Effects of Open and Forest Habitats on Distribution and Diversity of Bumblebees (Bombus) in the Małopolska Upland (Southern Poland): Case Study

Simple Summary Southern Poland represents one of the most diverse habitats for bumblebees (Bombus sp.); however, little is known about the abundance and distribution of many insect species in this region. Bumblebees are important for crop and wildflower pollination in different temperate latitudes because many plant species are only pollinated by them. Studies were conducted in natural and semi-natural habitats in southern Poland during the years 2003–2006 and compared with material collected from 2017–2020. During this eight-year-long study, more than 6214 bumblebee specimens of 25 species were found in the Małopolska Upland. The most frequently observed bumblebee species were: Bombus pascuorum, B. lapidarius, B. pratorum, and B. lucorum. The low-numbered bumblebees were: Bombus humilis, B. pomorum, B. veteranus, B. muscorum, and B. semenoviellus. There were also four rarely found species: Bombus confuses, B. ruderatus, B. soroeensis, and B. jonellus. Abstract Bumblebees are an important insect group occurring in different land ecosystems, but the number of these species has declined dramatically across Poland as well as in Europe in recent years. The fragmentation of bumblebee habitats influences the abundance and richness in community composition and trophic and competitive interactions. During the years 2003–2006 and 2017–2020, we studied the diversity and distribution of bumblebee species in two natural (boron-mixed Vaccinio-Piceetea and riparian forest Querco-Fagetea) and two semi-natural (segetal-ruderal Stellarietea mediae ruderal Artemisietea vulgaris) habitats in southern Poland. For that, we evaluated how habitats as well as local flowering communities influenced bumblebees’ abundance, richness, and community composition in 16 sites (which are located in four parks). Bumblebee communities responded to environmental factors in different ways according to the type of habitat. Vegetation factors were the most important drivers of bumblebee community structures. Forests showed the lowest bumblebee abundance, richness, and diversity, and the highest dominance levels of these parameters were found in the open ruderal-segetal habitats. The meadows from the Molinio arrhenatheretea class were characterized by bumblebee communities with a more complex structure. Species diversity was positively correlated with open ruderal-segetal habitats, and negatively with mixed forest cover, while abundance was positively correlated with forest cover. Studies like this are necessary to anticipate the impact of habitat fragmentation on bumblebee decline.


Introduction
Bumblebees play an important role as pollinators of many crop plants and wildflowers. Although the distribution of bumblebees encompasses a wide geographic range from Arctic tundra to lowland tropical forest, they are clearly most abundant in mountain habitats and the cold and temperate regions of the Northern Hemisphere [1,2]. Currently, more than 250 species are known within the genus Bombus to occur on all continents except

Collections
The bumblebee species were harvested every two weeks from mid-May to mid-Sep tember, between 9.00 a.m. and 4.00 p.m. on days without precipitation, with little or n wind, and with air temperature above 18 °C (Table S1). During each visit, all foragin bumblebees within a transect, as well as the plants on which they were observed, wer registered. We identified plants using the botanical keys [21][22][23]. Single specimens o bumblebee were caught with an entomological net and photographed live, free-foraging The specimens were prepared by using cyanide or ethyl acetate. All bumblebees wer placed in an airtight container with a few layers of tissue and the addition of a few drop of ethyl acetate. After the field sampling, the specimens were dry-mounted on standar insect pins for identification. In total, 1536 samples were collected from 2017-2020 an deposited in the Institute of Biology (Jan Kochanowski University, Kielce) entomologica collections following analysis. The community structure, species relationships, foragin activities, abundance, and phenology of every species were studied throughout the sea son. Species identifications made in the field were verified in the laboratory using the tax onomic keys of Krzysztofiak et al. [24], Pawlikowski [25], Dylewska and Flaga [26 Dylewska et al. [27], Williams et al. [28], and Banaszak [29].
