Tracking Ground Arthropod Diversity in Urban Forests: Lessons Learned From a New Undergraduate Long-Term Project
Brian Alfaro1*, Riley C. Porter1, Tyler G. Foote1, Mercedes J. Lohmann1, Caroline Herring1, Kaitlyn E. Blankley1, Julianne N. Anemone1, Kayla Madden1, David W. Unander1, and Rachael E. Alfaro1
1Biology Department, Eastern University, 308 McInnis Hall, 1300 Eagle Road, St. Davids, PA 19087- 3696. *Corresponding author: brian.alfaro@eastern.edu, (610) 225-5564.
Urban Naturalist, No. 74 (2024)
Abstract
The greater Philadelphia area contains a mosaic of forest fragments that can be used to track the health of urban ecosystems in educational settings. Here, we used ground and leaf litter dwelling arthropods as ecological indicators to assess the diversity of leaf litter communities in an old growth forest and a secondary regrowth forest. In this preliminary study that we conducted with a small team of undergraduate students, we found that the old growth forest samples contained more abundant and diverse arthropods than the secondary growth samples. We assert that increased arthropod diversity in the old growth forest is due to age and density of the foliage of the canopy. Conversely, the reduced diversity of arthropods in the secondary regrowth forest is likely due to clearing of the site almost 50 years ago, which have allowed early successional tree species to form monocultures that house fewer arthropod taxa. While our study shows that ground and leaf litter arthropod diversity in forest fragments in St. Davids, Pennsylvania can be sensitive to human disturbance, we need to improve the timing, frequency, spatial scale, and taxonomic resolution of our sampling. In this brief communication, we reflect on our initial results, and describe future directions for this long-term ecological research project that we designed for our undergraduate biology courses.
Download Full-text pdf
Site by Bennett Web & Design Co.
Urban Naturalist
B. Alfaro et al.
2024 No. 74
1
2024 Urban Naturalist 74:1–10
Tracking Ground Arthropod Diversity in Urban Forests:
Lessons Learned From a New Undergraduate Long-Term
Project
Brian Alfaro1*, Riley C. Porter1, Tyler G. Foote1, Mercedes J. Lohmann1,
Caroline Herring1, Kaitlyn E. Blankley1, Julianne N. Anemone1, Kayla Madden1,
David W. Unander1, and Rachael E. Alfaro1
Abstract - The greater Philadelphia area contains a mosaic of forest fragments that can be used to track
the health of urban ecosystems in educational settings. Here, we used ground and leaf litter dwelling
arthropods as ecological indicators to assess the diversity of leaf litter communities in an old growth
forest and a secondary regrowth forest. In this preliminary study that we conducted with a small team
of undergraduate students, we found that the old growth forest samples contained more abundant and
diverse arthropods than the secondary growth samples. We assert that increased arthropod diversity in
the old growth forest is due to age and density of the foliage of the canopy. Conversely, the reduced
diversity of arthropods in the secondary regrowth forest is likely due to clearing of the site almost 50
years ago, which have allowed early successional tree species to form monocultures that house fewer
arthropod taxa. While our study shows that ground and leaf litter arthropod diversity in forest fragments
in St. Davids, Pennsylvania can be sensitive to human disturbance, we need to improve the timing,
frequency, spatial scale, and taxonomic resolution of our sampling. In this brief communication,
we reflect on our initial results, and describe future directions for this long-term ecological research
project that we designed for our undergraduate biology courses.
