What Determines the Distribution of Forest Flightless Bush Cricket Pholidoptera griseoaptera in the Eastern Part of Its Range (The Kaluga Region, Russia)?

Abstract

(1) Pholidoptera griseoaptera (De Geer, 1773) (Orthoptera, Tettigoniidae) is a common and widespread inhabitant of forest edges in Europe and may therefore serve as a suitable model species for understanding past and future changes in forest wildlife. (2) We recorded the presence or absence of the species in 189 forest and forest-edge plots within the Kaluga Region using acoustic observations and pitfall trapping, and analysed the data using logistic regression. (3) Across the region, the main positive factor affecting species presence was the dominance of nemoral herbs in the herb layer. The main negative factors were habitat isolation caused by physical barriers and location within moraine plains formed during the late stage of the Moscow glaciation. The presence of coniferous tree species and spatial autocovariation were also significant factors, although their contributions were relatively small. The abundance of Ph. griseoaptera was higher in forests located within river valleys. Within Kaluga, the long-term persistence of tree vegetation and habitat isolation were the main significant factors affecting species occurrence. The smallest urban habitat occupied by the species covered approximately 13 ha, whereas the total area of unmown patches within this habitat was only about 0.2 ha. (4) Ph. griseoaptera may be used as an indicator of the long-term persistence of broadleaved deciduous (nemoral) forests. Under conditions of high urbanization, however, the species may become threatened.

1. Introduction

One core goal of ecology is to understand the distribution and abundance of living organisms in their environment [1,2]. To conserve species, their communities, and whole ecosystems, we need to know what factors affect the survival of a certain species. As it is impossible to investigate all species, surrogate or target species are often used as representatives of different ecological groups. Species distribution may be influenced by habitat quality, landscape matrix quality, and historical factors [3,4,5,6]. However, there is little information on distribution patterns for many species and their assemblages.

One of the priority species groups used to investigate factors affecting their distribution and to conserve their habitats comprises nemoral species, e.g., species associated with the broadleaved deciduous forest biome. In Europe, nemoral or broadleaved deciduous forests are widely distributed within the Continental biogeographical region. This region extends in a central east–west band across most of Europe and includes some of the continent’s most productive ecosystems. Forests in this region have been strongly fragmented and transformed because of long-term intensive human use. A high degree of habitat fragmentation due to transportation and urban infrastructure is one of the major pressures on biodiversity [7]. Therefore, research and conservation of nemoral species and ecosystems are highly important goals [8].

Orthopteran insects (Orthoptera) are a suitable model group for ecological and conservation research [9,10]. Their relatively large body size, moderate dispersal capacity, and dependence on both vegetation structure and abiotic environmental factors provide a good basis for investigating how various factors affect their distribution. Numerous studies have revealed both environmental and historical determinants affecting single orthopteran species and multispecies assemblages, e.g., [11,12,13,14,15]. However, most European orthopterans inhabit grasslands. Apparently, only two species can be regarded as widespread and common forest dwellers in Europe. They are the wood cricket Nemobius sylvestris (Bosc, 1792) [16] and the dark bush cricket Pholidoptera griseoaptera (De Geer, 1773) [17,18]. For Eastern Europe, only the latter species is typical.

Pholidoptera griseoaptera (Orthoptera, Tettigoniidae) is known as an inhabitant of forest edges in most parts of Europe [19,20]. In many regions, it is also found in tall grasses and herbaceous vegetation outside forests [21,22,23,24], although such sites may serve as supplementary feeding habitats [25]. Ph. griseoaptera is a uniformly brachypterous species. Therefore, several authors have studied its movement, dispersion, and distribution in fragmented landscapes [25,26,27,28,29]. It has been shown that this bush cricket inhabits a large proportion of suitable habitat patches and exhibits successful inter-patch dispersal in agricultural landscapes [28]. At the same time, movement characteristics may depend on landscape structure, and rapid fragmentation of formerly well-connected landscapes or habitat loss below a certain threshold may threaten this ground-dwelling species [27]. Extensive forest matrixes may also act as barriers to dispersal [25].

The cited papers describe the distribution of this species in several regions of Central and Western Europe, which appear to represent the optimal part of its range. However, insect species may change their habitat preferences, occupancy of habitat patches, and other ecological characteristics across their ranges [30,31]. Therefore, it is important to investigate how this species uses landscapes in Eastern Europe. Pholidoptera griseoaptera has been reported all across European Russia [32]. Recently, it was found to be widespread across the nemoral forest and forest-steppe zones [33], but detailed data on its habitats and ecological characteristics in these areas are still lacking. Nevertheless, some studies [24,34] suggest that this species is less typical of hemiboreal mixed coniferous–broadleaved forests.

