Tree cavity availability in urban cemeteries and city parks

Abstract

Tree cavities provide shelter, nesting sites and food storage for many species of birds, mammals and insects. While tree cavities are present in a variety of habitats, most prior research focuses on forests, with fewer studies completed in urban areas. However, city parks provide some habitat, and cemeteries may also provide adequate, or even better, habitat for cavity-nesters, especially in urban areas where cavities are a limiting resource. While city park and cemetery habitats have similarities, including lawn and tree maintenance, differences in tree characteristics resulting from management practices may impact tree cavity prevalence and characteristics. Our objective was to determine if parks and cemeteries were comparable habitat types in terms of tree cavity availability. We sampled 1007 trees across 10 cemeteries and 10 large parks for excavated and natural cavities throughout Cook County, IL, USA. We found that while there was no significant difference in natural cavities per tree in parks versus cemeteries, there were 3.4 times as many woodpecker-excavated cavities per 50 trees sampled in cemeteries and there was high variance in the number of excavated cavities in cemeteries. Trees in cemeteries tended to be larger and more decayed than those in parks. Trees with cavities (natural or excavated) were generally larger and more decayed than trees without cavities. This research shows the potential for cemetery and park management to promote cavity excavators and nesters.

Introduction

Urbanization has a wide range of effects on wildlife. In urban areas, negative effects on wildlife biodiversity and abundance are often attributed to the direct loss of available habitat (Chace and Walsh 2006; Evans, Newson, and Gaston 2009; Meffert and Dziock 2013), exclusion of species sensitive to human disturbance in areas with high anthropogenic modification despite the presence of remaining habitat (Batalha, Ramos, and Cardoso 2013; Charter, Leshem, and Izhaki 2013), and increased competition from non-native species (Bennett 1990; Koch, Martin, and Aitken 2012; Charter, Leshem, and Izhaki 2013). Urban habitats can also be favorable for wildlife species that can exploit resources provided in the modified habitat (Chace and Walsh 2006). Among the variety of habitat types occurring within cities, parks are thought of as biodiversity hotspots, with greater richness of bird species in larger parks, and tree-lined streets provide increased connectivity and additional habitat (Fernández-Juricic and Jokimäki 2001).

Tree cavities provide shelter and essential resources to not only diverse groups of birds (Newton 1998; Smith, Withgott, and Rodewald 2000) but also many species of mammals, insects and other organisms (Ranius and Jansson 2000; Gibbons and Lindenmayer 2002; Carpaneto et al. 2010; Bryant, Dundas, and Fleming 2012; Hennessey, Dubach, and Gehrt 2012; Fleming et al. 2013). Over 89 animal species in North America use tree cavities for roosting, nesting, breeding or food storage (Titus 1983). Tree cavities can form via the decay of wood by natural age and rot, where branch-breaks or cracks followed by colonization by fungi and then physical or insect damage form openings into the hollowed wood (‘natural cavities’), or by fungal colonization followed by the removal of softened heartwood by woodpeckers and other primary excavators (‘excavated cavities’) (Smith, Withgott, and Rodewald 2000; Gibbons and Lindenmayer 2002; Blewett and Marzluff 2005; Remm, Lõhmus, and Remm 2006). Primary excavators include both strong excavators that include most species of woodpeckers [e.g. yellow-bellied sapsucker (Sphyrapicus varius), northern flicker (Colaptes auratus)] that are well-adapted to creating holes in trees (Kilham 1971; Harestad and Keisker 1989) and weak excavators that tend to be smaller birds less well adapted to excavation (e.g. chickadees, nuthatches; Bunnell 2013).

