Gradient Of Landscape Complexity Research Articles

Intensive ground cover management and landscape simplification affect community-wide floral visitor-plant interactions in olive groves

Plant-pollinator interaction networks can provide insights about the impacts of agricultural practices on biodiversity. However, studies on the agricultural impact on these networks most often focus on specific pollinator guilds (i.e., wild bees), hindering our understanding of the effects at the entire insect visitor community level. Here, we evaluated the impact of intensive agriculture and the moderation effects of landscape context on the topology of community-wide plant-visitor interaction networks (381 insect species). We also compared the responses of wild bee-based networks and community-wide plant-floral visitor networks to agricultural intensification to examine whether the former can be a good proxy of the latter. To this aim, we sampled plant-pollinator interaction networks in 24 paired olive farms with contrasting herb cover management (low-intensity management, where ground herb cover is not removed during the flowering season, and intensive management, with herbs persistently removed by herbicides and/or recurrent tillage) across a gradient of landscape complexity in Andalusia (Spain). Results show that the topology of whole floral visitor interaction networks was affected by the herb cover management and the landscape simplification. Furthermore, we found that the effects of the herb cover management were often moderated through the interaction with the landscape context at both floral stand and landscape scale. However, the landscape scale at which these interaction effects occurred varied with the network metric. Non-bee floral visitors were fundamentally redundant for the network topology. We concluded that the implementation of low-intensity management of ground covers in olive groves, together with the enhancement of the flower stands at the local level (e.g., floral strips) and the maintenance and/or restoration of semi-natural habitats at the large landscape scale, would be beneficial to the preservation of healthy and robust floral visitor communities. Our study further supports that wild-bee based interaction networks may adequately reflect the response of the entire networks of plant-floral visitor, at least in this study system.

Field edge flower plantings have variable effects on wild bee abundance, richness, nesting success, and crop pollination, independent of the surrounding landscape

To mitigate the effects of habitat loss on bee conservation and crop pollination, farmers are encouraged to plant flowers along the edges of crop fields. However, it is unclear where and to what degree these flower plantings are effective. Recent studies suggest that flower plantings may be less effective if they are established in landscapes with either very low or very high amounts of non-crop habitat. Moreover, while field edge flower plantings often increase the abundance and richness of bees in field edges, it is unclear whether this improves crop pollination or bee nesting success, which could increase bee populations over time. In this study we examined how field edge flower plantings and the landscape context in which they are established affect wild bee conservation and crop pollination. Across two years we established flower plantings of 0.1–0.5 ha along the edges of 17 of 34 commercial cucumber fields arranged across a gradient of landscape complexity. We then measured wild bee abundance, richness, and crop pollination in and around cucumber fields with and without plantings and assessed the nesting success of stem-nesting bees. We found that both field edge flowers and landscape-scale non-crop habitat had independent, positive effects on bee abundance, richness, nesting, or crop pollination in one of the two years of the study, and that the effects of field edge flowers on bees and crop pollination were not influenced by the surrounding landscape. We also found that managing field edges in a way that maintains naturally-occurring flowers may matter as much or more than flower plantings for short-term increases in bee abundance and richness. However, certain species of planted flowers might be associated with improved bee nesting success, an outcome important for both bee conservation and sustained crop pollination.

Local and landscape context affects bee communities in mixed fruit orchards in Southern Thailand

Abstract Bees play an important role by contributing to the pollination of a diverse range of crops and wild plants. However, bees are threatened by deforestation and habitat degradation caused by urbanization and agricultural intensification, which reduce floral and nesting resources. In this study, the effects of local (orchard size, flowering plant abundance and richness and floral abundance) and landscape factors (surrounding agricultural, forest and urban cover) on bee richness and abundance in mixed fruit orchards were determined from 28 mature orchards interspersed along a gradient of landscape complexity. Bees were observed for 15 min in 2 × 2 m plots. We categorized all bees into one of the three groups: (a) honey bees, (b) solitary bees and (c) stingless bees. Several local and landscape factors were found to influence bee richness and abundance. Honey bee abundance was positively affected by orchard size, whereas honey bee richness was not. Stingless bee richness and abundance were positively correlated with floral abundance. Forest cover within a 3‐km radius positively affected solitary bee richness. Solitary bee abundance was positively affected by forest cover within a 3‐km radius and negatively impacted by urban cover within a 1.5‐km radius. In our study area, small forest patches are found in abundance; thus, the high proportion of forest edge, relative to the forest interior, greatly supports solitary bee abundance and richness in this landscape. Moreover, most solitary bees are soil‐nesting species that may be excluded from environments that are converted from natural to anthropogenically developed areas. Recommendations to protect pollinators included conserving existing forest patches, especially forest margins; restoring forest patches within farmland landscapes; maintaining a diversity of herbs and grasses within mixed fruit orchards; establishing live fences composed of diverse plant species and retaining dead trees within farmland. For urban areas, augmenting floral resources in green spaces is suggested.