Introduction
Human activity puts pressure on natural areas within urban ecosystems. Some of these
activities are severe enough to disturb intact natural habitats and change communities (Alberti
and Marzluff 2004). On the other hand, some habitats in urban areas are intentionally protected
from disturbance from human activities, so that their composition, ecological processes,
and ecosystem services can be preserved or restored. It is therefore necessary to examine ecological
components of these altered habitats to understand composition, function, and value
of these changing urban landscapes. Specifically, taxonomic abundance and biodiversity of
guilds can be tracked and compared so the status of habitats with different histories of humanmediated
disturbance can be assessed (Vačkář et al. 2012). One common type of human-mediated
disturbance in urban forests is land-use change, such as clearing, which consequently
results in secondary succession that can affect forest ecosystems via trophic downgrading
(Nytch et al. 2023). Specifically, removal of trees, grading, or sediment deposition can alter
understory communities, which can be true across trophic levels (Woodcock et al. 2013). In
previously cleared areas that are undergoing passive restoration, monitoring different components
of the forest ecosystem can assess changes in ecosystem health of the habitat.
One method of tracking health of disturbed forests is by monitoring abundance and
diversity of ecological indicators, such as ground-dwelling arthropods (Menta and Remelli
2020). Arthropod populations and communities can be particularly sensitive to changing
1 Biology Department, Eastern University, 308 McInnis Hall, 1300 Eagle Road, St. Davids, PA 19087-
3696. *Corresponding author: brian.alfaro@eastern.edu, (610) 225-5564
Associate Editor: Katalin Szlavecz, Johns Hopkins University
Urban Naturalist
B. Alfaro et al.
2024 No. 74
2
landscape conditions, such as those in urban ecosystems (Kotze et al. 2020). While some
arthropod taxa have advantageous innovations for urban environments, other groups decline
in urbanized areas due to loss of natural habitats. In many cases, this can directly affect
the abundance and distribution of organisms that are important for maintaining soil health
(Parker et al. 2023). To monitor the local effects of landscape change to an ecosystem or
community, an ongoing data set that describe arthropod communities can be used as ecological
indicators (Carvalho et al. 2020).
Here, we describe an ongoing study designed to train undergraduates to conduct field
work to track the effect of land-use change on arthropods in the understory of a mixeddeciduous
forest in the suburban areas of the greater Philadelphia area (PA, USA). Specifically,
we compared ground-dwelling arthropods in deciduous forest leaf litter between a
minimally disturbed, old growth forest, and a nearby secondary succession forest, recovering
from a major disturbance. Our field team of undergraduate students conducted this study
with the following questions:
1) Can we detect enough variability in taxonomic diversity and composition of arthropods
within and between 2 sets of 50 m transect lines in an old growth forest versus a secondary
regrowth forest? Addressing this question, even in a small pilot study, is critical for
two reasons. First, we intend to standardize an arthropod sampling protocol for undergraduate
students for educational training in field ecology, biostatistics, and natural history. Second,
we need to identify a minimum sampling unit for long-term monitoring of arthropod
composition and habitat change in our study areas on campus. Quantifying and describing
the frequency distribution of abundance and diversity estimates even in a limited survey
will be important information for scaling the sampling ef fort in future field seasons.
2) Because our 2 sites are near each other, there is the possibility that the 2 forest floors
share some of the same types of arthropods. We therefore asked: will our old growth and
secondary growth sampling sites have contrasting or similar composition and diversity of
ground-dwelling arthropods? We expected that the leaf litter from the old growth forest
would have higher taxonomic diversity for ground-dwelling arthropods than the leaf litter
from the forest floor of the young, secondary forest.
Materials and Methods
To answer our questions, we quantified and described the abundance and diversity of
ground-dwelling arthropods in the understory leaf litter of 2 adjacent, mixed deciduous forest
stands in greater Philadelphia, Pennsylvania (USA).
Study area
We sampled ground-dwelling arthropods in forest leaf litter among 2 sites in St. Davids
(Delaware County, Pennsylvania), which is in the Delaware Valley, in late fall (November)
of 2022. The 2 forest stands are within a 226 ha area that were naturally eastern deciduous
forests. Our campus and study areas are within the administrative boundaries of metropolitan
Philadelphia, and 24 km from the downtown financial district. Once occupied by the
Lenape peoples, the Delaware Valley have been logged and developed by European settlers
since the 1600s, and have been growing in population since the establishment of Philadelphia
as a city. Still, there are old growth and secondary forest fragments that are intact in the
valley. This area receives 1052 mm of rain per year, with an average of 121 days of rainfall
annually. In the last 10 years, Saint Davids has had an average minimum temperature of
-3°C (in January) and an average maximum temperature of 30°C (in July).
