A Deep Dive into the Waterbird Community of an Urban Oasis: Implications for Park Management

doi:10.21203/rs.3.rs-3793956/v1

Download PDF

Research Article

A Deep Dive into the Waterbird Community of an Urban Oasis: Implications for Park Management

https://doi.org/10.21203/rs.3.rs-3793956/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 03 Jul, 2024

Read the published version in Urban Ecosystems →

You are reading this latest preprint version

The rapid decline in avifauna populations across North America underscores the urgency of implementing effective conservation strategies. While waterfowl species have generally increased in abundance, due to investments in habitat restoration and species management, most other waterbird species have declined. In contrast to the widespread decline of natural habitat for waterbirds, urban greenspaces have proliferated. However, the extent to which urban greenspaces provide habitat for waterbirds is largely unknown. We measured waterbird abundance in Forest Park, Saint Louis, which contains a large expanse of restored waterways. Between 2020 and 2022, we used point counts to quantify alpha and beta diversity of waterbirds across five waterbodies. We also quantified several environmental factors hypothesized to influence waterbird presence including weather, bathymetry, water quality, and prey abundance. We assessed environmental drivers of alpha and gamma diversity using Generalized Linear Mixed Models, and beta diversity with Jaccard’s similarity indices and Mantel tests. Water depth and depth heterogeneity were the strongest determinants of alpha diversity. Measured environmental variables did not predict beta diversity. Forest Park is a regionally important urban habitat for waterbirds, hosting 62% of the total regional species pool, including nine species of conservation concern and two state endangered species. Urban parks can contribute to waterbird conservation in the face of habitat loss elsewhere if park managers prioritize habitat restoration. In addition to the total area of waterways, water depth and water depth homogeneity are important considerations to maximize waterbird species diversity in urban greenspaces.

Urban greenspace

habitat reconciliation

urban wetland

waterfowl

wading birds

The rapid decline in avifauna populations of North America in the past half-century is largely due to a variety of anthropogenic impacts (Hallmann et al. 2014, Stanton et al. 2018, Rosenberg et al. 2019). Population declines have occurred in all major taxa, except for wetland bird species which have generally increased in abundance, in part due to long term investment in conservation management (Anderson et al. 2018). The most effective management interventions include habitat restoration and regulation of unsustainable hunting; however, the recovery of wetland bird species has been inconsistent across taxa (Rosenberg et al. 2019). Since 1970, there has been a 56% increase in waterfowl populations— species in the Anatidae family including ducks, geese, and swan. Meanwhile, shorebirds, a subset of waterbirds in the order Charadriiformes and managed under the U.S. Shorebird Conservation Partnership Plan, declined by 37.4% and all other waterbirds, species ecologically dependent on water, declined by 21.5% (Wetlands International 2010, Rosenberg et al. 2019). This stark contrast coincides with continued wetland habitat loss both within North America and around the world (Davidson 2014, Sofaer et al. 2016). Many waterbird species are obligate wetland habitat specialists limited to habitats such as swamps and marshes (Gutzwiller and Flather 2011). In contrast, other waterbird species have a wider ecological breadth, exploiting additional habitats including artificial wetlands found in urban centers (McKinney et al. 2006, Baratti et al. 2009, Avilova 2016). Moreover, these taxonomic divisions do not reflect ecological functional diversity, and different functional groups display different habitat requirements which influence population trajectories in response to environmental change (Cadotte et al. 2011, Laureto et al. 2015, Stroud et al. 2015).

In the Midwestern United States, multiple states have lost over 80% of their historic wetlands, in large part due to agriculture and urbanization (Zedler 2004). The loss of existing habitat has resulted in declines for both overall avian species and functional diversity, as well as increased risk for currently stable populations of waterfowl, such as diving ducks (Sol et al. 2020, Donnelly et al. 2022). However, habitat retention or restoration in managed urban green space could potentially replace losses of some critical habitat for some avian species and communities (Snep et al. 2016). Reconciliation ecology encourages the redevelopment of artificial habitats to support and further increase biodiversity (Rosenzweig 2003, Lundholm and Richardson 2010, Muñoz-Pedreros and Dellacasa 2023). Identifying the features of artificial wetlands that influence the presence of waterbirds and waterfowl could allow managers to reinforce these influential factors and consequently encourage a greater presence of these birds. Ma et al. (2010) provided insights into critical factors to consider in this context, as well as the yet-to-be-explored interactions among these factors. These factors, which include water depth, connectivity, food and food availability, vegetation, salinity, topography, and water level fluctuation, are all suggested to be interconnected. Understanding the interactive effects of these and other factors on waterbird habitat use, could identify priorities for wetland management in urban greenspaces.

Urban greenspaces and the waterways they contain may provide suitable habitats for wetland bird species that can tolerate high levels of anthropogenic disturbance (Lundholm and Richardson 2010). Some species, such as the mallard (Anas platyrhynchos), thrive in urban areas, and urban greenspace may become increasingly more important for populations of these species (Lowry et al. 2013, Meissner et al. 2015, Singh et al. 2021, Wolf et al. 2022). However, the potential of urban greenspace to provide wetland bird habitat is poorly understood. Predictably, increasing park area and distance from city centers is correlated with increasing bird diversity (Huang et al. 2022). However, the impact of urban parks on the presence of threatened or endangered waterbird species has been little studied (Fernández-Juricic and Jokimäki 2001, Homayoun and Blair 2016, Yang et al. 2020). While rigorous studies of waterbird and waterfowl community ecology have been reported from Asia (Chang and Lee 2016, Xu et al. 2022) and South America (Corrêa and Melo 2013), few published data are available from the Midwestern United States (Andrade et al. 2018, Rahlin et al. 2022). While the dramatic rise in popularity of citizen science projects, such as eBird (Sullivan et al. 2009), may capture the general composition of bird communities found in given areas, including urban parks, the biases inherent in citizen science methods can skew accurate understanding of species abundance, richness, and diversity (Dickinson et al. 2010). These include spatial and temporal sampling bias (Tang et al. 2021) and observation bias toward large, easily identified, charismatic, and/or rare species (Callaghan et al. 2021, Stoudt et al. 2022, Randler et al. 2023).

