Revealing the biodiversity of wild bees (Hymenoptera: Apoidea: Anthophila) in flower strips in Mediterranean floodplains. Which monitoring method fits best?

doi:10.21203/rs.3.rs-4846902/v1

Download PDF

Research Article

Revealing the biodiversity of wild bees (Hymenoptera: Apoidea: Anthophila) in flower strips in Mediterranean floodplains. Which monitoring method fits best?

https://doi.org/10.21203/rs.3.rs-4846902/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Context

The ongoing pollinator decline may threaten and compromise the resilience of terrestrial ecosystems. Implementing conservation action requires monitoring pollinator populations' actual status, but this is particularly difficult for pronubes insects such as wild bees. Their monitoring is difficult and time-consuming but crucial for assessing their health status.

Objectives

Here we compared and evaluated the efficiency of three different monitoring methods to evaluate wild bee biodiversity in lowland areas sown with entomophilous flowers to support pollinating insects in a Mediterranean environment.

Methods

We sampled wild bees for two years by using hand netting, pan traps and artificial nests. We compared species richness and abundance among these methods with a particular focus on how flowering coverage affects the efficiency of walking transects and pan traps and discussed the attractiveness of the different colours of pan traps.

Results

Hand netting captured a higher abundance of wild bees than the other two methods but a similar number of species to pan traps. Artificial nests captured fewer specimens and species. Bee assemblages were significantly different between pan traps and hand netting, and pan traps had greater potential in capturing the whole bee biodiversity, but their attractiveness is negatively influenced by the flowering coverage contrary to hand netting sampling.

Conclusions

Like other studies, the three sampling methods are complementary regarding species assemblages. The juxtaposition of several monitoring methods is essential to assess the biodiversity status of species with such particularly different ecological traits.

biodiversity

Anthophila

pan traps

bee nests

monitoring schemes

Pollinator biodiversity and decline

The complexity and resilience of terrestrial ecosystems are highly related to pollinators (Bascompte et al. 2006). The dependence of angiosperm species on zoogamous pollination is estimated to vary from 78% in temperate systems to about 94% in tropical areas (Ollerton et al. 2011). At the same time, many crops important for healthy human nutrition largely depend on animal pollination (Klein et al. 2007; Potts et al. 2010). The steady decline of pollinators globally is well documented and the need for actions to counter it is clear and accepted (Dunn et al. 2009; Klien et al. 2007; Potts et al. 2010). This problem could be particularly relevant in regions with a Mediterranean climate, known to be important hotspots for pollinator biodiversity (Orr et al. 2021). The European Union is working on multiple measures to reverse the trend of pollinator decline both in natural and anthropized ecosystems (Moldoveanu et al. 2024). These are condensed in the recently published document “A new deal for pollinators” (EU Communication, 2023), highlighting that the conservation of pollinators also requires i. improving knowledge on the causes and consequences of their decline and; ii. implementing and coordinating policy decisions on their conservation (EU Communication, 2023). Within the Biodiversity Strategy 2030, the Nature Restoration Law focuses on the restoration of 20% of the degraded terrestrial and marine ecosystems by 2030 and of all damaged environments until 2050 (https://environment.ec.europa.eu/topics/nature-and-biodiversity/nature-restoration-law_en) and since habitat loss and fragmentation are among the major drivers of decline of pollinator decline (IPBES 2016; Nieto et al., 2014; Potts et al. 2010), these actions will also favour this functional group of insects. Among pollinator species, wild bees are considered to be the most efficient group due to their great species richness and functional diversity (Michener 2007). The European continent counts 2.138 species of wild bees (Ghisbain et al. 2023) and the Mediterranean region has the greatest species biodiversity (Quaranta et al. 2018). Yet, the IUCN Red List of bees classified more than 56% of the species as “Data Deficient” (Nieto et al. 2014). The lack of data makes it impossible any classification into risk categories and decisions about actions for their protection (Nieto et al. 2014, Potts et al. 2016). Despite this knowledge gap, it is estimated that about one in ten European wild bee species are threatened with extinction (Nieto et al. 2014; Status and Trends of European Pollinators, www.STEP-project.net). To overcome the lack of data on the trends of wild bee populations, it is necessary to improve the monitoring efforts. Still, unlike for the butterfly species (e.g. BMS, Sevilleja et al. 2019), there is no consensus among states on the best standard monitoring method. Yet, monitoring schemes for wild bees are important to understand how populations respond to stressors and how they may develop in time on local and global scales (Breeze et al. 2021).

Monitoring methods for wild bees

According to many assessments (IPBES 2016; IUCN 2014; Millenium Ecosystem Assessment 2005) and researchers (Bell et al. 2022; Hutchinson et al. 2021; Klaus et al. 2024; O’Connor et al. 2018), there is a clear need for standardized methods of monitoring pollinators through a wide range of habitats and many geographical areas. The monitoring methods that are usually adopted for pollinators, particularly suitable for assessing wild bees, are divided into passive and active methods (Westphal et al. 2008). Active methods include primarily walking transects and observation plots (Cane et al. 2006; Dafni et al. 2005) whereas passive methods are numerous: pan traps (Rhoades et al. 2017; Westphal et al. 2008), malaise traps (Geroff et al. 2014), vane traps (Hall et al. 2018), and trap nests (Staab et al. 2018). Walking transects are permanent corridors often 250m long and 4m wide, usually divided into 10 sections of 25m each (Westphal et al. 2008). They differ from those used for butterfly monitoring which are longer, usually up to 500m (Sevilleja et al. 2019). Observers are required to walk the transect for a maximum of 50 minutes, about 5 minutes for each section, and shall annotate or capture each specimen that cannot be determined in the field (Westphal et al. 2008). Observation plots consist of patches of various standard sizes in which the researcher actively observes generally for 30 minutes the organisms flying in (Banaszak 1980). These active methods are often time-consuming and heavily dependent on observers' expertise in species determination, but they allow the collection of much more information such as that about plant-pollinator interactions (Bell et al. 2023, Westphal et al. 2006).