Urban Naturalist
B. Alfaro et al.
2024 No. 74
3
One sampling site, situated in Cabrini University (40.0552° N, 75.3740° W), is adjacent
to a campus trail, and has a mature forest canopy. This old growth site (10 ha) is proximate
to a 2-lane road (Eagle Road). An ongoing tree survey started in 1974 in the old growth
stand has recorded more than 30 tree species that include Liriodendron spp. (Tulip Poplar),
Fagus grandifolia Ehrh. (American Beech), Acer rubrum L. (Red Maple), and several common
oak species: Quercus velutina Lam. (Black Oak), Quercus rubra L. (Northern Red
Oak), and Quercus alba L (White Oak). Others less common trees in this site were Carya
glabra Miller (Hickory), Cornus sanguinea L. (Common Dogwood), Betula spp. (Birch),
Prunus serotina Ehrh. (American black cherry), Nyssa sylvatica Marshall (Tupelo/Sourgum),
Acer negundo L. (Box Elder), Quercus montana Willd. (Chestnut Oak), and a naturalized
Japanese tree, Cercidiphyllum japonicum Siebold & Zucc. (Katsura). The old growth
stand also contains a small population of Rhododendron maximum L. in the understory.
The second area, located in Eastern University (40.0512° N, 75.3713° W), is a disturbed
secondary habitat (3 ha) that was cleared and then covered in sediment deposit from dredged
lakebed material in 1974 (Fig. 1). The disturbed forest area is part of an ongoing long-term
study on northeastern mixed-deciduous forest succession (Fig. 1). The young forest contained
Box Elder and Juglans nigra L. (Black Walnut) in 1992, but over time the Box Elders were
increasingly crowded out by Salix nigra marshall (Black Willow). In the last decade, Red
Maple has sprouted in the site, as well as hickory seedlings that likely were dispersed by small
mammals. Viburnum plicatum Thunb. (Japanese Snowball Bush) from temperate Eurasia, an
ornamental shrub in the U.S., is also currently present in the disturbed stand. The secondary
forest stand is primarily composed of trees that are less than 50 years old.
Figure 1 – Disturbed forest sampling site in Eastern University (St. Davids, PA, USA) in the years (a)
1974, (b) 1975, (c) 1980, and (d) 1988.
Urban Naturalist
B. Alfaro et al.
2024 No. 74
4
Sampling and arthropod processing
In each site, we delineated a 50 m transect, and then sampled from 10–13 random
locations per transect (n = 23); we collected leaf litter with a minimum distance of 5 m
between samples. We bagged approximately 2 L in volume per sample of leaf litter that
was within the boundaries of a 0.25 m2 quadrat, and stored this sample in a dark cabinet
until the arthropods were collected. To collect ground-dwelling arthropods from each leaf
litter sample, we implemented the Berlese funnel method. To set up the collection, we
placed funnels with a 4–6 mm mesh hardware cloth disc into mason jars that contained
soapy water. We put the funnel setup under fluorescent lights for 48 hours, and then stored
the arthropod specimens in 75% ethanol until processing. To sort and identify the sampled
arthropods, we used recognizable taxonomic units so that the undergraduate students can
rapidly assess the specimens. Recognizable taxonomic units, sometimes called RTUs, are
categories that can be used if technicians are expected to receive a day or less of training
in taxonomic sorting and species identification (Oliver and Beattie 1993). The major
recognizable taxonomic units we monitored were Coleoptera (beetles), Formicidae (ants),
Oniscidea (terrestrial isopods – designated Isopoda for this study), Oribatida (orbatid
mites), Collembola (springtails), and Chilopoda (centipedes). This sampling effort was
part of Eastern University’s Biology 309L (Ecology Lab) course in the Biology Department,
wherein one series of modules trained undergraduate students on ground arthropod
sampling and identification.