In each habitat, soil physical (density, structure, and texture) and chemical (inorgani and organic matters) properties were studied using a multifunctional probe YSI Profes sional Plus. For a detailed analysis of the chemical parameters, water samples (500 mL

Collections
The bumblebee species were harvested every two weeks from mid-May to mid-September, between 9.00 a.m. and 4.00 p.m. on days without precipitation, with little or no wind, and with air temperature above 18 • C (Table S1). During each visit, all foraging bumblebees within a transect, as well as the plants on which they were observed, were registered. We identified plants using the botanical keys [21][22][23]. Single specimens of bumblebee were caught with an entomological net and photographed live, free-foraging. The specimens were prepared by using cyanide or ethyl acetate. All bumblebees were placed in an airtight container with a few layers of tissue and the addition of a few drops of ethyl acetate. After the field sampling, the specimens were dry-mounted on standard insect pins for identification. In total, 1536 samples were collected from 2017-2020 and deposited in the Institute of Biology (Jan Kochanowski University, Kielce) entomological collections following analysis. The community structure, species relationships, foraging activities, abundance, and phenology of every species were studied throughout the season. Species identifications made in the field were verified in the laboratory using the taxonomic keys of Krzysztofiak et al. [24], Pawlikowski [25], Dylewska and Flaga [26], Dylewska et al. [27], Williams et al. [28], and Banaszak [29].
In each habitat, soil physical (density, structure, and texture) and chemical (inorganic and organic matters) properties were studied using a multifunctional probe YSI Professional Plus. For a detailed analysis of the chemical parameters, water samples (500 mL) were taken, which were conserved with chloroform (CHCl 3 ) and stored at −10 • C for further analysis.

Statistical Analysis
The Simpson diversity index was calculated based on the data on the occurrence of insects in the studied sites. One-way ANOVA was used to test for changes in Simpson diversity at sites across years. The calculations of the index were made in the PAST program [30]. One-way ANOVA and graphs considering the mean value of the indicator, standard error, and standard deviation were calculated using the Statistica 9.0 software (TIBICO Software Inc., Palo Alto, CA, USA) [31]. The data on the number and occurrence of insects were also used to perform a direct ordination analysis (RDA). The file with environmental data was constructed on the basis of information about the year of research, the community, and the facility where the research was conducted. This information was coded in the "0-1" system. In order to determine which of the resulting variables were statistically significant for the diversity in insect occurrence, the forward-selection and a Monte Carlo permutation test were performed during the RDA. The result of the RDA is an ordinance diagram in which both species and environmental variables are marked with vectors. Changes in the number of species in samples in the ordination space were presented using a diagram. In order to indicate the tendency of species occurrence in sites and time, indirect coding analyses (Principal Component Analysis, PCA) were also performed. Ordinance analyses were done using the Canoco 5.0 software (Ithaca, NY, USA) [32].  For each of the study samples, the Simpson diversity index varied from 0 (one species in one of the samples) to 0.895 ( Figure 3) for either the year, community, or study site. In no case were the differences in the index values statistically significant. The analysis of redundancy indicated that the study years (2005,2006, and 2020), the communities (seminatural A1 and A2) and the study site (ChKLP,ŚNP, and COLP) were important for the differentiation in the number of bumblebees ( Figure 4). These variables account for 40.8% of the total variability of the study set. The significance of the variables also results from the relationship of some species with the studied variables; for example, Bombus pascuorum was the most numerous in 2006. Some species were present only in specific sites; for example, B. pratorum and B. sylvestris were most numerous in the COLP. The number of species collected from these two parks was also significantly higher in 2005 and 2006, compared with the other sites (ANOVA, p < 0.05). More species also occurred in semi-natural habitats such as on the fallow. This regularity became apparent when isolines indicating the number of species were placed in the ordination space ( Figure 5). The variables attached with a large number of species in the samples and inŚNP, where the species are significantly lower, are statistically significant.       The analysis of the indirect ordinance was presented three times to indicate regularities in the occurrence of bumblebee species depending on the years (Figure 6), sites (Figure 7), and habitats (natural and semi-natural) where the study was performed (Figure 8). The B. pascuorum and B. hortorum were recorded more frequently in 2005 and 2006 than later years. It was observed that especially in 2004, the occurrence of these species was The analysis of the indirect ordinance was presented three times to indicate regularities in the occurrence of bumblebee species depending on the years (Figure 6), sites (Figure 7), and habitats (natural and semi-natural) where the study was performed (Figure 8). The B. pascuorum and B. hortorum were recorded more frequently in 2005 and 2006 than later years. It was observed that especially in 2004, the occurrence of these species was sparse ( Figure 6). Species from the Pseudobombus family appeared sporadically in the study years, while in 2020, B. campestris and B. sylvestris were more numerous in semi-natural habitats.      (Table 2). Meanwhile, in the year 2017, 369 specimens of bumblebee were identified, and in the following years of the study, a decline in the bumblebee communities was observed.  (Table 2). Meanwhile, in the year 2017, 369 specimens of bumblebee were identified, and in the following years of the study, a decline in the bumblebee communities was observed.