Pan traps are plastic bowls of 400–500 mL volume that are coloured with UV-reflecting varnishes, usually in blue, white and yellow, and contain water with a few drops of liquid soap to reduce the surface tension (Westphal et al. 2008). Vane traps are similar to pan traps, but they are made of a transparent glass jar topped with a plastic-coloured (UV-reflecting) funnel that traps insects in the jar (Stephen and Rao 2005). Usually, both types of traps are left in the field for up to 24–36 or 48 hours (Stephen and Rao 2005; Westphal et al. 2008). Malaise traps consist of a large tent-like structure that traps and funnels insects towards a container with a preserving liquid; the choice of positioning site is crucial to the proper functioning of the trap (Townes 1962).

Passive methods are less time-consuming, and they are independent of the sampler knowledge in the first instance, but they may be affected by vegetation height and distribution (Cane et al. 2001; Westphal et al. 2008). Despite their similarity, pan traps sample mainly small-sized bees while vane traps were first designed for the sampling of large-bodied bee species (Stephen and Rao 2005; Westphal et al. 2006); moreover, malaise traps are used mainly for non-Lepidoptera species but have been rarely used for wild bees (Townes 1962, Klaus et al. 2024). Trap nests are highly biased to the sampling of only cavity-nesting bee species (Dafni et al. 2005) and their design may vary among researchers (Klaus et al. 2024). Plastic tubes of 10 cm diameter and 20 cm length, filled with common reed (Phragmites australis) with diameters up to 10mm are the most employed trap nests (Gathmann et al. 1994) but book-like structures may be also employed (MacIvor 2017). Trap nests are placed in the field usually in the late winter season and then are analysed after the autumnal months; this allows researchers to collect data over a whole flight period with relatively low sampling effort and to gather additional information such as that from pollen analysis and trophic interactions with associated parasitic fauna (Prendergast et al. 2020; Staab et al. 2018).

Since bee species richness can be relatively high and they also can exhibit great ecological diversity, many studies (Chamorro et al. 2022; Hutchinson et al. 2021; Prado et al. 2017) recommend a combination of different sampling methods to achieve efficient monitoring and an exhaustive characterisation of the wild bee community. The most used combination of sampling methods is vane traps, pan traps and standard transects (Bell et al. 2023, Willson et al. 2008). Furthermore, several studies have compared monitoring methods to understand their pros and cons, and the most suitable to standardise results (Bell et al. 2023; Hutchinson et al. 2021; Nielsen et al. 2011; Thompson et al. 2021), but those methods are often difficult to confront above all when it comes to considering passive and active methods together.

Many studies have compared primarily pan traps and vane traps (Bell et al. 2023; Rhoades et al. 2017). Even if many researchers demonstrated that pan traps capture the greatest abundance of bees, mostly small-sized and ground-nesting species (Thompson et al. 2021; Westphal et al. 2008), others found that, when comparing pan traps and vane traps, the second ones capture the highest number of specimens (Bell et al. 2023; Rhoades et al. 2017; Prendergast et al. 2020). In general, it was demonstrated that vane traps, especially the blue-coloured ones, were more efficient than pan traps by capturing a higher abundance of species and exhibiting major reliability across different types of habitats (Bell et al. 2023; Hall 2018). The reasons for the diverse attractiveness of colours are still debated (Gibbs et al. 2017; Hall 2018). According to several studies (Chamorro et al. 2022; O’Connor et al. 2018; Krahner et al. 2021; Thompson et al. 2021), there are other concerns about the efficiency of pan traps especially because they seem to be appropriate only for open sampling areas and for habitats with low abundance of flowering. Some authors demonstrated that pan traps compete with flowers for pollinators' attractiveness showing that the number of captured bees decreased when the flowering surface increased but this result was not reflected in species richness results (Kuhlman et al. 2021). Ortiz-Sanchez and Aguirre-Segura (1992) also demonstrated that pan traps are biased by the vertical height of sampling, unlike vane traps. Moreover, trapped specimens are usually immersed in water or other preservative liquids, thus they must be washed and dried and this may be very time-consuming and damaging for the exemplars, hindering the identification by morphological characters (Prendergast et al. 2020). Additionally, if comparing passive and active methods, there is a need to consider multiple elements, including differences in sampling effort (Bell et al. 2023; Krahner et al. 2021). Hand netting is highly dependent on flowering richness and on samplers' expertise, but it may also allow to capture parasitic bees that are just sporadically patrolling flowers but usually are not detected by passive sampling methods (Popic et al. 2013; Thompson et al. 2021). These are also useful when assessing wild bee species richness or investigating the structure of wild bee communities (Nielsen et al. 2011; O’Connor et al. 2018; Thompson et al. 2021). Finally, many authors show agreement in suggesting that the choice of sampling methods must consider primarily the objectives of the experiment (Bell et al. 2023; Klaus et al. 2024) and the life-history traits of species (Hall 2018) but also how habitat structure may influence monitoring methods (de Assis et al. 2021; O’Connor et al. 2018).

Here we compared bee monitoring methods, among pan traps, walking transects and trap nests, to study species richness and abundance in floodplain areas recently subjected to massive earth-moving works and then to sowing of commercial mixtures of nectariferous and polliniferous plants to support wild pollinator insects. Alongside we tried to identify the most suitable monitoring method for these specific habitats by combining the use of pant traps, walking transects and trap nests.

2.1 Study sites and area description

The data was collected during two consecutive years, 2022 and 2023, in two areas in Tuscany (Central Italy). Both areas are located in floodplain zones of two major tributaries of the Arno River (Fig. 1). The first area is named “Castelletti”, and it is enclosed by an elbow of the Ombrone River in the municipality of Carmignano (PO) while the second area, named “Bramasole”, is found in the municipality of Montelupo Fiorentino (FI) along the Pesa River (Fig. 2). Both areas are within recently realised expansion basins, they are less than 9 km apart in a beeline and are ecologically similar. The first area covers a surface of about 6.5 ha, the second about 5 ha and both are interested in a major experimental three-year project of environmental amelioration for the conservation of pollinators. Five plots of 0.3 ha were sowed within both areas with three different mixtures of entomophilous plants, and then pollinators, specifically wild bees (Hymenoptera: Apoidea: Anthophila) were monitored. The sites are embedded in peri-urban and agricultural landscapes and may experience flooding during wet seasons but are also subjected to drought during summer. These types of landscapes may be very suitable for solitary bees due to a wide range of nesting sites and because the slopes of embankments are suitable for underground nesting bees. Despite these elements, these areas, due to recent large-scale hydraulic works to reshape and excavate land to construct detention basins, are poor in trophic resources since entomophilous wild flowering plants are limited and the ground is covered mainly by gramineous weeds.