A complex landscape favours the abundance and species richness of syrphids (Diptera: Syrphidae) in olive groves

Abstract Sustainability of agroecosystems depends on the ecosystem services, provided by beneficial organisms, such as the syrphids (Diptera: Syrphidae) acting as other‐insect predators and pollinators in different crops. In the case of olive groves, syrphids are predators of important pests and are endangered by the continuous intensification of agriculture, including landscape simplification. In this work, the effect of landscape structure on the abundance and species richness of syrphids in the olive agroecosystem of central Spain is assessed. Adult syrphids were sampled in 2015 and 2016 in the centre of circular areas of 15 olive grove landscapes (from 250 to 1500 m radius) representing a gradient of landscape complexity, in the province of Madrid, central Spain. The syrphid assemblage was made up of 57 species, although Eupeodes corollae, Eristalis tenax, Eristalis similis and Chrysotoxum intermedium made up 76% of the captures. Syrphid abundance and species richness were favoured in complex landscapes, with low olive grove area (CAO). Other landscape indices showed also this trend although their effect was not as strong and consistent as that of CAO. The area of scrublands (CAS) favoured syrphids in spring, but was unfavourable in autumn. The effect of Shannon's Evenness Index (SHEI) and mean shape index (MSI) was less consistent. The effect of CAO increased with increasing buffers radii, while the effect of CAS, SHEI and MSI was detected in general at smaller radii, indicating these indices affect different processes. Redundancy analysis (RDA) detected some differences between species regarding their abundance in spring versus autumn, and their relationship with landscape indices. This work provides new evidence that maintenance of complex landscapes is essential for maintaining functional biodiversity in the olive agroecosystem.

Small floral patches are resistant reservoirs of wild floral visitor insects and the pollination service in agricultural landscapes

Small floral patches that coexist with crops in agricultural landscapes can function as biodiversity reservoirs. However, the influence of the landscape context and agricultural management on the capacity of these small green infrastructures to support diversity is poorly understood. Here, we evaluate the effect of landscape simplification, agricultural intensification in the neighbourhood, and quality of the floral habitats on the success of these patches to support flower-visiting insect communities as well as the pollination service they provide.To this aim, we sampled floral patches located in 18 paired olive farms with contrasting herb cover management (intensive vs. low-intensity), distributed along a wide gradient of landscape complexity at the regional scale of South Spain. We conducted surveys of flower-visiting insects in 36 multi-floral stands and 36 mono-floral stands of Sinapis alba Linnaeus (1753) within these floral patches. Mono-floral stands were used to evaluate variations in the pollination service through number of viable seeds and seed set.Results revealed that the abundance and diversity of flower-visiting insects respond to the quality of the floral patch (diversity of flowers) but not to landscape context nor agricultural management around it. Moreover, the pollination service was similar and high (seed set ca. 100 %) in all floral patches regardless of their context.Our findings highlight the importance of even small floral patches that function as reservoirs of diversity of flower-visiting insects and the pollination service. They also show the high resistance of these patches to agricultural intensification and simplification in olive grove landscapes.