Although the species is not considered endangered in Russia, it is included in the Red Book of Moscow City [35]. Since Moscow City is a neighbour of the Kaluga Region, it can be expected that certain environmental factors limit the occurrence of Ph. griseoaptera in both regions. If these factors are not exclusively related to urbanisation, they may be assessed more clearly in the Kaluga Region, where landscapes are less transformed by urban development than in Moscow City.

Based on our observations from numerous localities in the Kaluga Region and Moscow City, we found that Ph. griseoaptera is absent from many forest edges surveyed. Therefore, the aim of our study was to identify the factors affecting the distribution of Ph. griseoaptera in the Kaluga Region, which represents the eastern part of the species’ range.

We tested the following hypotheses: (1) Pholidoptera griseoaptera has a continuous distribution across the region, but mainly occupies nemoral forests. (2) The presence of Pholidoptera griseoaptera is influenced by habitat characteristics and historical factors, such as the presence of old forests. (3) Habitat isolation negatively affects the presence of this species; therefore, it is expected to occur locally in urban areas.

By testing these hypotheses, we aimed to answer the following questions: (1) What are the prospects for conserving this species under ongoing landscape fragmentation, and could it become threatened in the future? (2) Can Ph. griseoaptera be used as an indicator of habitats with specific properties or as a marker of past habitat conditions?

2. Materials and Methods

2.1. Study Species

Pholidoptera griseoaptera (De Geer, 1773) (Figure 1) is a medium-sized (13–18 mm in length) grey-brown short-winged bush cricket (Orthoptera, Tettigoniidae) inhabiting mainly forest edges [25,26]. Adults can jump a horizontal distance that often exceeds 1 m [24]. Bush crickets can move up to 70 m in a day [26]. Compare to other orthopterans, its dispersal ability is assessed as low [36,37]. This species is polyphagous, feeding on both plants (e.g., Rubus sp., Taraxacum officinale F.H.Wigg., and Urtica dioica L.) and arthropods (e.g., caterpillars, flies, and spiders) [26]. It has a biennial life cycle; its eggs hatch in the spring of the second year. Eggs are laid into leaf axils, pithy stems, or rotten wood. There are seven nymphal stages. Adults usually appear in July. This species is active day and night, and males typically stridulate until the first signs of frost [25].

2.2. Study Area

The Kaluga Region is located in the centre of the East European Plain, south of Moscow, primarily within the upper basin of the Oka River, the largest right tributary of the Volga. The area of the region is 29.9 thousand km². Altitude varies from 108 to 279 m a.s.l. across the region. The southeastern part of the region forms the northern margin of the Central Russian Upland. The northwestern part of the region belongs to the Smolensk–Moscow Upland, while the Baryatino–Sukhinichi Plain is located in the central part of the region. Most geomorphological features were formed before the Quaternary Period. The climate is moderately continental. Mean annual temperatures vary from 4.5 to 4.9 °C. Annual precipitation ranges from 650 to 700 mm [38].

Despite the relatively small area of the region and its predominantly plain landforms, the territory of the Kaluga Region is highly differentiated (Figure 2). Its northern and western parts belong to the zone of mixed coniferous–broadleaved forests [39,40], or the hemiboreal biome [41], which is classified as part of the Boreal biogeoregion [7]. The primary vegetation in the watersheds consists of mixed forests dominated by Norway spruce (Picea abies (L.)), small-leaved lime (Tilia cordata Mill.), and pedunculate oak (Quercus robur L.).

The landforms are the result of the Moscow glaciation (corresponding to the Central European Early Würm glaciation), approximately 62 kyr BP [38]. The northwestern territories are located within the area affected by the late stage of the Moscow glaciation. They are mainly represented by flat moraine plains with numerous wetlands. The southwestern parts of the region are represented by fluvioglacial plains with sandy soils and vast pine forests dominated by Pinus sylvestris L. The southeastern parts of the hemiboreal biome are associated with landforms formed during the early stage of the Moscow glaciation (or along its margin). These areas are more strongly dissected by river networks, contain fewer swamps, and are dominated by deciduous trees.

The southern and eastern parts of the Kaluga Region belong to the zone of broadleaved forests, or the nemoral biome, considered part of the Continental biogeoregion. This area remained free from the effects of the Moscow glaciation. Its landforms were influenced by an older glaciation, identified either as the Dnieper (Riss) glaciation (190 kyr BP) [38] or as the Don (Günz) glaciation [42,43]. Typical landforms are erosional plains. The primary vegetation in the watersheds consists of broadleaved forests dominated by oak, lime, and European ash (Fraxinus excelsior L.). In present-day forests of both biomes, the dominant tree species are usually Betula pendula Roth and Populus tremula L.

