Research Article |
Corresponding author: Philipp Andreas Unterweger ( philipp.unterweger@biodiversitaetsplanung.de ) Academic editor: Josef Settele
© 2018 Philipp Andreas Unterweger, Jorinde Klammer, Manuela Unger, Oliver Betz.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Unterweger PA, Klammer J, Unger M, Betz O (2018) Insect hibernation on urban green land: a winter-adapted mowing regime as a management tool for insect conservation. BioRisk 13: 1-29. https://doi.org/10.3897/biorisk.13.22316
|
Insect conservation is challenging on various ecological scales. One largely neglected aspect is the quality of undisturbed hibernation sites. This study aims to fill a lack of knowledge concerning insect hibernation on uncut meadows persisting in urban green spaces during the winter season in a middle-sized town in south Germany. During two years of sampling, 13,511 insect specimens of the orders Heteroptera, Hymenoptera, Coleoptera and Diptera were caught from their winter stands. The specimens were assigned to 120 families and 140 taxonomic species were determined from the orders Heteroptera, Coleoptera and Diptera and 324 morphotypes from the orders Hymenoptera, Coleoptera and Diptera. The data indicate the importance of winter fallows for insect hibernation. Unmown meadows offer additional plant structures in winter (flower heads, stems, tufts and leaves) that are absent from mown ones. This increased structural diversity results in both higher species diversity and numbers of insect individuals during spring emergence. The results of this study thus emphasise the value of unmown structures for insect conservation and suggest a mosaic-like cutting maintenance of meadows, way- and river-sides and other green infrastructure in both the urban area and the open landscape.
Coleoptera , Diptera , green space, habitat protection, Heteroptera , hibernation, Hymenoptera , insect decline, meadow, mowing, urban ecology
The multifaceted threat and the extent of worldwide and massive insect loss (e.g.
In Central Europe, natural and anthropogenic grasslands can be seen as hotspots for plants and insects (
In Central Europe, the expansion of urban areas is rapidly increasing. In west Germany, the areas settled by humans have increased by about 140% in the past fifty years (
The general positive effect of management reduction in urban grasslands on insect diversity and numbers (including endangered species) has been evaluated in several previous research projects on a diversity of insect groups such as grasshoppers (e.g.
Insect protection measures must be primarily focused on the egg and larval habitat during the entire year. In the present study, previous investigations have been expanded by considering the entire life cycle of an insect (
In the present study, the authors focus on urban meadows as hibernation sites for insects. Meadows can be sub-divided into various structural layers (e.g.
Focusing on urban public green spaces, the present contribution explores the differences for insect populations between meadows unmown in winter versus mown meadows. Second, the significance of various herbal structures (i.e. flower heads, stems, tufts, leaves) of winter plant stands is evaluated. These results are analysed with respect to the collection of ecological data concerning the hibernation sites used by the representatives of various insect orders (i.e. Coleoptera, Heteroptera, Diptera, Hymenoptera), families and species that potentially provide insect-based ecosystem services (e.g. pollination, pest control). To test the effect of hibernation areas, emergence traps (photo-eclectors) were used in early spring in order to compare the hatching of insects. In particular, the authors focused on (1) the general effect of autumn mowing on the emergence in the following spring, (2) the temporal development of the total insect emergence from the winter stands on a weekly basis over the course of spring and early summer, (3) the evaluation of the value of the various herbal structures for insect hibernation, (4) the role of herbal structures for the hibernation of various insect orders, (5) the extent of insect hibernation within the soil and (6) the importance of stems and flower heads for the hibernation of insect species. These questions are addressed in two experiments, referred to as experiment 1 and 2 in this work.
The data collection for this study was divided into two main experiments that both aimed at analysing the effect of unmown urban green spaces as potential habitats for hibernating insects. Whereas experiment 1 investigated the use of various plant structures (flower heads, stems, tufts, leaves) existing in mown versus unmown winter stands of meadows for hibernation, experiment 2 additionally focused on the soil of unmown meadows as a hibernation site for insects. Data collection occurred in 2014 from mid February to mid June, with experiment 1 being repeated in 2015 for verification purposes.
Square-shaped emergence traps (photo-eclectors, see Fig.
The sampling sites (100–1000 m2) were urban meadows in the city of Tübingen, Germany (48°32'15.9"N, 9°2'28.21"E) that had been mown twice a year (end of June and end of September) for at least two years. In both experiments, only one half of each meadow was mown at the end of June (unmown in autumn), whereas the other half was mown at the end of June and the end of September (mown in autumn).
