Blühstreifen als Naturschutzmaßnahme zur Förderung der Avifauna und Tagfalterfauna in der Agrarlandschaft

The intensification of agriculture is one of the main causes of biodiversity loss, leading to considerable declines in flora and fauna. The plight of biodiversity in the agricultural landscape has been known for a long time, legal conditions for biodiversity conservation are available and various nature conservation measures have already been implemented. Nevertheless, it has so far not been possible to stop biodi-versity loss in the agricultural landscape. The contribution of agriculture to biodiversity conservation is essential. Agricultural regions, however, are in conflict with the interests of various users. This situation is aggravated by the long-lasting trend towards increased cultivation of renewable raw materials. Alt-hough amendments to the EEG since 2012 (EEG 2012, 2014 and 2017) have counteracted the "biogas boom", there is still a very high land requirement for the cultivation of energy crops. For example, 21% of Germany's farmland is used for the agricultural cultivation of energy crops, with maize as the domi-nant crop. These facts undeniably demonstrate that a more efficient design of conservation measures is absolute-ly necessary. Although an adequate evaluation of existing conservation measures is an indispensable prerequisite, it is not sufficiently implemented. Flower strips are considered as an appropriate measure for the promotion of biodiversity. However, they can be created in many variations, and there is a con-siderable lack of knowledge about the effects of the different types of flower strips. It is precisely the relevance of individual design variations that is decisive for deriving transferable recommendations. For this reason, a differentiated ecological efficiency control is required for flower strips. Even winter, with its extreme weather conditions, lack of coverage and food scarcity, is a particularly critical time for many animal species. However, year-round studies of flower strips are rare. In light of these issues, the overall aim of this doctoral thesis is to investigate the effects of different types of flower strips during summer and winter and on different species groups (vegetation, breeding birds, winter birds and butterflies). Field margins were examined as a reference. In this context, the following research questions arise: 1. How does the vegetation (food supplied by flowers, spatial structure) alter due to certain char-acteristics of flower strips (seed mixture and age) and due to seasonal changes? 2. How do certain characteristics of flower strips (seed mixture, location, width and age) affect the occurrence (number of species and individuals) of breeding birds, winter birds and butter-flies? For a solid knowledge base, two detailed aspects were decisive which were resolved exemplarily for one species group. On the one hand, certain species are often underrepresented due to traditional sur-vey methods with short examinations at study sites. So far, there were no observations of several hours in flower strips. On the other hand, previous studies indicate that various factors of different spatial scales are decisive for the occurrence of species. It can be assumed that the different types of flower strips are not the only factor that determines species diversity of flower strips. These deficits were ad-dressed in the following research questions: 3. What are the advantages and limits of using of time-triggered cameras (continuous shooting function with a short interval of 20 seconds) for bird surveys? 4. Which factors of different spatial scales (local habitat level and landscape context) determine the occurrence of butterflies in flower strips and field margins? What effect do the individual factors have? Based on the results, three further research questions were raised in order to evaluate the effect of flower strips on vegetation, birds and butterflies compared to field margins: 5. How do species diversity and vegetation of the different types of flower strips differ from those of the field margins? 6. Which cross-species and species-specific recommendations for the creation of flower strips can be derived? 7. What nature conservation value do the different types of flower strips have in comparison to field margins? A key component of the doctoral thesis are the multi-year field studies in the district of Rotenburg (Wümme), a region in which the problem of energy crop cultivation is intensifying. The administrative district Rotenburg (Wümme) is a district in the federal state of Lower Saxony and has the second high-est land requirement for energy crop cultivation for biogas production. It is also one of the main re-gions for maize cultivation in Germany. During two winters (2012/13 and 2013/14), two summers (2013 and 2014), one autumn (2013) and one spring (2014), vegetation characteristics, breeding birds, winter birds and butterflies were recorded on flower strips (n=20) and field margins (n=20). The flower strip variation included different seed mixtures (Rotenburger mixture 2012 and Rotenburger mixture 2013), different location (open landscape and along rows of trees), different width (6 m wide flower strips and 30-80 m wide flower strips) and differ-ent age (flower strips with a life span of one year (April/May to the end of February) and with a life span of 1.5 years (April/May to the end of September of the following year)). As references, field margins along maize fields and along flower strips in the first and second life spans were recorded. Transects or sample quadrats were used for the vegetation surveys (species lists, amount of flowering resources and vegetation structure). Line transects were used for faunistic surveys (number of species as a proxy for species richness and number of individuals as a proxy for abundance). The analysis of the effects of the flower strips on the