2.3. Data Sampling

The material for this paper was collected from samples and observations obtained between 2003 and 2025 by the authors and researchers from the Parks Directorate of the Kaluga Region. We identified the presence or absence of Ph. griseoaptera in any plot using two main sampling methods: acoustic observation and pitfall traps. Although these methods are not identical, observations from several model plots showed that whenever this bush cricket was collected in pitfall traps, it was also detected acoustically, and vice versa.

Acoustic observations were conducted during warm afternoons and evenings from July to September. Using this method, we recorded the absence of the bush cricket in plots where it was not detected, provided that the species was simultaneously detected in other sample plots under the same weather conditions and on the same days. We did not count singing individuals; instead, a sample plot was considered occupied when more than one bush cricket was detected. Soil pitfall traps consisted of 0.5 L transparent plastic cups with an opening diameter of 85 mm filled to approximately one-third of their volume (150 mL) with a 4% formalin solution and covered with transparent polyethylene film. The number of traps per study plot ranged from 10 to 30, depending on the size and complexity of the biotope, as well as the objectives of sampling other insects. Pitfall traps were operated from May to September. Using pitfall traps, we sampled 617 specimens. Additionally, nine records of iNaturalist research grade were obtained from GBIF [44].

In 17 sample plots, observations were conducted over multiple years, and the species was either recorded or not recorded every year (Supplementary File). Therefore, observations from different years can be compared.

As additional methods for detecting the bush cricket, we used window traps and sweep-netting. Window traps were installed in wooded habitats at heights of up to 1.5 m above the ground. Our material includes 19 specimens collected using window traps in nine sample plots representing different forest types. Thus, this method confirmed the presence of the species but was not used to confirm its absence.

Sweep-netting was usually conducted in grasslands; therefore, bush crickets inhabiting forest edges and near-ground vegetation were unlikely to be collected. Consequently, this method was also used only to confirm the presence of the species.

2.4. Properties of Sample Plots

As individual sample plots, we considered habitat patches with different combinations of environmental predictors (see below), or located at distances of at least 500 m from one another, or separated by habitat patches with unsuitable characteristics (fields, grasslands, major roads, or buildings). Each sample plot was characterized by a unique combination of geographical coordinates (decimal latitude and longitude). Geographical coordinates were determined using the WGS 84 datum based on GPS measurements or Yandex satellite imagery.

Based on published data, several types of sample plots were considered potential habitats of Ph. griseoaptera (

Table 1. Numbers of plots of different habitat types where we recorded the presence or absence of Pholidoptera griseoaptera in the Kaluga Region, Russia. There are significance differences between subtypes of forests (X-squared = 19.117, df = 3, p-value = 0.0003).

Type of Sample Plots	Total	Presence	Absence
Gardens	15	11	4
Grasslands	23	11	12
Wetlands	9	0	9
Forests and edges (mesic), incl.	189	117	75
deciduous (non-riparian)	101	72	29
Pine	47	18	29
Spruce	11	3	8
Riparian (willow or alder)	30	17	13

): (1) forests (mesic or dry) and their edges; (2) wetlands, particularly swampy woodlands and forests; (3) gardens (plots with ploughed soil and a mosaic of trees and herbs); and (4) grasslands (with tall grasses and herbs).

First, we found that the species was not recorded in any swampy woodland or wetland. Therefore, this type of habitat was excluded from the dataset. Second, grasslands were excluded as habitats of Ph. griseoaptera. Although the bush cricket was recorded in many grassland plots, all such plots were located close to tree-dominated habitats (10–20 m away) and usually contained elements of tree vegetation, such as sparse shrubs or young trees. Therefore, these observations were interpreted as forest-edge habitats. When the species was recorded both in a grassland and in a neighbouring forest, the grassland plot was excluded from the analysis of species presence. Finally, among garden plots, Ph. griseoaptera was recorded only in sites located close to forests (at distances of several tens or the first hundreds of metres, without barriers to movement such as heavily trafficked roads or urban buildings). In addition, we did not observe nymphs in any garden plot. Therefore, we cannot consider gardens to be a residential habitat of Ph. griseoaptera (although we do not exclude this possibility), and only forests and forest edges were analysed as habitats of this species.