Figure
The experimental design for data collection of experiment 1 (blue squares) and experiment 2 (red squares) from eight sampling sites (green circles) divided into a mown (M, mown in autumn) and an unmown (U, without autumn cut) part. Green arrows indicate the collection and sorting of the plant structures. Red arrows indicate the placement of the eclectors on the sampling sites. In experiment 2, only six sampling sites could be used because of vandalism.
Experiment 1 (Fig.
Experiment 2 (Fig.
In both experiments, the adult insect individuals were collected in the sampling pots every week from February to June. The insects were stored in small twist-top vials in 70 % ethanol for later identification and preparation.
True bugs and some large beetles from experiment 1 and all individuals from experiment 2 were pinned, whereas all the other insects were stored in 70% ethanol. The determination of the insect orders was performed based on
In order to consider as many taxonomic groups as possible within an acceptable time period, a frequently used taxonomic shortcut was applied by identifying all taxa only as detailed as required, occasionally termed taxonomic sufficiency (Ellis, 1985) or lowest practical taxonomic level (LPT) (e.g.
Levels of identification within each insect order considered in this study.
Insect order | Level of identification |
---|---|
Heteroptera | Species level (confirmed by Dr. Christian Rieger, Nürtingen) |
Hymenoptera (excluding ants) | Family level, all individuals of the various families were measured and these groups were counted as morphotypes |
Coleoptera | Species level if scientifically confirmed by external expert (Dr. Nadein (Tübingen) for Chrysomelidiae, Dr. Salamon (Hannover) for Staphylinidae). Otherwise determination at lowest practical taxonomic level (LPT). These determinations (species level, genus, family) were counted as morphospecies |
Diptera | Family level; all individuals of the various families were measured and these groups were counted as morphotypes; the scientific validation of the families by Dr. Sabine Prescher (Braunschweig), Dr. Anke Schäfer (Weitramsdorf) and Gerrit Öhm (Wasserhausen) led to some lower level identifications (genus, species) |
All statistical analyses were performed with regard to the Shannon index (
In the two years of sampling, a total sum of 13,511 insects of the orders Heteroptera, Hymenoptera, Coleoptera and Diptera were sampled in both experiments. The authors identified 120 families, 140 taxonomic species and 324 morphotypes (see Suppl. material
Table
Total numbers of individuals (I), families (F), the validated taxonomic species (S) and the morphotypes (T) itemised by insect order over both the pooled experiments. Counts are divided into mown versus unmown sites and into the various plant compartments (organs, Experiment 1 only) flower heads, stems, tufts, leaves. “–“ = not applicable. A detailed list of the sampled species is provided in the electronic supplement (Suppl. material
Heteroptera | Hymenoptera | Coleoptera | Diptera | |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
families | 10 | 38 | 37 | 35 | ||||||||||||
species / types | 34 / − | − / 135 | 68 / 141 | 38 / 48 | ||||||||||||
I | F | S | T | I | F | S | T | I | F | S | T | I | F | S | T | |
unmown | 78 | 9 | 27 | − | 2221 | 28 | − | 118 | 730 | 32 | 60 | 107 | 5723 | 35 | 33 | 43 |
mown | 12 | 5 | 9 | − | 869 | 20 | − | 82 | 420 | 31 | 30 | 87 | 3458 | 30 | 16 | 35 |
flower heads unmown | 10 | 3 | 4 | − | 579 | 18 | − | 53 | 76 | 17 | 17 | 27 | 579 | 15 | 6 | 18 |
flower heads mown | − | − | − | − | − | − | − | − | − | − | − | − | − | − | − | − |
stems unmown | 7 | 3 | 5 | − | 230 | 12 | − | 40 | 102 | 16 | 17 | 31 | 370 | 14 | 5 | 17 |
stems mown | − | − | − | − | − | − | − | − | − | − | − | − | − | − | − | − |
tuft unmown | 9 | 5 | 6 | − | 274 | 14 | − | 47 | 174 | 17 | 27 | 41 | 792 | 20 | 8 | 21 |
tuft mown | 3 | 2 | 3 | − | 184 | 13 | − | 42 | 86 | 20 | 15 | 34 | 717 | 23 | 6 | 21 |
leaves unmown | 4 | 2 | 2 | − | 419 | 18 | − | 62 | 85 | 16 | 18 | 29 | 687 | 22 | 11 | 24 |
leaves mown | 2 | 2 | 2 | − | 140 | 11 | − | 23 | 79 | 16 | 19 | 25 | 464 | 16 | 3 | 20 |
One of the major interests was the evaluation of the influence of the mowing regime on insect hibernation in the investigated meadows. Therefore, the means between autumnal mown versus unmown meadows were compared. The comparison was performed with regard to the Shannon index (p < 0.001), the number of individuals (p < 0.001) and the number of species including morphotypes (p = 0.60) (Figure
Impact of autumn mowing on the arithmetic mean (± 95 % CI) of A Shannon index B the number of individuals and C the number of species/types. Asterisks indicate significant effects of the factor mowing regime as derived from linear mixed models (* p < 0.05; ** p < 0.005; *** p < 0.001). Shannon: p < 0.001, DF: 1,392, F-ratio: 121.8; Number of individuals: p < 0.001, DF: 1,55, F-ratio: 21.9; number of species/morphotypes: p = 0.60, DF: 1,56, F-ratio: 0.2651 (n = 6).