Discussion
In this study, bumblebee diversity and distribution in the Małopolska Upland (southern Poland) in natural and semi-natural habitats during 2003-2006 and 2017-2020 were presented. The relationships between bumblebee community and habitat, study years, and species occurrence in parks were observed. During this eight-year study, we observed that the number of flowering plants and bumblebee abundance and richness decreased the composition of the communities changed, by reducing habitat-specialized species in favor of highly generalist ones. The greatest bumblebee species diversity was found in NLP (24 species), while in ChKLP, there were 23 species. Species such as: B. pascuorum, B. terestris, and B. lucorum occurred in natural and semi-natural habitats, in all studied years. B. hypnorum is considered a species typical from northern forests and, normally, nests above ground, mainly in tree cavities and in the wooden nesting boxes for birds. For bumblebee species, in study habitats, the number of flowering plants was important, as it is positively related to their abundance and richness. Bumblebees preferred plants such as: Caluna vulgaris, Centaurea jacea, Centaurea scabiosa, Lamium album, Rubus hirtus, Trifolium pratense and Trifolium alba, and Vicia cracca. Moreover, Westphal et al. [33] and Person et al. [34] described the relationship of bumblebee abundance and richness with flower density varied along the flowering season. The smallest quantity of species was captured in the COLP (21) and in the SNP (only 17). Natural and semi-natural habitats of SNP are poor in flowers and flower availability plays such an important role in bumblebee distribution. The number of described species from the Małopolska Upland also correlates with the research of Krzysztofiak [35], which studied 23 species of bumblebees in Wigierski National Park, whereas Pawlikowski [36] described only 21 species in pine stands in the Toruń Basin. It was observed that along with abundance and richness, the composition of the bumblebee communities also changed during the eight-year study. We concluded with Gomez-Martinez [37] that species adapted to forest habitats decreased in number with forest fragmentation, while species related to open areas become more abundant.
Nieto et al. [38] stated that, among European bees, the genus Bombus includes the highest percentage of species with an extinction risk according to IUCN criteria. Among the 68 species in Europe, 45.6% are stable, and 13.2% expose positive population trends and an expansion of their distribution [38]. The community of bumblebees occurring in the Małopolska Upland (southern Poland) needs permanent scientific monitoring. According to IUCN criteria [38,39], among the 30 bumblebee species found in Poland, 19 species were passed in the Polish Red List [40], 11 species labelled as vulnerable (VU), 6 as poorly recognized status (DD), and 2 as endangered (EN).
Sarospataki et al. [41] concluded that except for the UK, in most of Europe, detailed information on the abundance and distribution of bumblebee species is not available. The current status of pollinators in Central Europe, such as Carpathian Basin, is not well-known [38]; however, the fauna and ecology of bumblebees in the European North are quite well-studied [42]. More than half of the Małopolska Upland bumblebee species were found to be rare or moderately rare. The very similar situation is in Hungary [41], Belgium [42] and Bawaria [6]. Four bumblebee species from the Małopolska Upland were rare.