We considered all tree-dominated uncultivated habitats to be forests or forest edges, regardless of their size, from large forests to linear willow or alder stands along rivers. Some forest plots were excluded because of the absence of accurate geographical coordinates or other variables. Nine plots from Moscow were used to assess temporal dynamics of species presence across multiple years (Supplementary File), but they were excluded from the main analysis because of the different landscape structure of Moscow compared with the Kaluga Region and the scattered distribution of these plots within the city. In total, 189 sample plots were included in the analysis.

The sample plots covered most of the Kaluga Region, extending from N 55.2087 E 35.9894 in the north to N 53.3766 E 35.0679 in the south, and from N 54.2749 E 33.9226 in the west to N 54.7198 E 37.2012 in the east (Figure 2). For each plot, we recorded 12 independent variables (predictors) (

Table 2. Description of variables used to explain the presence of Pholidopteta griseoptera in the forests and their edges in the Kaluga Region, Russia.

Label	Name	Levels	Description
Dec_Tree	Deciduous trees in tree layer	0–1–2	0—adult deciduous trees are absent 1—sparse deciduous trees 2—dense tree layer dominated by deciduous species
Dec_shrubs_un	Deciduous shrubs or undergrowth	0–1	0—deciduous shrubs (such as Corylus avellana L., Euonymus verrucosus Scop., Lonicera xylosteum L.) or undergrowth are not seen 1—deciduous shrubs or undergrowth are conspicuous
Nem_herbs	Nemoral herbs	0–1	0—there are no nemoral species among dominants of the herb layer 1—there are any nemoral species among the dominants of the herb layer
Nitr_herb	Nitrophilous herbs	0–1	0—there are no nitrophilous species among dominants of the herb layer 1—there are any nitrophilous species among the dominants of the herb layer
Con_Tree	Coniferous trees	0–1	0—there are no coniferous trees in the tree stand. 1—there are any coniferous trees (Pinus sylvestris or Picea abies) in the tree stand.
Con_un	Coniferous undergrowth	0–1	0—there is no coniferous undergrowth. 1—there is coniferous undergrowth.
Hight_herb	Herbage height	0–3	0—there is no closed herb layer 1—low herb layer (to 20 cm) 2—moderate herb layer (20–60 cm) 3—tall herb layer (above 60 cm)
Cover_herb	Herbage coverage	0–2	0—no closed herb layer 1—sparse herb layer (coverage 10–70%) 2—dense herb layer (70–100%)
Deadwood	Deadwood	0–2	0—deadwood was not seen when we sampled the insects 1—moderate stock of deadwood (single stumps or trunks) 2—large stock of deadwood
Isolation	Isolation	0–1	0—plot is not isolated by roads, buildings, bogs or other physical barriers from neighbouring forests 1—plot is isolated by roads, buildings, bogs or other physical barriers from neighbouring forests
Old_forest	Presence of old forest	0–2	0—no forests within 5 km radius based on 1860 map 1—forests within 500 m to 5 km distance 2—forests on the same place or up to 500 m
Landscape	Landscapes	5 levels	Mos_mor—moraine plains of the last stage of the Moscow glaciation Edg_mos—plains of the early stage (marginal zone) of the Moscow glaciation Ers—erosional plains outside the boundaries of the Moscow glaciation Flvg—fluvioglacial plains of both Moscow and Don glaciations Rip—landscapes of river valleys

). Nine variables described the current properties of habitats; isolation characterized the landscape matrix; the presence of old forests represented a historical factor (modern history); and landscape type reflected both the landscape matrix and glacial history.

Nemoral and nitrophilous species were distinguished according to the concept of ecocoenotic groups in East European forests [8] using a published database [45]. Only dominant species in the upper herb layer were recorded. In our plots, the dominant nemoral herbs were Aegopodium podagraria L., Allium ursinum L., Carex pilosa Scop., Galeobdolon luteum Huds., and Mercurialis perennis L. Dominant nitrophilous species (or species typical of alder forests) included Chelidonium majus L., Filipendula ulmaria (L.) Maxim., Geum rivale L., and Urtica dioica L.

The presence of old forests was determined using the oldest accurate map of the region, published in 1860 and known as the Shubert map [46]. We examined whether forests were present at the locations of the sample plots, in their immediate surroundings (within 500 m), or within a radius of 5 km. To verify continuity of forest cover, we also examined maps published in 1919 [47] and 1942 [48].

To determine landscape (landform) types for each sample plot, we used an anonymous digital map corresponding to the scheme developed by Vitaly Esipov [38]. The digital map was provided by Alexey Streltsov from the Bioindication Laboratory at Kaluga State University. To reduce the number of categories, moraine–fluvioglacial and fluvioglacial plains of both glaciations were combined. Mapping was performed using QGIS.

