Urban meadows as an alternative to short mown grassland: effects of composition and height on biodiversity

Abstract

There are increasing calls to provide greenspace in urban areas, yet the ecological quality, as well as quantity, of greenspace is important. Short mown grassland designed for recreational use is the dominant form of urban greenspace in temperate regions but requires considerable maintenance and typically provides limited habitat value for most taxa. Alternatives are increasingly proposed, but the biodiversity potential of these is not well understood. In a replicated experiment across six public urban greenspaces, we used nine different perennial meadow plantings to quantify the relative roles of floristic diversity and height of sown meadows on the richness and composition of three taxonomic groups: plants, invertebrates, and soil microbes. We found that all meadow treatments were colonized by plant species not sown in the plots, suggesting that establishing sown meadows does not preclude further locally determined grassland development if management is appropriate. Colonizing species were rarer in taller and more diverse plots, indicating competition may limit invasion rates. Urban meadow treatments contained invertebrate and microbial communities that differed from mown grassland. Invertebrate taxa responded to changes in both height and richness of meadow vegetation, but most orders were more abundant where vegetation height was longer than mown grassland. Order richness also increased in longer vegetation and Coleoptera family richness increased with plant diversity in summer. Microbial community composition seems sensitive to plant species composition at the soil surface (0–10 cm), but in deeper soils (11–20 cm) community variation was most responsive to plant height, with bacteria and fungi responding differently. In addition to improving local residents’ site satisfaction, native perennial meadow plantings can produce biologically diverse grasslands that support richer and more abundant invertebrate communities, and restructured plant, invertebrate, and soil microbial communities compared with short mown grassland. Our results suggest that diversification of urban greenspace by planting urban meadows in place of some mown amenity grassland is likely to generate substantial biodiversity benefits, with a mosaic of meadow types likely to maximize such benefits.

Full text

Urban meadows as an alternative to short mown grassland: effects
of composition and height on biodiversity
BRIONY A. NORTON ,
1,2,10
GARY D. BENDING,
3
RACHEL CLARK,
1
RON CORSTANJE,
4
NIGEL DUNNETT,
5
KARL L. EVANS,
1
DARREN R. GRAFIUS,
1,6
EMILY GRAVESTOCK,
1
SAMUEL M. GRICE,
6
JIM A. HARRIS,
6
SALLY HILTON,
3
HELEN HOYLE,
5,7
EDWARD LIM,
1
THERESA G. MERCER,
6,8
MARK PAWLETT,
6
OLIVER L. PESCOTT,
9
J. PAU L RICHARDS,
1
GEORGINA E. SOUTHON,
5
AND PHILIP H. WARREN
1
1
Department of Animal and Plant Sciences, University of Sheffield, Sheffield S10 2TN United Kingdom
2
College of Life and Natural Sciences, University of Derby, Derby DE22 1GB United Kingdom
3
School of Life Sciences, University of Warwick, Coventry CV4 7AL United Kingdom
4
Centre for Environmental and Agricultural Informatics, Cranfield University, Cranfield MK43 0AL United Kingdom
5
Department of Landscape, University of Sheffield, Sheffield S10 2TN United Kingdom
6
Cranfield Soil and Agrifood Institute, Cranfield University, Cranfield MK43 0AL United Kingdom
7
Department of Architecture and Built Environment, UWE Bristol, Bristol BS16 1QY United Kingdom
8
School of Geography, University of Lincoln, Lincoln LN6 7TS United Kingdom
9
Centre for Ecology and Hydrology, Wallingford OX10 8BB United Kingdom
Citation: Norton, B. A., G. D. Bending, R. Clark, R. Corstanje, N. Dunnett, K. L. Evans,
D. R. Grafius, E. Gravestock, S. M. Grice, J. A. Harris, S. Hilton, H. Hoyle, E. Lim, T. G.
Mercer, M. Pawlett, O. L. Pescott, J. P. Richards, G. E. Southon, and P. H. Warren. 2019.
Urban meadows as an alternative to short mown grassland: effects of composition and height
on biodiversity. Ecological Applications 29(6):e01946. 10.1002/eap.1946
Abstract. There are increasing calls to provide greenspace in urban areas, yet the eco-
logical quality, as well as quantity, of greenspace is important. Short mown grassland
designed for recreational use is the dominant form of urban greenspace in temperate
regions but requires considerable maintenance and typically provides limited habitat value
for most taxa. Alternatives are increasingly proposed, but the biodiversity potential of these
is not well understood. In a replicated experiment across six public urban greenspaces, we
used nine different perennial meadow plantings to quantify the relative roles of floristic
diversity and height of sown meadows on the richness and composition of three taxonomic
groups: plants, invertebrates, and soil microbes. We found that all meadow treatments were
colonized by plant species not sown in the plots, suggesting that establishing sown mead-
ows does not preclude further locally determined grassland development if management is
appropriate. Colonizing species were rarer in taller and more diverse plots, indicating com-
petition may limit invasion rates. Urban meadow treatments contained invertebrate and
microbial communities that differed from mown grassland. Invertebrate taxa responded to
changes in both height and richness of meadow vegetation, but most orders were more
abundant where vegetation height was longer than mown grassland. Order richness also
increased in longer vegetation and Coleoptera family richness increased with plant diversity
in summer. Microbial community composition seems sensitive to plant species composition
at the soil surface (0–10 cm), but in deeper soils (11–20 cm) community variation was most
responsive to plant height, with bacteria and fungi responding differently. In addition to
improving local residents’site satisfaction, native perennial meadow plantings can produce
biologically diverse grasslands that support richer and more abundant invertebrate commu-
nities, and restructured plant, invertebrate, and soil microbial communities compared with
short mown grassland. Our results suggest that diversification of urban greenspace by
planting urban meadows in place of some mown amenity grassland is likely to generate
substantial biodiversity benefits, with a mosaic of meadow types likely to maximize such
benefits.
Key words: beetles; carbon; conservation planning; green infrastructure; insects; microbial diversity;
nitrogen; overwintering; plant richness; urban ecology; urban parks.
INTRODUCTION
Urban greenspace has the potential to support consid-
erable biodiversity (Aronson et al. 2014, Beninde et al.
2015) with potential benefits for human well-being
Manuscript received 25 September 2018; revised 13 March
2019; accepted 26 March 2019. Corresponding Editor: John M.
Marzluff.
10
E-mail: briony.a.norton@gmail.com
Article e01946; page 1095
Ecological Applications, 29(6), 2019, e01946
©2019 The Authors. Ecological Applications published by Wiley Periodicals, Inc. on behalf of Ecological Society of America.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited.

(Fuller et al. 2007, Dallimer et al. 2012, Pett et al. 2016),
ecosystem service provision (Tratalos et al. 2007, Radford
and James 2013, Schwarz et al. 2017), and local and glo-
bal conservation (Ives et al. 2016). With the growth of
urban land cover globally (Seto et al. 2012), the role of
cities in contributing to conservation and ecosystem ser-
vice provision is increasing. The potential of urban areas
to deliver these benefits is, however, being eroded by loss
of greenspace to redevelopment and densification (Haa-
land and van den Bosch 2015) and by typical approaches
to urban greenspace management (Aronson et al. 2017).
Particularly common is the maintenance of greenspace as
short mown grass in the form of lawns or amenity grass-
land (M€
uller et al. 2013). Amenity grassland is frequently
mown, short sward vegetation that is managed for human
recreational use, examples include lawns in public parks
and sports grounds. Short-mown grassland habitats dom-
inate temperate cities, in both public and private urban
greenspaces, for example they cover 22.5% of the land
area of Swedish cities, almost double the cover 50 yr ago
(Hedblom et al. 2017), and similar amounts in the UK
(25%; Evans et al. 2009) and United States (23%; Rob-
bins and Birkenholtz 2003), which equates to 1.9% of the
total land area of the continental United States (Milesi
et al. 2005).
Short-mown urban grasslands are popular due to their
assumed aesthetic value, well-established and widely
accepted management protocols, provision of recre-
ational space and associated social norms (Harris et al.
2013, Ignatieva et al. 2015, Hoyle et al. 2017). However,
they require intensive management, with UK local
authorities typically mowing every 2–3 weeks during the
growing season (March–September; Garbuzov et al.
2015) and the total number of annual cuts is increasing
with extended growing seasons under climate change
(Sparks et al. 2007). Many lawns and parks receive fre-
quent inputs of fertilizer, herbicide, and, depending on
local climate, irrigation (Alumai et al. 2009, Bertoncini
et al. 2012, Bijoor et al. 2014). This is financially and
environmentally costly (Smetana and Crittenden 2014),
and at odds with reduced funding for managing public
spaces in many developed regions (Walls 2009, Heritage
Lottery Fund 2014).
Cumulatively across urban areas, parks can harbor
significant numbers of plant species (Thompson et al.
2004, Stewart et al. 2009, Bertoncini et al. 2012), and,
per unit area, lawns support species richness similar to
those of seminatural grasslands, although composition
is often dominated by a small number of grass species
(Thompson et al. 2004, Bertoncini et al. 2012, Wheeler
et al. 2017). However, the limited vegetation structure
provided by short grass swards leads to reduced diversity
of many invertebrate taxa relative to more structurally
complex grasslands (Morris 2000, Jerrentrup et al.
2014). This results from direct effects of reduced habitat
availability and complexity and other effects such as
microclimate alteration (Gardiner and Hassall 2009),
trampling by humans (Duffey 1975), and mowing
limiting forb flowering and seed set (Garbuzov et al.
2015) and causing direct mortality (Humbert et al.
2010). As a result, there is growing interest around the
world in finding more structurally and botanically
diverse alternatives to mown amenity grassland (Bor-
mann et al. 2001, Klaus 2013, Blackmore and Goulson
2014, Hwang et al. 2017, Jiang and Yuan 2017).
Introducing areas of “meadow”vegetation, broadly
defined as infrequently mown grassland, usually with
flowering forbs, to replace park grass is thought to ame-
liorate some of these effects. Meadow-like areas can be
established by reducing mowing frequency, allowing the
existing plant community to increase in height and flower
cover (Garbuzov et al. 2015, Wastian et al. 2016, Lerman
et al. 2018). However, the outcome of this approach is
dependent on the diversity of the existing flora, and any
subsequent natural colonization. An alternative method
for establishing meadows is deliberate seeding or planting
of designed mixes of plant species. This latter approach to
meadow creation predominantly uses annual plant spe-
cies to enhance aesthetic value (Dunnett and Hitchmough
2007, Dunnett 2011), but may be complemented with
perennial species to reduce the need for re-sowing (Hoyle
2016). Urban meadow areas are widely advocated by con-
servation organizations (RSPB 2013, The Wildlife Trusts
2018). While the potential benefits to people and wildlife
are widely articulated, and there are some studies of
human responses (Jiang and Yuan 2017, Southon et al.
2017), there is little work quantifying the ecological
effects of different types of urban meadows in public
greenspaces (Klaus 2013). This contrasts with the more
extensive examination of the ecological effects of increas-
ing wildflower coverage in agricultural systems (Knop
et al. 2006, Haaland et al. 2011, Buri et al. 2016). Urban
meadows, however, warrant separate attention as condi-
tions, and constraints, differ substantially from agricul-
tural systems. Notably, soil conditions in urban and
agricultural areas differ due to numerous factors includ-
ing the absence of livestock or specialist management for
crops, pollutant concentrations and different exposure to
other management activities such as regular plowing
(Pouyat et al. 1995, Set€
al€
a et al. 2016). There are also
substantial differences in colonization potential as urban
grasslands are often poorly connected (Hejkal et al.
2017), fragmented by urban land covers or other vegeta-
tion (Williams et al. 2009). Furthermore, the need to
develop plant mixes that are acceptable to the public in
areas close to housing presents particular challenges in
introducing taller and “messier”vegetation (Hoyle et al.
2017).
Here, we use a replicated set of nine different peren-
nial meadow treatments, sown in six public urban green-
spaces in southern England, to quantify the relative
roles of floristic diversity and height on the diversity and
composition of plant, invertebrate and soil microbial
communities. These results form part of a wider assess-
ment of these meadow plantings, which include assess-
ments of the responses of local residents (Southon et al.
Article e01946; page 1096 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