Saint Louis, located in eastern Missouri, lies on the Mississippi River. The portion of the Mississippi Flyway that passes through Saint Louis once supported enormous tracts of wetland habitats. Today, following the massive canalization project of the Army Corps of Engineers in the early 1900s (Brown 2012) and the expansion of urban Saint Louis, wetland habitats have decreased to a few isolated remnant fragments (Maimaitijiang et al. 2015). Just west of the geographic center of the city, an artificial waterway in Forest Park has been under a robust conservation management plan for over 20 years that includes steps to increase biodiversity (Forest Park Forever 2020). This plan includes wetland restoration for both conservation and recreation purposes. However, the impact of this plan on waterbird species diversity, functional diversity, and conservation implications has not been evaluated.

With this background in mind, we sought to answer the following questions: 1) What are the spatial and temporal patterns of waterbird species alpha, beta and gamma diversity in Forest Park; 2) What are the spatial and temporal patterns of waterbirds by functional group; 3) How many species of conservation concern occur in the park; and 4) What ecological and anthropogenic factors influence the presence of waterbird and waterfowl species or functional groups. Answering these questions allowed us to consider the current conservation value of Forest Park for waterbirds and waterfowl and recommend management interventions. These questions provoked the following hypotheses: 1) Alpha and gamma diversity of species and functional groups will peak during migratory seasons; 2) Waterbird community similarity is positively correlated with environmental similarity between sites; and 3) Among tested environmental factors, alpha diversity will be best predicted by the foraging area available for wading and diving species.

Study Site

We studied the presence of waterfowl and waterbird species along the River Des Peres watershed in Forest Park, Saint Louis, Missouri (Fig. 1). The greater Saint Louis region (ca. 21,906 km²) is home to over 2.8 million people, spanning some 15 counties across the states of Missouri and Illinois (U.S. Census Bureau 2020). The region lies within both the Ozark highland and the interior river valleys and hills ecoregions (Omernik and Griffith 2014). Saint Louis City is located ca. 6 km south of the confluence of the Missouri and Mississippi Rivers, making it a major urban center along the Mississippi avian migration flyway (Oliver et al. 2011). Forest Park is surrounded by the urban landscapes of Saint Louis City to the north, south and east and Saint Louis County to the west (Fig. 1). The park has been identified as a part of the Saint Louis Urban Oases, a listed Important Bird Area (IBA) of global priority for waterbirds by the National Audubon Society (Wells et al. 2005, Jensen 2007). The 5.25 km² park receives 15 million visitors annually and has over 0.75 km² of nature reserve (Forest Park Forever 2020). This reserve includes 1.8 km of artificial waterway, sourced primarily via treated water, which was constructed over time following the burial of River Des Peres in the early 1900s (Brown 2012).

Methods

Point Count Surveys

Point count surveys were conducted to estimate alpha diversity of waterbirds and waterfowl at five sites during two survey periods. The first survey period was conducted from September 2020 through August 2021 and the second from January 2022 through December 2022. The sites were selected based on size and connectivity, and include: 1. Grand Basin (GB), a large ornamental lake surrounded by a concrete ledge and containing multiple fountains; 2. Post-Dispatch Lake (PD), a large artificial lake, semi-natural in appearance, containing three small islands; 3. Bullfrog Lake (BF), a small pond with many fallen trees along its shoreline; 4. Catfish Cove (CC), a small pond adjacent to BF and stocked with fish year-round; 5. Jefferson Lake (JL), a large artificial lake, semi-natural in appearance, stocked year-round (Fig. 1).

Point count surveys were conducted twice per week during the first survey period and once per week in the second (except in cases of severe thunderstorms when surveys were abandoned). At least three surveys were conducted each month. The survey duration at each site was 20 minutes. The survey cycle began between 30 and 45 minutes after local sunrise. The maximum observed abundance for a given species was recorded noting the sex of all individuals. Individuals were recorded if they, or their flock, were within ca. 10 meters of the site. On each survey day, every site was visited within a three-hour window. The starting site for each survey rotated in a set order to avoid temporal bias in count data. In addition, the number of pedestrians that were present within ca. 10 meters of the site was counted at the beginning of each site survey as an index of human activity.

eBird

All bird sightings in Forest Park reported on eBird from 2020 through 2022 were retrieved and filtered to species presence in the park as a complementary source of species richness data.

Definition of Functional Groups

We placed all species of waterbird and waterfowl into four distinct functional groups based primarily on foraging behavior (Reynolds and Cumming 2016, Duan et al. 2020). These functional groups included: 1. dabbling species (DAB), which primarily forage by swimming on top of the water; 2. diving species (DIV), which primarily forage by diving under the water; 3. wading species (WAD), which primarily forage by wading near the shore; 4. grazing species (GRZ), which primarily forage on land near the shore. Notably, shorebirds were excluded from analysis due to their limited abundance in the park and limited availability of foraging habitat.

Sampling of Environmental Conditions

Factors identified as possibly influencing species’ presence were measured at different scales throughout the duration of the study.

Weather

Weather data were obtained from the Saint Louis Science Center’s weather station located in the southeast corner of Forest Park. Monthly means of precipitation and air temperature, median ambient temperature, and total precipitation were calculated from these data. In addition, as surveys were conducted between 6:00 am and 11:00 am, the mean precipitation and air temperature, median temperature, and total precipitation between these hours on each survey date were also calculated.

Site-level Environmental Factors

Prey relative abundance

Wading bird prey relative abundance was measured via crayfish barrel traps deployed from April of 2022 through November of 2022. At JL, PD, and GB, 20 traps were deployed overnight in pairs around the perimeter approximately equidistant to one another and as close to the edge as possible while remaining completely submerged. At BF and CC, 10 traps in pairs were deployed due to the smaller perimeters of these sites. Traps were deployed at each site once a month for approximately 12 hours before retrieval. Prey species were identified and counted for each trap.