2.2 Bee sampling methods and data collection

Sampling was carried out from May to September in 2022 and from March to July in the year 2023. The two periods of sampling are mismatched due to the seeding works and maintenance requirements of the areas by the competent authority that manages them (Consorzio di Bonifica Medio Valdarno 3). We used three different methods of sampling: i. hand netting (or standardised walking transects); ii. pan traps; iii. artificial bee nests. Only bee nests were used as a sampling method during the first year in the “Bramasole” area. Transects were walked from once to twice a month during sunny and no-wind days. Each transect was 250 m long, divided into 10 sections of 25 m each, and was walked for a maximum of 50 minutes. For each area, we walked five transects per sampling session. During each transect, field forms were filled with information such as atmospheric temperature, cloud cover and wind force indexes, and flowering coverage. Specimens were determined up to genera in the field and along it, the visited plant species were recorded for each observation. Specimens were netted and placed into falcon tubes of 15 mL, anaesthetised with acetic ether then placed at -20° C in the laboratory. Subsequently, they were processed and conserved in entomological boxes. Pan traps were positioned on the same days of transects, in pairs of triplets placed 15 m away along the transect path. Each triplet was composed of blue, white, and yellow bowls and was left in the field for 24 hours. Each trap of the triplet was arranged as the vertices of an equilateral 5 m-side triangle (Westphal et al. 2008). Traps were made of reusable plastic bowls 15 cm in diameter, had a capacity of 500 mL and were coloured with UV-reflecting paints (Spray-Color GmbpH, Westphal et al. 2008)

Sampled specimens were put in 15 mL falcon tubes and conserved at -20° C in the laboratory. Later, they were quickly rinsed in water, dried in the air by gently shaking them in a small bag containing small pieces of blotting paper and then processed the same way as described for the netted specimens.

Trap nests were placed in the late spring and were removed in December for the year 2022 while for 2023, they were placed in the early spring and removed again in December. A total of six nests were placed, three for each study area, and they were constructed in such a way that they are also pleasing to sight since citizens can exploit the experimental areas for pleasure. Trap nests were made of a wood frame and filled with drilled wooden logs and river reeds (Arundo donax). Both wooden logs and river reeds were 20 cm long, the diameters of the reeds ranged from 0.5 cm to 2.5 cm, and the holes in the logs were 10 cm deep and 0.4 cm in diameter. During the second year, part of the truncheons was replaced by florist's sponges with transparent straws of compostable plastic inserted to facilitate pollen observation and extraction. After the dislocation, trap nests were taken to the laboratory waiting for cavity-nesting specimens to emerge, then cavities were opened to extract pollen and to identify the remaining wild bees. Pollen was used for studies on ecological networks and trophic preferences of bees, while specimens were processed the same way as described for the transect specimens.

2.3 Species identification

Only specimens of the superfamily Apoidea, clade Anthophila (Hymenoptera:Apoidea:Anthophila) (Michener 2000), were considered during sampling. Apis mellifera was recorded but was not considered for the analysis. In the present study, we identify at the species level all specimens belonging to the family Megachilidae; this taxon has been among the most represented groups in the two years of sampling, possibly because of the availability of several nesting microhabitats. All other specimens were identified up to the species level, or morphospecies for the Andrenidae and Halictidae families, unless necessary identification characters missing or unrecognizable. We used dichotomous keys (see References) both for genus and species identification.

2.4 Data analysis

We performed statistical analysis using R 4.4.0 (R Core Team 2024). Rarefaction curves were conducted with iNext package (Hsieh et al. 2016) to understand how hand netting and pan trapping captured the α-diversity of the sites as the number of individuals sampled increased. The number of sampled specimens and species were used as proxies to compare active and passive methods in terms of effectiveness and sampling effort. To compare the efficiency of these two methods for capturing species richness we performed a non-parametric Wilcoxon signed rank test with continuity correction, whereas a Kruskall- Wallis test adjusted with the Bonferroni method was used to compare the attractiveness of different colours of pan traps. A GLM model was computed to evidence the influence of flowering coverage on the two sampling methods. We used the vegan package to address the composition of wild bee communities between hand netting and pan traps and we used “adonis2” function with 999 permutations to perform a distance-based multivariate analysis of variance (PERMANOVA). We then evaluated multivariate dispersion, that is how different samples are from one another, for both sampling methods using the function “betadisper”. For all these analyses we used Bray–Curtis dissimilarities. VennDiagram package was used to show both the difference of captures among the three colours of pan traps and to display the distribution of Megachilidae species among the three sampling methods. ggplot2 package was used to construct all boxplots and barplots; finally, ggtern package was used to design the ternary plot for the Megachilidae species.

We collected wild bees from 30 genera (about 51% of all genera on the Italian territory) from all six families present in Italy (Table 1), for a total of 2,703 specimens. Hand netting captured 23 genera, approximately 76.7% of all genera collected, among which 8 genera were not present in the pan traps. Pan traps sampled 21 genera, about 70% of all collected genera and 6 that were not observed with hand netting. Artificial nests trapped 5 genera, about 16.7 % of all collected genera, with one genus that was not captured by the other two methods. Among all, we sampled 6 genera of parasitic bees (cuckoo bees) with the three methods: one genus (Coelioxys) with artificial nests, two genera (Nomada and Sphecodes) with pan traps and 5 genera (Epeolus, Melecta, Nomada, Pasites, and Sphecodes) with hand netting. Artificial nests and hand netting caught respectively 1 and 4 unique genera of parasitic bees.