2.5. Data Analysis

Data analysis was performed using R version 4.5.2 [49]. First, we examined correlations among the predictors. Herbage height and herbage cover were strongly correlated (Pearson’s r = 0.69, p < 0.0001) and were also significantly correlated with the presence of nitrophilous herbs (r = 0.53 and 0.53, respectively) and nemoral herbs (r = 0.24 and 0.39, respectively). In addition, these variables had a larger number of levels compared with the other predictors. Therefore, they were excluded from further analysis.

As the dependent variable was binary, we used logistic regression. At the first stage, logistic regression models were fitted using the glm function (family = binomial) from the base stats package. We tested different combinations of predictors with and without interaction terms. To assess model quality, we used AIC, the percentage of correctly predicted cases, and McFadden’s pseudo-R² from the pscl package [50]. For each predictor, we calculated the significance of coefficients (p-values) and partial R² values using the Anova function from the car package [51]. Non-significant predictors were removed. Some results are presented in the Supplementary File. Using different model combinations, we found that the significant predictors were the presence of coniferous trees, nemoral herbs, old forests, and isolation.

Because the set of sample plots represented a spatial dataset, we calculated a distance-weighted autocovariate using the autocov_dist function (type = “inverse”) from the spdep package [52]. The optimal results were obtained using a neighbourhood radius of 5 km: 55 points had zero autocovariation, the maximum autocovariation value was 0.031, and the mean value was 0.003. The final model was established using the Anova function from the car package [51]. The predictors included the presence of coniferous trees, nemoral herbs, old forests, isolation, landscape type, and autocovariation.

The next step was to identify factors affecting the presence of Ph. griseoaptera in urban areas and their surroundings. The largest urban area within the study region is Kaluga. We analysed a subset of the dataset consisting of 51 sample plots located within the urban area, suburban zone, and surrounding rural habitats of the Kaluga Gorsovet and Ferzikovsky District. These plots were situated within the plains formed during the early stage (marginal zone) of the Moscow glaciation, river valleys, and small fluvioglacial plains occupying a transitional position between two former landscape types. Since the previous analysis showed that the bush cricket avoided only moraine plains formed during the late stage of the Moscow glaciation, landscape type was not included as a predictor in this subset analysis. Instead, we analysed the effects of coniferous trees, nemoral herbs, old forests, isolation, and autocovariation.

To compare bush cricket abundance among different habitat types (including grasslands and gardens, which were not used in the analysis of factors affecting species presence), we used only pitfall-trap data. We calculated activity density as the number of specimens per 100 trap-days. Trap exposure duration was standardised to 90 days because adults were collected from July to September. We distinguished four habitat types: forests located in large river valleys (9 plots), other forests (18 plots), gardens (9 plots), and grasslands (6 plots).

3. Results

3.1. Distribution of Pholidoptera griseoaptera Across the Kaluga Region

We identified three major factors determining the presence or absence of Pholidoptera griseoaptera in forests and forest edges (

Table 3. Factors affecting the presence of Pholidoptera griseopatera in the forests of the Kaluga Region, Russia: results of a general linear model. McFadden R2 is 0.367, AIC = 183.46, accuracy is 81.9%.

Factor	Coeff.	Std. Error	z Value	Pr (>|z|)	LR_Chisq	Partial_R2
(Intercept)	−0.4570	1.3352	−0.3422	0.7322
Landscape:					25.0711	0.134
Flvg	−0.7789	1.2306	−0.6330	0.5268
Edg_mos	0.2994	1.2858	0.2329	0.8159
Mos_mor	−3.2771	1.3855	−2.3653	0.0180
Rip	−0.2310	1.1601	−0.1991	0.8422
Ers	15.7300	1446.1421	0.0109	0.9913
Con_Tree	−1.3725	0.4366	−3.1434	0.0017	10.5609	0.061
Old_forest	0.2022	0.2999	0.6742	0.5002	0.4556	0.003
Nem_herbs	2.1991	0.4553	4.8296	<0.0001	27.2026	0.144
Isolation	−3.4308	0.7535	−4.5533	<0.0001	26.5071	0.141
Autocovariation	121.1268	49.6305	2.4406	0.0147	6.5204	0.039

; see the highest partial R² values). Pholidoptera griseoaptera preferred habitats containing nemoral herbs and avoided habitats located on moraine plains formed during the late stage of the Moscow glaciation. Habitat isolation caused by physical barriers strongly reduced the probability of the species’ presence. The presence of coniferous tree species in the canopy was also a significant negative factor, although its contribution was relatively small. Autocovariation was significant as well, but its contribution was minor. The persistence of old forests was not a significant predictor.