2017, 2018) and greenspace managers (Hoyle et al.
2017).
METHODS
Meadow establishment and experimental design
Meadow plots were established in areas of urban
mown amenity grassland at sites adjacent to residential
housing, on clay-loam soils in five areas in Bedford
(Chiltern Avenue, Jubilee Park, Goldington Green,
Brickhill Heights) and Luton (Bramingham Road), in
Southern England (Appendix S1: Fig. S1, Table S1).
Meadows were also established on clay soils adjacent to
campus residential housing at Cranfield University, situ-
ated in the countryside but with urban development fea-
tures (high- and low-rise buildings and housing, roads,
airport; Appendix S1: Fig. S1). Meadows were hand
sown in early May 2013 in plots rotovated to a depth of
100–150 mm (rotovating breaks the ground up and
achieves a fine tilth for sowing, similar to tilling) after
being treated with glyphosate herbicide. Some hand
weeding was done on all plots in July 2013, targeting
four species that became sources of complaints from
local residents (all sites: Chenopodium album,Sonchus
oleraceus, and Helminthotheca echiodes; Jubilee Park
only: Potentilla reptans). Weeding was done across all
sites, and although weeding effort was not quantified, it
was not systematically related to treatments. In addition,
to ensure successful establishment, all low diversity
(grass) plots (except two; the tall plots at Goldington
Green and Bramingham, which established adequately)
were sprayed with herbicide and reseeded at higher den-
sity in autumn 2013, and bare patches in the medium
and high diversity plots were over-sown at the original
density (Appendix S1: Table S2). One site (Jubilee Park)
was rotovated and resown in April 2014 due to poor
establishment. Plots were sampled in their second grow-
ing season (Jubilee Park, 2015; all other sites, 2014). Due
to their smaller size, the Cranfield plots were only used
when assessing soil properties.
Nine meadow treatments spanned two axes of varia-
tion: plant species richness (low, medium, and high) and
height (short, medium, and tall; Fig. 1). All sown species
were perennials and native to southern England. Seed
mixes for each treatment were randomly allocated to
standardized rectangular plots, with at least 5-m gaps of
original short mown grass between plots. The arrange-
ment of the plots in relation to each other varied
between sites depending on site shape and existing
infrastructure. Plots were 250 m
2
(12.5 920 m), except
at Cranfield where, due to space constraints, plots were
50 m
2
(5 910 m). At all sites, an area of the original
short mown grass equal in area to the treatment plots
was identified and surveyed, but was subject to no
preparatory cultivation and continued to be managed
identically to the surrounding mown amenity grassland
(referred to as the unmanipulated control).
Plant species richness was manipulated by sowing seed
mixes (Appendix S1: Table S2) varying in total species
richness and ratio of grass to forbs (broad-leaved herba-
ceous plants; Fig. 1; Appendix S1: Table S2). The low
plant species richness seed mixes contained only grasses
and the short plots containing this mix thus simulated
newly sown, mown, amenity grassland. Species composi-
tion of plots was chosen to achieve the target heights
and flower cover for the different treatments under the
proposed mowing regimes (see below paragraph), mean-
ing that the composition had to vary somewhat between
treatments (i.e., maintenance of high floral diversity in a
tall plot requires different species from a short, regularly
mown, high diversity plot).
Vegetation height was determined by choice of plant
species and different cutting regimes: short plots were
cut to a target height of 0.05 m every 4–6 weeks depen-
dent on staffing and weather (Hoyle et al. 2017), med-
ium height plots were cut twice a year (April and
September), and tall plots were cut once a year (Febru-
ary). Medium plots had a target height of approximately
0.50 m during the growing season, and tall plots reached
an average maximum height of 1.50 m. At most sites, all
nine treatments were established (Appendix S1:
Table S1). The exceptions were Goldington Green where,
due to space constraints, only the low and high richness
treatments were implemented, and Brickhill Heights,
where four plots were discontinued due to feedback from
residents (Hoyle et al. 2017), leaving two short and all
tall treatments (Appendix S1: Table S1).
Botanical surveys
Botanical surveys were conducted in July, in the sec-
ond year after establishment (Jubilee Park, 2015; all
other sites, 2014). Surveys were undertaken in five repli-
cate 1-m
2
quadrats in all plots, arranged in a quincunx
and at least 2 m from the plot edge. The percentage
cover of each species was recorded on the Domin scale
(Rodwell 2006) and an average calculated for each plot
using the midpoints of the Domin categories. In each
plot, species recorded were separated into sown and
non-sown, the latter classified as any species not in the
seed mix for that plot (although it may occur in one of
the other treatment seed mixes).
Invertebrates
Aboveground invertebrates were sampled in all plots
in summer (June 2014, July 2015) and early autumn
(September 2014, 2015) using sweep-nets and vacuum
sampling. Overwintering invertebrates were sampled at
four sites in February 2015, prior to tall plots being cut
but after the medium plots were cut (September). Winter
sampling involved time standardized searches that com-
prised a sequence of beating, cutting, collecting and siev-
ing vegetation. A pooter/aspirator was used to extract
invertebrates from the ground surface or sieved samples,
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1097

by orally sucking them into collecting tubes. For all sam-
ples, we quantified the number of individuals and, for
summer and autumn samples, the biomass in each inver-
tebrate order (Appendix S1: Table S3). We use both
abundance and biomass as these are not necessarily
strongly correlated, for example, if high abundances are
driven by large numbers of small-bodied individuals.
Coleoptera from both summer and autumn samples
were identified to family at the three sites with a full set
of treatments (Chiltern Avenue, Bramingham Road,
Jubilee Park). Summer, autumn and winter samples were
analysed separately as responses of some taxa to the
meadow treatments may vary seasonally. Further details
of the sampling methods are provided in the
Appendix S1: Section S1.
Soils
Soil sampling.—All soil sampling and measurements
were undertaken at Chiltern Avenue, Bramingham
Road, and Cranfield, providing examples of all
treatments, across the broadest range of soil types at
sites with the full suite of treatments. Soils were sampled
(33 mm internal diameter, gouge-style auger) in Febru-
ary 2015 in each plot, including the unmanipulated con-
trol. Soil sampling locations were ascertained by
splitting each plot into three subplots of equal size.
Within each subplot, three samples were taken from ran-
domly generated coordinates. These were then bulked at
each of two depths (0–10 cm and 11–20 cm), providing
three bulked cores for each plot and depth.
Soil total nitrogen and total carbon.—The soil samples
were prepared for analysis by homogenizing and sieving
(2 mm) each sample. Total carbon (BS 7755-3.8:1995)
and total nitrogen (BS EN 13654-2:2001) were assessed
using the Elementar Vario III EL analyzer (Elementar,
Langenselbold, Germany).
Soil biological community.—Microbial biomass C was
determined using the fumigation-extraction procedure
(Jenkinson and Powlson 1976) using a K
EC
(extraction
D1
Low richness (3-4 spp.)
Grass only
D2
Medium richness (9-13 spp.)
Grass + forb mix
D3
High richness (16-21 spp.)
Grass + forb mix
H1
Short.
Cut once every
4-6 weeks
H2
Medium.
Cut twice a year
(April,
September)
H3
Tall.
Cut once a year
(February)
H1.D1 H1.D2
H2.D3
H1.D3
H2.D2
H2.D1
H3.D1 H3.D2 H3.D3
Diversity
Height
FIG. 1. The nine experimental treatments shown across two axes of variation; height (H1, H2, H3) and diversity (D1, D2, D3),
with example photographs taken in early summer of the second year after establishment. Diversity treatments differed in total spe-
cies richness and relative proportion of forb and grass. Height treatments differed in mowing regimes as well as plant selection.
Details of the nine seed mixes are in Appendix S1: Table S2. At each site, an area of the original mown amenity grassland equal in
area to the treatment plots but without special management (the unmanipulated control plot) was also surveyed.
Article e01946; page 1098 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

efficiency coefficient) of 0.45 (Vance et al. 1987). Micro-
bial community phenotypic characteristics were deter-
mined by analyzing cellular phospholipids (based on
Frosteg

ard et al. 1993) and the relative abundance of
indicator fatty acids for arbuscular mycorrhizal fungi and
bacteria. Total DNA was extracted from 250 mg of
homogenized soil and sequenced on the Illumina MiSeq
(Illumina, inc., San Diego, California, USA) platform for
fungal (ITS3 and ITS4) and bacterial (16S—515f and
806r) primers. Sequences were clustered to operational
taxonomic units (OTUs) with a 97% minimum identity
threshold (after excluding sequences that only occurred
once). Taxonomy was assigned using quantitative insights
into microbial ecology (QIIME 1.8; Caporaso et al. 2010)
and the Greengenes reference database for 16S (McDon-
ald et al. 2011), or the UNITE database for ITS (K~
oljalg
et al. 2013). Full details of the extraction and sequencing
methods are provided in the Appendix S1: Section S2.
Data analysis
All analyses were performed using R (version 3.5.0; R
Core Team 2018) unless otherwise stated.
Plants and invertebrates: comparison of unmanipulated
controls and short, low diversity treatments.—We com-
pared the response of plants and invertebrates to the
short, low diversity (H1.D1) plots (that simulated newly
sown mown amenity grassland) and the unmanipulated
control plots with paired ttests using the PairedData
package (Champely 2018). We compared the richness
and cover of plants, the richness of invertebrate orders
(all seasons) and Coleoptera families (summer and
autumn), and total invertebrate abundance (all seasons).
Plants and invertebrates: effects of vegetation height and
diversity treatments.—Linear mixed effects models were
constructed using the package lme4 (Bates et al. 2015)
with height (three levels) and diversity (three levels)
treatments included as fixed effects and site as a random
intercept, using maximum likelihood parameter estima-
tion. No interaction term was included as there was no
within-site replication. These analyses exclude the unma-
nipulated control plots. Response variables were trans-
formed where necessary (Table 1). Significance of the
main fixed effects was assessed with an ANOVA of the
model output using the package lmerTest (Kuznetsova
et al. 2017), with degrees of freedom based on Satterth-
waite’s approximation. Within-treatment pairwise com-
parisons were determined with post-hoc tests using least-
squares means in package lsmeans (Lenth 2016).
Models were constructed for three plant responses:
richness of all plant species, richness of non-sown spe-
cies, and cover of non-sown species. The response vari-
ables for the invertebrate models for all seasons
(summer, autumn, winter) were richness (order level)
and total abundance. For summer and autumn, inverte-
brate sample models were also run for Coleoptera
richness (family level), total biomass (dry mass), and
abundance of each order with more than 1,000 individu-
als (Appendix S1: Table S4). Data from summer,
autumn, and winter were analysed separately.
Plants and invertebrates: ordinations.—To assess the
effect of height and diversity treatments on the composi-
tion of the biological communities, data from the nine
experimental treatments were ordinated using nonmetric
multidimensional scaling using metaMDS in the vegan
package (Oksanen et al. 2017). Ordinations were run for
the non-sown plant community, the order-level inverte-
brate community (summer and autumn) and family-level
Coleoptera community (summer and autumn). Inverte-
brate data from summer and autumn were assessed sepa-
rately. Data were square-root transformed and
submitted to Wisconsin double standardization (Oksa-
nen et al. 2017). Pairwise dissimilarities were calculated
using the Bray-Curtis index. A maximum of 50 random
starting configurations were used to find a stable solu-
tion. If stress was greater than 0.2 with two axes, a third
axis was added. The function adonis in vegan was used
to compare the location of the centroid of each tested
group statistically, applying nonparametric permuta-
tional ANOVA using dissimilarity matrices (Oksanen
et al. 2017). Data met the adonis test assumptions of
homogeneity of variance (tested using betadisper in
vegan) with the exception of the non-sown plants and
Coleoptera community in summer (Appendix S1:
Table S5). Models were constructed including height and
diversity blocked by site, with Bray-Curtis dissimilarities.
As explanatory variables are entered sequentially, each
model was run twice, reversing the order that height and
diversity treatments were entered into the model, and
results for each variable are reported for the model
where the other variable has been taken into account.
Soils: effects of vegetation height and diversity treatments.—
Linear mixed effects models of total nitrogen and total
carbon and four measures of the microbial community
(see below paragraph) were constructed in JMP (Version
13.0; SAS Institute Inc., Cary, North Carolina, USA).
Height treatments (three levels), diversity treatments
(three levels) and an interaction term between height
and diversity were included as fixed effects. Replicates
within each plot were nested within site as a random
effect, corresponding to the three cores obtained from
each plot. This approach reflects differences in the sam-
pling design used for soil microbes and the invertebrate
and plant communities. Parameter estimation was
undertaken with restricted maximum likelihood
(REML). Each response variable was analysed sepa-
rately for 0–10 cm and 11–20 cm depths.
The microbial community was characterized with
phospholipid fatty acid analysis (PLFA) and DNA. The
microbial community of each plot, including the unma-
nipulated control plots, as characterized by PLFA was
ordinated using principal components analysis, in JMP
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1099