The insect emergence from the middle of February to the middle of June of the individuals of the various insect orders is summarised in Figure
In contrast with the unmown meadows, flower heads and stems were absent on the meadows with an autumn cut (see Table
The following results are based on the linear mixed model analyses (Fig.
Significance of plant compartments as insect hibernation sites with respect to the arithmetic mean (± 95 % CI) of A the Shannon index B number of individuals and C number of species / morphotypes subdivided by mowing regime. Flower heads and stems are not found on mown meadows. Different capital letters indicate pairwise significant differences between plant compartments of unmown (dark grey) sampling sites, whereas asterisks indicate significant differences (* p < 0.05; ** p < 0.005; *** p < 0.001) between both the mowing regimes based on Tukey’s HSD post-hoc comparisons after linear mixed models. Shannon index (only significant results reported): p < 0.001; DF: 3,5; F-ratio: 8.3 (flower head – stem: p < 0.005, flower head – tuft: p < 0.001, flower head – leaves: p < 0.005). Unmown – mown testing: tuft: p < 0.005; DF: 1,873; F-ratio: 8.8. Number of Individuals (only significant results reported): p < 0.001; DF: 1,97; F-ratio: 4.1 (flower head – stem: p < 0.05, stem – tuft: p < 0.05). Unmown – mown testing: tuft: p < 0.05; DF: 1,301; F-ratio: 4.0. Leaves: p < 0.05; DF: 1,20; F-ratio: 5.0. Species/morphotypes: p < 0.05; DF: 1,93; F-ratio: 1.6. Unmown – mown testing: tuft: p < 0.05; DF: 1,303; F-ratio: 4.0. Leaves: p < 0.05; DF: 1,20; F-ratio: 5.0. n = 8.
Shannon index: The highest Shannon index values were found in tufts with no mowing in autumn. Flower heads were used less for hibernation compared with the other plant compartments.
Number of individuals: The number of individuals shows, in comparison with the plant compartments, the fewest numbers in stems (absent in mown meadows) and in the structures of the mown areas.
Number of species/morphotypes: On unmown meadows, the species/morphotype comparison did not show any significant differences between the plant compartments. The differences between the compartments of mown versus unmown areas are significant for tufts and leaves. The important fact that flower heads and stems are missing as hibernation spaces on mown areas stresses the ecological value of these plant compartments on unmown meadows.
To test the hypothesis that, in the studied insect orders, special preferences occur regarding hibernation in the various plant compartments, linear mixed models were applied to the data from the unmown meadows that showed all plant compartments (Fig.
The distribution of the insect orders with respect to arithmetic mean (± 95 % CI) on the various plant organs showed no significant differences with respect to the Shannon index (A). The number of individuals (B) is distributed across the plant compartments with no significant trend in the orders of Heteroptera and Diptera, whereas a strong tendency to significance can be seen in some hibernation places of Hymenoptera (linear mixed model p < 0.05; DF: 3,4; F-ratio: 2.8; flower head – stem: p = 0.065; flower head – tuft: p = 0.095). Significantly more coleopterans hibernate in tufts compared with flower heads (linear mixed model p < 0.05; DF: 3,376; F-ratio: 3.5; flower head – tuft: p < 0.05, tuft – leaves: p = 0.074). The number of species (C) was not significantly different across plant compartments for Heteroptera and Coleoptera. For Hymenoptera, significantly more species occurred in leaves compared with stems (linear mixed model p = 0.075; DF: 3,173; F-ratio: 2.4; leaves – stem: p < 0.05). For Diptera, a strong trend showed that more individuals hibernated in tufts compared with stems (linear mixed model p < 0.05; DF: 3,76; F-ratio: 2.8; stem – tuft: p = 0.072). Significant differences between the plant compartments within each insect order, as based on Tukey’s HSD post-hoc comparisons after linear mixed models, are indicated by different capital letters (n = 8).