Water Quality

Water quality was sampled at each site once per week between April and November 2022, except in cases of severe thunderstorms. Turbidity was recorded using a Hach 2100Q Portable Turbidimeter. Dissolved oxygen and water temperature were recorded using a YSI Pro20i Dissolved Oxygen Meter, and salinity, total dissolved solids, conductivity, and pH were recorded using a Hach Pocket Pro + Multi 2 Tester. The multi tester was calibrated daily following manufacturer’s instructions and the turbidimeter monthly to ensure accuracy of measurements. Water level was also recorded using a grade rod to the nearest tenth of a foot, which was then converted to metric.

Bathymetry

Bathymetry data for each site, except JL, were obtained using the Deeper Sonar PRO + 2 fish finder with a spatial distribution of < 15 m between sample points. ArcGIS Pro (Esri, version 3.1, 2023) was used for all bathymetric calculations. A bathymetric raster for each site was interpolated using the ArcGIS Pro Topo to Raster tool, with a pixel size of ca. 0.8 meters. The bathymetry map for JL was generated from data collected from a Saint Louis City survey conducted in 2020. These data were rasterized in ArcGIS Pro, also using the Topo to Raster tool. From each raster, the interpolated average depth, max depth, depth standard deviation and variance were obtained using the Create Chart tool in ArcGIS Pro of the raster histogram. In addition, the extent of each site that was usable for wading species to forage was calculated both as an absolute area (m²) and as a percentage of the total area of each site. The usable extent was defined based on an estimated depth of 0.6 meters, as the largest wading bird observed, the great blue heron (Ardea herodias), has a maximum total leg length of ca. 0.58 meters (Riegner and Bassar 2023).

Waterbody Surface Area and Connectivity

The surface area of each site was calculated by multiplying the number of pixels of each site’s bathymetry raster by the cell size found in the raster properties. The connectivity between each site was calculated as the minimum distance between the closest edges of each site using the Near tool in ArcGIS Pro.

Statistical Analyses

Statistical analyses were performed using R software (R Core Team, version R-4.2.2, 2022). We evaluated spatial and temporal patterns in species richness and Hill-Simpson diversity within the park (Roswell et al. 2021). The Hill-Simpson diversity represents the count of equally common species necessary to yield the Simpson index observed at each study site (Jost 2006). Hill-Simpson was selected for use due to its emphasis on common species, since mallard and Canada geese (Branta canadensis) dominated community composition. We tested hypothesis 1 (species and functional diversity varied across months) with a two-way Analysis of Variance (ANOVA) with the index value as the dependent variable and month and site as independent variables. The Hill-Simpson diversity data were transformed by taking the square root to meet the assumptions of the test. A Tukey’s HSD post-hoc test was then performed. We tested hypothesis 2 (community similarity is correlated with environmental similarity) by using Jaccard’s similarity indices (Koleff et al. 2003), followed by a series of Mantel tests (Legendre et al. 2005) to estimate community similarity between sites (beta diversity) by species and functional groups. We assessed temporal change in community metrics across seasons, defined as months grouped in three, with spring starting in March to reflect early migration.

We tested hypothesis 3 by reassessing the effects of site-level factors measured for both study years on the species and functional Hill-Simpson diversities using generalized linear mixed models (GLMMs). Site characteristics, both static (e.g., area and depth), and temporal variables (e.g., month and year, air temperature, precipitation, daylength, number of pedestrians), were included in models as fixed effects. Site was included as a random effect, while year was tested as both a fixed and random effect. We analyzed additional site factors measured from spring to fall 2022 using GLMMs to understand their effects on species and functional Hill-Simpson diversities. This included previous parameters, excluding the year, and incorporated significant water quality principal components (PCs) taken from a principal components analysis (PCA). The PCA was performed on all contiguous water quality data and a broken stick model was then used to determine which PCs were statistically significant (Bro and Smilde 2014). Significant PCs were then used in modeling. We also modeled wading bird abundance as a function of predator related environmental factors, specifically the percentage of area available to wade at each site and the monthly number of prey items caught. This analysis also considered previous parameters, excluding year, as well as water quality PCs. For GLMMs, we rescaled each continuous parameter via Z-score normalization to facilitate comparison of coefficient sizes of independent variables. We used Akaike information criteria (AIC) (Burnham et al. 2011) to test the different predictors on species and functional diversities, as well as wading bird abundance.

Of the 61 native waterbird and waterfowl species present in the Saint Louis region within the functional groups outlined (Alwood 2022), we observed 21 species from point count surveys (Supplementary Table 1). Between point count observations and eBird submissions during the same timeframe, 38 of the 61 native species present in the region were recorded in Forest Park.

Spatial and Temporal Patterns of Species Richness, Gamma, and Alpha Species Diversity

Species richness varied significantly across both month (F_{(11, 43)} = 3.65, p = 0.001) and site (F_{(4, 43)} = 24.71, p < 0.001) (Fig. 2). Temporally, species richness was greatest in April (15), followed by May, October, and November (11), and smallest in January (6), followed by February, March, June, July, and December (9). Spatially, PD had the greatest species richness (18), followed by JL (15), BF (12), CC (10), and GB had the lowest (8). Testing effects of season and site, species richness varied significantly across season (F _{(3, 12)} = 3.55, p = 0.05) and site (F _{(4, 12)} = 7.96, p = 0.002). Gamma diversity varied significantly by month, with a bimodal distribution peaking during the spring and fall migrations, supporting hypothesis 1 (Fig. 3). Gamma functional diversity remained consistent throughout the study period. Alpha diversity varied significantly across both month (F _{(11, 45)} = 5.58, p < 0.001) and site (F _{(4, 45)} = 10.42, p < 0.001) but there was no evidence of an interaction between month and site (F _{(43, 45)} = 1.19, p = 0.34), indicating that the effect of site level factors on diversity were consistent across seasons. Among seasons, alpha diversity was highest during the spring (F _{(3, 20)} = 15.14, p < 0.001) and among sites, at CC (F _{(4, 20)} = 3.73, p = 0.02). There was little support for an interaction between season and site (F _{(12, 20)} = 2.09, p = 0.07) suggesting that spatial patterns of species diversity remain relatively consistent over time.