avifauna and butterflies was carried out by statistical pairwise com-parisons to the species and individual numbers of the different types of flower strips and references, the field margins. In both winters, summer and autumn of 2013, birds were additionally recorded by time-triggered cameras using a continuous shooting function with a short interval of 20 seconds. In order to assess the results of camera traps and improve this method for use in bird surveys, the results of the camera traps were calibrated with those of the traditionally used line transects, which were con-ducted on the same study areas and at the same time period. The comparison of methods was carried out using statistical pairwise comparisons and linear mixed-effect models. For factor analysis, environmental variables likely to be decisive for the occurrence of butterflies were identified. They were identified on the basis of a literature research and a principal component analysis: year of the investigation, width of the study site, age/ life span of flower strips, open-ground propor-tion, dominant height of vegetation, maximum height of vegetation, number of flowering herbaceous plant species, number of flowering herbaceous plant species of the seed mixture, adjacent areas, habi-tat diversity and connectivity of the surrounding landscape. These variables were recorded during the field studies or determined from topographic maps and from orthophotos. Linear mixed models were used to test the effect of these variables for both flower strips and field margins. The dissertation clearly proved that flower strips make a decisive contribution to the promotion of but-terflies and avifauna in the agricultural landscape. All 29 bird species detected by line transects (in flow-er strips and field margins; during summer and winter) were recorded on the flower strips, with the exception of one species. In contrast, only half of all recorded bird species could be observed on the field margins in summer and in winter only about a third. In addition, winter birds were observed much more frequently on the flower strips than on the field margins. The statistical pairwise comparisons between the different types of flower strips and field margins showed significant differences for many combinations, both in summer and winter and for all species as well as for farmland birds. The combi-nation of methods, camera traps and line transects, confirmed this result: Both camera traps and line transects were able to record significantly more species in the flower strips than in the field margins (on average over the level of season; camera trap p=0.0471, line transects p=0.0005). All 20 in total (flower strips and field margins) recorded butterfly species could be observed on the flower strips, five of them solely there. No butterfly species, however, was observed solely on the field margins. Butterflies were also recorded more frequently on flower strips than on field margins. Statisti-cal pairwise comparisons between the different types of flower strips and the field margins also re-vealed significant differences for the butterflies (number of species and individuals). The different widths of the flower strips mainly affected winter birds. The wider flower strips (width 30-80 m) had higher species diversity and were used much more frequently than the 6 m wide flower strips. For farmland birds and predominantly herbivorous bird species, the differences in the frequency index were close to significant (p=0.056). Red List species were also observed more frequently on the wider flower strips. In addition, bird observations on wider flower strips decreased less than on the other types of flower strips over the course of the winter, in some cases with significant differences. Butter-flies were also detected much more frequently on the 6 m wide flower strips than on the wider flower strips. However, with regard to the number of butterfly species, the width had no effect. The age of the flower strips mainly affected the butterfly fauna and the vegetation. The butterfly occur-rences were considerably higher on the flower strips in the first growing season than on those in the second growing season. The differences were significant for the frequency index (p=0.04) and close to significant for the number of species (p=0.074). The flower strips in the first growing season had a high-er number of flowering plants than those in the second growing season. The good growth of the seed mixture in the first year proved to be important for the appearance of flowering plants in the following year. In addition, the analyses of vegetation development showed that the flower strips had different structural and floristic characteristics depending on their life span and complement each other during seasonal change. Since the emergence of the vegetation takes a certain amount of time, the newly created flower strips can only make a contribution to the supply of flowers and structures from around July onwards. At this time, the flower strips in the second growing season can provide a good supply of flowers and structure. However, since many of the flower strips herbaceous plants are already withered at the end of summer in the second growing season, flower strips in their first year with good flower and structure variety can compensate for the lack. For the number of flowering herbaceous plant species, a highly significant effect on butterfly occur-rences (number of species and individuals) was demonstrated both in the linear mixed model for flower strips and in the model for field margins. Furthermore, an additional significant positive effect of the seed mixture’s flowering species on butterflies (number of species and individuals) for the flower strips could be demonstrated, as could a significant negative effect on the habitat diversity in the surrounding landscape. In order to improve the positive effects of flower strips, various