Analysis
We counted the number of individuals per recognizable taxonomic unit for each
sample point for each site so we can quantitatively survey and compare the taxa present
in both sites. We know that there was a possibility of non-random sampling because of
the limited sampling areas, so we used the Brillouin diversity index (Boyle et al. 1990)
instead of Shannon-Weaver index. To quantify how dissimilar the composition of the
sampling points and the 2 sites were, we measured Bray-Curtis dissimilarity between
sampling points tallied in old growth and secondary growth sampling points (Ricotta
and Podani 2017). We used the package vegan in the R statistical software (Oksanen et
al. 2023, R Core Team 2023) to calculate the Brillouin H’ and Bray-Curtis dissimilarity
values for each sampling point. To analyze variability of recognizable taxonomic unit
diversity among the 2 sampling sites, we tested possible differences in mean taxonomic
diversity of ground arthropods between the old growth and secondary regrowth sampling
sites via Welch’s t-test for H’ (alpha diversity) and for Bray-Curtis dissimilarity (beta
diversity).
Results
The forest floor in the old growth stand had the most arthropod specimens (396)
compared to the secondary regrowth stand (133). We did not find any Coleoptera in the
secondary regrowth stand, but beetles were abundant in the old growth forest (Table 1).
With respect to recognizable taxonomic units, the old growth forest floor was more diverse
than the secondary regrowth forest; this result approached significance (d.f. = 16, p = 0.09,
Figure 2a). Ground arthropod composition among sampling points in the old growth forest
were more dissimilar than sampling points in the secondary regrowth site, this result is not
statistically significant (d.f. = 17, p = 0.25, Figure 2b).
Urban Naturalist
B. Alfaro et al.
2024 No. 74
5
Table 1 – Mean (standard deviation) of individuals counted per sample for each recognizable taxonomic
unit for ground-dwelling arthropods in old growth (396 specimens, n = 13 sampling points)
and secondary regrowth (133 specimens, n = 10 sampling points) forest sites, St. Davids, PA, (USA).
Araneae (1 specimen) not shown.
Recognizable taxonomic unit Old growth forest Secondary growth forest
Coleoptera 6 (7.3) 0 (0)
Acari 7 (8.2) 5 (3.8)
Formicidae 2 (2.0) 2 (3.7)
Collembola 9 (12.0) 6 (4.6)
Isopoda 4 (5.9) 1 (0.9)
Chilopoda 1 (0.3) 1 (0.3)
Figure 2 – Box plots for
(a) alpha diversity of old
growth (purple, n = 10
sampling points) and secondary
regrowth (green,
n = 10 sampling points)
forest floors in St. Davids,
PA, (USA). Box plots for
(b) beta diversity of old
growth (purple, n = 10
sampling points) and secondary
regrowth (green,
n = 10 sampling points)
forest floors in St. Davids,
PA, (USA).
Urban Naturalist
B. Alfaro et al.
2024 No. 74
6
Discussion
After analyzing our initial data, we recognize that our sampling strategy needs to be
modified with respect to scale, timing, and taxonomic resolution. Further, we reframed our
initial three questions into one new question that we can test in the future: Will differences
in habitat and environmental features in disturbed versus intact forest floors translate into
differentiation of arthropod abundance and diversity? To test this in the future, we sought
to perform ongoing and expanded sampling efforts, based on our initial approach and new
recommendations, to confirm if potential patterns are consistent or changing over time. We
outline lessons we learned, and future directions we take, for this long-term study.
Even though our sampling points were limited to 2 transects, we detected enough
variability in taxonomic diversity of ground arthropods in our old growth and secondary
regrowth forest sampling sites. To answer our first question, we can use the methods and
sampling design in this study as a standardized arthropod sampling protocol for undergraduate
students for educational training in field ecology, biostatistics, and natural history.