Bombus confusus Schenck is protected in Poland [40]. The distribution of this species is Europe and Central Asia [43]. This species does not reach the Mediterranean peninsulae, nor the Mediterranean sea coast. It is assessed as vulnerable in the IUCN Red List of European Bees [38]. In Poland, B. confusus is rare, mainly both on the edges of forests, warm meadows, and fields. It nests mainly in the ground. The species is endangered by mechanized forest management, the use of chemicals to combat pests of crops, burning and ploughing of balks, roadsides, removal of brushwood, fallow land, and increased tourism in Poland. The only observation of this species is from Nadnidziański Landscape Park.
Bombus ruderatus (Scopoli) is a large garden or ruderal bumblebee species, common in western Europe, especially in its Mediterranean zone. This species is very rare in Poland [40] and it was reported from three sites inŚwiętokrzyski National Park and Chęcińsko-Kielecki Landscape Park. The eastern limit of its range is less clear, though it includes Poland, after Hungary and Slovakia, and seems to reach Ukraine [44]. Bombus soroeensis (Fabricius) Palearctic is a species widespread in Europe (from Spain to the southern part of Central Siberian Plateau and Turkey, and the Carpathian), but highly localized [45]. This species is vulnerable in Poland [40] and known from single sites. It was found on the smaller-flowered legumes, such as Melilotus sp., and Campanula sp. It was collected from two sites inŚwiętokrzyski National Park and Nadnidziański Landscape Park. During our studies, B. soroeensis was found on xerothermic grasslands and in forest environments.
Bombus jonellus (Kirby) is a small species, widespread and common in Europe: from Iceland and the Sierra Cantabrica in the west, to the Anadyr on the Pacific. In the south of Europe, the range of this species is restricted to montane biotopes, reaching the highest mountains of the Iberian Peninsula where it is very rare [38]. This species occurs on the moorland (on the Ericaceae) and it is vulnerable in Poland [40]. During our studies, we observed that the most frequently visited plant species by B. jonellus were wild thyme (Thymus praecox), marsh cinquefoil (Comarum palustre), water avens (Geum rivale), and tufted vetch (Vicia cracca). This species was collected from two sites in Nadnidziański Landscape Park and Cisowsko-Orłowiński Landscape Park.
Rasmont and Mersch [13] and Rollin et al. [42] suggested that among European bumblebees, the severely declining species tend to be those with a low genetic diversity, a short flight season, a late emergence, a small number of habitat types, a long tongue, and a restricted dietary breadth associated a narrow pollen diet or with flowers with long corolla such as Fabaceae, such as B. humilis, B. ruderatus, and B. subterraneus. However, other studies [4,6,39,42] indicated that the climate may have an influence on complex interactions between ecological traits and environmental factors that may be associated with higher susceptibility of bumblebee declines. A similar situation has occurred with Hungarian bumblebees, where seven species are critically endangered (CR), three are endangered (EN), and two species are vulnerable (VU). B. soroeensis as a vulnerable species and B. subterraneus shows a frequency trend, but both are rare, and present in the IUCN Red List of European Bees [38].
Rollin et al. [42] concluded that traits correlated with higher rates of species extinction are a narrow geographic distribution, slower reproductive rate, low population density, and ecological specialization. Protective measures concerning bumblebees should mainly be the protection of the natural habitats and natural resources, which are the optimal place for the development of many species. The preservation of bumblebees is possible due to supplementing the food base by sowing attractive plants and planting pollen-bearing shrubs and trees, protection of forest islands, woodlands, and roadsides.
Several studies [46,47] documented that many environmental factors, such as temperature, wind, sun exposure, and humidity, affect the activity of bumblebees and their occurrence in different habitats. Our results indicate that the diversity of flowering plants occurring in natural and semi-natural studied habitats, as well as local features of the site itself are important for bumblebee diversity and abundance. Westphal et al. [33] implicated the importance of the habitats in maintaining bumblebee abundance and finding a correlation between the availability of flowering crops and bumblebee density. Rasmont et al. [27] suggested that climate change poses a threat to many bumblebee species worldwide and to change the structure of their functioning, including access to host plants. On the other hand, the expansion of some bumblebee species into new areas has been observed [48,49]. Distribution and occurrence data of bumblebees in new habitats indicates the ecological plasticity of these species and the possibilities of adaptation in the context of ongoing changes.