(Version 13.0; SAS Institute) on the correlation matrix
using a standard least squares estimator. The first and
second principal components were extracted for further
analysis. The community, as characterized by its DNA,
was first separated into fungal (Kingdom Fungi) and
bacterial (Domain Bacteria) components. Each commu-
nity was analysed at the OTU level. Fisher’s alpha diver-
sity (Fisher et al. 1943) of the bacterial and fungal
communities was calculated for each plot. The commu-
nity in each treatment plot was characterized by its dif-
ference from the unmanipulated control plot at the same
site using Bray-Curtis dissimilarities (Bray and Curtis
1957). These six variables (PLFA principal components
one and two; bacterial and fungal DNA alpha diversity;
bacterial and fungal DNA Bray-Curtis dissimilarity),
measured at both depths, were used as response vari-
ables in linear models.
RESULTS
Plants
A total of 106 plant species were detected across all
the surveyed plots. Of these, only 33 were sown, and two
were unique to unmanipulated control plots. Five sown
species were not detected: Anthriscus sylvestris,Hyper-
icum perforatum,Primula veris,Ranunculus acris, and
Tanacetum vulgare. The three species with the greatest
total cover were Lolium perenne,Achillea millefolium,
and Leucanthemum vulgare. At the three sites with the
full nine set of treatments, there were 35 (Chiltern Ave-
nue), 47 (Jubilee Park), and 61 (Bramingham Road)
non-sown taxa detected. Averaging across all the sites,
there were between four and eight non-sown species per
treatment plot. At the sites with the full nine set of treat-
ments, most (between 79% and 84%) of the non-sown
taxa were not found in the unmanipulated control plot;
similar proportions (74% and 78%) were observed at
sites with fewer plots.
The unmanipulated control plot and short, low diver-
sity, treatment (H1.D1) did not differ significantly in
total richness (t
4
=8.85, P=0.441) or cover (t
4
=0.96,
P=0.387). Overall, total plant richness follows the pat-
tern established by the diversity treatments (Tables 1, 2).
Richness of non-sown plants varied significantly with
height treatments, where tall plots had significantly
lower non-sown plant richness than medium or short
plots (Fig. 2a; Table 1). Cover of the non-sown species
varied with diversity treatment, with the high diversity
treatments having significantly lower cover of non-sown
species (Fig. 2b; Table 1).
Analysis with adonis showed that plant community
composition, across all plant species, was significantly
affected by height (F
9,37
=2.02, P=0.001, R
2
=0.32)
and diversity treatments (F
9,37
=1.77, P=0.001, R
2
=
0.28). Community composition of non-sown plant spe-
cies was significantly affected by diversity treatments
(F
9,37
=1.24, P=0.045, R
2
=0.23) and height (F
9,37
=
1.32, P=0.018, R
2
=0.24), although the result for non-
sown plants for height should be treated cautiously as
there is heterogeneity in variance across the height treat-
ments (P=0.01; Fig. 3a, b; Appendix S1: Table S5).
Invertebrates
In summer and autumn, over 138,000 invertebrates
were collected from the treatment plots (excluding the
unmanipulated control plots). The most abundant taxa
were Collembola (45,151), Acari (24,502), Hemiptera
(20,253), Diptera (13,662), and Coleoptera (11,048 of
which 9,477 were adults). A total of 7,172 Coleoptera
adults were sorted to family. Twenty-three families were
identified, of which the Nitidulidae (3,690), Latridiidae
(781), and Staphylinidae (681) were most abundant. In
winter, over 4,400 invertebrates were collected from a
subset of plots (Appendix S1: Tables S1, S4). The most
abundant taxa were Collembola (1,724), Diptera (705),
and Gastropoda (455).
Community-level richness and abundance.—The inverte-
brate communities in the unmanipulated control plots
and short, low diversity treatments (H1.D1) were not
TABLE 1. Results of linear mixed models for plant community variables.
Contrasts
Fixed effects variable df FPLow–medium Low–high Medium–high
All plants richness, square-root transformed
Diversity 2,34.0 5.80 0.007 M > L,0.044 H > L,0.009 0.942
Height 2,33.9 3.84 0.031 0.507 0.218 M > H, 0.026
Not-sown plants richness, untransformed
Diversity 2,34.0 1.70 0.197 0.991 0.263 0.283
Height 2,33.9 7.07 0.003 0.649 L>H,0.022 M > H,0.003
Not sown plants cover, 1/4-power transformed
Diversity 2,34.1 3.45 0.043 0.882 L>H,0.043 0.189
Height 2,33.9 2.38 0.108 0.224 0.914 0.107
Notes: Significant effects (P<0.05) are indicated in boldface type. Contrasts are the results of least-squares means (see subsec-
tion Plants and invertebrates: effects of vegetation height and diversity treatments for details). Where there was a significant difference
between treatments, the direction of the effect is indicated. Refer to Fig. 2 for the differences between treatment means.
Article e01946; page 1100 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

significantly different in summer (order richness:
t
4
=1.58, P=0.189; total abundance: t
4
=2.30, P=
0.083; Coleoptera family richness: t
2
=2.14, P=0.166).
In autumn, richness and abundance were higher in
H1.D1 than the unmanipulated control plots (order
richness: t
4
=3.65, P=0.022; total abundance: t
4
=
4.12, P=0.014; Coleoptera family richness: t
2
=8.00,
P=0.015). In winter, abundance was not significantly
different between H1.D1 and the unmanipulated control
(t
2
=2.82, P=0.106) but order richness was higher in
H1.D1 plots (t
2
=4.91, P=0.040).
Invertebrate responses varied between autumn and
summer. Order-level richness was affected by height
treatment in both seasons (Fig. 2c; Tables 2, 3); tall plots
had significantly higher richness than either short or
medium height plots in summer, while in autumn, short
plots had significantly lower richness than other height
treatments. Coleoptera family richness responded only to
diversity in summer, when the high diversity treatments
had the highest richness (Fig. 2d). Invertebrate abun-
dance (Fig. 2e) and biomass (Fig. 2f) both varied with
height treatment in summer, as did total invertebrate bio-
mass in autumn. The short plots drove differences in
invertebrate biomass, having lower invertebrate biomass
than either medium height or tall plots, while abundance
in summer was significantly lower in the short than the
tall plots. Invertebrate abundance was significantly
higher in low than high diversity plots in autumn
(Table 3). In winter, both order-level richness and total
abundance were significantly higher in the tall plots com-
pared to the short plots (Table 3; Appendix S1: Fig. S2).
Community composition.—The composition of the inver-
tebrate community at the order level was significantly
affected by height treatment in both summer (F
9,37
=
1.70, P=0.026, R
2
=0.25) and autumn (F
9,37
=1.76,
P=0.023, R
2
=0.26) but not by diversity treatments in
either season (Fig. 3c, d; Appendix S1: Table S5).
Coleoptera community composition was only signifi-
cantly affected by height treatment in autumn (F
6,26
=
2.02, P=0.007, R
2
=0.33; Fig. 3e, f; Appendix S1:
Table S5). There was no effect of plant diversity on com-
munity composition in summer or autumn, or of plant
height on composition in summer, although the latter
result should be treated cautiously as there is heterogene-
ity in variance across the height treatments (P=0.01;
Appendix S1: Table S5).
The effect of height and plant diversity on invertebrate
abundance varied across taxonomic groups (Tables 2, 3).
Differences in abundance were most pronounced
between height treatments. Almost all groups had higher
abundance in medium or tall plots than short plots, in at
least one season (summer or autumn). The exceptions
were the two taxa predominately found in soil, Acari
and Collembola, which did not respond to vegetation
height. Some taxa were more abundant in both medium
and tall plots than short plots (both seasons: Coleop-
tera; summer: Hemiptera; autumn: Psocodea), some
TABLE 2. Summary of the response of plant and invertebrate
richness, abundance, and composition and of individual
invertebrate orders to the two axes of experimental meadow
treatments: diversity (three levels) and height (three levels).
Response variable
Direction of effect
Diversity Height
Plants
All plants richness ↑,L <M/H ∩,M >H
All plants composition yes, NA yes, NA
Not-sown plants richness ↓,H<L/M
Not-sown plants cover ↓,L>H
Not-sown plants composition yes†, NA yes†,NA
Invertebrates, Summer
Order richness ↑,H>L/M
Order composition yes, NA
Coleoptera richness ↑,H >L/M
Coleoptera composition
Invertebrate abundance ↑,H>L
Invertebrate biomass ↑,L<M/H
Invertebrate orders, Summer
Acari
Araneae ↑,H >L
Coleoptera ↑,H >L↑,L <M/H
Collembola
Diptera ↑,H >L↑,H >L/M
Hemiptera ↓,L >H↑,L <M/H
Hymenoptera ↑,H >L
Thysanoptera ↑,H>L
Invertebrates, Autumn
Order richness ↑,L<M/H
Order composition yes, NA
Coleoptera richness
Coleoptera composition yes, NA
Invertebrate abundance ↓,L >H
Invertebrate biomass ↑,L<M/H
Invertebrate orders, Autumn
Acari
Araneae
Coleoptera ↑,L <M/H
Collembola
Diptera
Hemiptera ↓,L >M/H ∩,M >L/H
Hymenoptera ∩,M >L/H
Psocodea ↑,H >M>L
Thysanoptera ↓,L>M/H ↑,H >L
Invertebrates, Winter
Order richness ↑,H>L
Invertebrate abundance ↑,H >L
Notes: Results summarize linear model outputs (full results in
Tables 1 and 3) and adonis results (full results in Appendix S1:
Table S5). Here, significant effects for compositional change are
indicated by “yes”and the direction effects are not applicable
(NA). Significant effects for diversity and abundance measures are
designated with a symbol indicating the direction of effect and text
with detail of which treatments were higher or lower. An up arrow
indicates the response variable increases with increasing meadow
height/diversity. A down arrow indicates the response variable
decreases with increasing meadow height/diversity. A ∩symbol
indicates that the response variable was highest in the medium
height/diversity treatments. Non-significant responses are left
blank and indicate no detectable response to the treatment.
†Note that data did not meet the adonis test assumptions of
homogeneity of variance (tested using betadisper).
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1101