The distribution map (Figure 3) shows that Pholidoptera griseoaptera is widespread in the eastern and southern parts of the Kaluga Region, where it occupies both river valleys and watershed habitats. In the western part of the region, the species was recorded only in river valleys. The boundary of the species’ continuous distribution coincides neither with the boundary of the Moscow glaciation nor with that of the Continental biogeoregion (or nemoral biome). From a physical–geographical perspective, the species inhabits the Central Russian Upland and adjacent low plains, but does not occur in the Baryatino–Sukhinichi Plain or the Smolensk–Moscow Upland. However, the landscape map based on Quaternary sediments appears to be a good predictor of the species’ distribution.

3.2. Distribution of Pholidoptera griseoaptera in Urban Areas

In Kaluga and its surroundings, the presence of Pholidoptera griseoaptera was primarily affected by the presence of old tree vegetation and habitat isolation (

Table 4. Factors affecting the presence of Pholidoptera griseoaptera in the forests and other tree-dominated habitats in Kaluga City and its surroundings. McFadden R2 is 0.789, AIC = 25.75, accuracy is 96.1%.

Factor	Estimate	Std. Error	z Value	Pr (>|z|)	LR_Chisq	Partial_R2	Pr (>Chisq)
(Intercept)	−3.020	2.541	−1.189	0.2346
Con_Tree	−9.151	6.112	−1.497	0.1343	6.106	0.308	0.0135
Old_forest	6.479	3.430	1.889	0.0589	22.408	0.620	<0.0001
Nem_herbs	8.968	6.506	1.378	0.1681	6.716	0.329	0.0096
Isolation	−7.504	3.792	−1.979	0.0478	12.296	0.473	0.0005
Autocovariation	−446.485	336.271	−1.328	0.1843	4.344	0.241	0.0372

). Nemoral herbs had only a minor contribution because they were present in almost all sample plots within this area. The contribution of autocovariation was also very low. These results suggest that populations of the bush cricket are highly fragmented in urban environments.

The species inhabits almost all suburban forests, whereas within the urban area it is distributed mainly along river valleys. We identified two habitats of the species surrounded by dense urban development (Figure 4). The first habitat is Grove Komsomolskaya, a nemoral pine forest massif covering approximately 30.4 ha. In Grove Komsomolskaya, Ph. griseoaptera was recorded along forest edges but was absent from patches with a dense tree canopy.

The second habitat is the territory of Khlustin City Hospital, covering approximately 13 ha and consisting of buildings, lawns, ruderal vegetation, shrubs, tree lines, and solitary trees. According to the 1860 map, this area was originally surrounded by unforested land, with the nearest forest located approximately 1 km away along the Kievka River. In 1919 and 1942, the territory was already surrounded by buildings, although large wastelands still separated the area from the river. By 1942, the area between the hospital and the river had been fully built up, and traffic intensity on the surrounding roads had increased substantially, preventing insect dispersal. Therefore, the population of Ph. griseoaptera has likely been isolated for at least the last half-century.

Within the territory of Khlustin City Hospital, Pholidoptera griseoaptera was recorded only in patches with unmown ruderal vegetation and mature trees or their undergrowth, and was absent from areas where the herb layer had been mown (Figure 5). Therefore, we suggest that mowing of the herb layer may influence the occurrence of the bush cricket. This hypothesis has not yet been statistically confirmed because all other sites inhabited by Ph. griseoaptera also have unmown herb vegetation. The current total area of unmown patches within this habitat is approximately 0.2 ha.

3.3. Abundance of Pholidoptera griseoaptera

In forests located within large river valleys, the abundance of Pholidoptera griseoaptera was significantly higher than in sample plots representing other habitat types (Figure 6). The plot with the highest abundance of the species was a small oak forest situated on the slope of the Oka River valley (N 54.4506 E 36.5162), while the second highest abundance was recorded in an alder forest located in the floodplain of the Oka River (N 54.2373 E 36.2588).

4. Discussion

4.1. Current Habitat Characteristics Affect the Distribution of Pholidoptera griseoaptera

As in Central Europe [28,53], in the Kaluga Region, within the nemoral biome, Pholidoptera griseoaptera is a common species inhabiting most forests and their edges. Outside the nemoral biome (or the Continental biogeoregion), its distribution is scattered, as well as in Central and Western Europe on the margins of this biome [19,54,55]. These results confirm that Ph. griseoaptera is associated with broadleaved deciduous forests.