This study conducted in Poland indicates new locations of bumblebees and their dispersal, especially in the southern part of the country. Michołap et al. [50] suggested that the increase of continentalism in Europe could also be likely to affect the current status of bumblebee species expansiveness, while the biodiversity of plants occurring in Central and Eastern Europe may help these species spread throughout Europe.
In our research, we noticed that forest specialist bumblebee species exist, where Scoble [51] also found forest bees and forest butterflies which are considered creatures of open habitats [52], although in temperate zones, butterflies tend to be associated with open areas. We have suggested that the type of habitats might be an important determinant of pollinator responses to land use change with, for example, forest to open transitions having different effects from open-to-open transitions, as described by Winfree et al. [53]. Comparing habitat types, pollinator abundance and species richness in natural habitats, such as the forest, were often lower than in anthropogenic habitats. Steffan-Dewenter et al. [54], Sjodin et al. [55], and Winfree et al. [53] concluded that many studies use seminatural habitats, such as grazed grasslands, fallow agriculture, and suburban gardens as the good bumblebees' occurrence habitat. These semi-natural habitats are not then compared to natural habitats, such as the forest. However, the loss of semi-natural habitats has negative effects on pollinators of various species [56]. Tews et al. [57] concluded that low-level semi-natural land use may increase the heterogeneity of habitats and resources, thus increasing niche diversity. Klein et al. [58] observed that pollinators could nest in forest habitats, but forage in semi-natural habitats. Zajdel et al. [59] concluded that the size of parks, percentage of area covered by trees, and characteristics of the areas surrounding the parks were not significant for the diversity and abundance of bumblebees, however indicating the importance of the semi-natural habitats for the species diversity. Meanwhile, Diaz-Forero et al. [60] found that the presence of forest is very important for bumblebees, even for those species that seem to prefer open areas, because forest habitats may provide overwintering sites and nesting places.
Despite the overall decline in bumblebees, not all the species may respond similarly to landscape fragmentation. Studies have shown that while some species have considerably declined in fragmented habitats, others have remained relatively abundant. Habitat fragmentation can affect the diversity of bumblebee species and communities, and may depend, among other aspects, on their habitat preferences, foraging ranges, and behavioural patterns [37].
Bumblebees do not show so-called floral fidelity to all flowers encountered on their flight path and do not focus, like bees, on just one species that blooms en masse at a time [61]. Sikora et al. [18] suggested that bumblebee species can also be treated as bioindicators of the state of the natural environment, and natural and anthropogenic habitats (e.g., urban space) that can be a refuge for bumblebees and many other species of insects involved in the process of pollination of plants. In general, bumblebee species diversity, and especially common species, can be promoted in different kinds of landscapes by ensuring a variety of good-quality local habitats [62]. Rollin et al. [42] concluded that long-term records are necessary to estimate population trends accurately and to propose appropriate mitigation strategies. The Polish Bombus fauna needs much more protection and higher conservation efforts than they are benefiting from today.

Conclusions
The paper presented the results of scientific monitoring of bumblebees carried out within the Małopolska Upland in the years 2003-2006 and 2017-2020. Twenty-five species of bumblebees were found on 16 sites. Based on an original data set of 6214 specimens, we also assessed a high proportion of species declining, analyzing both richness changes and species range sizes during the last 8 years. Our results indicated that the natural and semi-natural resources habitats in park areas are important for bumblebee diversity and abundance in the Małopolska Upland (southern Poland). Providing flowering areas might enhance the diversity and abundance of bumblebees, as well as other insect pollinators. Plant species diversity and composition are the most important factors determining bumblebee abundance and diversity.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.