Article e01946; page 1102 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

taxa were more abundant in medium plots than short
plots (autumn: Hemiptera and Hymenoptera) and other
taxa were more abundant only in tall plots compared to
short plots (both seasons: Thysanoptera; summer: Ara-
neae, Diptera, Hymenoptera). Significant effects of the
diversity treatment on the abundance of individual
orders occurred only for Coleoptera (positive; summer),
Hemiptera (negative; summer and autumn), Thysanop-
tera (negative; autumn) and Diptera (positive; summer).
Soils
Total nitrogen and total carbon.—Total N varied with
height treatments at both depths (P=0.037 at 0–10 cm;
P=0.034 at 11–20 cm), where the medium plots had
the highest total N (0–10 cm, 
x=0.42, SE =0.02; 11–
20 cm, 
x=0.25, SE =0.01), followed by the short plots
(0–10 cm, 
x=0.4, SE =0.01; 11–20 cm, 
x=0.23,
SE =0.01) and tall plots had the lowest total N (0–
10 cm, 
x=0.34, SE =0.01; 11–20 cm, 
x=0.22,
SE =0.01; Appendix S1: Fig. S3). At both depths, total
C was highest in high diversity plots (0–10 cm, 
x=5.13,
SE =0.28, P=0.024; 11–20 cm, 
x=3.6, SE =0.28,
P=0.01). The response of total C to the low and med-
ium diversity plots varied with depth: at 0–10 cm, total
C was higher in low diversity plots (
x=4.41, SE =0.13,
P=0.005) compared to medium diversity plots
(
x=4.48, SE =0.13); at 11–20 cm, medium diversity
plots had higher (
x=2.79, SE =0.12, P=0.004) total
C than low diversity plots (
x=2.85, SE =0.09;
Appendix S1: Fig. S3).
Microbial community.—The soil microbial community
had different structure in shallow (0–10 cm) and deep
(11–20 cm) layers. The first principal component repre-
senting the PLFA data was significantly affected by
height treatment at both depths and there was a signifi-
cant interaction with diversity treatment at both depths
(Fig. 4, Tables 4, 5). The second principal component
representing the PLFA data was significantly affected by
diversity treatment at both depths and there was a signif-
icant interaction with height treatment at both depths
(Fig. 4, Table 4).
The bacterial community composition was character-
ized both as a function of alpha diversity and divergence
of community composition from the unmanipulated con-
trol, measured as Bray-Curtis dissimilarity. In terms of
the alpha diversity, there was no significant effect on bac-
teria of either diversity or height treatment at either depth
with the exception of a significant effect of height at the
11–20 cm depth (Fig. 2g, Tables 4, 5), which was driven
by higher bacterial biodiversity for the short treatments.
For fungal alpha diversity, there was a weak significant
effect of height treatment at 0–10 cm, and taller plots were
associated with higher fungal diversity. At 11–20 cm, there
was a significant effect of diversity, and the low plant rich-
ness treatment was associated with higher fungal diversity
(Fig. 2h, Table 4). For bacteria, Bray-Curtis dissimilarity
analysis showed that height had a significant effect on
community composition at both depths (0–10 cm and
11–20 cm; Fig. 2i, j, Table 4). Fungal community compo-
sition was significantly affected by diversity at 0–10 cm
depth, and height at 11–20 cm depth, and furthermore
there was an interactive effect of height and floristic diver-
sity on fungal community composition at 0–10 cm.
DISCUSSION
We replaced mown amenity grassland with a range of
meadow-type vegetation in six public greenspaces to
assess the response of biological diversity to urban mea-
dow habitat creation. We found that increasing the
height and sown species diversity of the plant community
generally altered the composition of soil microbial and
aboveground invertebrate communities, increased inver-
tebrate biomass, abundance, and richness and reduced
incursion by non-sown plant species. Increasing plant
height was associated with lower richness of non-sown
plants and higher richness, biomass, and abundance of
invertebrates, although responses of individual taxa var-
ied. Increased height also changed composition of the
soil bacterial community at 0–10 cm and 11–20 cm
depth, and of soil fungal communities at 11–20 cm.
Increasing plant diversity and forb to grass ratio was
associated with lower cover of non-sown plants, greater
beetle richness in summer, greater abundance of some
invertebrate orders and changed the composition of the
soil fungal community at 0–10 cm depth. Overall, the
meadows increased biodiversity but the varying
responses of the different taxonomic groups to the treat-
ments suggests that maintaining a diverse range of mea-
dow types within a site or across a network of sites would
be most beneficial for urban green space biodiversity.
Plants
Non-sown species contributed substantially to the
plant communities of all treatment plots. The vast
FIG. 2. Plant and invertebrate community properties by treatment and by season for invertebrates and depth for the soil com-
munity. Treatment combinations along the x-axis correspond to Fig. 1. Bars are organized from short (left) to tall (right) treat-
ments, with small gaps between the height groups, and diversity treatment is indicated by gray shading; light gray, low diversity;
medium gray, medium diversity; black, high diversity. White bars represent the unmanipulated control. Bars are the mean per treat-
ment combination with standard deviation bars. Plants are represented by (a) non-sown plant richness and (b) percent cover. The
invertebrate community is represented by (c) order-level richness, (d) Coleoptera family richness, (e) total community abundance,
and (f) estimated total community biomass. The soil taxonomic community is represented by alpha diversity of (g) bacterial and (h)
fungal operational taxonomic units (OTUs) and the difference of the (i) bacterial and (j) fungal DNA communities from the com-
position of the unmanipulated control. Error bars show SD.
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1103

majority of non-sown species (approximately 75%) were
not found in the unmanipulated control plots at the
same site. While this may in part be related to survey
area (species found in single 250-m
2
control plots are
unlikely to include all those present in this site), it does
suggest that many of the non-sown species have not
colonized directly from a site’s mown amenity grassland
species pool. A large proportion (67%) of the non-sown
species had ruderal tendencies (Grime 1979) and within
these, 41% had annual lifecycles (Appendix S1: Table S6;
Grime 1979, Grime et al. 1995). Many of these produce
large quantities of seed that disperse widely (Grime
FIG. 3. Nonmetric multidimensional scaling ordinations of (a) the whole plant community, (b) the non-sown plant community,
(c) the order-level invertebrate community in summer and (d) in autumn, and (e) the Coleoptera family-level community in summer
and (f) in autumn. Points represent communities in each plot, coded by height (red, H1 [short]; blue, H2 [medium]; black, H3 [tall])
and diversity (square, D1 [low]; triangle, D2 [medium]; circle, D3 [high]). Stress values are in the bottom left of the plot space. Only
the first two axes (NMDS1 and NMDS2) are shown, although all ordinations required three axes to reduce stress to <0.20, except
for the order and Coleoptera communities in autumn.
Article e01946; page 1104 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

TABLE 3. Results of linear mixed models for invertebrate community variables in summer, autumn, and winter and for individual
orders
†
in summer and autumn.
Fixed effects variable df FP
Contrasts
Low-medium Low-high Medium-high
Order richness
Summer, ln(x+1)-transformed
Diversity 2,41.5 0.97 0.388 1.000 0.433 0.508
Height 2,41.5 16.22 <0.001 0.891 H>L,<0.0001 H > M,<0.001
Autumn, untransformed
Diversity 2,41.2 2.86 0.069 0.340 0.059 0.750
Height 2,41.2 17.41 <0.001 M > L,<0.001 H > L,<0.0001 0.921
Winter, ln(x+1)-transformed
Diversity 2,25.0 0.88 0.426 0.476 0.836 0.795
Height 2,25.0 6.73 0.005 0.979 H>L,0.019 0.068
Coleoptera, family richness
Summer, untransformed
Diversity 2,24.0 6.65 0.005 0.984 H>L,0.009 H > M,0.014
Height 2,24.0 0.14 0.874 0.938 0.867 0.984
Autumn, untransformed
Diversity 2,27.0 0.88 0.426 0.536 0.990 0.457
Height 2,27.0 2.62 0.091 0.098 0.206 0.914
Invertebrate abundance
Summer, ln(x+1)-transformed
Diversity 2,41.4 0.11 0.892 0.992 0.931 0.897
Height 2,41.4 5.35 0.009 0.538 H>L,0.007 0.128
Autumn, square-root-transformed
Diversity 2,42.1 4.05 0.025 0.107 L>H,0.028 0.927
Height 2,42.1 2.93 0.064 0.059 0.792 0.198
Winter, 1/4-power-transformed
Diversity 2,25.0 0.85 0.440 0.510 0.719 0.917
Height 2,25.0 6.80 0.004 0.983 H>L,0.018 0.065
Invertebrate biomass
Summer, ln(x+1)-transformed
Diversity 2,33.6 2.66 0.085 0.187 0.871 0.079
Height 2,33.6 6.82 0.003 M > L,0.006 H > L,0.016 0.822
Autumn, ln(x+1)-transformed
Diversity 2,38.0 0.00 0.996 0.997 0.997 1.000
Height 2,38.0 5.34 0.009 M > L,0.040 H > L,0.012 0.941
Acari
Summer, 1/4-power-transformed
Diversity 2,33.4 0.03 0.972 0.983 0.998 0.971
Height 2,33.4 0.73 0.490 0.781 0.461 0.897
Autumn, square-root-transformed
Diversity 2,33.6 0.08 0.924 0.967 0.920 0.994
Height 2,33.6 0.36 0.701 0.990 0.704 0.813
Araneae
Summer, 1/4-power-transformed
Diversity 2,33.5 0.73 0.492 0.940 0.470 0.753
Height 2,33.5 4.18 0.024 0.610 H>L,0.020 0.213
Autumn, square-root-transformed
Diversity 2,38.0 0.76 0.473 0.501 0.607 0.964
Height 2,38.0 2.09 0.138 0.236 0.165 0.995
Coleoptera
Summer, ln(x+1)-transformed
Diversity 2,33.0 3.62 0.038 0.784 H>L,0.034 0.226
Height 2,32.9 9.28 0.001 M > L,0.013 H > L,0.001 0.678
Autumn, 1/4-power-transformed
Diversity 2,33.2 1.78 0.184 0.648 0.159 0.709
Height 2,33.1 15.99 <0.001 M > L,<0.0001 H > L,<0.001 0.640
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1105

1977), and they may have colonized by this route; how-
ever, ruderals may also have seeds that remain viable in
seed banks for many decades (Thompson et al. 1997)
and soil disturbance during plot cultivation may have
stimulated their germination. Higher cover of ruderal
species has also been observed early in the restoration of
agricultural areas with meadow-like vegetation (Pywell
et al. 2011).
The number of colonizing species was lower in taller
plots, and their cover was lowest in the plots with the
highest diversity treatment. In our study system, ecologi-
cal conditions at the time of meadow establishment were
similar across all treatment plots, and there were consis-
tently low levels of bare ground during the survey period
(Appendix S1: Section S3, Figure S4). In addition, the
potential pool of incoming species should be relatively
consistent across plots within each site. This suggests
that a combination of the diversity of competitors, and
factors associated with vegetation structure (light avail-
ability being the most likely), determine colonization
TABLE 3. (Continued)
Fixed effects variable df FP
Contrasts
Low-medium Low-high Medium-high
Collembola
Summer, 1/3-power-transformed
Diversity 2,33.9 1.21 0.311 0.977 0.318 0.517
Height 2,33.8 1.42 0.255 0.392 0.282 0.992
Autumn, ln(x+1)-transformed
Diversity 2,34.3 2.22 0.124 0.362 0.118 0.893
Height 2,34.2 2.22 0.124 0.870 0.119 0.343
Diptera
Summer, ln(x+1)-transformed
Diversity 2,38.0 4.37 0.020 1.000 H>L,0.032 0.054
Height 2,38.0 4.85 0.013 0.994 H>L,0.023 H > M,0.039
Autumn, ln(x+1)-transformed
Diversity 2,34.1 1.55 0.227 0.271 0.347 0.950
Height 2,34.1 1.19 0.316 0.879 0.547 0.309
Hemiptera
Summer, 1/5-power-transformed
Diversity 2,38.0 2.83 0.071 0.622 L>H,0.058 0.442
Height 2,38.0 16.92 <0.001 M > L,0.001 H > L,<0.0001 0.351
Autumn, 1/5-power-transformed
Diversity 2,33.9 7.41 0.002 L > M,0.026 L > H,0.003 0.867
Height 2,33.8 5.24 0.010 M > L,0.020 1.000 M>H,0.018
Hymenoptera
Summer, ln(x+1)-transformed
Diversity 2,33.8 2.65 0.086 0.930 0.087 0.264
Height 2,33.8 3.5 0.041 0.447 H>L,0.032 0.435
Autumn, ln(x+1)-transformed
Diversity 2,34.3 0.08 0.924 0.970 0.982 0.917
Height 2,34.1 10.46 <0.001 M > L,0.001 0.965 M>H,0.001
Psocodea
Autumn, 1/5-power-transformed
Diversity 2,38.0 0.74 0.483 0.451 0.869 0.732
Height 2,38.0 34.33 <0.001 M > L,<0.001 H > L,<0.0001 H > M,0.004
Thysanoptera
Summer, ln(x+1)-transformed
Diversity 2,33.4 2.86 0.071 0.937 0.074 0.231
Height 2,33.4 13.52 <0.001 0.056 H>L,<0.0001 0.051
Autumn, 1/4-power transformed
Diversity 2,33.8 17.17 <0.001 L > M,<0.001 L > H,<0.0001 1.000
Height 2,33.8 5.58 0.008 0.330 H>L,0.006 0.228
Notes: Significant effects (P<0.05) are indicated in boldface type. Contrasts are the results of least-squares means (see subsec-
tion Plants and invertebrates: effects of vegetation height and diversity treatments for details). Where there was a significant difference
between treatments, the direction of the effect is indicated. For analyses of separate orders, only results for taxa with more than
1,000 individuals are presented (Appendix S1: Table S4). Refer to Fig. 2 for the differences between treatment means.
†Taxa with fewer than 1,000 individuals in total and not included here: Chilopoda, Diplopoda, Gastropoda, Dermaptera,
Diplura, Ephemeroptera, Isopoda, Lepidoptera, Neuroptera, Odonata, Opiliones, Orthoptera, Psocodea (summer), Zygentoma.
Article e01946; page 1106 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