The most abundant genera, in terms of sampled specimens, were Bombus (23.2%) followed by Eucera (17.9 %), Lasioglossum (13.7%), Halictus (11.5%) and Andrena (8.8%). Parasitic genera were the less abundant. Hand netting detected 2121 specimens from both observations and captures which is about 78.5% of all specimens observed. Pan traps captured 395 specimens (about 14.6% of all specimens), more than five times fewer specimens than hand netting. Finally, artificial nests trapped 187 specimens (about 6.9% of all specimens), more than ten times fewer specimens than hand netting. We collected 155 species (and morphospecies) over the two years, 96 species (61.9%) from hand netting, 95 species (61.3%) from pan traps and 14 species (7.1%) from artificial nets. We focused only on data from hand netting and pan traps to compare the monitoring methods' efficiency because, as expected, the data from artificial nests showed that they only captured a small fraction of the species. The number of species (and morphospecies) collected over the two years (Figure 3) by hand netting was significantly higher (Wilcoxon rank test, p < 0.05). Species (and morphospecies) rarefaction curve (Figure 4) cumulative on both sampling years showed that pan traps and hand netting collected almost the same number of species but pan traps have a greater potential to detect species as the number of individuals increases (see extrapolation curve of Figure 4 and Figure S1, Supplementary Materials). Also, the number of singletons (species for which a single specimen was collected) greatly differs between pan traps (49) and hand netting (20).

The attractiveness of pan traps and the efficiency of walking transects are significantly dependent on the flower coverage of the monitored site (Table 2 a and b). While the effectiveness of pan traps is high at low flowering coverage, the opposite is true for walking transects.

Wild bee assemblage differs between the two sampling methods (Figure 5, R2 = 0.093, F = 2.88, p <0.001) but the distance between- samples for each group is not significant (Beta Dispersion Analysis, p > 0.05), although higher for pan traps.

Pan traps of different colours attracted a similar number of individuals (124, 144 and 127 respectively for blue, white and yellow pan traps). All three colours attracted the same number of genera (14 genera for each colour), but the blue ones attracted a unique genus more than the other two colours. Yellow pan traps attracted more species (and morphospecies) among the three colours but there is no significant difference (Figure 6, Kruskall-Wallis, p > 0.05). Blue pans attracted Amegilla, Dasypoda and Tetralonia as unique genera; the yellow ones trapped Hoplitis and Panurgus as unique genera; the white ones collected Chelostoma and Hylaeus as exclusive genera (Figure 7a.). The relative proportion of the abundance of each genus among the three pans colours (Figure 7b.) showed some unique patterns: bees of the genus Systropha were mostly attracted by the white colour as most of the Andrena specimens; Eucera individuals were most abundant in the blue pans, Lasioglossum preferred the yellow traps and finally, Halictus bees were equally distributed among the three colours. The three colours attracted a similar number of species and morphospecies (respectively 54, 48 and 38 of all collected species for the white, yellow and blue pan traps).

Focusing on the species of the Megachilidae family, we collected 9 genera (30% of all sampled genera) and 38 species (24.5% of all species collected) with one genus, represented by one single species, the cuckoo bee Coelioxys mandibularis. Hand netting (20) and artificial nests (14) collected a greater number of species than pan traps (12). Only one species, Megachile analis, was captured by all three sampling methods, while hand netting and artificial nests shared 6 species, hand netting and pan traps 2 and artificial nests and pan traps shared only one species. Hand netting captured (Figure 8a) the highest number of unique species (n=13, 34.2%), followed by pan traps (n=10, 26.3%) and artificial nests (n=8, 21.1%). Visualization by the ternary plot of the Megachilidae species assemblage collected with each method (Figure 8b) showed that species tend to cluster for each monitoring method, as many species tend to be sampled exclusively or mostly by hand netting and artificial nests.

Table 1. List of all collected genera by each of the three sampling methods: Hand netting, Pan traps and Artificial nests. Numbers indicate the number of specimens for each genus and the total individuals collected by each method. * indicates the parasitic genera.

Table 2. Parameter estimates of the Generalized Linear Models (GLMs) with Poisson distribution and log-link function assessed as most parsimonious according to the Akaike information criterion (AIC). The model explains respectively the significative negative and positive influence of high flowering coverage on the number of captures by pan traps (a) and hand netting (b).

This study demonstrates how different monitoring methods, regarding wild bee communities, are complementary, with each method capable of collecting mutually exclusive species. Active and passive methods are often difficult to compare, and, to the best of our knowledge, only a few studies report comparison based on extensive sampling and a high effort in species resolution (Nielsen et al. 2011, Rhoades et al. 2017, O'Connor et al. 2018, Krahner et al. 2021, Kuhlman et al. 2021, Thompson et al. 2021). Furthermore, the strong dependence of these methods on habitat types, vegetation composition and bee fauna can weaken the comparison. Many studies point out the use of trapping, especially pan traps, as the most efficient monitoring method, as these can attract both more specimens and more species than other methods (Stephen and Rao 2005, Westphal et al. 2008, Nielsen et al. 2011, Krahner et al. 2021). However, the use of pan traps raises doubts regarding the effect of their repetitive and extensive use on the bee communities and bycatches of other insect groups (Gezon et al. 2015); as well as on the relative lack of standardisation due to the influence of vegetation (LeBuhn et al. 2012, Kuhlman et al. 2021). The efficiency of pan traps is highly dependent on the height of the vegetation and thus on the different heights at which the traps are placed, so that they are competitive with the surrounding vegetation (Westphal et al. 2008). Furthermore, as we also found, pan traps can lose efficiency when flowering coverage is abundant since, in addition to colour, many pollinators are attracted by other factors that characterise flowers but not pans (O’Connor et al. 2018). Finally, pan traps are subject to weather phenomena for at least 24 hours, thus resulting in high variability of the results, which may be affected by sudden violent weather phenomenons or to large animals tipping over, with consequent loss of sampling data (LeBuhn et al. 2012, O’Connor et al. 2018, Kuhlman et al. 2021). Our results are consistent with those of Cane et al. (2000) and Thompson et al. (2021) who reported that generally, hand netting captured more specimens and species than pan traps. In our case, however, the lower efficiency of pan traps could be due to other factors, and this is visible in the rarefaction and extrapolation curves divided by year. This study took place in an experimental field where areas were sown with strips of entomophilous plants specifically designed to attract pollinating insects. The strips were sown in the early spring of 2022 and monitoring took place immediately at their flowering, a couple of months after sowing (see Supplementary Materials). The flowering and vegetation coverage in that year was rather low due to the work carried out to sow the seeds and the fact that most plants generally have greater blooming in their second year (Kowalska et al. 2022). In addition, the areas are subject to high hydric stress during the dry summer season, so in the first year, vegetation cover and flowering did not reach high values. The traps were always placed on the ground and captured more species than transects, which in turn depended heavily on the presence of blooms on which to observe and net specimens. The flower coverage dependence of pan traps and transects is inversely proportional (O’Connor et al. 2018) and this is validated also in our study. Whereas as flower cover increases, walking transects are more efficient in attracting specimens, pan traps lose efficiency. Pan traps were less attractive to bees and transects captured more species in 2023. However, the total rarefaction and extrapolation curves are in line with the results of other studies in indicating the great potential of pan traps in attracting a high diversity of bee species at the same sample size (Nielsen et al. 2011, Hall 2018, O’Connor et al. 2018, Krahner et al. 2021). Pan traps catch many more singletons than transects, offering the possibility of attracting more species based on fewer captured specimens. However, an ideally exhaustive sampling of all species at a given site, based solely on the use of pan traps, would require a greater sampling effort than hand netting in terms of several sampling sessions and higher hours of placement of traps. Finally, pan traps tend to catch a more diverse and variable community of bees than hand netting which are inclined to sample much more similar communities each time. Our results are consistent with the study by Kuhlman et al. (2021) who also investigated seasonal and flower abundance patterns on the efficiency of pan traps and hand netting. In our case, hand netting was also crucial for the collection of parasitic genera that can sometimes be found feeding on nectar on some flowering species but are rarely attracted to pans. The use of artificial nests as an exclusive monitoring method entails, in the case of wild bees, the monitoring of only a fraction of diversity, that is confined to cavity-nesting species. As in Nielsen and colleagues (2011), nests captured the fewest specimens and species, with Osmia (5 species and 146 specimens) and Megachile (5 species and 31 specimens) being the most abundant, but they were crucial for capturing 8 species that had not been observed otherwise (5.2% of total species and 21.1% of Megachilidae species). Among Osmia, most individuals belonged to the species O. caerulescens, in contrast to the other species of Osmia that were mostly observed through walking transects and pan traps.