We can use abundance and diversity estimates of ground arthropods to describe forest
floor fauna in future field seasons. Still, we can improve the resolution of our diversity and
abundance estimates by increasing the taxonomic resolution of our arthropod specimens.
While we found variation in arthropod abundance and diversity with our limited sampling
size, we realized that our original method is inefficient. Instead of collecting a volume
of 2 L of leaf litter, we can sample from a 25 x 25 cm area to standardize our sample
collection. This standardized collection method will yield absolute density of ground arthropods
(individuals/m2) that can be comparable to other studies. Furthermore, we intend
to sample more plots from each forest type for replication. This can potentially provide
multiple samples per student that each one can analyze for each semester. Compared to
studies that include sampling in warmer periods of the year (e.g., Blair et al. 1994, Myers
and Marshal 2021), we did not collect a lot of individuals, and it is likely that the relative
abundance of some taxa were misrepresented. We therefore cannot make conclusions about
community structure with our current data. The sensible approach in the future is to sample
in warmer months when more arthropods will be active.
In future sampling, we intend to target potential indicator species. Specifically, we can
use litter dwelling spiders and ground spiders as potential indicator species for forest leaf
litter community, as there is strong support in the literature for long term studies using spiders
as indicators of ecosystem health (Argañaraz et al. 2020). While one of our principal
investigators work mostly in spider and scorpion identifications, we have connections to
research museums (i.e. University of New Mexico and California Academy of Sciences)
that could help with potential identification questions. We can also further refine this study
by including sensitive species, such as beetles and ants (Carvalho et al. 2020). Our taxonomist
for the project is also developing a non-insect Arthropods course for future academic
years, which will improve the ability of potential students in sample identification. Having
an arthropods field and lab course would also enable us to have quarterly sampling with
students who are better versed in taxa of interest. Our department is also in the early stages
of developing a reference collection for research and teaching. Nonetheless, we recognize
that this is a baseline study to establish proof of concept. Our future plan is to take all arthropod
specimens to family level and focus on the indicator taxa above to genus or species
level.
We found that isopods were abundant within our sampling points in the old growth forest,
but the authors and their students have seen them in the secondary growth forest durUrban
Naturalist
B. Alfaro et al.
2024 No. 74
7
ing warm periods. While isopods are not our focal species, it would be a relevant research
project for a senior thesis student or summer research student to explore whether isopods
are beneficial or not to the habitats in question. The main author is an invasion biologist,
and there are intriguing questions outside the scope of this project that we can answer
with isopods. For example, we can examine if potential variation body size of non-native
isopods in disturbed versus intact forests are due to differences in forest ground arthropod
diversity and abundance, and habitat characteristics.
While we detected mites in all of the sites, we know that this taxon will be the most
difficult for us to identify to the family level without extensive reference to a dichotomous
key (i.e. Dindal 1991). While we do have contact with mite specialists who could assist
with specimen identification, we have developed eDNA and specimen DNA extraction
and PCR protocols that can work on field-collected mite samples (A. Martinez, Eastern
University, Saint Davids, PA, unpubl. data; M. Weeks, Eastern University, Saint Davids,
PA, unpubl. data). In addition to microscopy, we can identify mites to genus and species
using the ITS2 rDNA barcoding gene (Ben-David et al. 2007). We can potentially refine
this study by standardizing and scaling up our sampling protocols to enable us to have a
long-term dataset tracking the impacts of urbanization and development on arthropod soil/
litter taxa.
To answer our second question, the old growth and secondary growth sampling sites
have contrasting abundance and diversity of ground-dwelling arthropods. The sampled leaf
litter from the old growth understory contained more arthropods than the disturbed secondary
forest. Overall, the old growth forest differed in arthropod abundance and diversity
from the secondary growth forest, and this could be due to the differences in site conditions
of the 2 forest floors. However, we collected leaf litter in November, when daily temperatures
have started to drop to freezing conditions, which could be a reason why we found
fewer arthropods in the secondary regrowth forest and the old growth forest (Fitzgerald
et al. 2021). As we mentioned, we will launch a new arthropods field and lab course that
would allow us to sample forest sites quarterly with trained students. This would let us collect
a better representation of arthropod abundance and diversity by including field work
in the warmer periods of the year (late summer/early fall).