Community Composition over Time and Space (Beta Diversity)

Community similarity measured via Jaccard’s Similarity Index was highest between JL and PD (JI = 0.4776) and lowest between JL and CC, indicating high species turnover between sites (Fig. 4A). The only species observed at all sites were the mallard, wood duck (Aix sponsa), Canada goose, and great egret (Ardea alba), of which the mallard and Canada goose were most abundant. Species richness was greatest at PD. Species that were in greater abundances at other sites outside PD included wood duck and green heron (Butorides virescens), which were in greatest abundance at BF, and American coot (Fulica americana), pied-billed grebe (Podilymbus podiceps), and Canada goose, which were in greatest abundance at JL. Gadwall (Mareca strepera) were the only species that were not observed in greater abundance at any one site. Additionally, ruddy duck (Oxyura jamaicensis) and snowy egret (Egretta thula) were only observed at PD, while yellow-crowned night heron (Nyctanassa violacea) was only observed at BF. Similarity among sites varied temporally between tested summer and winter months. Over summer months, species beta diversity ranged from very low to moderate, with the highest similarity between JL and PD (JI = 0.5408) and the lowest similarity between JL and GB (JI = 0.0193) (Fig. 4B). Site similarities during the winter ranged from very low to moderately similar, with highest similarity between BF and CC (JI = 0.4348) and lowest similarity between JL and BF (JI = 0.0161) (Fig. 4C). Between all species comparisons, most sites exhibited low similarity regardless of season.

The similarity of functional groups between sites was highest between JL and PD (JI = 0.6828) and lowest between JL and CC (JI = 0.0498), consistent with species trends (Fig. 4D). Site similarity over summer months was highest between JL and PD (JI = 0.5816) and lowest between JL and GB (JI = 0.0193) (Fig. 4E), following an increase in wading species and decrease in diving species at every site. The pattern of summer functional similarity was consistent with patterns of species similarity. Over winter months, functional similarity between sites was highest between BF and CC (JI = 0.6500) and lowest between JL and CC (JI = 0.0172), following a decrease in wading species and increase in diving species at every site. While the lowest winter functional diversity differed from the lowest winter species similarity, the differences between the JL to CC and JL to BF similarities were less than 0.002, and therefore followed the same trend as the winter species. Mantel tests indicated that none of the environmental drivers hypothesized to influence community similarity had significant impacts on beta diversity.

The Influence of Environmental Factors on Patterns of Species Diversity and Community Composition

Of the measured differences in area and depth, JL had the largest area (32,328 m²) and a maximum depth of 2.5 meters, followed by GB, PD, CC, and BF with an area of 2,593 m² and a maximum depth of 2.28 meters (Fig. 5A, Supplementary Table 2). Jefferson Lake (JL) had the highest wading bird percent forging area (29.69%) followed by PD, CC, BF, and finally GB with less than 0.001% of its area available for wading. GB had the highest total number of pedestrians, 646, counted over both observation periods, followed by JL with 232 pedestrians, PD with 60 pedestrians, BF with 28 pedestrians, and CC with 26 pedestrians (Supplementary Table 3).

Individual water quality metrics varied between sites and over time. Dissolved oxygen, total dissolved solids, salinity, and conductivity all followed a unimodal distribution reaching a low in the summer apart from JL, which had extremely high values in the spring for each metric. BF had ca. 1 meter water level fluctuation, followed by PD, CC, JL, and GB. Measures of conductivity, salinity and total dissolved solids showed JL with the greatest average value for each, followed by GB, PD, BF and CC. Mean pH ranged between 7 and 10, with PD having the highest followed by GB, CC, BF and JL. Dissolved oxygen mean values followed a similar trend between sites. The water temperature was nearly identical between sites for any given day, and mean turbidity was highest at GB, followed by PD, JL, CC, and BF. A principal component analysis of water quality metrics revealed the first and second PCs (PC1 and PC2) were statistically significant, with PC1 accounting for 42% of total variance and PC2 accounting for 23%. The PCA indicated there were significant differences in overall water quality between sites, except between GB and PD, the only study ponds connected by a physical channel (Fig. 1; Fig. 5B). Ordination lines of each water quality metric demonstrated that PC1 was closely associated with change in conductivity, salinity, and total dissolved solids, while changes in PC2 were closely associated with dissolved oxygen and to a partial extent turbidity and water temperature.

The highest prey abundance was recorded at BF with a mean of 51 prey items caught per trap, followed by PD (7), CC (6), GB and JF (4). Prey species caught included various sunfish species, bass, and catfish, as well as crayfish and tadpoles. The presence of tadpoles at BF on two trap sessions greatly inflated the mean trap prey count. If these two outlier trap sessions are excluded, mean trap catch was 9 prey items per session.

All GLMMs supported month, day length, average depth, and depth variance as important environmental determinants of species and functional diversity (Table 3). Moreover, the best models with species and functional diversity as response variables across both study periods also included precipitation and the number of pedestrians present. The best models with species and functional diversity response variables between spring and fall 2022 also included average temperature, while the best model with wading bird abundance also included percent foraging area for wading birds. Higher diversity occurred in spring and fall months, correlating with lower precipitation and temperature. Finally, lower temperatures between spring and fall were associated with higher wading bird abundance.