The wild bee communities sampled by hand netting and by pan traps were distinctive between the two monitoring methods, confirming, as already seen in other studies (Nielsen et al. 2011, O’Connor et al. 2018), that these two methods are highly complementary to each other. Some studies (Stephen and Rao, 2005; Hall 2018) support the hypothesis of a strong preference for the blue colour of traps for most of the wild bee fauna. In our study, there is no significant overall preference for one of the three colours, although, the greatest abundance of individuals and the greatest diversity of species was captured by the white and yellow traps. The blue traps, on the other hand, attracted both lower abundance and lower species diversity. According to Hall (2018) in the northern hemisphere the blue-coloured traps have greater attractiveness because of a greater abundance of bumblebees that prefer this colour. This study took place in a Mediterranean and lowland habitat and although bumblebees are, from an abundance point of view, the dominant group in the walking transects, very few specimens were caught with both blue and white traps. A possible explanation may also be that we used pan traps and not vane traps, unlike Hall (2018). The lower abundance of bumblebees sampled with pan traps is consistent with the results of Bell and colleagues (2023) who demonstrated that vane traps are more efficient at catching bumblebees than pan traps. The blue traps, however, attracted more exclusive genera than the others, in particular the genera Amegilla, Dasypoda and Tetralonia. These three are genera of long ligula bees, that have a strong preference for Boraginaceae and Malvaceae, all of which are flowers in the blue-purple range (Michener 2007). The blue traps also had greater attractiveness towards bees of the genus Eucera, which also have strong preferences towards Boraginaceae and blue purple Cichorioideae (Michener 2007). The white traps, on the other hand, attracted high abundances of Hylaeus (exclusive genus), and Systropha. These short ligula genera show preferences for white-coloured flowers such as Umbelliferae and Convolvulaceae (Michener 2007, https://www.beewatching.it/). Finally, yellow traps attracted abundant specimens of the genus Lasioglossum. Although the species of this genus are largely polylectic, they are often found foraging on mostly yellow-coloured Asteraceae (Michener 2007). These patterns indicate how indeed, according to the literature (Nielsen et al. 2011, Rhoades et al. 2017, Kuhlman et al. 2021), the most suitable sampling method is extremely influenced by the studied habitat and by the local composition of the bee fauna. Therefore, if interested in monitoring a particular group of bees, it is necessary to assess beforehand what might be the best method or combination of methods to capture the diversity of the entire group. In the Megachilidae family, we find genera and species with very diverse trophic and nesting habits (Michener 2007, Danforth et al. 2019). For example, within the genus Osmia or Megachile, we find species that nest in hollow stems and holes in wood, but also many species that prefer to nest on the ground or in snails (Michener 2007, Danforth et al. 2019). The trophic habits are also different, while the genus Megachile is known to be related to Fabaceae, the genus Chelostoma prefers Ranunculaceae and Asteraceae species (https://www.beewatching.it/). This makes it difficult to identify a single sampling method. Our results showed that all three monitoring methods are strongly complementary. We did not find any Megachilidae species of conservation concern in our monitoring, but they are very important species concerning the pollination of fodder crops and fruit trees, and therefore of some importance to human activities (Ollerton et al. 2021). Furthermore, the use of artificial nests also makes it possible to investigate parasites, parasitoids and hyperparasites of these species, which make up a not-so-insignificant fraction of the biodiversity of an ecosystem (Klaus et al. 2024). Indeed, the only parasitic species among the Megachilidae was captured thanks to the use of artificial nests. In addition to this, it is also possible, through the use of nests, to investigate the species and quantity of pollen preferred by wild bees as well as the presence of pesticides and other pollutants in the stored pollen (Staab et al. 2018). These kinds of studies show that we should not underestimate the aspects of “standardised” monitoring, especially in the case of organisms as diverse as pollinators, and that we should always include more than just one method in our monitoring to best capture the biodiversity of a site, habitat type or a certain kind community.

Mediterranean areas are a hot spot for the biodiversity of wild bees. We found that bee communities have distinct compositions depending on the monitoring method. In our experimental areas, hand netting caught a higher abundance and species richness of wild bees but pan traps had greater attractiveness when flowering coverage was low. Artificial nests only sampled a smaller fraction of species but were fundamental to capturing unique Megachilidae species, including parasitic ones. In conclusion, we recommend the use of multiple highly complementary methods when we want a comprehensive assessment of a wild bee community's biodiversity.