In this study, we did not differentiate between trophic levels in assessing species diversity,
and there are possible issues with this approach. Specifically, there can be sampling
artefact from chilopods vs. mites, as mites are typically higher in abundance than centipedes
in a given area (e.g., Napierała et al. 2015). In future work, with taxonomic refining
to genus or species level, we plan to have more statistically sound and informative groupings
of taxa and useful species diversity indices. We can reduce this artefact by performing
rarefaction in our entire study area, and by using corrected richness and diversity indices,
such as Chao1 (Chao and Chiu 2016) and the Brillouin index that we used in this study.
For comparative analysis among sites, we can perform nonmetric multidimensional scaling
analysis and PERMANOVA in the future (e.g., Argañaraz et al. 2020). We may use the
Ecology class to gross sort the samples, and then the Entomology or non-insect Arthropods
class to take the specimens down to lower taxonomic levels. Our summer researchers and
thesis students can lead data extraction from field and lab sheets, data management, and
statistical analysis of both arthropod, environmental, and leaf litter data.
One reason that the secondary regrowth understory is less diverse in arthropods is because
the site is still recovering from the forest clearing of 1974. That is, the disturbed site
likely does not have enough plant biomass and detritus on the forest floor to support a more
diverse arthropod community (Schaffers et al. 2008). The secondary regrowth site curUrban
Naturalist
B. Alfaro et al.
2024 No. 74
8
rently contains mostly black willow and box elder. This near monoculture of the secondary
forest site can be a factor in reduced ground arthropod diversity, as there is evidence
that plant diversity can promote arthropod diversity (Dinnage et al. 2012). Nonetheless,
we interpret this as 2 forest understories potentially having enough differences in forest
composition to form differentiated arthropod composition between the 2 sites.
In previous field courses, students at Eastern University have observed distinct soil
characteristics in the old growth versus regrowth forest sites. Old growth soils are shallow
with well-defined horizons due to metamorphic rock parent material and nearby bedrock.
In contrast, regrowth forest soils are deeper and potentially enriched by well-aged goose
feces deposited from initial dredged material. The soil in the regrowth forest lack clear
horizons in the deeper layer. These differences in soil characteristics, when combined with
forest age and vegetation composition, may have resulted in enough habitat differences to
affect ground-arthropod composition (Melliger et al. 2018).
While the studies that describe changes in soil biota after deposition of lakebed dredging
are lacking, there are studies that describe community response of ground-dwelling
arthropods to changes in forest composition and clearing practices. For example, deforestation
can lead to a decrease in Coleoptera compared to natural forests (Gunnarsson et
al. 2004, Wang et al. 2020), but this can be complicated by timing of sampling and habitat
type. We saw differences in abundance and diversity of arthropods between the old growth
and secondary growth forest sites, and there is potential for this pattern to be consistent in
similar forest types in our study area. However, in addition to sampling size, one limitation
of our current methods is that we cannot make direct associations of leaf litter or forest
floor characteristics to arthropods. This is because we did not measure variables that
describe forest leaf litter characteristics to be able to quantify forest floor differences, and
possible association of forest floor features with ground arthropod diversity. Therefore, to
explain possible differences among the 2 types of forest communities, we intend to include
data collection of habitat variables. In particular, we intend to collect soil samples in each
sampling point, and then analyze soil nutrient content for each sample. We can collect leaf
litter data, such as litter depth, dry leaf mass of leaf litter collected, and species richness/
diversity of leaf litter sample to explicitly describe and quantify leaf litter. We also intend
to include soil variables, in particular soil temperature, soil moisture, and nutrient content,
as well as weather data (amount of precipitation before sampling, and ambient temperature),
that can affect our forest sampling sites.