rates. One caveat, as noted in Methods, is that hand
weeding that targeted three or four species (depending
on the site) was carried out in the plots’first season,
which, by definition, removed some individuals of par-
ticular non-sown species. However, the combination of
the facts that the weeding was carried out across all sites,
was only conducted in the first year, was targeted at a
limited number of key weed species, and was not system-
atically related to treatment in a way that would produce
the observed effects, inclines us to the view that the
result is a genuine reflection of treatment effects on colo-
nization and establishment.
The loss of sown species and the establishment of
non-sown species clearly have implications for the long-
evity of urban meadow plantings and therefore their
utility for enhancing urban grassland diversity. Our
experiment indicates the potential for change in species
composition in the short term, but does not allow
longer-term trajectories to be assessed directly. A num-
ber of processes are likely to be important over longer
time scales: the disappearance of initially colonizing
ruderal species in the absence of further disturbance to
the soil (Hofmann and Isselstein 2004, Pywell et al.
2007), the capacity of planted biennial species, e.g., tea-
sel Dipsacus fullonum, to self-seed into the new sward
(van der Meijden et al. 1992), the colonization of addi-
tional perennial species through dispersal from other
sites (Tilman 1997), and the longer term outcomes of
competitive interactions between currently established
species (Harrison and Bardgett 2010, Maynard et al.
2017). External drivers may also interact with these
processes to affect longer-term outcomes. The plant
community will interact with soil microbial and inverte-
brate communities, and there is emerging evidence that
soil microbial communities may facilitate, not simply
follow, vegetation development (Harris 2009, van der
Putten et al. 2013). Establishing seminatural grasslands
in the longer term is, however, also highly dependent on
environmental conditions (Stuble et al. 2017), and
future management interventions will impact the com-
munity trajectory, particularly the frequency of mow-
ing, the choice of whether vegetation clippings are
removed (see next paragraph), and any management
designed to enhance plant diversity or density, for
example re-seeding or scarifying (Westbury et al. 2006,
Trowbridge et al. 2017).
High nitrogen levels of urban soils (Ladd 2016)
potentially limit plant diversity. Indeed, two of the five
sown plant species (Primula vulgaris and Ranunculus
acris) that failed to become established are very sensi-
tive to high fertility soils, as indicated by their Ellen-
berg N scores (Hill et al. 1999). The tallest plots
consistently had the lowest soil total N, presumably
due to greater uptake by growing vegetation, combined
with removal of vegetation clippings. Management that
includes arising removal should gradually lower soil
nitrogen (Walker et al. 2004) and increase the site’s
suitability for long-term maintenance of diverse mea-
dow vegetation. However, implementation of such
management at larger scales is seen as a major chal-
lenge (Hoyle et al. 2017). The tall plots here produced
approximately 67 kg (dry mass) per 100 m
2
of meadow
in an annual cut (Appendix S1: Section S4). While
there is in principle potential for land managers to use
this material in composting or as biomass, this is chal-
lenging in urban areas due to concerns regarding con-
tamination with litter and dog feces (Hoyle et al.
2017).
FIG. 4. Biplots of the principal component scores of the
phospholipid fatty acid profiles of the soil microbial community
at (a) 0–10 cm depth and (b) 11–20 cm depth. The percentages
on the axes refer to the variance described by the component
represented on that axis. Points represent communities in each
plot, coded by height (red, H1 [short]; blue, H2 [medium];
black, H3 [tall]) and diversity (square, D1 [low]; triangle, D2
[medium]; circle, D3 [high]).
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1107

Invertebrates
Urban meadows typically supported invertebrate
communities that were more abundant, diverse, and had
greater biomass than those in the unmanipulated ame-
nity grassland. While our measures of invertebrate diver-
sity are based on coarse taxonomic data, higher taxon
diversity can indicate trends at finer taxonomic
TABLE 4. Results of linear mixed models for soil organism community variables at two depths (0–10 cm, 11–20 cm).
Random effects Wald PFixed effects df FP
0–10 cm depth
PLFA principal component 1
Plot 0.89 diversity 2,2 1.28 0.48
Site 0.32 height 2,2 6.94 0.01
diversity 9height 4,4 4.08 0.01
PLFA principal component 2
Plot <0.0001 diversity 2,2 21.33 <0.0001
Site 0.35 height 2,2 5.63 0.01
diversity 9height 4,4 17.55 <0.0001
Bacterial DNA alpha diversity
Plot 0.31 diversity 2,2 0.65 0.53
Site 0.21 height 2,2 1.03 0.36
diversity 9height 4,4 0.84 0.50
Fungal DNA alpha diversity
Plot 0.35 diversity 2,2 2.60 0.082
Site 0.58 height 2,2 3.34 0.04
diversity 9height 4,4 0.7 0.59
Bacterial DNA Bray-Curtis
Plot 0.965 diversity 2,2 0.55 0.582
Site 0.381 height 2,2 4.77 0.012
diversity 9height 4,4 1.32 0.277
Fungal DNA Bray-Curtis
Plot 0.04 diversity 2,2 3.41 0.04
Site 0.32 height 2,2 0.86 0.43
diversity 9height 4,4 3.18 0.02
11–20 cm depth
PLFA principal component 1
Plot 0.32 diversity 2,2 1.28 0.28
Site 0.27 height 2,2 6.94 0.001
diversity 9height 4,4 4.08 0.004
PLFA principal component 2
Plot 0.37 diversity 2,2 5.76 0.004
Site 0.32 height 2,2 1.58 0.21
diversity 9height 4,4 2.93 0.02
Bacterial DNA alpha diversity
Plot 0.32 diversity 2,2 0.78 0.46
Site 0.39 height 2,2 3.65 0.03
diversity 9height 4,4 0.95 0.44
Fungal DNA alpha diversity
Plot 0.39 diversity 2,2 4.29 0.02
Site 0.45 height 2,2 =0.13 0.13 0.87
diversity 9height 4,4 0.84 0.50
Bacterial DNA Bray-Curtis
Plot 0.58 diversity 2,2 4.14 0.77
Site 0.33 height 2,2 0.29 0.02
diversity 9height 4,4 1.67 0.17
Fungal DNA Bray-Curtis
Plot 0.04 diversity 2,2 0.21 0.81
Site 0.34 height 2,2 4.91 0.01
diversity 9height 4,4 2.01 0.10
Notes: Significant effects (P<0.05) are indicated in boldface type. Where there was a significant difference between treatments,
the direction of the effect is indicated. Refer to Fig. 2 for the differences between treatment means.
Article e01946; page 1108 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

resolutions (Timms et al. 2013, van Rijn et al. 2014),
and maintaining a range of orders also has conservation
significance. These effects of meadow creation occurred
within the meadows’second growing season suggesting
that invertebrate communities could respond rapidly to
habitat creation despite potential colonization barriers
arising from the fragmented nature of urban greenspace
(Braaker et al. 2014, Vergnes et al. 2014).
Effects on invertebrate communities arose more fre-
quently in response to differences in vegetation height
rather than plant richness. This is consistent with effects
seen in more rural grasslands (Blaauw and Isaacs 2012,
Buri et al. 2013, Andrey et al. 2014) and with previous
studies of a range of invertebrate groups in which both
abundance (Garbuzov et al. 2015) and species diversity
(Unterweger et al. 2017) increased with reduced mowing
and resultant longer vegetation. We find, however, that
taxa varied in their response to medium and tall vegeta-
tion with some groups responding more favorably to
vegetation of intermediate height in autumn. This could
be a response to specific abiotic conditions such as tem-
perature and humidity that vary with sward height and
density and will attract and retain species with different
requirements than those occurring in longer vegetation
(Crist and Ahern 1999, Gardiner and Hassall 2009). Pre-
vious work in urban areas focused on shorter swards has
found that even relatively small changes in mowing fre-
quency (e.g., every 3 weeks rather than every week (Ler-
man et al. 2018) can lead to meaningful increases in
resources for invertebrates (Shwartz et al. 2013, Lerman
et al. 2018). Our work suggests that further increases in
vegetation height are likely to yield additional biodiver-
sity benefits. In addition, we note that taller plots were
important in winter as well as summer and autumn. It is
difficult to tease apart the effects of taller vegetation cre-
ating more favorable conditions than short mown grass
in winter, from the positive effect of taller vegetation in
summer on invertebrate abundance carrying over to
influence the winter community. However, the inverte-
brate abundances in the medium height plots (which
were cut low in the winter) were very similar to those of
the short plots, and less than the tall plots (uncut;
Appendix S1: Fig. S2), suggesting that maintaining
longer vegetation during winter is critical. These results
emphasize the generally neglected point that considera-
tion of resource availability throughout the year is
important for invertebrates (Unterweger et al. 2018),
particularly if these areas are to support sustainable pop-
ulations rather than relying on annual summer recolo-
nization (Leather et al. 1995).
Effects of plant diversity on invertebrate communities
were more limited. This may reflect the greater impor-
tance of structure, or simply that the structural variation
captured a more ecologically significant range of condi-
tions. It may also be a consequence of the taxonomic res-
olution of the data: at coarse taxonomic resolutions it is
harder to detect more specialized responses to particular
plant species. The low plant diversity treatment did,
however, lack sown forbs and increasing forb cover in
higher plant diversity treatments could have contributed
to some of the observed responses through the provision
of suitable food plants, structural features, and flower-
related resources (nectar, pollen, and seeds). Pollinators,
or flower associated species, occur in a number of the
Coleoptera families, and other orders that responded to
diversity treatments (notably Diptera; Orford et al.
2015). While an equivalent response was not docu-
mented in Hymenoptera, which includes a number of
pollinators, this order is dominated by parasitic taxa that
rarely visit flowers (Shaw and Hochberg 2001). In early
autumn, when many flowers were no longer in bloom,
the Hemiptera and Thysanoptera were most abundant in
the low diversity, grass-only treatments. This may be due
to large numbers of common UK Hemiptera associated
with grasses as food plants (Chinery 2012), and that
while many thrips are flower dependent, some taxa live
in and feed on grass seed heads (Mound and Palmer
1972, Stevenson et al. 1997); indeed, we observed very
high abundances of thrips within grass seed heads in our
samples.
Declines in invertebrate biomass (Hallmann et al.
2017) and abundance (Conrad et al. 2004, Brooks et al.
TABLE 5. Summary of the response of soil organism diversity
and composition at two depths (0–10 cm, 11–20 cm) to the
two axes of experimental meadow treatments: diversity and
height.
Measure
Effect
Diversity Height Interaction
Soils, 0–10 cm
PLFA composition PC1 yes yes
PLFA composition PC2 yes yes
Bacterial DNA alpha
diversity
Fungal DNA alpha
diversity
↑
Bacterial DNA
composition BC
yes
Fungal DNA
composition BC
yes yes
Soils, 11–20 cm
PLFA composition PC1 yes yes
PLFA composition PC2 yes yes
Bacterial DNA alpha
diversity
↓
Fungal DNA alpha
diversity
↓
Bacterial DNA
composition BC
yes
Fungal DNA
composition BC
yes
Notes: Significant effects for compositional change are indi-
cated by “yes.”Significant effects for alpha diversity are desig-
nated with a symbol indicating the direction of effect; an up
arrow indicates the response variable increases with increasing
meadow height/diversity, a down arrow indicates the response
variable decreases with increasing meadow height/diversity. Full
results are in Table 4.
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1109