Acknowledgements

Not applicable.

Funding

Open access funding was provided by Università degli Studi di Firenze within the CRUI-CARE Agreement. O.C.M. acknowledges the support of Programma Operativo Nazionale Ricerca e Competitività. F.R.D. and M.M. acknowledge the support of the NBFC to the University of Florence, funded by the Italian Ministry of University and Research, PNRR, Missione 4 Componente 2, “Dalla ricerca all’impresa”, Investimento 1.4, Project CN00000033.

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Author contributions

O.C.M., M.M. and F.R.D. authors contributed to the monitoring work; D.V. supported the agronomic fieldwork; O.C.M. conceived the study; O.C.M. and M.M. identified the specimens, performed data analysis and wrote the paper; all authors read, commented and approved the manuscript.

Data availability

All data analysed for this paper are available on request.

Banaszak J (1980) Studies on methods of censusing the numbers of bees (Hymenoptera, Apoidea). Polish ecological studies, 6: 355-365.
Bascompte J, Jordano P, Olesen J M (2006) Asymmetric coevolutionary networks facilitate biodiversity maintenance. Science, 312: 431–433. https://doi.org/10.1126/science.1123412
Bell C, Tronstad L, Hotaling S (2023) Tailoring your bee sampling protocol: Comparing three methods reveals the best approaches to capturing bees. Agricultural and Forest Entomology, 25(3): 477–488. https://doi.org/10.1111/afe.12569
Breeze TD, Bailey AP, Balcombe KG, Brereton T, Comont R, Edwards M, Garratt MP, Harvey M, Hawes C, Isaac N, Jitlal M, Jones CM, Kunin WE, Lee P, Morris RKA, Musgrove A, O'Connor RS, Peyton J, Potts SG, Roberts SPM, Roy DB, Roy HE, Tang CQ, Vanbergen AJ, Carvell C (2021) Pollinator monitoring more than pays for itself. Journal of Applied Ecolology, 58: 44–57. https://doi.org/10.1111/1365-2664.13755
Cane J H, Minckley RL, Kervin LJ, Roulston TH, Williams NM (2006) Complex responses within a desert bee guild (Hymenoptera: Apiformes) to urban habitat fragmentation. Ecological applications: a publication of the Ecological Society of America, 16(2): 632–644. https://doi.org/10.1890/1051-0761(2006)016[0632:crwadb]2.0.co;2
Cane JH (2001). Habitat Fragmentation and Native Bees: a Premature Verdict? Conservation Ecology, 5(1). http://www.jstor.org/stable/26271798
Cane JH, Minckley RL, Kervin LJ (2000) Sampling Bees (Hymenoptera: Apiformes) for Pollinator Community Studies: Pitfalls of Pan-trapping. Journal of the Kansas Entomology Society, 73(4), 225–231.
Chamorro FJ, Faria CMA, Araújo FS, Freitas BM (2023) Elevated pan traps optimise the sampling of bees, including when the availability of floral resources is high. Insect Conservation and Diversity, 16(1): 16–32. https://doi.org/10.1111/icad.12621
Dafni A, Kevan PG, Husband BC (2005) Practical Pollination Biology. Enviroquest Ltd.
Danforth BN, Minckley RL, Neff JL (2019) The Solitary Bees. Biology, Evolution, Conservation. Princeton University Press. ISBN: 9780691168982.
de Assis JC, Toppa RH, Martines MR, Arena MVN, da Silva Souza M, Carneiro LT, Oliveira LF, da Silva-Zacarin ECM (2021) The influence of climate and seasonality on bee communities: a complementary method for bee sampling in forest patches of an anthropic matrix. International Journal of Tropical Insect Science, 41: 711–723. https://doi.org/10.1007/s42690-020-00261-1
Dormann CF, Schweiger O, Arens P, Augenstein I, Aviron S, Bailey D, Baudry J, Billeter R, Bugter R, Bukácek R, Burel F, Cerny M, Cock RD, Blust GD, DeFilippi R, Diekötter T, Dirksen J, Durka W, Edwards PJ, Frenzel M, Hamersky R, Hendrickx F, Herzog F, Klotz S, Koolstra B, Lausch A, Coeur DL, Liira J, Maelfait JP, Opdam P, Roubalova M, Schermann-Legionnet A, Schermann N, Schmidt T, Smulders MJM, Speelmans M, Simova P, Verboom J, Wingerden W, Zobel M (2008) Prediction uncertainty of environmental change effects on temperate European biodiversity. Ecology Letters, 11: 235-244. https://doi.org/10.1111/j.1461-0248.2007.01142.x
Dunn RR, Harris NC, Colwell RK, Koh LP, Sodhi NS (2009) The sixth mass coextinction: are most endangered species parasites and mutualists? Proceedings. Biological Sciences, 276(1670): 3037–3045. https://doi.org/10.1098/rspb.2009.0413
Gathmann A, Greiler HJ, Tscharntke T (1994) Trap-Nesting Bees and Wasps Colonizing Set-Aside Fields: Succession and Body Size, Management by Cutting and Sowing. Oecologia, 98(1): 8–14. http://www.jstor.org/stable/4220660
Geroff RK, Gibbs J, McCravy KW (2014) Assessing bee (Hymenoptera: Apoidea) diversity of an Illinois restored tallgrass prairie: methodology and conservation considerations. Journal of Insect Conservation, 18: 951–964. https://doi.org/10.1007/s10841-014-9703-z
Gezon ZJ, Wyman ES, Ascher JS, Inouye DW, Irwin RE (2015) The effect of repeated, lethal sampling on wild bee abundance and diversity. Methods in Ecology and Evolution, 6: 1044-1054. https://doi.org/10.1111/2041-210X.12375
Ghisbain G, Rosa P, Bogusch P, Flaminio S, Divelec RL, Dorchin A, Kasparek M, Kuhlmann M, Litman JR, Mignot M, Müller A, Praz CJ, Radchenko VG, Rasmont P, Risch S, Roberts SP, Smit J, Wood TJ, Michez D, Reverté S (2023) The new annotated checklist of the wild bees of Europe (Hymenoptera: Anthophila). Zootaxa, 5327(1):1-147. https://doi.org/10.11646/zootaxa.5327.1.1