In the years to come, local governments of urban areas can potentially mandate longterm
biodiversity studies to monitor ecosystem health (Rall et al. 2015). Consequently,
cities will need workers that can analyze the ever-increasing local and global biodiversity
data sets. It is therefore important to establish long-term ecological research plots in urban
areas at different scales for biodiversity sampling. Having long-term monitoring plots can
allow tracking of ecosystem health of fragmented habitats interspersed in cities and suburban
areas over time and space (Fa and Luiselli 2023). Additionally, the next generation of
scientists and professionals can use these sites for hands-on training in field, lab, and data
work for science-driven urban land management.
By conducting this potentially long-term study, we showed that our campus at Eastern
University in St. Davids, PA (USA) has the potential to serve as a local observatory
for natural heritage data to track how eastern mixed-deciduous forests behave in urban
regions. More importantly, subsequent annual sampling efforts can benefit students, as
this can provide a training opportunity for learning field sampling design, forest ecology,
arthropod identification, and ecological data analysis.
Urban Naturalist
B. Alfaro et al.
2024 No. 74
9
Acknowledgements
We thank Cabrini University for allowing our students to access the old growth forest understory
site. The authors thank the Associate Editor and 2 anonymous reviewers for their helpful comments
that greatly improved the manuscript.
Literature cited
Alberti, M., and J.M. Marzluff. 2004. Ecological resilience in urban ecosystems: Linking urban patterns
to human and ecological functions. Urban Ecosystems 7:241–265.
Argañaraz, C.I., G.D. Rubio, M. Rubio, and F. Castellarini. 2020. Ground-dwelling spiders in agroecosystems
of the Dry Chaco: A rapid assessment of community shifts in response to land use
changes. Biodiversity 21:125–135.
Ben-David, T., S. Melamed, U. Gerson, and S. Morin. 2007. ITS2 sequences as barcodes for identifying
and analyzing spider mites (Acari: Tetranychidae). Experimental and Applied Acarology
41:169–181.
Blair, J.M., R.W. Parmelee, and R.L. Wyman. 1994. A comparison of the forest floor invertebrate communities
of four forest types in the northeastern US. Pedobiologia 38:146–146.
Boyle, T. P., G.M. Smillie, J.C. Anderson, and D.R. Beeson. 1990. A sensitivity analysis of nine diversity
and seven similarity indices. Research Journal of the Water Pollution Control Federation
62:749–762.
Carvalho, R.L., A.N. Andersen, D.V. Anjos, R. Pacheco, L. Chagas, L. and H.L Vasconcelos. 2020.
Understanding what bioindicators are actually indicating: Linking disturbance responses to ecological
traits of dung beetles and ants. Ecological Indicators 108:105764.
Chao, A. and C.H. Chiu. 2016. Species richness: Estimation and comparison. Wiley StatsRef: Statistics
Reference Online 1:26.
Dindal, D. L. 1990. Soil Biology Guide. Wiley-Interscience. Hoboken, NJ, USA. 1376 pp.
Dinnage, R., M.W. Cadotte, N.M. Haddad, G.M. Crutsinger, and D. Tilman. 2012. Diversity of plant
evolutionary lineages promotes arthropod diversity. Ecology Letters 15:1308–1317.
Fa, J.E., and L. Luiselli. 2023. Community forests as beacons of conservation: Enabling local populations
monitor their biodiversity. African Journal of Ecology 62 e13179.
Fitzgerald, J.L., K.L. Stuble, L.M. Nichols, S.E. Diamond, T.R. Wentworth, S.L. Pelini, N.J. Gotelli,
N.J. Sanders, R.R. Dunn, and C.A. Penick. 2021. Abundance of spring and winter active arthropods
declines with warming. Ecosphere 12:e03473.
Gunnarsson, B., K. Nittérus, and P. Wirdenäs, P. 2004. Effects of logging residue removal on groundactive
beetles in temperate forests. Forest Ecology and Management 201:229–239.