Cavity-nesting birds depend on both living trees with dead or decaying sections, as well as dead trees and snags (Martin, Aitken, and Wiebe 2004; Anderson and LaMontagne 2014) because the soft wood substrate is easy to excavate and prone to forming cavities by natural wood rot (Bender et al. 2016). Cavities can be durable, remaining accessible to animals for long periods of time (Cockle, Martin, and Wesołowski 2011; Edworthy, Wiebe, and Martin 2012; Anderson and LaMontagne 2014). However, in cities, trees are managed to minimize liability related to the danger to citizens by removing cracked, broken or dead limbs and decaying trees at risk of falling (Terho and Hallaksela 2008). Because these limbs are often removed in urban areas for public safety and protection of property, habitat space for urban cavity nesters may be limited, and as a result trees and cavities are predominantly located in designated green spaces within cities. Despite cavity longevity in natural areas, snags (standing dead trees) are rare in urban landscapes and the removal of dead and decaying trees and individual tree limbs, negatively impact woodpeckers (Kilham 1971; Blewett and Marzluff 2005; Koenig, Walters, and Rodewald 2017). Thus, the overall abundance of cavity nesters in urban areas is decreased as competition for the limited number of cavities increases (Blewett and Marzluff 2005). With a decline in populations of primary cavity excavators that produce a supply of tree cavities, urban wildlife may experience increased competition for cavity sites (Jokimäki 1999; Davis, Major, and Taylor 2013; Remacha and Delgado 2009), which may reduce less dominant species and decrease overall biodiversity (Dodaro and Battisti 2014).

In Chicago, excavated tree cavities are most abundant in forest preserves, followed by large city parks, and least abundant in residential areas (LaMontagne et al. 2015). Many cities do not have large areas of forest within their boundaries, but other land-use types can support urban wildlife including residential areas, city parks, cemeteries and golf courses (Adams 2005; Gallo et al. 2017). Here, we compare cavity availability and cavity characteristics between urban cemeteries and large city parks. In urban areas, both cemeteries and large parks tend to have open greenspaces with manicured areas and lawns with grass cover, trees and sometimes freshwater ponds. Large parks and cemeteries may have similar levels of total plant cover and diversity of plant types (Livingston, Shaw, and Harris 2003), and both can contain high biodiversity (Cornelis and Hermy 2004; Kowarik et al. 2016). However, differences in land management of cemeteries and parks influence habitat structure, for instance, many historic cemeteries contain ‘champion trees’, massive old trees (Barrett and Barrett 2001). Given the presence of older trees (Gilbert 1989), and lower levels of human activity, cemeteries have the potential to be wildlife refuges, particularly for birds (Pearson 1915; Lussenhop 1977; Wexler 2008).

We examined the question: How does tree cavity availability and the characteristics of cavities and trees compare between cemeteries and large parks? We addressed this question in three parts: (1) the abundance of cavities (including different cavity types, excavated or natural) in each habitat type, (2) characteristics of each cavity type (cavity size, location of cavities on trees, orientation of cavity openings in the cardinal direction and height of cavity opening from the ground) and (3) tree-level characteristics [diameter at breast height (dbh), decay class and evidence of fungus] between the two habitat types, and between cavity types. We hypothesized that there may be differences in overall tree cavity availability between urban cemeteries and city parks, reflecting management practices for dead branches and trees; we predicted that urban cemeteries will have a greater overall number of both natural and excavated cavities due to higher tree decay. Trees containing cavities tend to be more decayed than trees without cavities (LaMontagne et al. 2015; Anderson and LaMontagne 2016), and primary excavators select trees that are live and unhealthy or dead trees (Smith, Withgott, and Rodewald 2000; Blewett and Marzluff 2005; Blanc and Martin 2012). We predicted differences in tree characteristics between urban cemeteries and city parks, with more decayed trees in cemeteries, due to differences in maintenance. Management techniques in city parks often result in the removal of dead, diseased or otherwise decayed wood (Carpaneto et al. 2010). We therefore predicted that more excavated cavities would be found on trees of a higher decay class, that trees in city parks would be smaller and less decayed than trees in urban cemeteries, and that there would be fewer cavities per tree in city parks. We also predicted that excavated cavities would be located at greater heights on the tree than natural cavities, due to the selection of this niche space by excavators (e.g. woodpeckers and other piciiformes) and the potential ability to avoid predators at greater heights (Li and Martin 1991; LaMontagne et al. 2015). Additionally, natural cavities can be lower on trees due to the lower sections being older and larger, providing more time and space for the natural decay process to occur (Aitken and Martin 2007).