Pholidoptera griseoaptera is regarded as an inhabitant of dense (and tall) herb layers [15,54], particularly of coarse herbage [21] or bushes [23] such as Rubus sp. [22,32]. Our results suggest that in the Kaluga Region, as in other areas located at the northern margin of the Continental biogeoregion [23], it is a forest species occupying only near tree vegetation. Therefore, tree vegetation (adult or young trees) is a key factor for the survival of Ph. griseoaptera. It is probable that tree vegetation moderates daily changes in temperature and other environmental parameters [23]. Bush crickets may get warm when they climb on the trunks [23]. We once observed this species using deadwood in the Kaluga Region. Anyway, as broadleaved forests originally have multi-layered and multispecies vegetation [8], they provide the highest structural heterogeneity of the environment, which may favour the insects [56]. Also, Ph. griseoaptera can consume insect carcasses [57], and broadleaved forests are rich in insect biomass [58].

However, based on our results, the most important habitat feature that determines the presence of the bush cricket is the presence of nemoral herbs as dominants of the herb layer. This result can have both ecological and historical explanations. Nemoral herbs, especially Aegopodium podagraria, Mercurialis perennis, and Allium ursinum, create a relatively stable microenvironment in the lower vegetation layer, characterized by higher humidity, smaller daily temperature fluctuations, and other specific microclimatic conditions [8]. This dense herbage can hide the bush crickets from predators, like birds. At the same time, shoots of nemoral plants are not too dense in the lower part to hinder the movement of the bush cricket (compare to grasses). The leaves of nemoral grasses can provide a good surface for nymphs to rest on if they need to bask in the sun. We often saw the nymphs of Ph. griseoaptera sitting on the leaves of A. podagraria, more rarely on other species. Sometimes we observed adult bush cricket sitting on the leaves of different plants, even Carex pilosa. Additionally, the bush cricket can feed on small insects living on nemoral herbs. Historically, nemoral herbs may reflect the existence of broadleaved forests in the past, even if now the area is occupied by small-leaved or coniferous forests [8].

We did not find a significant effect of the shrubs and undergrowth on our species. The shrubs may protect the bush cricket from occasional grazing or trampling by the livestock [23], but in observed plots, there is currently no livestock, and favourable microclimatic conditions may be formed by the herb layer.

We did not detect an effect of deadwood on the presence of Ph. griseoaptera. It may be assessed as indirect evidence of the ability of this species to complete the egg stage not only in rotten wood and dead bark [59] but also into leaf axils and pithy stems [26] and into the soil [25].

We found that Ph. griseoaptera disperses into the Boreal biogeoregion (or hemiboreal mixed forest biome) along large rivers. In the Continental biogeoregion, the forests in river valleys are optimal habitats for this species based on its activity density. It is possible that the rivers can facilitate the dispersion of this species by transport of eggs in wood during high water [55]. However, we found activity density to be high in non-flooded as well as flooded forests along the river. Thereby, we propose that the river valleys favour this species mainly due to the edge effect. Forest ecotones provide more suitable feeding and breeding (microclimate) conditions for this species [25], and forests in river valleys have been more fragmented due to human use than forests on watersheds [60]. In general, river valleys are warmer than watersheds [61]. So, species of the broadleaved forest disperse into the north along rivers [62]. The microclimate of the river valley can directly affect orthopterans [63]. Local disturbances from floods or landslides, which create patches with unvegetated soil, may facilitate the dispersion and reproduction of the bush crickets. It was a surprise that the medium-sized rivers, such as the Luzha and the Shanya Rivers, did not provide suitable habitats for the bush cricket. Probably, broadleaved patches along these rivers are too small and isolated. As Figure 3 shows, Ph. griseoaptera colonizes mainly the ancient (pre-quaternary) valleys of rivers.

4.2. Modern History Affects the Distribution of Pholidoptera griseoaptera

A long existence of the forests in places of sample plots was found not to be an important factor driving the presence of Ph. griseoaptera in the scale of the Kaluga Region. It may be explained by a sufficient dispersal ability of the bush cricket in fragmented landscapes if habitat patches are not isolated by any strong physical barriers. It corresponds to the results of research in Central Europe, which found that Ph. griseoaptera exhibits successful inter-patch dispersal in a fragmented landscape and thereby succeeds in maintaining viable populations in almost all suitable habitat patches [28]. A sufficient dispersal ability of this species is confirmed by regular capture of jumping specimens in window traps in our research, as well as in Central Europe [64]. Another explanation of this fact may be the long existence of small fragments of tree vegetation, which were not shown in any maps. However, the first explanation looks to be more probable, because examining old landscape photographs [65], we did not observe tree vegetation along rivers. Apparently, the livestock prevented the growth of tree vegetation in such places.

