitors
and predators (Lowe et al. 2011, McDonnell and Pickett 1993). It has been shown that the
disturbance caused by urbanization may negatively affect some species (urban avoiders),
benefit some others (urban exploiters), and benefit the rest up to a moderate level before
negatively affecting them (suburban adaptable) (Blair 1996, Pauw and Louw 2012).
Urbanization may cause generalist species (including non-native species) to exploit urban
resources more effectively and replace native specialist species that can no longer survive in
the altered habitats (Gagne and Fahrig 2011, McKinney and Lockwood 1999). While overall
species diversity of birds decrease in urban areas, overall abundance and biomass tend to
increase due to generalist species that are able to thrive in an urban setting (Chase and Walsh
2006, Crooks 2003, Smith and Chow-Fraser 2010). More urbanized habitats provide a wide
array of roosting and nesting structures from man-made building cavities to bird houses,
thus favoring cavity-nesting bird species such as non-native European Starlings and House
Sparrows (Chase and Walsh 2006). These species-specific interactions with the environment
lead to an overall biodiversity loss and biotic homogenization (McKinney and Lockwood
1999), creating a much more simplified bird community within urb anized areas.
Our results also showed significant differences in feeding guild composition between
urban and non-urban areas. Both field data and eBird data showed similar patterns for
feeding guild composition with insectivore-dominated community composition in nonurban
areas, while omnivore-dominated community structure in urban areas. These results
were most likely due to species-specific traits associated with urbanization sensitivity
(McKinney and Lockwood, 1999). Previous research has shown that urbanized areas tend
to select for generalist, omnivores and granivores because of their ability to exploit the
resources in urban areas much effectively (Crooks 2003, McKinney and Lockwood 1999,
Silva 2016). By contrast, urban settings discourage insectivores, which is the leading
feeding guild in non-urban areas, possibly due to poor insect availability and altered habitat
structure (MacGregor-Fors et al. 2012, McKinney and Lockwood 1999).
Although urbanization generally poses a negative impact on native bird species diversity, not
all bird species show consistent declines in urban settings; hence, urban habitats are not always
present ecological sinks (Warren and Lepczyk 2012). Our field surveys showed that native
Chimney Swifts thrive in urban settings due to the availability of man-made nesting and roosting
sites making them the third most abundant (16.2%) bird species in urban sites. Moreover, we
reported a single Falco peregrine (Peregrine Falcon) in Richmond city center during our field
surveys endorsing the positive impacts of man-made nest boxes in tall buildings for this once
imperiled, cliff nester (Barclay and Cade 1983). We also noted that native scavengers such as
Turkey Vultures thrive in urban habitats representing the largest biomass (39.2%) possibly due
to the availability of novel food items such as road kills (Henkanaththegedara, unpubl. data).
Other studies have shown that backyard bird feeding in urban areas may also have positive
effects on some adaptable native bird species resulting in stable or increasing populations and
increased reproductive success (Robb et al. 2008). Fuller et al. (2009) reported that urban areas
support high densities of nationally threatened bird species, such as Song Thrush in the United
Kingdom, making urban habitats important as refuge populations for such species.
Data gathered from eBird generated very similar outcomes for bird species richness
and feeding guild composition for both non-urban and urban sites compared to our field
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
10
survey results. The correlation between eBird data and field survey data for the narrow-scale
analysis yielded a strong positive linear relationship validating the utility of eBird data to
assess bird species richness and allowing us to expand our analysis to a broader geographic
area. However, eBird consistently generated higher values for species richness compared to
our field survey data. This may be due to data saturation (Krebs 1999) in eBird data base,
since we have set the parameters to extract data covering past 10 years for all 12 months for
every site (vs. our field surveys conducted in May-June during a single season). Broad-scale
analysis covering 300 sites across five southeastern states consistently resulted in low bird
species richness for urban sites, compared to non-urban sites, which suggests widespread
negative impacts of urbanization on bird species richness. Our results also suggest that
state park systems (i.e., non-urban sites) have a positive impact on conservation of native
bird assemblages. Overall, our findings using citizen-collected data from eBird database
generated results consistent with previous work (Mills et al. 1989, Blair 1996, McKinney
2002, MacGregor-Fors et al. 2012) suggesting the feasibility of wider application of citizen
science data such as eBird to assess broad-scale impacts of urbanization.