Methods

Study area

We compared the characteristics of trees and the availability of natural and excavated cavities in cemeteries and large city parks in and around the city of Chicago, IL, USA (41°53′N, 87°38′W; Fig. 1). This area was historically dominated by oak savannahs that were categorized by open grassland expanses and stands of trees, providing habitat for a wide variety of cavity nesting species (Lanham et al. 2002). Both cemeteries and parks used in the study were developed between the late-1800s and the mid-1900s, after the Great Chicago Fire of 1871.

Large city parks (minimum 23 ha) were characterized by large grassy areas and numerous trees, distributed across neighborhoods in Chicago, and managed by the City of Chicago (n = 10, mean area = 134.31 ha; data from LaMontagne et al. 2015). This size is similar to that of cemeteries included in the study and eliminated smaller neighborhood parks, or ‘pocket parks’, that support few species (Morrison and Chapman 2005; LaMontagne et al. 2015). The primary species of trees in Chicago city parks include Silver Maple (Acer saccharinum), Box Elder (Acer negundo) and Slippery Elm (Ulmus rubra; Anderson and LaMontagne 2016). Cemeteries (n = 10, mean area = 53.77 ha) were either managed privately (1 of 10 cemeteries), managed by a company responsible for multiple cemeteries (2 of 10 cemeteries) or by the Archdiocese of Chicago (7 of 10 cemeteries). All cemeteries had mowed grassy lawns with trees distributed in a relatively uniform pattern within the cemetery (Bovyn, pers. comm.). Tree species in cemeteries included predominantly maples (Acer spp.), oaks (Quercus spp.) and various other species, and one cemetery (Graceland Cemetery) is a certified arboretum, with high species diversity.

Tree cavity sampling methods

To enable the highest level of tree cavity visibility possible, we assessed tree cavities in cemeteries in winter 2014 and early spring 2015 before leafing-out (following LaMontagne et al. 2015); data for parks were collected in winter-spring 2012–2013. In both habitats, we included only trees with a dbh > 12 cm, because this is the minimum dbh used by black-capped chickadees (Poecile atricapillus), one of the smallest cavity nesting birds found in our study area (Brewer 1963). We selected an area for sampling within cemeteries or parks a priori using Google Earth images, with the goal of having an area with at least 50 trees to sample for each site. In large parks, we sampled areas with trees and excluded areas of open fields, areas designated for athletics (e.g. baseball diamonds) and all trees were at least 4 m from the edge of roads. Surveyed by observers on the ground and using binoculars, 50 trees were sampled at each site. The first, last and every 10th tree without any cavities were assessed as ‘control trees’. We defined cavities as holes that penetrated into the heartwood of a tree without any visible interior wood blocking the hole (to differentiate them from excavations from feeding or shallow holes), where ‘excavated’ cavities were created by a woodpecker pecking into the tree bark, forming a hole of a consistent circular shape, and a ‘natural cavity’ was formed by a broken branch with a scar forming around the opening or a crack that formed a tree hollow with a more irregular shape than an excavated cavity. Observers were trained together, and sampling occurred in pairs or as a group of three observers to avoid biases.

For each cemetery and city park, we recorded the total number of sampled trees with cavities. For trees containing one or more cavities, we recorded the type of cavity (natural or excavated), cavity opening size [based on the largest dimension; small (3 cm), medium (3–8 cm), large (8–40 cm) and extra-large (>40 cm)], location (trunk, primary branch, secondary branch, tertiary branch), orientation (in eight categories based on cardinal directions) and height (meters from ground) for each cavity. Cavity size and height were estimated by the observer from the ground. For both control trees (that had no cavities) and cavity trees, we measured tree size (dbh), the presence or absence of visible fungal growths and level of decay. We recorded the level of decay based on a modified Maser et al. (1979) scale ranging from 1 to 5 (1, healthy and the crown area is 100% alive; 2, the crown is 1–49% dead; 3, crown is 50–99% dead; 4, the crown is dead but the branches are still intact; 5, snag with a broken top; LaMontagne et al. 2015).