Gibbs J, Joshi NK, Wilson JK, Rothwell NL, Powers K, Haas M, Gut L, Biddinger DJ, Isaacs R (2017) Does Passive Sampling Accurately Reflect the Bee (Apoidea: Anthophila) Communities Pollinating Apple and Sour Cherry Orchards? Environmental Entomology, 46(3): 579–588. https://doi.org/10.1093/ee/nvx069
Goulson D, Lye GC, Darvill B (2008) Decline and conservation of bumble bees. Annual review of entomology, 53: 191–208. https://doi.org/10.1146/annurev.ento.53.103106.093454
Hall M (2018) Blue and yellow vane traps differ in their sampling effectiveness for wild bees in both open and wooded habitats. Agricultural and Forest Entomology, 20: 487-495. https://doi.org/10.1111/afe.12281
Hsieh TC, Ma KH and Chao A (2016) iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution, 7: 1451-1456. https://doi.org/10.1111/2041-210X.12613
https://www.beewatching.it/
Hutchinson LA, Oliver TH, Breeze TD, O'Connor RS, Potts SG, Roberts SPM, Garratt MPD (2022) Inventorying and monitoring crop pollinating bees: Evaluating the effectiveness of common sampling methods. Insect Conservation and Diversity, 15(3), 299–311. https://doi.org/10.1111/icad.12557
IPBES (2016). The assessment report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services on pollinators, pollination and food production. Secretariat of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, Bonn, Germany. 552 pages. https://doi.org/10.5281/zenodo.3402856
Kevan PG, Greco CF, Belaoussoff S (1997) Log- Normality of Biodiversity and Abundance in Diagnosis and Measuring of Ecosystemic Health: Pesticide Stress on Pollinators on Blueberry Heaths. Journal of Applied Ecology, 34(5): 1122–1136. https://doi.org/10.2307/2405226
Klaus F, Ayasse M, Classen A, Dauber J, Diekötter T, Everaars J, Fornoff F, Greil H, Hendriksma HP, Jütte T, Klein AM, Krahner A, Leonhardt SD, Lüken DJ, Paxton RJ, Schmid-Egger C, Steffan-Dewenter I, Thiele J, Tscharntke T, Pistorius J (2024) Improving wild bee monitoring, sampling methods, and conservation. Basic and Applied Ecology, 75: 2-11. https://doi.org/10.1016/j.baae.2024.01.003
Klein AM, Vaissière BE, Cane JH, Steffan-Dewenter I, Cunningham SA, Kremen C, Tscharntke T (2007) Importance of pollinators in changing landscapes for world crops. Proceedings. Biological Sciences, 274(1608): 303–313. https://doi.org/10.1098/rspb.2006.3721
Kowalska J, Antkowiak M, Sienkiewicz P (2022) Flower Strips and Their Ecological Multifunctionality in Agricultural Fields. Agriculture, 12(1470). https://doi.org/10.3390/agriculture12091470
Krahner A, Schmidt J, Maxiner M, Porten M, Schmitt T (2021) Evaluation of four different methods for assessing bee diversity as ecological indicators of agro-ecosystems. Ecological Indicators, 125:107573. https://doi.org/10.1016/j.ecolind.2021.107573
Kuhlman MP, Burrows S, Mummey DL, Ramsey PW, Hahn PG (2021) Relative bee abundance varies by collection method and flowering richness: Implications for understanding patterns in bee community data. Ecological Solutions and Evidence, 2: e12071. https://doi.org/10.1002/2688-8319.12071
LeBuhn G, Droege S, Connor EF, Gemmill-Herren B, Potts SG, Minckley RL, Griswold T, Jean R, Kula E, Roubik DW, Cane J, Wright KW, Frankie G, Parker F (2013) Detecting insect pollinator declines on regional and global scales. Conservation Biology, 27(1): 113–120. https://doi.org/10.1111/j.1523-1739.2012.01962.x
MacIvor JS (2017) Cavity-nest boxes for solitary bees: a century of design and research. Apidologie 48: 311–327. https://doi.org/10.1007/s13592-016-0477-z
Michener CD (2000) The Bees of the World. Baltimore and London (The John Hopkins University Press). ISBN 0-8018-6133-0
Michener CD (2007) The bees of the world. The Johns Hopkins University Press. https://doi.org/10.56021/9780801885730
Moldoveanu OC, Maggioni M, Dani FR (2024) Environmental ameliorations and politics in support of pollinators. Experiences from Europe: A review. Journal of Environmental Management, 362:121219. https://doi.org/10.1016/j.jenvman.2024.121219
Neumann P, Carreck LN (2010) Honeybee colony losses. Journal of Apicultural Research, 49(1):1-6. https://doi.org/10.3896/IBRA.1.49.1.01
Nielsen A, Steffan-Dewenter I, Westphal C, Messinger O, Potts SG, Roberts SPM, Settele J, Szentgyörgyi H, Vaissière BE, Vaitis M, Woyciechowski M, Bazos I, Biesmeijer JC, Bommarco R, Kunin WE, Tscheulin T, Lamborn E, Petanidou T (2011) Assessing bee species richness in two Mediterranean communities: importance of habitat type and sampling techniques. Ecology Research, 26: 969-983. https://doi.org/10.1007/s11284-011-0852-1
Nieto A, Roberts SPM, Kemp J, Rasmont P, Kuhlmann M, García Criado M, Biesmeijer JC, Bogusch P, Dathe HH, De la Rúa P, De Meulemeester T, Dehon M, Dewulf A, Ortiz-Sánchez FJ, Lhomme P, Pauly A, Potts SG, Praz C, Quaranta M, Radchenko VG, Scheuchl E, Smit J, Straka J, Terzo M, Tomozii B, Window J, Michez D (2014) European Red List of bees. Luxembourg: Publication Office of the European Union. https://data.europa.eu/doi/10.2779/51181
O'Connor RS, Kunin WE, Garratt MPD, Potts SG, Roy HE, Andrews C, Jones CM, Peyton JM, Savage J, Harvey MC, Morris RKA, Roberts SPM, Wright I, Vanbergen AJ, Carvell C (2019) Monitoring insect pollinators and flower visitation: The effectiveness and feasibility of different survey methods. Methods in Ecology and Evolution, 10: 2129–2140. https://doi.org/10.1111/2041-210X.1329