Citizen science data has a significant utility in environmental and conservation research
such as environmental pollution, climate change, species occurrence, distribution and
composition (Cohn 2008, Bonney et al. 2009, Devictor et al. 2010, Dickinson and Bonney
2012). Furthermore, recent extensive engagement of citizen scientists with scientific research
coupled with the ready availability of data due to advances in information technology makes
citizen science data a powerful tool for environmental and conservation research. Although
some citizen science databases such as USGS Breeding Bird Survey (Sauer et al. 2003) and
Audubon Society’s Christmas Bird Count (Bock and Root 1981) has generated hundreds of
ecological and conservation publications, eBird is still underutilized, despite its untapped
potential as an extensive data base with quality data and ready access. A limited number of
researchers have utilized eBird data for avian research including modeling large scale bird
distributions (Fink et al. 2013, Humphries and Huettmann 2014), assessing regional bird
diversity (Akresh and King 2015, More et al. 2015) and bird migration patterns (La Sorte et
al. 2014). However, we realize much broader application of eBird data for a wide array of
avian research such as monitoring of species with conservation concerns, tracking spatial
and temporal dynamics of bird populations/communities, and assessing responses of bird
populations to anthropogenic stressors (e.g., this study) as well as natural stressors. More
importantly, eBird provides extensive bird occurrence data covering a vast geographic area
allowing researchers to conduct national and/or global scale analyses.
Acknowledgments
We thank The Longwood University Perspectives on Research In Mathematics & Science (PRISM)
summer undergraduate research program and the Margaret Watson Bird Club of Farmville, Virginia for research
funding; Emily Crawford, Brady Donovan, and Rivoningo Hlophe for preliminary research; Bodini
Herath, Mihin, and Sanuthi Henkanaththegedara, Warren Rofe, and Mary Elliott for assistance in the field;
the Cornell Lab of Ornithology for making all the data used in this study available through ebird.org, and
reviews by Anett Richter and two anonymous reviewers of an earlier draft of this manuscript.
Literature Cited
Akresh, M. E., and D.I. King. 2015. Observations of new bird species for San Salvador Island, the
Bahamas. Caribbean Naturalist 26:1–10.
Barclay, J. H., and T.J. Cade. 1983. Restoration of the Peregrine Falcon in the eastern United States. Bird
Conservation 1:3–40.
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
11
Bock, C. E., and T.L. Root. 1981. The Christmas bird count and avian ecology. Pp. 17–23, In C.J. Ralph
and J.M. Scott (Eds.). Estimating Numbers of Terrestrial Birds. Studies in Avian Biology 6. Cooper
Ornithological Society. Allen Press, Inc., Lawrence, KS, USA. 630 pp.
Bonney, R., C.B. Cooper, J. Dickinson, S. Kelling, T. Phillips, K.V. Rosenberg, and J. Shirk. 2009. Citizen
Science: A developing tool for expanding science knowledge and scientific literacy. Bioscience
59:977–984.
Blair, R.B. 1996. Land use and avian species diversity along an urban gradient. Ecological Applications
6:506–519.
Cam, E., J.D. Nichols, J.R. Sauer, J.E. Hines, and C.H. Flather. 2000. Relative species richness and
community completeness: birds and urbanization in the Mid-Atlantic States. Ecological Applications
10:1196–1210.
Chase, J.F., and J.J. Walsh. 2006. Urban effects on native avifauna: A review. Landscape and Urban
Planning 74:46–69.
Cohn, J.P. 2008. Citizen science: Can volunteers do real research? Bioscience 58:192–197.