Table 1

The number of species observed at five sites (BF, CC, GB, JL, PD) in Forest Park from 2020 through 2022, separated by functional group. Groups included DAB (dabbling species), DIV (diving species), GRZ (grazing species), and WAD (wading species). Species not in any of the designated functional groups or are non-native were excluded from the table.
Functional Group	Number of Species							N Species Observed	Total Species in Saint Louis Region
Functional Group	All Sites	BF	CC	GB	JL	PD	eBird Only	N Species Observed	Total Species in Saint Louis Region
DAB	6	4	3	3	5	5	2	8	11
DIV	6	1	2	2	5	6	7	13	25
GRZ	2	1	1	1	1	2	4	6	7
WAD	7	6	4	2	4	6	4	11	18
Total	21	12	10	8	15	19	17	38	61

Table 2

Species alpha diversity trends differed between study periods, possibly due to increased evenness at survey sites BF and CC. The highest value is bolded.
Site	Species Alpha Diversity
Site	Overall	Study Period 1	Study Period 2	Change
BF	2.59	1.89	5.61	+ 3.72
CC	3.50	1.91	4.43	+ 2.52
GB	2.89	1.71	3.34	+ 1.63
JL	3.01	3.12	2.34	-0.76
PD	2.48	2.46	2.45	-0.01

Table 3

The predictor variables used in each best fit GLMM. ‘2022’ indicates the response variable being tested was between spring and fall 2022. ‘Y’ indicates the predictor variable significantly improved model AIC. ‘–’ indicates the predictor was not tested for the given response variable.
Predictor Variable	Response Variable
Predictor Variable	Species Diversity	Functional Diversity	Species Diversity (2022)	Functional Diversity (2022)	Wad Area %
Month	Y	Y	Y	Y	Y
Day Length	Y	Y	Y	Y	Y
Depth Variance	Y	Y	Y	Y	Y
Avg Depth	Y	Y	Y	Y
Avg Temperature (Hour)			Y	Y	Y
Avg Precipitation (Hour)	Y	Y
N Pedestrians	Y	Y
PC2			Y
WAD Area Percent	–	–	–	–	Y
Top Model AIC Score	164.43	104.79	-112.35	-114.02	-123.8
Top Model Conditional R²	0.51	0.55	0.71	0.67	0.99

Gamma diversity and temporal variations in species presence

Of a total regional waterfowl and waterbird richness of 61 species, we observed 21 species (34.4%) in Forest Park. An additional 15 species were reported via eBird between 2020 and 2022, resulting in a combined total richness of 38 species, or 62.3% of the regional species pool. We conclude that the isolated waterways of Forest Park provide an important contribution to waterbird and waterfowl habitat in the heart of urban Saint Louis. In support of hypothesis 1, we observed a bimodal distribution in alpha and gamma diversity for both species and functional groups over time, with peaks occurring during spring and fall migrations. Of the 13 species of waterbird and waterfowl identified for conservation concern in Missouri (Missouri Department of Conservation 2022), we observed four, including the state endangered snowy egret. An additional five species were reported via eBird, including the state endangered American bittern (Botaurus lentiginosus). These endangered species were only observed rarely, and the significance of their presence within the park is difficult to evaluate due to limited knowledge on their overall population trends in the greater Saint Louis region. Previous work has identified that the presence of avian species of conservation concern decreases as urbanization increases, particularly for raptors, Galliformes (land fowl) and certain passerine species (Sorace and Gustin 2010). This stands in contrast to our findings that, of the native waterbird and waterfowl to the Saint Louis region, only four species of conservation concern were not recorded in Forest Park via our data or eBird: the least bittern (Ixobrychus exilis), yellow rail (Coturnicops noveboracensis), king rail (Rallus elegans), and common gallinule (Gallinula galeata). This may suggest the presence of waterbird and waterfowl species of conservation concern may not be as negatively impacted by urbanization in the Saint Louis region as with other avian groups, however more research is needed. The presence of these species within Forest Park underscores the Park’s importance as a prime habitat within the urbanized region for both resident and migratory species and reinforces the role of urban greenspaces in conserving biodiversity and supporting threatened avian populations.

Some of the more common species showed highly variable distributions over time and space. For example, wood ducks were consistently abundant at BF in the fall of the first study period. However, during the second study period, they were almost completely absent with no obvious change in habitat characteristics. Hooded mergansers (Lophodytes cucullatus), consistently present in the first winter at multiple sites but at low abundance, avoided all sites in the second winter despite being reported elsewhere in the park via eBird. The temporal distributions of migratory species, with these examples in mind, appear to be highly variable and consequential in habitat selection preference. Mallards and Canada geese gathered in huge numbers (> 150 individuals in some flocks) during the fall and into the winter both years but abundance of both crashed in the spring and summer. The functional gamma diversity followed similar temporal trends as species diversity, with a bimodal distribution peaking in the spring and fall. Forest Park provides habitat for both migrating and resident species, promoting overall avian diversity and ecosystem function. Spatial differences in diversity within the park, however, suggest some environmental factors have a greater effect on avian diversity than others.

Environmental Drivers of Diversity

Beta Diversity

Refuting hypothesis 2, none of the tested environmental factors were significantly correlated with beta diversity. This is despite some site comparisons demonstrating moderate similarity via the Jaccard’s Similarity Index, and the temporal variation found when isolating seasons. Clear patterns of habitat selection were observed over both study periods, however, our analyses failed to provide context for these observed patterns. This may be due to a compounding effect that multiple variables which cannot be detected from univariate mantel tests. We may also have failed to include important environmental determinants of community structure or failed to adequately measure environmental variables as they are perceived by the bird species within communities.

Alpha Diversity

The environmental drivers of species and functional alpha diversity supported by our analysis included month, daylength, average water depth, variance in water depth, average precipitation, and the number of pedestrians, partially supporting hypothesis 3. Waterbird diversity by site was positively correlated with decreasing mean water depth and increasing water depth variation. Correlations between depth and depth variance and species diversity have been inconsistently reported (Colwell and Taft 2000, Bara and Segura 2019). Average precipitation, meanwhile, along with the number of pedestrians had no clear temporal patterns. Both precipitation and pedestrian numbers were negatively correlated with diversity. In contrast to our results, Li et al. (2022) found that, along the Liaohe Estuary in northeastern China, while waterfowl functional diversity was positively correlated with temperature and negatively correlated with precipitation (which our results follow), for the wading bird functional diversity, the opposite trends were observed. This difference observed in precipitation and temperature correlation with diversity among wading bird species may be due to differences in patterns of migration of the individual species being compared.