Bat Activity in Organic Rice Fields Is Higher Than in Conventional Fields in Landscapes of Intermediate Complexity

The extent to which organic farming can support biodiversity has been extensively studied. However, most of the research has been conducted on organic farms in temperate regions, with the focus mainly being on birds, insects, and plants and rarely on insectivorous bats, especially in Southeast Asia. We studied pairs of matched organic and conventional rice fields along a gradient of landscape complexity in the Songkhla Lake Basin and conducted acoustic surveys using bat detectors to analyze the influence of farming system and landscape characteristics on bat activity and prey availability. We also tested the “intermediate landscape complexity” hypothesis, which states that local conservation efforts are most effective in landscapes of intermediate complexity compared to extremely simple or extremely complex landscapes. We detected no difference in bat species richness, total bat activity, feeding activity, and insect prey abundance between organic fields and conventional fields. Even though organic farming did not increase bat activity on its own, it was most beneficial to bat activity in landscapes of intermediate complexity. Our findings suggest that landscape traits contribute more to bat activity than farm management and that insectivorous bats have species- and guild-specific responses to various landscape contexts. We also found that disturbance caused by tropical storms negatively impacts the activity of insectivorous bat.

Spatial configuration and landscape context of wildflower areas determine their benefits to pollinator α- and β-diversity

Wildflower areas have become a staple tool within agro-environmental schemes (AES) to counteract pollinator declines. While their role in providing food resources to resident flower-visiting insects is unambiguous, the conservation effectiveness in a landscape context is less clear. Particularly, how multiple vs. single wildflower area utilization differs between simple and complex landscapes is understudied. We examined colonisation and community dynamics of wild bees and hoverflies in 33 newly established wildflower areas across a gradient of landscape complexity (amount of semi-natural habitat) and connectivity (presence of additional wildflower areas) for seven weeks during three consecutive years (one year during and two after establishment). We recorded more than 25% of the wild bee and hoverfly species of Hesse in an area of approx. 10 ha, substantiating the general benefit of wildflower areas to pollinators. While alpha-diversity increased with landscape complexity in isolated wildflower areas, the opposite pattern was observed for connected areas. The low alpha-diversity in complex landscapes indicated a dilution effect between connected sites. The inverse relationship between alpha and beta diversity among wildflower areas within landscapes suggests interspecific trade-offs between local resource conditions and landscape context. Accordingly, the establishment of multiple wildflower areas within AES is advisable to increase connectivity of suitable habitats in simple landscapes. Moreover, adjusting local conditions (plant diversity) to landscape context likely optimizes conservation effectiveness in modern agroecosystems.

Landscape complexity benefits bumble bee visitation in faba bean (Vicia faba minor L.) but crop productivity is not pollinator-dependent

Faba bean (Vicia faba minor L.) is partially dependent on insects for pollination, but the degree of pollinator-dependency and whether crop productivity is limited by insufficient insect pollination remain unknown. We monitored insect pollinators and their foraging behaviors (i. e., legitimate flower visitation, nectar robbing and extra-floral nectary visitation) in a total of 20 faba bean fields of a single cultivar (Tiffany) 2018–2019. Focal fields were situated along a gradient of landscape complexity. In each field, a pollination experiment was established, where plants were either bagged to exclude pollination by insects or remained open for pollinator visits. In addition, all flowers on half of the bagged and open-pollinated plants were pollinated by hand to measure the degree of pollen limitation. We found that bumble bee abundance was higher, and legitimate flower visitor abundance tended to be higher, in complex landscapes with more semi-natural habitat, indicating that the faba bean pollination potential is higher in complex landscapes. The pollination experiment showed that the number of beans per pod was lower in bagged plants compared with other treatments in one of the years, but the lower number of beans per pod was compensated for by a higher individual bean weight, such that there was no pollinator-dependency or effect of hand pollination on total bean mass per plant. We conclude that to be able to characterize the value of insect pollination services in faba bean we need an improved understanding of how pollinator-dependency varies across modern cultivars, and how environmental conditions can increase or, as in our case, cancel this pollinator-dependency.

Landscape and agri-environmental scheme effects on ant communities in cereal croplands of central Spain