design features must be considered on the basis of these results: Flower strips with a life span of 1.5 years can be recommended across species groups, even though this was not directly apparent in the occurrence of birds and butterflies in this study. However, flower strips can only provide a continuous supply of flowers and structures in the landscape (and thus food, cover or breeding sites) from a life span of 1.5 years onwards. As the seasonally differentiated analysis of vegetation showed, the different age stages vary in terms of plant and structural diversity and comple-mented each other most optimally over the course of the season. Therefore, it is optimal that flower strips with different life spans are arranged as a mosaic within a section of landscape, so that different age stages are present in spatial-functional proximity. In addition, only flower strips with a minimum life span of 1.5 years allow many insects to pass through a complete reproduction cycle, as the upheav-al in February can completely destroy the offspring of insects. Since no work is carried out during the life span of the flower strips, the disturbance on perennial flower strips is less. With regard to the age of flower strips, a high supply of flowering resources must be ensured for sub-sequent years. For the improvement of the habitat quality for butterflies, a high supply of flowering resources was the most fundamental factor. Nevertheless, in the second growing season, the supply of flowers had already been greatly reduced on some flower strips, which also resulted in considerably fewer butterflies. As explained in the previous paragraph, however, the time between sowing and ploughing of the flower strips should not be too short so as not to disturb the life cycle of butterflies. Thus, it must not be concluded that a short life span of flower strips is generally the best variant for butterflies or insects. Rather, the supply of flowering resources in flower strips must be ensured over several years. Since a good growth of the seed mixture can slow down the rapid progress of succession, a good basis is essential for the creation of flower strips. In order to be able to give recommendations on the specific life span of flower strips in combination with any necessary maintenance operations, there is a need for further research as the maximum life span of flower strips in this study was con-strained to 1.5 years. With regard to width, species-specific requirements should be taken into account when improving flower strips. Wider flower strips are particularly suitable for promoting birds, as the field studies on winter birds particularly clearly show. In summer, no decisive influence of width could be demonstrated on the basis of the bird data of this study. However, other studies have shown that the predation risk increases in linear, narrow habitat structures, and thus a width of 10 m to 40 m is recommended for flower strips or field margins. In the context of the dissertation, it could be proven that a width of 6 m is sufficient for the butterflies. In addition, a good connectivity of the landscape, which can be fostered by a high number of flower strips in a landscape section, has a positive effect on butterflies. With the same land use and the same use of funds, a much better network can be created with 6 m wide flower strips rather than with much wider flower strips e.g. on whole fields. Ultimately, a habitat network of 6 m wide flower strips is more efficient. The comprehensive factor analysis for butterflies showed that the effect of flower strips is highest in structurally simple landscapes. So, flower strips should be applied specifically in such landscapes. The following can be summarised for the optimisation of flowering strips. 1.5 year-old flower strips have a higher value for fauna than annual flower strips. 6 m wide flower strips are more efficient in improving habitat connectivity. Much wider flower strips are recommended in order to specifically promote the avifauna. A high supply of flowering resources can additionally increase the positive effect of flower strips for all species groups. In order to be able to cope with the landscape context and the species-specific requirements, target areas must be developed and these priority objectives (target species) must be defined. In addition, for birds and butterflies, it was clearly shown that flower strips have a higher value for na-ture conservation than field margins. The value for nature conservation of field margins, however, de-pends strongly on their respective characteristics. The field margins investigated were only poorly struc-tured. In order to counteract the loss of species in the agricultural landscape, the improvement of field margins is therefore also an option. In light of the fact that the supply of flowering resources is the fundamental factor for the occurrence of butterflies, field margins can be upgraded in this direction. The main advantage of field margins over flower strips is their continuous, permanent presence in the landscape. But even despite the rotary principle of flower strips and their related fauna, their value for nature conservation can be classified above that of poorly structured field margins on nutrient-rich sites, but it is lower than that of structurally-rich and species-rich field margins on nutrient-poor sites. In order to record the bird communities more reliably, the combination of line transects (repetitive over many days with short examinations of the study sites) and camera traps (a few days over many hours) proved to be well suited. When using both methods, the number of recorded species increased signifi-cantly in flower strips as well as in field margins. However, the benefits of multi-methods depend on both the habitat type and the applied method. In flower strips, significantly more species were recorded by line transect than by camera trapping and (p