2012, Ewald et al. 2015) are of increasing conservation
concern, in addition to the loss of species diversity (Fon-
seca 2009). While it is not possible with these data to
tease apart the mechanisms linking changes in plant
height and diversity to the changes in the invertebrate
community, it is clear that replacing mown amenity
grassland with urban meadows can increase the diver-
sity, biomass, and abundance and alter the composition
of urban invertebrate communities. It is also clear that
invertebrate responses vary according to the particular
height and diversity characteristics of the meadows cre-
ated. This suggests that, at the scale of individual sites or
across a network of urban greenspaces, beneficial
impacts will be maximized by creating a diverse range of
meadow types.
Soils
The disturbance to the soils during plot establishment,
particularly from tilling, combined with the relatively
short duration of the experiment compared to the
response time for some soil properties, means it is not
possible to draw conclusions about long-term effects of
the meadows on the soils. Despite this caveat, soil prop-
erties and microbial communities exhibited a number of
notable responses to the meadow treatments. The PLFA
analysis indicated that, at both soil depths, the composi-
tion of soil microbial communities was influenced by
plant height and diversity. The DNA results indicated
changes to community composition of bacteria and
fungi between treatments, although diversity was largely
unchanged. The bacterial community composition
responded to changes in plant height treatment at both
depths. Fungal community composition at 0–10 cm
depth responded to plant diversity, and fungi at 11–
20 cm depth responded to plant height. The contrasting
responses of bacteria and fungi to conditions at different
depths, and the closer association of fungal composition
with changes in plant diversity, are consistent with previ-
ous findings although the driving factors are still poorly
understood, ranging from pH, through nitrogen, to
antecedent use (Newbound et al. 2012, Xu et al. 2014,
Sarah et al. 2015, Yan et al. 2016, Hui et al. 2017). In
our study system, recent antecedent use is uniform
across treatments suggesting that divergent effects of
plant communities on soil nitrogen may contribute to
differences in microbial communities. A key change in
plant diversity in the treatments was from domination
by grasses to domination by forbs, with concomitant
changes in the structure and depth of the rooting zone.
While bacteria and fungi have complex interactions
within the rhizosphere (de Boer et al. 2005), our results
suggest a more pronounced shift in the fungal commu-
nity at shallower depths where the majority of change in
the root structure will be observed. Tilling during site
preparation may have contributed to the variation in
community composition between depths by creating new
microhabitats (Bruns 1995), although disturbances
similar to tilling can reduce the ability of fungi to estab-
lish interactions with host plants (Jasper et al. 1989,
McGonigle and Miller 1996). Despite these potential
adverse impacts of site management it is clear that
replacing mown amenity grassland with meadow style
vegetation can alter soil microbial communities, and
enhance microbial diversity especially when meadows
contain a greater number of plant species.
Management recommendations for establishing meadows
in public greenspaces
We introduced urban meadows in collaboration with
local authority partners to investigate the scope for
enhancing the attractiveness and quality of sites for peo-
ple, and enhancing biodiversity and ecological function.
Practitioner orientated management guidelines for creat-
ing such urban meadows are provided by Hoyle (2016).
Our results demonstrate that maintaining meadow com-
munities that are taller and botanically more diverse
than short mown grasslands in urban public parks can
increase the abundance and richness of invertebrate
communities throughout the year, while also altering soil
microbial communities. Crucially, meadows with differ-
ent height and diversity characteristics supported differ-
ent communities of invertebrates and microbes, and thus
a mosaic of meadow types is likely to enhance biodiver-
sity to a greater extent than habitat creation that focuses
on replacing mown amenity grassland with just a single
type of meadow.
Long-term maintenance and retention of urban mead-
ows will require acceptance by local residents and willing-
ness on the part of local authorities to maintain them.
Previous work on these experimental plots has suggested
that creating urban meadows typically increases local res-
idents’appreciation of the site and perception of site
quality, although not all not residents respond favorably
(Southon et al. 2017). Local residents generally gave
plots with high plant richness and medium height vegeta-
tion higher scores for aesthetic preference, compared to
short low diversity vegetation that represents mown-ame-
nity grassland (Southon et al. 2017). Medium height
plots do frequently enhance the diversity or abundance of
the taxa examined here compared to short treatments,
although taller plots were most consistently associated
with increased richness and abundance, particularly of
invertebrates (Table 3). While the tall treatments were
not generally favored by people and were even less attrac-
tive to them during winter, Southon et al. (2017) found
that people were more prepared to tolerate them when
provided with information about their benefits to biodi-
versity. Consequently, even though there is a perception
among some land managers that the public dislike more
natural vegetation (
€
Ozg€
uner et al. 2007), our work
demonstrates that this is not necessarily the case and
there is potential to generate win-win scenarios for biodi-
versity and people by introducing biodiverse urban
meadows in place of short mown grassland.
Article e01946; page 1110 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

Maximizing the potential benefits of urban meadows
will require careful consideration of a number of factors.
First, our results suggest that maintaining a diversity of
meadow types (with varying vegetation height and plant
richness) is likely to maximize cumulative biodiversity
benefits and the resulting landscape heterogeneity may
also further increase the aesthetic appeal of urban mead-
ows (Dramstad et al. 2001). Achieving this is likely to
require careful landscape design, yet integrating design
principles with the needs of biodiversity is rarely done
(Wang et al. 2017). It will also require urban managers
to coordinate novel and more complex mowing regimes,
which can be challenging, especially when this task is
subcontracted to a third party (Hoyle et al. 2017). Sec-
ond, an important step in introducing taller non-woody
vegetation is communicating intent to the public, for
example through on-site signage. In part, this will
increase acceptance among people who are sympathetic
to biodiversity conservation, but more generally will help
generate the “cues to care”that may be important in
increasing public acceptance of wilder vegetation (Nas-
sauer 1995). Creating mown paths that “frame”longer
vegetation may further help indicate that meadow areas
are under active management and thus cared for. Third,
the context of the site was also a factor in public accep-
tance, with sites directly overlooked by residential hous-
ing coming under greater scrutiny (Hoyle et al. 2017)
and such sites are thus perhaps less suitable than loca-
tions that are less visible from houses or may only be
suitable for the most preferred types of urban meadows.
Another important aspect of site context is likely to be
the proportion of mown amenity grassland converted to
urban meadows. Amenity grassland provides important
recreational space for exercise and sporting activities, for
which urban meadows are not suitable, suggesting it will
be best to establish urban meadow vegetation at sites
where some short grassland can also be retained. A final
challenge in realizing the potential dual benefits of urban
meadows for people and biodiversity concerns biomass
removal. Reducing mowing and the associated energy
and labor costs is one of the assumed benefits of manag-
ing green spaces as a perennial meadow rather than as
mown grass (Hoyle et al. 2017). At the end of the season,
though, all biomass has to be removed, and concerns
regarding contamination with litter, dog feces and other
material can increase the challenges of using the biomass
for energy production of composting although some con-
tractors appear willing to take this material.
We demonstrate that sown perennial meadows can
support plant, invertebrate and soil microbial communi-
ties both different and more diverse than those of mown
amenity grassland. Creating and maintaining urban
meadows is not without its challenges but our results
suggest that, with careful management and implementa-
tion, replacing some of the mown amenity grassland that
currently dominates many towns and cities with a range
of different types of meadow vegetation can generate
positive outcomes for both biodiversity and people.
ACKNOWLEDGMENTS
The authors are grateful to Anna Jorgensen for input on
the design and management of the experiment; Erik Button
and Meghann Mears for contributing to the invertebrate bio-
mass estimates; Paul Barton, Maria Biskupska, and Nigel
Janes for their work processing soil samples; and to many
undergraduate project students for processing invertebrate
samples. We would like to thank Bedford and Luton Borough
Councils for their collaboration on, and support for, this pro-
ject, particularly for hosting and mowing the plots. We thank
two anonymous reviewers and John Marzluff for comments
that considerably improved the manuscript. This research
(Grant Numbers NE/J015369/1 and NE/J015067/1) was con-
ducted as part of the Fragments, Functions and Flows in
Urban Ecosystem Services (F
3
UES) Project as part of the lar-
ger Biodiversity and Ecosystem Service Sustainability (BESS)
framework. BESS is a six-year programme (2011–2017) funded
by the UK Natural Environment Research Council (NERC)
and the Biotechnology and Biological Sciences Research
Council (BBSRC) as part of the UK’s Living with Environ-
mental Change (LWEC) programme. This work presents the
outcomes of independent research funded by NERC and the
BESS programme, and the views expressed are those of the
authors and not necessarily those of the BESS Directorate or
NERC. Order of authorship is alphabetical, excepting the lead
author. The experiment was designed by N. Dunnett, K. L.
Evans, P. H. Warren, H. Hoyle, and G. E. Southon and estab-
lished and managed in the field by H. Hoyle, N. Dunnett, J.
A. Harris, T. G. Mercer, J. P. Richards, and B. A. Norton.
Plant and invertebrate sampling was designed by P. H. War-
ren, K. L. Evans, B. A. Norton, G. E. Southon, and J. P.
Richards, and soil sampling by R. Corstanje, J. A. Harris, T.
G. Mercer, S. M. Grice, and D. R. Grafius. Invertebrate and
plant samples/data were collected, processed, and curated by
B. A. Norton, J. P. Richards, G. E. Southon, O. L. Pescott, R.
Clark, E. Lim, and E. Gravestock and soil samples/data by S.
M. Grice, S. Hilton, G. D. Bending, T. G. Mercer, R. Cor-
stanje, M. Pawlett, and D. R. Grafius. Data analyses were
designed and carried out by B. A. Norton, R. Corstanje, P. H.
Warren, K. L. Evans, J. A. Harris, M. Pawlett, S. Hilton, and
G. D. Bending. The manuscript was initially drafted by B. A.
Norton with additional writing and data interpretation from
K. L. Evans, P. H. Warren, J. A. Harris, R. Corstanje, G. E.
Southon, G. D. Bending, S. Hilton, and M. Pawlett. All
authors contributed to the critical review of subsequent drafts.
LITERATURE CITED
Alumai, A., S. O. Salminen, D. S. Richmond, J. Cardina, and P.
S. Grewal. 2009. Comparative evaluation of aesthetic, biolog-
ical, and economic effectiveness of different lawn manage-
ment programs. Urban Ecosystems 12:127–144.
Andrey, A., J.-Y. Humbert, C. Pernollet, and R. Arlettaz. 2014.
Experimental evidence for the immediate impact of fertiliza-
tion and irrigation upon the plant and invertebrate communi-
ties of mountain grasslands. Ecology and Evolution 4:2610–
2623.
Aronson, M. F. J., C. A. Lepczyk, K. L. Evans, M. A. Goddard,
S. B. Lerman, J. S. MacIvor, C. H. Nilon, and T. Vargo. 2017.
Biodiversity in the city: key challenges for urban green space
management. Frontiers in Ecology and the Environment
15:189–196.
Aronson, M. F. J., et al. 2014. A global analysis of the impacts
of urbanization on bird and plant diversity reveals key
anthropogenic drivers. Proceedings of the Royal Society B
281:20133330.
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1111