Statistical analyses

We analyzed the mean number of cavities per 50 trees sampled between habitat types using two-sample t-tests; one t-test was performed for natural cavities and another one was performed for excavated cavities. For excavated cavities, analysis was done on ln-transformed values in order to meet the assumption of equal variance between groups. We compared the number of excavated and natural cavities across habitat types using chi-squared tests. We tested for a difference in abundance of cavity-bearing trees (per 50 trees) between cemeteries and city parks using two-sample t-tests; one t-test for natural-cavity trees and another t-test for excavated-cavity trees. The influence of habitat type and, separately, cavity type on cavity opening size (small, medium, large or extra-large), location (trunk, primary branch, secondary branch, etc.) and orientation (eight cardinal directions) was tested using chi-squared tests. We determined the effects of habitat type, cavity type and their interaction on tree cavity height by conducting a two-way ANOVA. To assess differences in the trees that contain cavities, we used a two-way ANOVA to compare the diameter of trees between cavity types [no-cavity (controls), excavated-cavity tree and natural-cavity tree] and habitats; a tree was identified as an ‘excavated-cavity tree’ if at least one excavated cavity was present (LaMontagne et al. 2015). To compare size distributions of control, excavated and natural cavity trees between cemeteries and city parks, we used two sample Kolmogorov–Smirnov tests, because the distributions were not normally distributed. We tested for a difference in distribution of cavity-tree decay classes between parks and cemeteries using a chi-square test. All statistical analyses were conducted in R version 2.15.2 (R Core Team 2015).

Results

Cavity availability and characteristics

We surveyed a total of 1007 trees for cavities across the two urban habitat types (506 in parks and 501 in cemeteries). We found a similar number of total cavities in both habitats (183 in parks and 196 in cemeteries) and there was no significant difference between habitat types for the mean number of natural cavities, nor for the mean number of natural-cavity trees (Table 1). There was high variation in the number of excavated cavities across cemeteries, ranging from 0 up to 16 cavities per site, and notably, there were on average 3.4 times more excavated cavities in cemeteries than in city parks, but comparisons of the ln-number of excavated cavities was not statistically significant (Table 1; note that the variability in the data for cemeteries was large). There was no significant difference in the number of trees with at least one excavated-cavity, despite the trend for more cavity trees in cemeteries (Table 1). On average, 22.8% of trees had at least one cavity in cemeteries, and this was similar in parks, with 21.5% of trees having at least one cavity.

Cavities were mostly medium-sized (3–8 cm diameter), and between habitat types there was no significant difference in the size categories of all the cavities pooled together (χ² = 4.778, df = 3, P = 0.189). When pooled across cemeteries and parks, excavated cavities were significantly smaller than natural cavities (χ² = 10.51, df = 3, P = 0.015; Fig. 2a). There was also no significant difference between habitat types for cavity location (χ² = 3.556, df = 3, P = 0.314) in both habitats, tree cavities were mostly located on primary branches. Excavated cavities were mostly present on primary and secondary branches, where natural cavities which were predominantly on tree trunks and primary branches (χ² = 12.40, df = 3, P = 0.006; Fig. 2b). Mean excavated cavity height was 8.30 ± 2.65 m from the ground, significantly higher than natural cavities at 5.60 ± 2.79 m (F = 43.01, df = 1, P < 0.001; Fig. 3). There was not a significant difference in mean cavity height due to habitat type (F = 2.039, df = 1, P = 0.154), nor was the two-way interaction between habitat type and cavity type significant (F = 0.016, df = 1, P = 0.899; Fig. 3). There was no significant difference in cavity opening orientation between habitat types (χ² = 6.353, df = 7, P = 0.499) or cavity type (χ² = 9.483, df = 7, P = 0.220; Fig. 2c and d).