Water quality was not included in our best supported model explaining alpha diversity. In contrast, Miotto et al. (2023) did find that waterbird species richness was positively correlated with water temperature and dissolved oxygen while negatively correlated with turbidity. As these three metrics are closely associated with seasonal change and plant productivity (Carr et al. 1997, Barznji 2014), it is possible that PC2 is indirectly associated with site vegetation cover as well. PC2 values followed a unimodal distribution, albeit weakly. This pattern mirrored the diversity trends, which also displayed a unimodal pattern, reaching a local minimum during the summer. However, additional investigation is warranted, as it is plausible that incorporating site vegetation measurements could yield a stronger correlation with diversity.

Finally, our GLMM of wading birds also found the wading percent area predictor significantly improved model fit and therefore was included in the best fit model. Sites with a greater estimated wadable area had an increased number of wading birds observed. However, most wading species could not wade as deep as the maximum depth selected to estimate wadable area, and the wadable area did not account for perches above the water such as fallen tree limbs and artificial structures. Regardless, the estimated wading area did improve the model significantly. There are several potential reasons as to why mean trap prey count did not improve model fit. For example, we were unable to assess prey quality and may have incurred bias in our estimates of total prey abundance due to small trap sizes. Short pulses of extremely high abundance may also have distorted underlying trends.

Implications for Management and Conservation of Waterbirds

Forest Park has ca. 0.35 km² of open water available for waterbird use, of which we sampled ca. 0.09 km². The closest open water to Forest Park of over 0.01 km² is over 4 km away, and only six other waterbodies within Saint Louis City are larger than 0.01 km². Forest Park as the largest of few viable habitats for waterbirds in this urbanized region. Understanding the drivers of avian diversity in urban parks can inform planning and management of aquatic habitats generally (Ma et al. 2010), and in urban greenspaces (Fontana et al. 2011). Forest Park supports 38 species (62% of regional species richness). The presence of 9 of 13 state species of conservation concern highlights the park's role in conserving biodiversity, indicating a positive impact this urban greenspace has on waterbird and waterfowl populations. We have shown that waterbody depth and depth heterogeneity have a strong positive impact on waterbird and waterfowl species richness, likely because they increase overall habitat heterogeneity. Park managers can directly manipulate waterbody bathymetry structure to increase bird diversity, a likely indicator of overall community biodiversity. While the impact of additional habitat features- such as vegetation cover and habitat structure at waterbody margins- remains to be evaluated, our research has shown that active management of urban greenspace waterways can support waterbird and waterfowl diversity by contributing novel habitat to the conservation landscape for these species.

Acknowledgements

We thank Sharon L. Deem, Jason Knouft, Stella Uiterwaal, Nina Fogel and Cristy Portales-Reyes for their technical support, Samantha Serra, Cathy Taylor, Vasuki Rao, David Gosser, Nikhil Samayam, Isabella Camba, Matthew Montgomery, and Saima Farook for their assistance in data collection. Christopher Brzozski and Collin Fricke for their assistance with data entry. We also thank the numerous eBird citizen scientists of the Saint Louis community for conducting their own surveys in Forest Park. Forest Park Forever supported the study administratively and provided bathymetry data. The Saint Louis Science Center and University of Missouri-Columbia provided local weather data. Megan Johnson, Fabio Giacomelli, Natalia Agostini Schneider, Nelson Membreno, Alison Breakfield and Kiana Lindblad for all their additional support in completing this research.

Ethics Approval

All fish and aquatic invertebrate trapping and handling were covered by a Missouri Department of Conservation permit.

Author Contributions

August Wise and Stephen Blake designed the study. All authors contributed to data collection. August Wise analyzed the data and wrote the first draft of the manuscript. All authors edited and developed the final draft. All authors approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Availability of Data and Materials

Point count survey, prey abundance, water quality, and bathymetric data were deposited into the Open Science Framework repository and are available via the following URL: https://osf.io/z2kam/?view_only=4dc05b2075e74b0587445f036b26992a