Connell, J.H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302–1310.
Crooks, K.R., A.V. Suarez, and D.T. Bolger. 2003. Avian assemblages along a gradient of urbanization in
a highly fragmented landscape. Biological Conservation 115:451–462.
Czech, B., P.R. Krausman, and P.K. Devers. 2000. Economic associations among causes of species
endangerment in the United States. Bioscience 50:593–601.
Devictor, V., R.J. Whittaker, and C. Beltrame. 2010. Beyond scarcity: Citizen science programmes as
useful tools for conservation biogeography. Diversity and Distributions 16:354–362.
Dickinson, J.L., and R. Bonney. 2012. Citizen science: Public participation in environmental research.
Comstock Publishing Associates, Ithaca, NY, USA. 279 pp.
Dickinson, J. L., J. Shirk, D. Bonter, R. Bonney, R.L. Crain, J. Martin, T. Phillips, and K. Purcell. 2012.
The current state of citizen science as a tool for ecological research and public engagement. Frontiers
in Ecology and the Environment 10:291–297.
Fink, D., T. Damoulas, and J. Dave. 2013. Adaptive spatio-temporal exploratory models: Hemispherewide
species distributions from massively crowdsourced eBird data. Proceedings of the Twenty-
Seventh AAAI Conference on Artificial Intelligence. Bellevue, DC, USA.
Fuller, R.A., J. Tratalos, and K.J. Gaston. 2009. How many birds are there in a city of half a million
people? Diversity and Distributions 15:328–337.
Gagne, S.A., and L. Fahrig. 2011. Do birds and beetles show similar responses to urbanization? Ecological
Applications 21:2297–2312.
Humphries, G. R. and F. Huettmann. 2014. Putting models to a good use: A rapid assessment of Arctic
seabird biodiversity indicates potential conflicts with shipping lanes and human activity. Diversity
and Distributions 20:478–490.
Ikin, K., E. Knight, D.B. Lindenmayer, J. Fischer, and A.D. Manning. 2013. The influence of native versus
exotic streetscape vegetation on the spatial distribution of birds in suburbs and reserves. Diversity and
Distributions 19:294–306.
Krebs, C.J. 1999. Ecological Methodology, 2nd Ed. Addison-Wesley Educational Publishers, Inc., Menlo
Park, CA, USA. 620 pp.
La Sorte, F.A., M.W. Tingley, and A.H. Hurlbert. 2014. The role of urban and agricultural areas during
avian migration: An assessment of within-year temporal turnover. Global Ecology and Biogeography
23:1225–1234.
Lepczyk, C.A., and P.S. Warren. 2012. Urban bird ecology and conservation, Studies in avian biology No.
45, Cooper Ornithological Society and University of California Press, Berkeley, CA, USA. 326 pp.
Lowe, K.A., C.E. Taylor, and R.E. Major. 2011. Do Common Mynas significantly compete with native
birds in urban environments? Journal of Ornithology 152:909–921.
MacGregor-Fors, I., L. Morales-Perez, and J.E. Schondube. 2012. From forests to cities: Effects of
urbanization on tropical birds. Pp. 33–48, In C.A. Lepczyk and P.S. Warren (Eds.). Urban bird
ecology and conservation, Studies in avian biology (No. 45), University of California Press, Berkeley,
CA, USA. 326 pp.
Urban Naturalist
C. Ross and S. Henkanaththegedara
2019 No. 30
12
McDonnell, M.J. and S.T.A. Pickett. 1993. Humans as components of ecosystems: The ecology of subtle
human effects and populated areas. Springer-Verlag, NY, New York. USA. 316 pp.
McKinney, M. L. 2002. Urbanization, biodiversity, and conservation. Bioscience 52:883–890.
McKinney, M.L., and J.L. Lockwood. 2001. Biotic homogenization: A sequential and selective process.
Pp. 1–17, In Lockwood, J.L, and M.L. McKinney (Eds.). Biotic Homogenization. Kluwer Academic/
Plenum Publishers, New York, NY, USA. 278 pp.