The Shannon index did not show any significant patterns (Fig.
Number of individuals: By trend, significantly more hymenopterans hibernated in flower heads compared with stems and tufts. Coleopterans preferred tufts significantly more than flower heads with an additional trend to significance between tufts and leaves.
Number of species/morphotypes: The highest species/morphotype number of Hymenoptera can be found in leaves and tufts for Diptera.
To investigate the importance of soil for insect hibernation, the species lists were compared with respect to the single occurrence of the species in experiment 2 (Table
Species that are presumed to have hibernated in the soil (see Fig.
Order | Family | Species |
---|---|---|
Heteroptera | Cymidae | Cymus glandicolor |
Lygaeidae | Beosus maritimus | |
Drymus ryeii | ||
Rhyparochromus pini | ||
Scolopostethus thomsoni | ||
Miridae | Capsus ater | |
Dicyphus annulatus | ||
Plagiognathus arbustorum | ||
Plagiognathus cf. chrysanthemi | ||
Podops inunctus | ||
Stenodema calcarata | ||
Nabidae | Nabis ferus | |
Nabis rugosus | ||
Pentatomidae | Palomena prasina | |
Rhyparochromidae | Megalonotus emarginatus | |
Tingidae | Dictyla humuli | |
Coleoptera | Chrysomelidae | Cassida vibex |
Galeruca tanaceti | ||
Oulema erichsonii | ||
Staphylinidae | Philonthus carbonarius | |
Quedius boops | ||
Quedius curtipennis | ||
Quedius maurus | ||
Quedius nitipennis | ||
Scopaeus minutus | ||
Staphylinus dimidiaticornis | ||
Diptera | Opomyzidae | Geomyza tripunctata |
Geomyza venusta | ||
Opomyza florum | ||
Opomyza germinationes | ||
Stratiomyiidae | Beris geniculata | |
Chloromyia formosa | ||
Syrphidae | Dasysyrphus albostriatus | |
Platycheirus cf. scutatus | ||
Syrphus ribesii | ||
Syrphus torvus | ||
Tephritidae | Chaetorellia stylata |
The results of experiment 1 were used to evaluate the importance of flower heads and stems as hibernation sites for insects. Species that only occurred in flower heads and stems are listed in Table
Species that only occurred in flower heads and stems. Only species with confirmation by a taxonomic expert are listed (see Suppl. material
Order | Family | Species | Flower head | Stem |
---|---|---|---|---|
Heteroptera | Anthocoridae | Cardiastethus fasciiventris | 2 | − |
Lygaeidae | Peritrechus geniculatus | 1 | 1 | |
Scolopostethus affinis | − | 2 | ||
Nabidae | Himacerus mirmicoides | 1 | − | |
Nabis brevis | − | 1 | ||
Rhyparochromidae | Megalonotus antennatus | − | 1 | |
Coleoptera | Chrysomelidae | Hispella atra | 1 | − |
Oulema melanopus | 1 | − | ||
Psylliodes napi | 1 | − | ||
Staphylinidae | Sepedophilus testaceus | − | 1 | |
Rugilus rufipes | − | 1 | ||
Scopaeus gracilis | − | 1 | ||
Lathrobium dilutum | − | 1 | ||
Diptera | Tephritidae | Chaetorellia cylindrica | 1 | − |
Chaetorellia jaceae | 14 | − |
The present contribution aims to provide new data for optimising green space management for insect hibernation. Although this study has been performed on public urban green spaces, its results are also relevant for rural grassland. In the cultural landscape, such structures are usually present on waysides, field margins, river slopes, meadow orchards and fallows but not on intensive meadows, because these are usually mown until autumn. These structures are threatened in Europe by intense mowing (e.g.
These results show that meadows without autumn cut offer a huge potential for hibernating insects. The value of unmaintained structures (e.g. field margins) for the number of both species and individuals of animals has been demonstrated in numerous studies (e.g.
This study has revealed higher insect hatchings from autumnal unmown meadows (Fig.