Ollerton J (2021) Pollinators & Pollination. Nature and Society. Pelagic PublishingISBN: 9781784272289.
Ollerton J, Winfree R and Tarrant S (2011) How many flowering plants are pollinated by animals? Oikos, 120: 321-326. https://doi.org/10.1111/j.1600-0706.2010.18644.x
Orr MC, Hughes AC, Chesters D, Pickering, J, Zhu CD, Ascher JS (2021) Global patterns and drivers of bee distribution. Current Biology, 31(3): 451–458. https://doi.org/10.1016/j.cub.2020.10.053
Ortiz-Sanchez FJ, Aguirre-Segura A (1992) Comparison of the effectiveness of different heights in order to capture bees with the employ of Moericke traps (Hymenoptera, Apoidea). Graellsia, 48, 35-43.
Popic TJ, Davila YC, Wardle GM (2013) Evaluation of Common Methods for Sampling Invertebrate Pollinator Assemblages: Net Sampling Out-Perform Pan Traps. PLoS ONE 8(6): e66665. https://doi.org/10.1371/journal.pone.0066665
Potts S, Imperatriz-Fonseca V, Ngo H, Aizen MA, Biesmeijer JC, Breeze TD, Dicks LV, Garibaldi LA, Hill R, Settele J, Vanbergen AJ (2016) Safeguarding pollinators and their values to human well-being. Nature 540: 220–229. https://doi.org/10.1038/nature20588
Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, Kunin WE (2010) Global pollinator declines: trends, impacts and drivers. Trends in Ecology and Evolution, 25(6):345-353. https://doi:10.1016/j.tree.2010.01.007
Prendergast KS, Menz MH, Dixon KW, Bateman PW (2020) The relative performance of sampling methods for native bees: an empirical test and review of the literature. Ecosphere, 11. https://doi.org/10.1002/ecs2.3076
Quaranta M, Cornalba M, Biella P, Comba M, Battistoni A, Rondinini C, Teofili C (2018) Red list IUCN of the Italian threatened bees [Lista Rossa IUCN delle api italiane minacciate]. Comitato Italiano IUCN e Ministero dell’Ambiente e della Tutela del Territorio e del Mare, Roma.
Reid W, Mooney H, Cropper A, Capistrano D, Carpenter S, Chopra K (2005) Millennium Ecosystem Assessment. Ecosystems and human well-being: synthesis.
Rhoades P, Griswold T, Waits L, Bosque-Pérez NA, Kennedy CM, Eigenbrode SD (2017) Sampling technique affects detection of habitat factors influencing wild bee communities. J Insect Conservation, 21: 703–714. https://doi.org/10.1007/s10841-017-0013-0
Sevilleja CG, van Swaay CAM, Bourn N, Collins S, Settele J, Warren MS, Wynhoff I, Roy DB (2019) Butterfly Transect Counts: Manual to monitor butterflies. Report VS2019.016, Butterfly Conservation Europe & De Vlinderstichting/Dutch Butterfly Conservation, Wageningen. https://butterfly-monitoring.net/
Staab M, Pufal G, Tscharntke T, Klein A-M (2018) Trap nests for bees and wasps to analyse trophic interactions in changing environments—A systematic overview and user guide. Methods in Ecology and Evolution, 9: 2226–2239. https://doi.org/10.1111/2041-210X.13070
Stephen WP and Rao S (2005) Unscented Color Traps for Non-Apis Bees (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society, 78(4): 373–80. JSTOR, http://www.jstor.org/stable/25086288
Stout JC, Morales CL (2009) Ecological impacts of invasive alien species on bees. Apidologie 40: 388–409. https://doi.org/10.1051/apido/2009023
Thompson AH, Frenzel M, Schweiger O, Musche M, Groth T, Roberts SP, Kuhlmann M, Knight TM (2021) Pollinator sampling methods influence community patterns assessments by capturing species with different traits and at different abundances. Ecological Indicators, 132. https://doi.org/10.1016/j.ecolind.2021.108284
Townes HK (1962) Design For A Malaise Trap. Proc. Ent. Soc. Wash., 64(4).
Westphal C, Bommarco R, Carré G, Lamborn E, Morison N, Petanidou T, Potts SG, Roberts SPM, Szentgyörgyi H, Tscheulin T, Vaissière BE, Woyciechowski M, Biesmeijer JC, Kunin WE, Settele J, Steffan-Dewenter I (2008) Measuring bee diversity in different European habitats and biogeographical regions. Ecological Monographs, 78: 653-671. https://doi.org/10.1890/07-1292.1
Westphal C, Steffan‐Dewenter I, Tscharntke T (2006) Foraging trip duration of bumblebees in relation to landscape-wide resource availability. Ecological Entomology, 31(4): 389-394. https://doi.org/10.1111/j.1365-2311.2006.00801.x
Wilson JS, Griswold T, Messinger OJ (2008) Sampling Bee Communities (Hymenoptera: Apiformes) in a Desert Landscape: Are Pan Traps Sufficient? Journal of the Kansas Entomological Society 81(3): 288-300. https://doi.org/10.2317/JKES-802.06.1
www.STEP-project.net

No competing interests reported.

SupplementaryMaterialsMoldoveanuetal.2024.docx

Download PDF

Reviewers agreed at journal
29 Aug, 2024
Reviews received at journal
15 Aug, 2024
Reviewers agreed at journal
07 Aug, 2024
Reviewers invited by journal
06 Aug, 2024
Editor assigned by journal
02 Aug, 2024
Submission checks completed at journal
02 Aug, 2024
First submitted to journal
02 Aug, 2024

You are reading this latest preprint version

Revealing the biodiversity of wild bees (Hymenoptera: Apoidea: Anthophila) in flower strips in Mediterranean floodplains. Which monitoring method fits best?

Status:

Version 1

Abstract

Context

Objectives

Methods

Results

Conclusions

Figures

1. Introduction

2. Material and Methods

2.1 Study sites and area description

2.2 Bee sampling methods and data collection

2.3 Species identification

2.4 Data analysis

3. Results

4. Discussion

5. Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1