Tree characteristics

The size distribution of all trees sampled ranged from 13.1 to 145.4 cm dbh, with a mean dbh ± standard deviation of 59.87 ± 22.03 cm. Trees were larger in cemeteries than in parks (cemetery mean dbh = 66.01 ± 21.19 cm; park mean dbh = 53.48 ± 21.12 cm; t = –5.469, P < 0.001). The size distribution of control (noncavity) trees was significantly different between habitat types, and shifted toward larger sizes in cemeteries (D = 0.410, P < 0.001; Fig. 4a and b). The size distribution of cavity trees was shifted toward larger trees compared to control trees, and the size distributions of natural-cavity trees differed between habitat types, with a wider range of values overall and a shift toward more larger trees in cemeteries (D = 0.255, P = 0.005; Fig. 4c and d). The range of sizes of excavated-cavity trees were similar in size to natural-cavity trees, and size distribution did not differ significantly between cemeteries and parks (D = 0.425, P = 0.167; Fig. 4e and f).

All trees surveyed in both habitats were living (i.e. there were no decay class 4 or class 5 trees). Trees with higher decay were found in cemeteries more than parks, and only decay classes 1 and 2 were found in city parks (χ² = 44.97, df = 2, P < 0.001; Fig. 5). Control (noncavity) trees in parks were the least decayed overall (Fig. 5b), while excavated-cavity trees in cemeteries were the most decayed (Fig. 5e). Decay class was not related to cavity type (χ² = 4.128, df = 2, P = 0.127), and most cavities in general were found in decay class 2 trees. Overall, larger trees were more decayed (r_s = 0.28, P < 0.001). In cemeteries, the prevalence of trees with fungus increased with decay class rank (8.7% of decay class 1 trees, 13.2% of decay class 2 trees and 21.4% of decay class 3 trees had fungus present), while a lower percentage of trees in city parks had visible fungus (7.0% of class 1 trees and 5.15% of class 2 trees had fungus present).

Discussion

We compared tree cavity availability between two common urban habitat green spaces, city parks and cemeteries, and found that on average there were 1.09 excavated cavities per 50 trees in city parks compared to 3.70 in cemeteries, a difference of 3.4×. However, comparisons of these means were not statistically significant, due to the much higher variability in the number of excavated cavities, and in tree decay, across cemeteries compared to that in city parks. This likely reflects the variability in cemetery tree maintenance and management; seven out of the ten cemetery sites were maintained by the Archdiocese of Chicago. Generally, these cemeteries had less tree decay, fewer excavated cavities (as well as natural cavities) and fewer woodpecker sightings than the other cemeteries, suggesting that differences in tree care and maintenance among cemeteries impact tree cavity availability. From casual observations during our site assessments, we saw more woodpeckers in cemeteries compared to park sites, including hairy (Leuconotopicus villosus), downy (Picoides pubescens) and red-bellied (Melanerpes carolinus) woodpeckers. Therefore, levels of tree maintenance in cemeteries and urban parks may represent different levels of habitat quality for wildlife, and in particular for tree cavity users who rely on dead branches and trees. The trend for more excavated cavities in cemeteries is despite similarities in cemeteries and city parks for the total number of tree cavities and the characteristics of tree cavities.

Consistent with our results, other studies have found that larger trees tend to be more highly decayed (Remm, Lõhmus, and Remm 2006), and we found that larger trees were more likely to have cavities. In addition to birds, mammals and insects tend to prefer large, decayed or damaged trees for nesting (Ranius 2002; Weir, Phinney, and Lofroth 2012). Overall, more fungi is found on whole, dead trees than smaller coarse wood debris (Heilmann-Clausen and Christensen 2004). Decayed wood is softer and preferred by woodpeckers and other cavity excavators (Bull 2002; Anderson and LaMontagne 2016), supporting that more decayed trees in cemeteries would lead to more excavated cavities. Most excavated cavities were found higher in trees and on primary and secondary branches, consistent with previous research on primary cavity nesters in this region (LaMontagne et al. 2015); however, there is variation in cavity heights depending on the species of excavator (Martin, Aitken, and Wiebe 2004). Cavity-nesters tend to prefer cavities high above ground, where they are less prone to predation (Nilsson 1984; Carlson, Sandstroem, and Olsson 1998). The availability of large trees with some level of decay may influence the attractiveness of an area to cavity excavators, and thereby the prevalence of excavated tree cavities.