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Alwood C, Rowe W, Uffman J (2022) St. Louis Circle Checklist. St. Louis Audubon
Anderson MG, Alisauskas RT, Batt BDJ, Blohm RJ, Higgins KF, Perry MC, Ringelman JK, Sedinger JS, Serie JR, Sharp DE, Trauger DL, Williams CK (2018) The Migratory Bird Treaty and a Century of Waterfowl Conservation. J Wildl Manage 82:247–259
Andrade R, Bateman HL, Franklin J, Allen D (2018) Waterbird community composition, abundance, and diversity along an urban gradient. Landsc Urban Plann 170:103–111
Avilova KV (2016) The Life Cycle and Number Dynamics of the Urban Mallard Population (Anas platyrhynchos, Anseriformes, Aves) in Moscow. Biology Bull 43:1212–1224
Bara M, Segura LN (2019) Effect of Air Temperature and Water Depth on Bird Abundance: A Case Study of Rallidae and Anatidae in the Northeastern Algerian Garaet Hadj Tahar. Pakistan J Zool 51:211–217
Baratti M, Cordaro M, Dessi-Fulgheri F, Vannini M, Fratini S (2009) Molecular and ecological characterization of urban populations of the mallard (Anas platyrhynchos L.) in Italy. Italian J Zool 76:330–339
Barznji DAM (2014) Role of aquatic plants in improving water quality. Unique J Pharm Biol Sci 02:12–16
Bro R, Smilde AK (2014) Principal component analysis. Anal Methods 6:2812–2831
Brown J (2012) Protecting St. Louis: The River des Peres Sewerage and Drainage Works. Civil engineering (New York, N.Y. 1983) 82:44–45
Burnham KP, Anderson DR, Huyvaert KP (2011) AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons. Behav Ecol Sociobiol 65:23–35
Cadotte MW, Carscadden K, Mirotchnick N (2011) Beyond species: functional diversity and the maintenance of ecological processes and services. J Appl Ecol 48:1079–1087
Callaghan CT, Poore AGB, Hofmann M, Roberts CJ, Pereira HM (2021) Large-bodied birds are over-represented in unstructured citizen science data. Sci Rep 11:1–11
Carr GM, Duthie HC, Taylor WD (1997) Models of aquatic plant productivity: a review of the factors that influence growth. Aquat Bot 59:195–215
Chang HY, Lee YF (2016) Effects of area size, heterogeneity, isolation, and disturbances on urban park avifauna in a highly populated tropical city. Urban Ecosyst 19:257–274
Chesser RT, Billerman SM, Burns KJ, Cicero C, Dunn JL, Hernández-Baños BE, Jiménez RA, Kratter AW, Mason NA, Rasmussen PC, Remsen JV, Winker K (2023) Sixty-fourth Supplement to the American Ornithological Society's Check-list of North American Birds. Ornithology 140:1–11
Colwell MA, Taft OW (2000) Waterbird communities in managed wetlands of varying water depth. Waterbirds 23:45–55
Corrêa LMC, Melo C (2013) Diversity, abundance and activity of avifauna in urban lakes of Uberlândia, Minas Gerais, Brazil: the influence of water quality and physical degradation of the surrounding area. Int J Sci Commer Humanit 1:88–98
Davidson NC (2014) How much wetland has the world lost? Long-term and recent trends in global wetland area. Mar Freshw Res 65:934–941
Dickinson JL, Zuckerberg B, Bonter DN (2010) Citizen Science as an Ecological Research Tool: Challenges and Benefits. Annual Review of Ecology, Evolution, and Systematics, Vol 41 41:149–172
Donnelly JP, Moore JN, Casazza ML, Coons SP (2022) Functional Wetland Loss Drives Emerging Risks to Waterbird Migration Networks. Front Ecol Evol 10:1–18
Duan HL, Xia SX, Yu XB, Liu Y, Teng JK, Dou YH (2020) Using citizen science data to inform the relative sensitivity of waterbirds to natural versus human-dominated landscapes in China. Ecol Evol 10:7233–7241
Fernández-Juricic E, Jokimäki J (2001) A habitat island approach to conserving birds in urban landscapes: case studies from southern and northern Europe. Biodivers Conserv 10:2023–2043
Fontana S, Sattler T, Bontadina F, Moretti M (2011) How to manage the urban green to improve bird diversity and community structure. Landsc Urban Plann 101:278–285
Forest Park F (2020) Forest Park Natural Resources Management Plan. Forest Park Forever
Gutzwiller KJ, Flather CH (2011) Wetland features and landscape context predict the risk of wetland habitat loss. Ecol Appl 21:968–982
Hallmann CA, Foppen RPB, van Turnhout CAM, de Kroon H, Jongejans E (2014) Declines in insectivorous birds are associated with high neonicotinoid concentrations. Nature 511:341–343
Homayoun TZ, Blair RB (2016) Value of park reserves to migrating and breeding landbirds in an urban important bird area. Urban Ecosyst 19:1579–1596
Huang PL, Zheng DL, Yan YJ, Xu WZ, Zhao YJ, Huang ZL, Ding YH, Lin YX, Zhu ZP, Chen ZR, Fu WC (2022) Effects of Landscape Features on Bird Community in Winter Urban Parks. Animals 12:1–15
Jensen WE (2007) Missouri Waterbirds and Working Lands. National Audubon Society, Inc
Jost L (2006) Entropy and diversity. Oikos 113:363–375
Koleff P, Gaston KJ, Lennon JJ (2003) Measuring beta diversity for presence-absence data. J Anim Ecol 72:367–382
Laureto LMO, Cianciaruso MV, Samia DSM (2015) Functional diversity: an overview of its history and applicability. Natureza & Conservacao 13:112–116
Legendre P, Borcard D, Peres-Neto PR (2005) Analyzing beta diversity: Partitioning the spatial variation of community composition data. Ecol Monogr 75:435–450
Li XZ, Zeng Q, Lei GC, Sun GQ (2022) Effects of Meteorological Factors on Waterbird Functional Diversity and Community Composition in Liaohe Estuary, China. Int J Environ Res Public Health 19:1–19
Lowry H, Lill A, Wong BBM (2013) Behavioural responses of wildlife to urban environments. Biol Rev 88:537–549
Lundholm JT, Richardson PJ (2010) Habitat analogues for reconciliation ecology in urban and industrial environments. J Appl Ecol 47:966–975
Ma ZJ, Cai YT, Li B, Chen JK (2010) Managing Wetland Habitats for Waterbirds: An International Perspective. Wetlands 30:15–27
Maimaitijiang M, Ghulam A, Sandoval JSO, Maimaitiyiming M (2015) Drivers of land cover and land use changes in St. Louis metropolitan area over the past 40 years characterized by remote sensing and census population data. Int J Appl Earth Obs Geoinf 35:161–174
McKinney RA, McWilliams SR, Charpentier MA (2006) Waterfowl-habitat associations during winter in an urban North Atlantic estuary. Biol Conserv 132:239–249
Meissner W, Rowinski P, Polakowski M, Wilniewczyc P, Marchowski D (2015) Impact of temperature on the number of mallards, Anas platyrhynchos, wintering in cities. North-Western J Zool 11:213–218
Miotto ML, Gusmao JB, Domit C, Di Domenico M (2023) Environmental drivers of waterbird diversity in a world heritage subtropical estuarine system. Estuar Coastal Shelf Sci 288:1–10
Missouri Department of Conservation (2022) Missouri Species and Communities of Conservation Concern. SCI077
Muñoz-Pedreros A, Dellacasa V (2023) Diversity of diurnal raptors in a suburban area of a city in southern Chile. Urban Ecosyst 26:767–777
Oliver AJ, Hong-Wa C, Devonshire J, Olea KR, Rivas GF, Gahl MK (2011) Avifauna richness enhanced in large, isolated urban parks. Landsc Urban Plann 102:215–225
Omernik JM, Griffith GE (2014) Ecoregions of the Conterminous United States: Evolution of a Hierarchical Spatial Framework. Environ Manage 54:1249–1266
Rahlin AA, Saunders SP, Beilke S (2022) Spatial drivers of wetland bird occupancy within an urbanized matrix in the Upper Midwestern United States. Ecosphere 13:1–14
Randler C, Staller N, Kalb N, Tryjanowski P (2023) Charismatic Species and Birdwatching: Advanced Birders Prefer Small, Shy, Dull, and Rare Species. Anthrozoos 36:427–445
Reynolds C, Cumming GS (2016) Defining functional groups using dietary data: Quantitative comparison suggests functional classification for seed-dispersing waterfowl. Basic Appl Ecol 17:333–343
Riegner MF, Bassar RD (2023) Morphological Covariance and Onset of Foot Prehensility as Indicators of Integrated Evolutionary Dynamics in the Herons (Ardeidae). Integr Organismal Biology 5:1–19
Rosenberg KV, Dokter AM, Blancher PJ, Sauer JR, Smith AC, Smith PA, Stanton JC, Panjabi A, Helft L, Parr M, Marra PP (2019) Decline of the North American avifauna. Science 366:120–125
Rosenzweig ML (2003) Reconciliation ecology and the future of species diversity. Oryx 37:194–205
Roswell M, Dushoff J, Winfree R (2021) A conceptual guide to measuring species diversity. Oikos 130:321–338
Singh N, Price C, Downs CT (2021) Aspects of the ecology and behaviour of a potential urban exploiter, the southern tree agama, Acanthocercus atricollis. Urban Ecosyst 24:905–914
Snep RPH, Kooijmans JL, Kwak RGM, Foppen RPB, Parsons H, Awasthy M, Sierdsema HLK, Marzluff JM, Fernandez-Juricic E, de Laet J (2016) and Y. M. van Heezik. Urban bird conservation: presenting stakeholder-specific arguments for the development of bird-friendly cities. Urban Ecosystems 19:1535–1550
Sofaer HR, Skagen SK, Barsugli JJ, Rashford BS, Reese GC, Hoeting JA, Wood AW and B. R. Noon. 2016. Projected wetland densities under climate change: habitat loss but little geographic shift in conservation strategy. Ecol Appl 26:1677–1692
Sol D, Trisos C, Murria C, Jeliazkov A, Gonzalez-Lagos C, Pigot AL, Ricotta C, Swan CM, Tobias JA, Pavoine S (2020) The worldwide impact of urbanisation on avian functional diversity. Ecol Lett 23:962–972
Sorace A, Gustin M (2010) Bird species of conservation concern along urban gradients in Italy. Biodivers Conserv 19:205–221
Stanton RL, Morrissey CA, Clark RG (2018) Analysis of trends and agricultural drivers of farmland bird declines in North America: A review. Agric Ecosyst Environ 254:244–254
Stoudt S, Goldstein BR, de Valpine P (2022) Identifying engaging bird species and traits with community science observations. Proc Natl Acad Sci USA 119:1–7
Stroud JT, Bush MR, Ladd MC, Nowicki RJ, Shantz AA, Sweatman J (2015) Is a community still a community? Reviewing definitions of key terms in community ecology. Ecol Evol 5:4757–4765
Sullivan BL, Wood CL, Iliff MJ, Bonney RE, Fink D, Kelling S (2009) eBird: A citizen-based bird observation network in the biological sciences. Biol Conserv 142:2282–2292
Tang B, Clark JS, Gelfand AE (2021) Modeling spatially biased citizen science effort through the eBird database. Environ Ecol Stat 28:609–630
Census Bureau US (2020) RACE.in T. P1, editor., Decennial Census, DEC Redistricting Data (PL 94–171)
Wells JV, Niven DK, Cecil J (2005) The Important Bird Areas Program in the United States: Building a Network of Sites for Conservation, State by State. PSW-GTR-191. USDA Forest Service
Wetlands International (2010) Guidance on waterbird monitoring methodology: Field Protocol for waterbird counting. Wetlands International
Wolf JM, Jeschke JM, Voigt CC, Itescu Y (2022) Urban affinity and its associated traits: A global analysis of bats. Glob Change Biol 28:5667–5682
Xu Q, Zhou L, Xia S, Zhou J (2022) Impact of Urbanisation Intensity on Bird Diversity in River Wetlands around Chaohu Lake, China. Anim (Basel) 12:1–14
Yang XR, Tan XW, Chen CW, Wang YP (2020) The influence of urban park characteristics on bird diversity in Nanjing, China. Avian Res 11:1–9
Zedler JB (2004) Compensating for wetland losses in the United States. Ibis 146:92–100