The comparison of these two experiments allows the authors to discriminate insect species that hibernate in the soil from those using the vegetation further above the ground. In experiment 2, it has been shown that 134 species/morphotypes (254 individuals) can only be found in the soil, whereas 338 species/morphotypes (3530 individuals) use hibernation sites in the vegetation such as flower heads, stems, tufts and leaves (see Tables
The structural diversity of the vegetation (including the various plant organs) can largely influence insect diversity (e.g.
With regard to the Shannon index, the four analysed insect orders showed a homogeneous distribution across the four plant compartments, i.e. no significant predominance of one order or even differences between the organs were detected (Fig.
However, in terms of numbers of individuals and species/morphospecies of the various insect orders, differences were found between plant compartments as overwintering sites (Fig.
Heteroptera: Only a few heteropterans hibernated as adults and so their numbers were low until the adults of the egg hibernating species emerged in late June (
Hymenoptera: For Hymenoptera (mainly parasitoid families of Braconidae and Ichneumonidae), the distribution of individuals amongst the plant compartments shows that they prefer flower heads and leaves compared with stems and tufts for hibernation (Fig.
Coleoptera: Many beetles prefer tufts for hibernation, a finding that can possibly be explained by their preference for walking instead of flying (
Diptera: According to this study, tufts form attractive hibernation places for dipterans. In the present investigation (Table
Natural meadows with dead winter vegetation and its complex vertical structure have become a rare and endangered biotope in Central European landscapes (
In the following, practical recommendations that follow from the results of these studies are provided.
The results of this study in combination with the authors’ previous results (
The studies of these previous experimental setups (
Scheme of an optimised management plan in consideration of the previous entomological studies and the present results. The authors distinguish between three mowing concepts, (1) the “flower concept” for optimising flower diversity and nutrient balance, (2) “autumnal mowing” for reproduction and larval development of late summer insects and (3) “summer mowing” for providing hibernation areas. The green line symbolises a possible course of vegetation height. A and B symbolise mowing periods rather than set mowing dates. The heights of mowing vary randomly (relation to constant (brown) soil baseline) to provide ecologically different situations (from open soil to longer vegetation as hiding places during mowing). The vegetation height of the “summer mowing” starts at a higher level in spring as a result of the vegetation that has remained on the site during winter. Blue arrows show the possible fluctuation between source and sink habitats and the metapopulations in the mosaic mowing system. In all three regimes, the disposal of the cut hay is recommended to avoid overfertilisation and grass dominance.
Even if, in some cases, a once a year mowing regime seems to be too intensive (
The resulting different types of meadows support insects by allowing them to escape, migrate and find suitable habitats at anytime of the year (
The combination of the “flower concept”, the “autumn mowing” and the “summer mowing” and the occurrence of elongated periods during which mowing is possible, instead of the specification of fixed mowing dates, supports metapopulation dynamics and the recolonisation of mown meadows (cf. blue arrows in Fig.
We thank Konstantin Nadein, Gerrit Öhm, Sabine Prescher, Volker Puthz, Christian Rieger, Jörg Salamon, Anke Schäfer, Karin Wolf-Schwenninger and Erich Weber for helping with the determination of the sampled insects. We are also grateful to the Zoological Collection of the University of Tübingen for providing their insect collections and to the Botanical Garden of the University of Tübingen (Alexandra Kehl and Brigitte Fiebig), the City of Tübingen (Martina Betaks) and the Amt für Vermögen und Bau (Rainer Boeß) for providing green spaces as experimental plots. Permission to collect insects for scientific purposes was given by the Regional Council of Tübingen. We thank Karen Bussmann, Alexander Eninger, Maria Georgi, Michaela Hauke, Stefanie Herbst, Sigrun Herrmann, Nina Kneule, Rune Michaelis, Fabian Roser, Liesa Schnee and Esther Schrode for technical assistance and valuable discussions. Theresa Jones corrected the English. We thank an anonymous reviewer for useful comments on a previous version of our manuscript.
Table with all captured species / morphotypes sorted by order, family and species / morphotype.
Data type: species data
Explanation note: Collection: University of Tübingen, Evolutionary Biology of Invertebrates, Auf der Morgenstelle 28, 72076 Tübingen, Germany. Individuals with scientific species name that were checked by a taxonomic expert were counted as taxonomic species (s); all the other determinations were counted as morphotypes (m). Morphotypes are defined by the lowest practical taxonomic level (e.g.