Bates, D., M. Maechler, B. Bolker, and S. Walker. 2015. Fitting
linear mixed-effects models using lme4. Journal of Statistical
Software 67:1–48.
Beninde, J., M. Veith, and A. Hochkirch. 2015. Biodiversity in
cities needs space: a meta-analysis of factors determining
intra-urban biodiversity variation. Ecology Letters 18:581–
592.
Bertoncini, A. P., N. Machon, S. Pavoine, and A. Muratet.
2012. Local gardening practices shape urban lawn floristic
communities. Landscape and Urban Planning 105:53–61.
Bijoor, N., D. Pataki, D. Haver, and J. Famiglietti. 2014. A com-
parative study of the water budgets of lawns under three man-
agement scenarios. Urban Ecosystems 17:1095–1117.
Blaauw, B. R., and R. Isaacs. 2012. Larger wildflower plantings
increase natural enemy density, diversity, and biological con-
trol of sentinel prey, without increasing herbivore density.
Ecological Entomology 37:386–394.
Blackmore, L. M., and D. Goulson. 2014. Evaluating the effec-
tiveness of wildflower seed mixes for boosting floral diversity
and bumblebee and hoverfly abundance in urban areas. Insect
Conservation and Diversity 7:480–484.
Bormann, F. H., D. Balmori, and G. T. Geballe. 2001.
Redesigning the American lawn: a search for environmental
harmony. Yale University Press, London, UK.
Braaker, S., J. Ghazoul, M. K. Obrist, and M. Moretti. 2014.
Habitat connectivity shapes urban arthropod communities:
the key role of green roofs. Ecology 95:1010–1021.
Bray, J. R., and J. T. Curtis. 1957. An ordination of the upland
forest communities of Southern Wisconsin. Ecological Mono-
graphs 27:325–349.
Brooks, D. R., J. E. Bater, S. J. Clark, D. T. Monteith, C.
Andrews, S. J. Corbett, D. A. Beaumont, and J. W. Chapman.
2012. Large carabid beetle declines in a United Kingdom mon-
itoring network increases evidence for a widespread loss in
insect biodiversity. Journal of Applied Ecology 49:1009–1019.
Bruns, T. D. 1995. Thoughts on the processes that maintain
local species diversity of ectomycorrhizal fungi. Plant and
Soil 170:63–73.
Buri, P., R. Arlettaz, and J.-Y. Humbert. 2013. Delaying mow-
ing and leaving uncut refuges boosts orthopterans in exten-
sively managed meadows: evidence drawn from field-scale
experimentation. Agriculture Ecosystems & Environment
181:22–30.
Buri, P., J.-Y. Humbert, M. Sta
nska, I. Hajdamowicz, E. Tran,
M. H. Entling, and R. Arlettaz. 2016. Delayed mowing pro-
motes planthoppers, leafhoppers and spiders in extensively
managed meadows. Insect Conservation and Diversity 9:536–
545.
Caporaso, J. G., et al. 2010. QIIME allows analysis of high-
throughput community sequencing data. Nature Methods
7:335–336.
Champely, S. 2018. PairedData: Paired data analysis. R package
version 1.1.1. https://cran.r-project.org/web/packages/Paired
Data/index.html
Chinery, M. 2012. Insects of Britain and Western Europe. Third
edition. Domino Books, Jersey, Channel Islands.
Conrad, K. F., I. P. Woiwod, M. Parsons, R. Fox, and M.
S. Warren. 2004. Long-term population trends in wide-
spread British moths. Journal of Insect Conservation
8:119–136.
Crist, T. O., and R. G. Ahern. 1999. Effects of habitat patch size
and temperature on the distribution and abundance of
ground beetles (Coleoptera: Carabidae) in an old field. Envi-
ronmental Entomology 28:681–689.
Dallimer, M., K. N. Irvine, A. M. J. Skinner, Z. G. Davies, J. R.
Rouquette, L. L. Maltby, P. H. Warren, P. R. Armsworth,
and K. J. Gaston. 2012. Biodiversity and the feel-good factor:
understanding associations between self-reported human
well-being and species richness. BioScience 62:47–55.
de Boer, W., L. B. Folman, R. C. Summerbell, and L. Boddy.
2005. Living in a fungal world: impact of fungi on soil bacte-
rial niche development. FEMS Microbiology Reviews
29:795–811.
Dramstad, W. E., G. Fry, W. J. Fjellstad, B. Skar, W. Helliksen,
M. L. B. Sollund, M. S. Tveit, A. K. Geelmuyden, and E.
Framstad. 2001. Integrating landscape-based values—Nor-
wegian monitoring of agricultural landscapes. Landscape and
Urban Planning 57:257–268.
Duffey, E. 1975. The effects of human trampling on the fauna
of grassland litter. Biological Conservation 7:255–274.
Dunnett, N. 2011. Urban meadows: an ecological discussion.
Aspects of Applied Biology 108:11–17.
Dunnett, N., and J. Hitchmough. 2007. The dynamic landscape:
design, ecology and management of naturalistic urban plant-
ing. Taylor & Francis, London, UK.
Evans, K. L., S. E. Newson, and K. J. Gaston. 2009. Habitat
influences on urban avian assemblages. Ibis 151:19–39.
Ewald, J. A., C. J. Wheatley, N. J. Aebischer, S. J. Moreby, S. J.
Duffield, H. Q. P. Crick, and M. B. Morecroft. 2015. Influ-
ences of extreme weather, climate and pesticide use on inver-
tebrates in cereal fields over 42 years. Global Change Biology
21:3931–3950.
Fisher, R. A., A. S. Corbet, and C. B. Williams. 1943. The rela-
tion between the number of species and the number of indi-
viduals in a random sample of an animal population. Journal
of Animal Ecology 12:42–58.
Fonseca, C. R. 2009. The silent mass extinction of insect herbi-
vores in biodiversity hotspots. Conservation Biology
23:1507–1515.
Frosteg

ard,

A., A. Tunlid, and E. B

a

ath. 1993. Phospholipid
fatty acid composition, biomass, and activity of microbial
communities from two soil types experimentally exposed to
different heavy metals. Applied and Environmental Microbi-
ology 59:3605–3617.
Fuller, R. A., K. N. Irvine, P. Devine-Wright, P. H. Warren, and
K. J. Gaston. 2007. Psychological benefits of greenspace
increase with biodiversity. Biology Letters 3:390–394.
Garbuzov, M., K. A. Fensome, and F. L. W. Ratnieks. 2015.
Public approval plus more wildlife: Twin benefits of
reduced mowing of amenity grass in a suburban public
park in Saltdean, UK. Insect Conservation and Diversity
8:107–119.
Gardiner, T., and M. Hassall. 2009. Does microclimate affect
grasshopper populations after cutting of hay in improved
grassland? Journal of Insect Conservation 13:97–102.
Grime, J. P. 1977. Evidence for the existence of three pri-
mary strategies in plants and its relevance to ecological
and evolutionary theory. American Naturalist 111:1169–
1194.
Grime, J. P. 1979. Plant strategies and vegetation processes.
Wiley and Sons, Chichester, UK.
Grime, J. P., J. G. Hodgson, R. Hunt, and K. Thompson. 1995.
The electronic comparative plant ecology. Chapman & Hall,
London, UK.
Haaland, C., R. E. Naisbit, and L.-F. Bersier. 2011. Sown wild-
flower strips for insect conservation: a review. Insect Conser-
vation and Diversity 4:60–80.
Haaland, C., and C. K. van den Bosch. 2015. Challenges and
strategies for urban green-space planning in cities undergoing
densification: a review. Urban Forestry & Urban Greening
14:760–771.
Hallmann, C. A., et al. 2017. More than 75 percent decline over
27 years in total flying insect biomass in protected areas.
PLoS ONE 12:e0185809.
Article e01946; page 1112 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

Harris, J. 2009. Soil microbial communities and restoration
ecology: facilitators or followers? Science 325:573–574.
Harris, E. M., D. G. Martin, C. Polsky, L. Denhardt, and A.
Nehring. 2013. Beyond “Lawn People”: The role of emotions
in suburban yard management practices. Professional Geog-
rapher 65:345–361.
Harrison, K. A., and R. D. Bardgett. 2010. Influence of plant
species and soil conditions on plant–soil feedback in mixed
grassland communities. Journal of Ecology 98:384–395.
Hedblom, M., F. Lindberg, E. Vogel, J. Wissman, and K.
Ahrn
e. 2017. Estimating urban lawn cover in space and time:
case studies in three Swedish cities. Urban Ecosystems
20:1109–1119.
Hejkal, J., T. K. Buttschardt, and V. H. Klaus. 2017. Connectiv-
ity of public urban grasslands: implications for grassland con-
servation and restoration in cities. Urban Ecosystems
20:511–519.
Heritage Lottery Fund. 2014. State of UK public parks 2016.
Research report. Heritage Lottery Fund, London, UK.
Hill, M. O., J. O. Mountford, D. B. Roy, and R. G. H. Bunce.
1999. Ellenberg’s indicator values for British plants. ECO-
FACT Volume 2 Technical Annex. Institute of Terrestrial
Ecology, Huntingdon, UK. 46 pp. (ECOFACT, 2a).
Hofmann, M., and J. Isselstein. 2004. Seedling recruitment on
agriculturally improved mesic grassland: the influence of dis-
turbance and management schemes. Applied Vegetation
Science 7:193–200.
Hoyle, H. 2016. Improving urban grassland for people and
wildlife. The Living With Environmental Change Partnership,
Newcastle University, Newcastle upon Tyne, UK.
Hoyle, H., A. Jorgensen, P. Warren, N. Dunnett, and K. Evans.
2017. “Not in their front yard”The opportunities and chal-
lenges of introducing perennial urban meadows: a local
authority stakeholder perspective. Urban Forestry & Urban
Greening 25:139–149.
Hui, N., A. Jumpponen, G. Francini, D. J. Kotze, X. Liu, M.
Romantschuk, R. Str€
ommer, and H. Set€
al€
a. 2017. Soil micro-
bial communities are shaped by vegetation type and park age
in cities under cold climate. Environmental Microbiology
19:1281–1295.
Humbert, J. Y., J. Ghazoul, G. J. Sauter, and T. Walter. 2010.
Impact of different meadow mowing techniques on field
invertebrates. Journal of Applied Entomology 134:592–599.
Hwang, Y. H., Z. E. J. Yue, and Y. C. Tan. 2017. Observation of
floristic succession and biodiversity on rewilded lawns in a
tropical city. Landscape Research 42:106–119.
Ignatieva, M., et al. 2015. Lawn as a cultural and ecological
phenomenon: a conceptual framework for transdisciplinary
research. Urban Forestry & Urban Greening 14:383–387.
Ives, C. D., et al. 2016. Cities are hotspots for threatened spe-
cies. Global Ecology and Biogeography 25:117–126.
Jasper, D. A., L. K. Abbott, and A. D. Robson. 1989. Soil distur-
bance reduces the infectivity of external hyphae of vesicular—
arbuscular mycorrhizal fungi. New Phytologist 112:93–99.
Jenkinson, D. S., and D. S. Powlson. 1976. The effects of bioci-
dal treatments on metabolism in soil—I. Fumigation with
chloroform. Soil Biology and Biochemistry 8:167–177.
Jerrentrup, J. S., N. Wrage-M€
onnig, K. U. R€
over, J. Isselstein,
and A. McKenzie. 2014. Grazing intensity affects insect
diversity via sward structure and heterogeneity in a long-term
experiment. Journal of Applied Ecology 51:968–977.
Jiang, Y., and T. Yuan. 2017. Public perceptions and preferences
for wildflower meadows in Beijing, China. Urban Forestry &
Urban Greening 27:324–331.
Klaus, V. H. 2013. Urban grassland restoration: a neglected
opportunity for biodiversity conservation. Restoration Ecol-
ogy 21:665–669.
Knop, E., D. Kleijn, F. Herzog, and B. Schmid. 2006. Effective-
ness of the Swiss agri-environment scheme in promoting bio-
diversity. Journal of Applied Ecology 43:120–127.
K~
oljalg, U., et al. 2013. Towards a unified paradigm for
sequence-based identification of fungi. Molecular Ecology
22:5271–5277.
Kuznetsova, A., P. B. Brockhoff, and R. H. B. Christensen.
2017. lmerTest package: tests in linear mixed effects models.
Journal of Statistical Software 82:1–26.
Ladd, B. 2016. Nitrogen pollution and the meltdown of urban
ecosystems. Land 5:23.
Leather, S. R., K. F. Walters, and J. S. Bale. 1995. The ecology
of insect overwintering. Cambridge University Press, Cam-
bridge, UK.
Lenth, R. V. 2016. Least-squares means: the R package lsmeans.
Journal of Statistical Software 69:1–33.
Lerman, S. B., A. R. Contosta, J. Milam, and C. Bang. 2018. To
mow or to mow less: lawn mowing frequency affects bee
abundance and diversity in suburban yards. Biological Con-
servation 221:160–174.
Maynard, D. S., T. W. Crowther, and M. A. Bradford. 2017.
Competitive network determines the direction of the diver-
sity–function relationship. Proceedings of the National Acad-
emy of Sciences USA 43:11464.
McDonald, D., M. N. Price, J. Goodrich, E. P. Nawrocki, T. Z.
DeSantis, A. Probst, G. L. Andersen, R. Knight, and P.
Hugenholtz. 2011. An improved Greengenes taxonomy with
explicit ranks for ecological and evolutionary analyses of bac-
teria and archaea. ISME Journal 6:610.
McGonigle, T. P., and M. H. Miller. 1996. Development of
fungi below ground in association with plants growing in dis-
turbed and undisturbed soils. Soil Biology and Biochemistry
28:263–269.
Milesi, C., S. W. Running, C. D. Elvidge, J. B. Dietz, B. T. Tut-
tle, and R. R. Nemani. 2005. Mapping and modeling the bio-
geochemical cycling of turf grasses in the United States.
Environmental Management 36:426–438.
Morris, M. G. 2000. The effects of structure and its dynamics
on the ecology and conservation of arthropods in British
grasslands. Biological Conservation 95:129–142.
Mound, L. A., and J. M. Palmer. 1972. Grass-flower infesting
thrips of the genus Chirothrips Haliday in Australia. Aus-
tralian Journal of Entomology 11:332–339.
M€
uller, N., M. Ignatieva, C. Nilon, P. Werner, and W. Zipperer.
2013. Patterns and trends in urban biodiversity and landscape
design. Pages 123–174 in T. Elmqvist, et al., editors. Urban-
ization, biodiversity and ecosystem services: challenges and
opportunities. Springer, Dordrecht, The Netherlands.
Nassauer, J. I. 1995. Messy ecosystems, orderly frames. Land-
scape Journal 14:161–170.
Newbound, M., L. T. Bennett, J. Tibbits, and S. Kasel. 2012.
Soil chemical properties, rather than landscape context, influ-
ence woodland fungal communities along an urban-rural gra-
dient. Austral Ecology 37:236–247.
Oksanen, J., et al. 2017. vegan: Community ecology package. R
package version 2.4-2. https://cran.r-project.org/web/package
s/vegan/index.html
Orford, K. A., I. P. Vaughan, and J. Memmott. 2015. The for-
gotten flies: the importance of non-syrphid Diptera as polli-
nators. Proceedings of the Royal Society B 282:20142934.
€
Ozg€
uner, H., A. D. Kendle, and R. J. Bisgrove. 2007. Attitudes
of landscape professionals towards naturalistic versus formal
urban landscapes in the UK. Landscape and Urban Planning
81:34–45.
Pett, T. J., A. Shwartz, K. N. Irvine, M. Dallimer, and Z. G.
Davies. 2016. Unpacking the people-biodiversity paradox: a
conceptual framework. BioScience 66:576–583.
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1113