Kotze, D.J., E.C. Lowe, J.S. MacIvor, A. Ossola, B.A. Norton, D.F. Hochuli, L. Mata, M. Moretti, S.A
Gagné, I.T. Handa, T.M. Jones, C.G. Threlfall, and A.K. Hahs. 2022. Urban forest invertebrates:
How they shape and respond to the urban environment. Urban Ecosystems 25:1589–1609.
Melliger, R.L., B. Braschler, H-P Rusterholz, and B. Baur. 2018. Diverse effects of degree of urbanisation
and forest size on species richness and functional diversity of plants, and ground surfaceactive
ants and spiders. PLoS ONE 13: e0199245.
Menta, C., and S. Remelli. 2020. Soil health and arthropods: From complex system to worthwhile
investigation. Insects 11:54.
Myers, A.L., and J.M. Marshall. 2021. Influence of forest fragment composition and structure on
ground-dwelling arthropod communities. The American Midland Naturalist 186:76–94.
Napierała, A., Z. Książkiewicz, M. Leśniewska, D.J. Gwiazdowicz, A. Mądra, and J. Błoszyk. 2015.
Phoretic relationships between uropodid mites (Acari: Mesostigmata) and centipedes (Chilopoda)
in urban agglomeration areas. International Journal of Acarology 41:250–258.
Nytch, C.J., J. Rojas-Sandoval, A. Erazo Oliveras, R.J. Santiago García, and E.J. Meléndez-Ackerman.
2023. Effects of historical land use and recovery pathways on composition, structure,
ecological function, and ecosystem services in a Caribbean secondary forest. Forest Ecology and
Management 546:121311.
Urban Naturalist
B. Alfaro et al.
2024 No. 74
10
Oksanen, J., G. Simpson, F. Blanchet, R. Kindt, P. Legendre, P. Minchin, R. O’Hara, P. Solymos, M.
Stevens, E. Szoecs, H. Wagner, M. Barbour et al. 2022. vegan: Community Ecology Package. R
package version 2.6–4.
Oliver, I., and A.J. Beattie. 1993. A possible method for the rapid assessment of biodiversity. Conservation
Biology 7:562–568.
Parker, D.M., K. Stears, T. Olckers, and M.H. Schmitt. 2023. Vegetation management shapes arthropod
and bird communities in an African savanna. Ecology and Evolution 13 e9880.
R Core Team. 2023. R: A Language and Environment for Statistical Computing. R Foundation for
Statistical Computing, Vienna, Austria. Avaliable online at https://www.R-project.org/.
Rall, E.L., N. Kabisch, and R. Hansen. 2015. A comparative exploration of uptake and potential application
of ecosystem services in urban planning. Ecosystem Services 16:230–242.
Ricotta, C., and J. Podani. 2017. On some properties of the Bray-Curtis dissimilarity and their ecological
meaning. Ecological Complexity 31:201–205.
Schaffers, A.P., I.P. Raemakers, K.V. Sýkora, and C.J. Ter Braak. 2008. Arthropod assemblages are
best predicted by plant species composition. Ecology 89:782–794.
Vačkář, D., B. ten Brink, J. Loh, J.E. Baillie, and B. Reyers. 2012. Review of multispecies indices for
monitoring human impacts on biodiversity. Ecological Indicators 17:58–67.
Wang, J., Y.S. Jin, Y.J. Huang, H.R. Li, F.R. Liu, X.S. Liu, L.Z. Wang, D.D. Liu, and Y.H. Lin. 2020.
Long-term effects of cutting on ground-dwelling arthropod community in coniferous and broadleaf
mixed forests in the Daxing’anling mountains. Scientia Silvae Sinicae 56:177–186.
Woodcock, P., D.P. Edwards, R.J. Newton, C. Vun Khen, S.H. Bottrell, and K.C. Hamer. 2013. Impacts
of intensive logging on the trophic organization of ant communities in a biodiversity hotspot.
PLoS One 8 e60756.