The isolation of habitats by roads, buildings, or wetlands affects the presence of Ph. griseoaptera significantly. The influence of such barriers on orthopterans is known [66]. Our results confirm the importance of inter-patch movements to support the metapopulation structure of the species [25,26,27,28]. Thereby, in urban areas, a long existence of tree vegetation with an unmown herb layer is a crucial factor affecting the survival of the species. It seems as if these patches might be small, and very small isolated populations of Ph. griseoaptera may survive for a long time. However, the traffic around two of the most isolated habitats of this species in Kaluga City (the Grove Komsomolskaya and Khlustin City Hospital) became intense only during the last several decades (based on observations of the first author, personal communications of citizens, and data on the dynamics of motorization in Russia [67]). So, we do not exclude gene flow into these patches 50 years ago and earlier. Therefore, we cannot guarantee long-term persistence of such small populations for a long time. In light of this, we should consider Ph. griseoaptera a threatened species in highly urbanized regions as has already been recognised in Moscow City [35].

4.3. Glacial and Postglacial History as a Probable Determinant of the Distribution of Pholidoptera griseoaptera

The most surprising result of our research is the absence of Ph. griseoaptera in the moraine plains of the last stage of the Moscow glaciation. Undoubtedly, this area became suitable for the forest insects later than other parts of the Kaluga Region. Based on the traditional concept of surface ice glacier in this area [38] or on any models of catastrophes in the Pleistocene [8] which anyway changed directions of the large rivers of the region [64], we should accept that about 70–60 kyr BP there were no suitable habitats of our species. As even narrow (10–50 m) geomorphological barriers may prevent dispersion of orthopterans [66], lowlands with postglacial waters could prevent colonization of areas which became suitable later than others. The persistence of glacial and postglacial geomorphological barriers prevented the dispersal of European orthopterans for a long time [68]. As current landforms mainly inherit pre-quaternary ones in the Kaluga Region [69], the lowlands now frame the continuous range of Ph. griseoaptera in this region. These areas may be regarded as an example of historical boundaries that were described for orthopterans [66]. A similar effect was found among Carabus ground beetles, whose species richness is affected by the distance of the site from the last glaciation border [70]. After the last glaciation, insects colonized Eastern Europe from the southern Carpathians or Balkans, the Caucasus refugia [71], and probably from forest patches situated along some rivers in the south of the East European Plain [72]. Thus, it seems reliable that Ph. griseoaptera colonized or recolonized the Kaluga Region from the southeast. However, the time since the last glaciation (about 60 thousand years BP) seems to be sufficient even for species with relatively low dispersal ability to colonize all suitable habitats.

At least in some areas, Ph. griseoaptera may have disappeared during the Holocene due to some changes in the vegetation and landscape structure. In the Kaluga Region, broadleaved deciduous forests decreased during the Valdai (late Würm) time (about 20 kyr BP) [73], and again after 2600 years BP [74]. Thereby, Ph. griseoaptera can disappear in named landforms at this time. Fluvio-glacial plains and valleys are usually dry and mineral-poor due to sandy grounds and growth by pine forests [8]. In historical time, fluvio-glacial plains were transformed by ploughing and fire [8]. Hence, these landscapes might serve as barriers for our species. So, postglacial barriers for our species may have existed relatively recently.

If we cannot reconstruct the history of the colonization or the extinction of Ph. griseoaptera in studied areas, we propose that the current distribution of the species across the Kaluga Region is affected both by historical barriers and by the current unfavourable landscape matrix of some areas. For example, the moraine plains of the last stage of the Moscow glaciation have a high percentage of swamps [73], so some suitable broadleaved and mixed habitats can be currently isolated by a low-quality matrix, and the relatively low mobility of Ph. griseoaptera might limit its distribution in these areas.

5. Conclusions

Our hypotheses were partially confirmed. Pholidoptera griseoaptera inhabits the edges of nemoral forests, and its distribution outside the nemoral biome is scattered. The long existence of the forests determines the distribution of this species only in urban areas where dispersion is suppressed. Across the Kaluga Region as a whole, the bush cricket is absent in the areas of the maximal development of the last glaciation. We propose that glacial history determined subsequent history and current specificity of these areas. Despite Ph. griseoaptera prefers fragmented landscapes, a high level of urbanization may threaten this species. So, all observations of this species in the cities are important. Ph. griseoaptera can live in a wide range of environmental conditions, but it probably marks historical areas of the broadleaved deciduous forests, at least in our study region.