Habitat quality is also influenced by intraspecific relationships and human activity. Levels of human activity differ between cemeteries and parks, with increased activity in parks (e.g. sporting events, picnics, walking dogs, etc.), compared to cemeteries that are more quiet. Human recreational activities have been shown to negatively influence breeding success of birds in urban parks (reviewed in Jokimäki 1999), and birds in cemeteries have been found to have decreased escape behavior toward human activity compared to birds in parks (Morelli et al. 2018). Recently a study on mammal communities in the Chicago region found that species diversity was higher in cemeteries than city parks (Gallo et al. 2017), which may suggest that cemeteries provide better habitat for animals than parks. Cemeteries are often enclosed in fences, which may exclude some predators found in large, open parks, and more species of predatory mammals have been observed in parks than cemeteries (Zorenko and Leontyeva 2003); however, the Gallo et al. (2017) study found more species richness in cemeteries. More studies on the use of habitats in urban areas by animal communities are needed to understand the potential role of predation on habitat use by woodpeckers.

In this study we observed more excavated cavities in cemeteries, suggesting that cemeteries have the potential to be refuges for primary and secondary cavity nesters in urban environments. Cavities created by primary excavators have impacts on other species as well, as they are also used by other species of birds, mammals and insects for a period of time after their creation (Martin, Aitken, and Wiebe 2004). Although more excavator activity was observed in cemeteries, there was no significant difference in total cavity number when natural cavities were included, because relative to natural cavities, excavated cavities are rare. A number of secondary cavity nesters rely on natural cavities which can be much larger than excavated cavities that are limited by the size and energy of the excavator. Natural cavities may be even more important than excavated cavities because they are more prevalent and persistent than excavated cavities (Cockle, Martin, and Bodrati 2017), as, in our study, excavated cavities were most likely found on decayed branches that fall down quickly or are removed according to best management practices compared to natural cavities that were found mostly on trunks. Similarly, Anderson and LaMontagne (2014) found that trees with lower levels of decay had excavated cavities that were used by red-headed woodpeckers for more years than trees with higher levels of decay. There was no significant difference in the mean number of natural cavities between habitat types, but there was still a large amount of variation between individual cemeteries. Preserving and promoting the development of both natural and excavated cavities should be a priority for conservation for urban habitat maintenance. The retention of large trees is crucial and if prioritized in modified landscapes, such as urban areas, may promote population sizes of cavity-nesting birds and other species that use tree cavities. While artificial nest-boxes are often used as supplementary nesting sites for many species of birds (Jokimäki 1999), the attachment of artificial nest boxes to smaller trees cannot mimic the overall value of larger trees (Le Roux et al. 2016).

A caveat of note when discussing differences between cavity types and habitat types is the influence of tree species, which was not recorded in our study. Trees were sampled in the winter and early spring months when leaves were absent in order to increase visibility of cavities by observers on the ground. We were unable to identify trees with confidence by bark and buds alone; however, all of the trees sampled were deciduous. Further research could be done to compare tree species between different urban land use areas. Additionally, because cavities were identified by observers on the ground with binoculars, it was not noted whether cavities contained nests previously, were used for other purposes or were unused.

This research promotes the recognition of green spaces like cemeteries, and the potential for green spaces to be managed for wildlife (Fernández-Juricic and Jokimäki 2001; Munshi-South 2012; Kowarik et al. 2016). Maintenance in parks and cemeteries could be encouraged to promote the management of trees for primary and secondary cavity nesters, e.g. selecting less branches to trim and/or snags to remove while continuing to uphold public safety standards. Our study adds to the growing body of knowledge on habitat availability for cavity-dwelling species in urban areas.

Funding

We thank the DePaul University College of Science and Health and the Department of Biological Sciences for funding this research.

Acknowledgments

We thank H. Anderson and K. Morris for their contributions to data collection and analysis, and to J. Hacker and M. Wade for constructive comments for this paper and presentations of this research. Thank you to the Chicago Park District and to those who manage the cemeteries used in this study: the Archdiocese of Chicago, Concordia Association, Oak Woods Cemetery Association and the Trustees of the Graceland Cemetery Improvement Fund.

References