McKinney, M.L. and J.L. Lockwood. 1999. Biotic homogenization: A few winners replacing many losers
in the next mass extinction. Trends in Ecology and Evolution 14:450–453
Miller, J.R., and J.R. Hobbs. 2002. Conservation where people live and work. Conservation Biology
16:330–337.
Mills, G., J. Dunning, and J. Bates. 1989. Effects of urbanization on breeding bird community structure
in southwestern desert habitats. The Condor 91:416–428.
More S. C., A. More, I. Franke, and C. Mendoza. 2015. First documented records of Red-faced Parrot
(Hapalopsittaca pyrrhops) from Peru. Boletín de la Unión de Ornitólogos del Perú (UNOP) 10:69–74.
Ormond, S.A., R. Whatmough, L.L. Hudson, and C.B. Daniels. 2014. Environmental and anthropogenic
impacts on avifaunal assemblages in an urban parkland. Animals (Basel) 4:119–130.
Partecke, J., I. Schwabl, and E. Gwinner. 2006. Stress and the city: Urbanization and its effects on the
stress physiology in European blackbirds. Ecology 87:1945–1952.
Pauw, A., and K. Louw. 2012. Urbanization drives a reduction in functional diversity in a guild of nectarfeeding
birds. Ecology and Society 17:27.
Poole, A. 2005. The Birds of North America. Cornell Laboratory of Ornithology, Ithaca, NY. Available
online at: http://bna.birds.cornell.edu/BNA/. Accessed May 2014.
R Core Team 2016. R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna, Austria. Available online at: https://www.R-project.org/. Accessed September
2016.
Robb, G. N., R.A. McDonald, D.E. Chamberlain, and S. Bearhop. 2008. Food for thought: Supplementary
feeding as a driver of ecological change in avian populations. Frontiers in Ecology and the
Environment 6:476–484.
Sauer, J. R., J. E. Hines, and J. Fallown. 2003. The North American Breeding Bird Survey: Results and
analysis 1966–2003. Version 2003.1. USGS, Patuxent Wildlife Research Center, Laurel, Maryland,
USA. Available online at: www.mbr-pwrc.usgs.gov/bbs/bbs.html. Accessed December 2016.
Silva, C.P., R.D. Sepúlveda, and O. Barbosa. 2016. Nonrandom filtering effect on birds: Species and
guilds response to urbanization. Ecology and Evolution 6(11):3711–3720.
Smith, L.A., and P. Chow-Fraser. 2010. Impacts of adjacent land use and isolation on marsh bird
communities. Environmental Management 45:1040–1051.
Sullivan, B.L., C.L. Wood, M.J. Iliff, R.E. Bonney, D. Fink, and S. Kelling. 2009. eBird: A citizen-based
bird observation network in the biological sciences. Biological Conservation 142:2282–2292.
Sushinsky, J.R., J.R. Rhodes, H.P. Possingham, T.K. Gill, and R.A. Fuller. 2013. How should we grow
cities to minimize their biodiversity impacts. Global Change Biology 19:401–410.
Warren, P.S. and C.A. Lepczyk. 2012. Beyond the gradient: Insights from new work in the avian ecology
of urbanizing lands. Pp. 1–6, In Lockwood, J.L, and M.L. McKinney (Eds.). Urban bird ecology and
conservation, Studies in avian biology (No. 45), University of California Press, Berkeley, CA, USA.
326 pp.
Zhou, D., and L.M. Chu. 2012. How would size, age, human disturbance, and vegetation structure affect
bird communities of urban parks in different seasons? Journal of Ornithology 153:1101–1112.
Zuckerberg, B., D.N. Bonter, W.M. Hochachka, W.D. Koenig, A.T. DeGaetano, and J.L. Dickinson. 2011.
Climatic constraints on wintering bird distributions are modified by urbanization and weather. Journal
of Animal Ecology 80:403–413.