Conservation of wild bees in agricultural landscapes through intercropping and habitat management

Bees, important providers of vital environmental functions such as pollination and soil bioturbation, are currently declining both on a local and global scale. Two main causes of this decline are habitat loss and fragmentation and floral scarcity in modern agricultural landscapes. Effective measures are needed to counteract these factors so that bees are preserved and their functions are not impaired. The aim of my dissertation is to investigate the potential of on-field and off-field measures for bee conservation, i.e. measures that target productive arable land and non-productive areas. Chapter 1 targets the on field-measure intercropping with focus on legumes and non-legume mixtures. Bees can benefit from this cultivation system through the provision of pollen and nectar. Characteristics of intercrops between legumes and non-legumes are investigated that may influence the value of this cropping system for bees, but also for agricultural production (i.e. pollination services, legume yields, crop genotype traits), as both aspects need to go hand in hand to create desirable "win-win" situations. Bees were surveyed in 2019 during time-standardized observation walks in a field experiment consisting of small-scale plots (21 sqm) with sole crop stands of six genotypes of the grain legume Vicia faba (L.) and intercrops of V. faba and wheat (replacement design). The number of flower visits by bees in intercrops and in sole crops was compared, assuming that intercrops are less attractive due to a lower density of pollen and nectar-providing flowers. Further, the foraging behavior of bees was observed, i.e. legal (= frontal) flower visits and nectar robbing. Differences between the two cropping systems and V. faba genotypes in terms of the yield per V. faba plant and the number of V. faba inflorescences, leaf area index (LAI) and mean V. faba plant height per plot were analyzed. The latter two traits are important for determining the performance of intercrops of V. faba and wheat and possibly for determining attractiveness for pollinators by moderating flower perceptibility. V. faba-wheat intercrops were characterized by a significantly lower number of inflorescences, a lower LAI and a lower V. faba plant height than V. faba sole crops. Further, V. faba yields were significantly higher in intercrops. LAI, plant height, number of inflorescences and yields also differed significantly between V. faba genotypes. No significant difference in the number of bee flower visits per plot was found between intercrops and sole crops. The total number of flower visits and cases of nectar robbing were positively associated with higher V. faba yields. The results show that intercrops with legumes, even though less rich in inflorescences, can be as attractive as legume sole crops. The lower inflorescences density in intercrops is possibly compensated by higher flower perceptibility or higher investments in floral rewards. However, it needs to be verified whether these findings also apply at larger spatial scales (e.g. field scale, landscape scale). With increasing total number of flower visits, including cases of nectar robbing, higher yields per V. faba plant was observed, suggesting positive effects of bee pollination. Yet it is difficult to distinguish to what extent this increase is due to pollination or to other differences between the genotypes / cropping systems. V. faba genotypes differed in LAI and average plant height per plot, but these differences were not linked to the number of flower visits or foraging behavior of bees. Thus, other traits (e.g. floral architecture and display, scent, and quality and quantity of floral rewards) are probably more important for bees. Higher yields per V. faba plant in intercrops compared to sole crops indicate agronomic advantages of this cropping system. Intercrops were mainly visited by domesticated honeybees and widespread ubiquitous bumblebees. Other wild bee species, including threatened species, were rare. As an on-field measure, intercropping with legumes therefore appears to be most suitable for supporting common bee species and thus promoting the provision of important ecosystem services. To foster rare, threatened and specialized species, off-field measures aiming at preserving semi-natural habitats are required. Limestone quarries are semi-natural habitats that could be targeted by off-field measures to promote demanding bee species. Chapter 2 addresses the question which habitat and landscape characteristics determine habitat quality of quarries for bees and which management measures are needed to conserve and restore them. In 2020, wild bees were surveyed in 19 limestone quarries near Göttingen using time-standardized transects. 114 wild bee species, including 35 Red List species, were found. Richness and abundance increased with increasing habitat age, but only at sites where woody plant encroachment was limited. Further, positive interactions between quarry age and landscape diversity and/or habitat connectivity to neighbouring dry grasslands, another valuable and threatened habitat type, were found. In particular, high connectivity between both habitats ensured stable richness of endangered species in old quarries. The results suggest that large, flower-rich and old quarries should be prioritized by conservation and restoration measures. To ensure positive effects of quarry age, woody plant encroachment must be controlled (e.g. by extensive mowing, grazing, scrub clearance), as otherwise important open patches with favorable microclimate, rich in food and nesting resources, could be lost. Furthermore, high connectivity between old limestone quarries and calcareous grasslands was found to be beneficial. Measures aimed at preserving and restoring both habitat types at a landscape scale could therefore be most successful. In Chapter 3, bee data from 2020 was compared with historical data from 1999 collected at the same quarries as in Chapter 2. This was done to determine how bee community composition and beta diversity (incidence-based Sørensen dissimilarity, abundance-based Bray-Curtis dissimilarity and their turnover and nestedness components) changed over time and to assess whether quarry conservation and restoration could be a suitable tool to ensure long-term stability of bee communities in these habitats. Changes in community composition and beta diversity were linked to local habitat characteristics (age, area, woody vegetation cover) and changes in functional characteristics that make bees particularly sensitive or adaptive to changing environmental conditions (nesting behavior, lecty, thermophily, body size). The composition of the bee communities in the quarries changed between the study years, resulting in higher beta diversity in 2020. This was indicated by increasing incidence-based Sørensen dissimilarity and its turnover component. However, no significant changes in the nestedness component of Sørensen dissimilarity, or of the abundance-based Bray-Curtis dissimilarity and its turnover and nestedness components could be found between study years. In terms of functional characteristics, the richness of thermophilic bee species was higher in 2020 than in 1999. Other functional characteristics did not differ significantly between the years, but were consistently influenced by habitat characteristics. High woody vegetation cover in quarries resulted in a decline of the proportions of bee species that nest obligatory underground in self-dug nests and in a tendency towards bee communities with a higher average body size. Proportions of oligolectic bee species (i.e. bees specialized on a narrow range of pollen sources) and the richness of thermophilic bee species were higher large quarries. These results show that bee communities changed between 1999 and 2020, becoming more dissimilar due to species turnover between quarries. This was caused by the spread of thermophilic bee species that migrated and successfully established in the quarries over the last two decades due to climate change. With regard to the most common bee species, little difference was found between 1999 and 2020, as the bee communities in both years were mainly dominated by widespread and non-threatened bumblebees. Further, the findings highlight the value of large quarries with limited woody vegetation for the two vulnerable functional groups of ground-nesting and oligolectic wild bee species, but also for thermophilic bees. However, large-bodied bees appear to be more attracted to wood-dominated quarries, which could be due associated advantages for bumblebees (e.g. favorable microclimate, nesting opportunities). I conclude that intercrops between legumes and non-legumes as well as the preservation and restoration of limestone quarries are suitable measures for bee conservation. Intercrops with legumes primarily support unthreatened and ubiquitous bee species such as bumblebees and domesticated honeybees and are therefore a suitable on-field measure for supporting ecosystem service providers. For threatened, specialized and rare wild bee species, the conservation and restoration of limestone quarries is a suitable off-field measure when in particular large, flower-rich and old quarries with limited amounts of woody vegetation and high connectivity to nearby calcareous grasslands are considered.

2025-03-19