No competing interests reported.

SupplementaryMaterials.docx

Download PDF

Journal Publication

published 03 Jul, 2024

Read the published version in Urban Ecosystems →

Editorial decision: Revision requested
10 Feb, 2024
Reviews received at journal
05 Feb, 2024
Reviewers agreed at journal
13 Jan, 2024
Reviewers invited by journal
13 Jan, 2024
Editor assigned by journal
12 Jan, 2024
Submission checks completed at journal
12 Jan, 2024
First submitted to journal
22 Dec, 2023

You are reading this latest preprint version

A Deep Dive into the Waterbird Community of an Urban Oasis: Implications for Park Management

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Study Site and Methods

Study Site

Methods

Point Count Surveys

eBird

Definition of Functional Groups

Sampling of Environmental Conditions

Weather

Site-level Environmental Factors

Prey relative abundance

Water Quality

Bathymetry

Waterbody Surface Area and Connectivity

Statistical Analyses

Results

Spatial and Temporal Patterns of Species Richness, Gamma, and Alpha Species Diversity

Community Composition over Time and Space (Beta Diversity)

The Influence of Environmental Factors on Patterns of Species Diversity and Community Composition

Discussion

Gamma diversity and temporal variations in species presence

Environmental Drivers of Diversity

Beta Diversity

Alpha Diversity

Implications for Management and Conservation of Waterbirds

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1