Pouyat, R. V., M. J. McDonnell, and S. T. A. Pickett. 1995. Soil
characteristics of oak stands along an urban-rural land-use
gradient. Journal of Environmental Quality 24:516–526.
Pywell, R. F., J. M. Bullock, J. B. Tallowin, K. J. Walker, E. A.
Warman, and G. Masters. 2007. Enhancing diversity of spe-
cies-poor grasslands: an experimental assessment of multiple
constraints. Journal of Applied Ecology 44:81–94.
Pywell, R. F., W. R. Meek, R. G. Loxton, M. Nowakowski, C.
Carvell, and B. A. Woodcock. 2011. Ecological restoration
on farmland can drive beneficial functional responses in plant
and invertebrate communities. Agriculture, Ecosystems &
Environment 140:62–67.
R Core Team. 2018. R: a language and environment for statisti-
cal computing. R Foundation for Statistical Computing,
Vienna, Austria. www.R-project.org
Radford, K. G., and P. James. 2013. Changes in the value of
ecosystem services along a rural-urban gradient: a case study
of Greater Manchester, UK. Landscape and Urban Planning
109:117–127.
Robbins, P., and T. Birkenholtz. 2003. Turfgrass revolution:
measuring the expansion of the American lawn. Land Use
Policy 20:181–194.
Rodwell, J. S. 2006. National vegetation classification: users’
handbook. Joint Nature Conservation Committee, Peterbor-
ough, UK.
RSPB. 2013. Urban advice pack. RSPB, London, UK.
Sarah, P., H. M. Zhevelev, and A. Oz. 2015. Urban park soil
and vegetation: effects of natural and anthropogenic factors.
Pedosphere 25:392–404.
Schwarz, N., M. Moretti, M. N. Bugalho, Z. G. Davies, D.
Haase, J. Hack, A. Hof, Y. Melero, T. J. Pett, and S. Knapp.
2017. Understanding biodiversity-ecosystem service relation-
ships in urban areas: a comprehensive literature review.
Ecosystem Services 27:161–171.
Set€
al€
a, H. M., G. Francini, J. A. Allen, N. Hui, A. Jumpponen,
and D. J. Kotze. 2016. Vegetation type and age drive changes
in soil properties, nitrogen, and carbon sequestration in urban
parks under cold climate. Frontiers in Ecology and Evolution
4:93.
Seto, K. C., B. G€
uneralp, and L. R. Hutyra. 2012. Global fore-
casts of urban expansion to 2030 and direct impacts on biodi-
versity and carbon pools. Proceedings of the National
Academy of Sciences USA 109:16083–16088.
Shaw, M. R., and M. E. Hochberg. 2001. The neglect of para-
sitic Hymenoptera in insect conservation strategies: the Bri-
tish fauna as a prime example. Journal of Insect
Conservation 5:253–263.
Shwartz, A., A. Muratet, L. Simon, and R. Julliard. 2013. Local
and management variables outweigh landscape effects in
enhancing the diversity of different taxa in a big metropolis.
Biological Conservation 157:285–292.
Smetana, S. M., and J. C. Crittenden. 2014. Sustainable plants
in urban parks: a life cycle analysis of traditional and alterna-
tive lawns in Georgia, USA. Landscape and Urban Planning
122:140–151.
Southon, G. E., A. Jorgensen, N. Dunnett, H. Hoyle, and K. L.
Evans. 2017. Biodiverse perennial meadows have aesthetic
value and increase residents’perceptions of site quality in
urban green-space. Landscape and Urban Planning 158:105–
118.
Southon, G. E., A. Jorgensen, N. Dunnett, H. Hoyle, and K. L.
Evans. 2018. Perceived species-richness in urban green spaces:
cues, accuracy and well-being impacts. Landscape and Urban
Planning 172:1–10.
Sparks, T. H., P. J. Croxton, N. Collinson, and D. A. Grisenth-
waite. 2007. The grass is greener (for longer). Weather
60:121–125.
Stevenson, M. J., L. K. Ward, and R. F. Pywell. 1997. Re-creat-
ing semi-natural communities: vacuum harvesting and hand
collection of seed on calcareous grassland. Restoration Ecol-
ogy 5:66–76.
Stewart, G., M. Ignatieva, C. Meurk, H. Buckley, B. Horne,
and T. Braddick. 2009. URban Biotopes of Aotearoa New
Zealand (URBANZ) (I): composition and diversity of tem-
perate urban lawns in Christchurch. Urban Ecosystems
12:233–248.
Stuble, K. L., S. E. Fick, T. P. Young, and L. Brudvig. 2017.
Every restoration is unique: testing year effects and site
effects as drivers of initial restoration trajectories. Journal of
Applied Ecology 54:1051–1057.
The Wildlife Trusts. 2018. Towards a wilder Britain: creating a
Nature Recovery Network to bring back wildlife to every
neighbourhood. The Wildlife Trusts, London, UK.
Thompson, K., J. P. Bakker, J. P. Bakker, and R. M. Bekker.
1997. The soil seed banks of North West Europe: methodol-
ogy, density and longevity. Cambridge University Press, Cam-
bridge, UK.
Thompson, K., J. G. Hodgson, R. M. Smith, P. H. Warren, and
K. J. Gaston. 2004. Urban domestic gardens (III): composi-
tion and diversity of lawn floras. Journal of Vegetation
Science 15:373–378.
Tilman, D. 1997. Community invasibility, recruitment limita-
tion, and grassland biodiversity. Ecology 78:81–92.
Timms, L. L., J. J. Bowden, K. S. Summerville, C. M. Buddle,
R. Didham, and Y. Basset. 2013. Does species-level resolu-
tion matter? Taxonomic sufficiency in terrestrial arthropod
biodiversity studies. Insect Conservation and Diversity
6:453–462.
Tratalos, J., R. A. Fuller, P. H. Warren, R. G. Davies, and K. J.
Gaston. 2007. Urban form, biodiversity potential and ecosys-
tem services. Landscape and Urban Planning 83:308–317.
Trowbridge, C. C., A. Stanley, T. N. Kaye, P. W. Dunwiddie,
and J. L. Williams. 2017. Long-term effects of prairie restora-
tion on plant community structure and native population
dynamics. Restoration Ecology 25:559–568.
Unterweger, P. A., J. Klammer, M. Unger, and O. Betz. 2018.
Insect hibernation on urban green land: a winter-adapted
mowing regime as a management tool for insect conservation.
Biodiversity and Ecosystem Risk Assessment 13:1–29.
Unterweger, P. A., C. Rieger, and O. Betz. 2017. The influence
of urban lawn mowing regimes on diversity of heteroptera
(Hemiptera). Heteropteron 48:7–21.
van der Meijden, E., P. G. L. Klinkhamer, T. J. de Jong, and C.
A. M. van Wijk. 1992. Meta-population dynamics of biennial
plants: How to exploit temporary habitats. Acta Botanica
Neerlandica 41:249–270.
van der Putten, W. H., et al. 2013. Plant–soil feedbacks: the
past, the present and future challenges. Journal of Ecology
101:265–276.
van Rijn, I., T. M. Neeson, and Y. Mandelik. 2014. Reliabilityand
refinement of the higher taxa approach for bee richness and
composition assessments.Ecological Applications 25:88–98.
Vance, E. D., P. C. Brookes, and D. S. Jenkinson. 1987. An
extraction method for measuring soil microbial biomass C.
Soil Biology and Biochemistry 19:703–707.
Vergnes, A., V. Pellissier, G. Lemperiere, C. Rollard, and P. Cler-
geau. 2014. Urban densification causes the decline of ground-
dwelling arthropods. Biodiversity and Conservation 23:1859–
1877.
Walker, K. J., P. A. Stevens, D. P. Stevens, J. O. Mountford, S. J.
Manchester, and R. F. Pywell. 2004. The restoration and re-
creation of species-rich lowland grassland on land formerly
managed for intensive agriculture in the UK. Biological Con-
servation 119:1–18.
Article e01946; page 1114 BRIONY A. NORTON ET AL. Ecological Applications
Vol. 29, No. 6

Walls, M. 2009. Parks and recreation in the United States: local
park systems. Resources for the Future, Washington, D.C.,
USA.
Wang, J. W., C. H. Poh, C. Y. T. Tan, V. N. Lee, A. Jain, and E.
L. Webb. 2017. Building biodiversity: drivers of bird and but-
terfly diversity on tropical urban roof gardens. Ecosphere 8:
e01905.
Wastian, L., P. A. Unterweger, and O. Betz. 2016. Influence of
the reduction of urban lawn mowing on wild bee diversity
(Hymenoptera, Apoidea). Journal of Hymenoptera Research
49:51–63.
Westbury, D. B., A. Davies, B. A. Woodcock, and N. P. Dun-
nett. 2006. Seeds of change: the value of using Rhinanthus
minor in grassland restoration. Journal of Vegetation Science
17:435–446.
Wheeler, M. M., et al. 2017. Continental-scale homogenization
of residential lawn plant communities. Landscape and Urban
Planning 165:54–63.
Williams, N. S. G., et al. 2009. A conceptual framework for pre-
dicting the effects of urban environments on floras. Journal
of Ecology 97:4–9.
Xu, H.-J., S. Li, J.-Q. Su, S. Nie, V. Gibson, H. Li, and Y.-G.
Zhu. 2014. Does urbanization shape bacterial community
composition in urban park soils? A case study in 16 represen-
tative Chinese cities based on the pyrosequencing method.
FEMS Microbiology Ecology 87:182–192.
Yan, B., J. Li, N. Xiao, Y. Qi, G. Fu, G. Liu, and M. Qiao. 2016.
Urban-development-induced changes in the diversity and
composition of the soil bacterial community in Beijing. Scien-
tific Reports 6:38811.
SUPPORTING INFORMATION
Additional supporting information may be found online at: http://onlinelibrary.wiley.com/doi/10.1002/eap.1946/full
DATA AVAILABILITY
Data are available from the Natural Environment Research Council (NERC) at the Environmental Information Data Centre
(EIDC): https://doi.org/10.5285/d0741544-cdf3-497d-996b-e30b4b7373c1. DNA data are available as a BioProject from the
National Center for Biotechnology Information (NCBI) (accession number PRJNA531648): http://www.ncbi.nlm.nih.gov/biopro
ject/531648.
September 2019 BIODIVERSITY OF NEW URBAN MEADOWS Article e01946; page 1115