Agri-environmental schemes (AES) of the Common Agricultural Policy (CAP) aims at reversing the negative effects of agricultural intensification on biodiversity and ecosystem services. Landscape context may modulate, and even constraint, AES effectiveness. We evaluate AES effectiveness on ant abundance, diversity and community composition. Ants are an ecologically dominant group whose response to conservation efforts in farmland has been rarely evaluated, despite its role in weed control, particularly in Mediterranean farmland.Ants were sampled in the edge and in the centre of paired cereal fields, managed with and without AES in three study areas along a landscape complexity gradient. AES application had no significant effects on ant species richness or ant community composition. Richness increased in fields and landscapes with higher amounts of complex edges and decreased towards the centre of the fields. Specialist granivorous ants (harvester ants, Messor spp.) were the most abundant. Abundance of foraging ants increased with the amount of complex edges around fields and in the landscape. AES application increased ant abundance close to field edges but not in field centers. AES fields had less specialist granivorous foraging in their centers than in control field centers.Ant communities in Mediterranean cereal cropland were mostly constrained by the availability of complex edges, needed for nest building. AES increased the abundance of foraging ants, mostly specialist harvester ants, and its potential service of weed control, but close to field edges mainly. Measures promoting the abundance of stable edges rather than of ephemeral headlands in the landscape are essential to enhance the potential of AES for increasing ant-mediated ecosystem services of weed control.

Mutual and Opposing Responses of Carabid Beetles and Predatory Wasps to Local and Landscape Factors in Vineyards.

Simple SummaryThe aim of this study was to contribute to closing knowledge gaps on managing vineyards and viticultural landscapes in order to support insect diversity and abundance. We studied two different groups of predating insects, carabid beetles and cavity-nesting wasps, in organically and conventionally managed vineyards in Germany. Effects of surrounding landscapes and vegetation structure within vineyards were evaluated. No differences in species richness and abundance of carabid beetles and cavity-nesting wasps were found between organic and conventional management. Enhanced vegetation cover was positively correlated with carabids and negatively with wasps. High covers of annual crops in the surrounding landscape led to fewer species and individuals of both groups. The results underline the importance of insect-friendly management, especially in intensely farmed landscapes.Preserving agro-biodiversity is one of the main means at the moment to counteract the global biodiversity crisis. Vineyard inter-rows offer vegetation covers which could function as foraging grounds for arthropods. Furthermore, organic management and enhanced landscape complexity often support biodiversity. Here, species richness and abundance of two groups of arthropod predators in vineyards were studied. Fifteen pairs of organically and conventionally managed vineyards were chosen along a gradient of landscape complexity in Rhine-Hesse, Germany. Carabid beetles were sampled using pitfall traps and cavity-nesting wasps with trap nests, respectively. Proportions of different land-use types surrounding the vineyards were calculated and inter-row vegetation cover was characterized. Species richness and abundances of both predator groups were not significantly affected by the management system. Likewise, increased cover of semi-natural habitats in the surrounding landscape did not promote their diversity or abundance. Instead, the increasing cover of annual crops diminished both groups. Cavity-nesting wasps profited from dense inter-row vegetation cover, while carabids were disadvantaged. The results indicate that distinct taxa within the same trophic group can respond oppositely to vineyard management. Thus, inter-row vegetation management with densely and sparsely vegetated elements might be best to support predator diversity. Overall, our results suggest that organic viticulture alone is insufficient to assist the studied insect groups, and that other local and landscape management options are needed for their protection.

Contrasting effects of host crops and crop diversity on the abundance and parasitism of a specialist herbivore in agricultural landscapes

Landscape complexity affects herbivores in agroecosystems, but consequences on pest control services are variable. Carryover effects of landscape composition in previous years on herbivore control may be important, but have been seldom assessed. Understanding landscape complexity effects at different temporal and spatial scales is important to improve sustainable pest control services. We examined the effect of agricultural landscape complexity (e.g., the percentage of semi-natural habitats) on cereal leaf beetle (CLB), Oulema melanopus L., and its parasitism by Tetrastichus julis (Walker). From 2014 to 2015, we assessed CLB abundance and parasitism in 54 fields along a gradient of landscape complexity (2–70% of semi-natural habitats) in southern Alberta. We used generalized linear models to test the effects of percentage of crops and semi-natural habitats, and landscape and crop diversity on CLB abundance and parasitism at 0.5, 1, 1.5 and 2 km spatial scales. CLB abundance decreased with higher crop diversity at the 0.5 km scale and increased with CLB host crops in the current and previous years at multiple scales, supporting the resource concentration hypothesis. CLB parasitism increased with CLB abundance and in landscapes with increased canola and alfalfa during a year of low CLB abundance. CLB abundance had contrasting associations with semi-natural habitats: positive with woodlands and negative with pastures. Our study suggests that crop diversity reduces the abundance of this specialist pest in agricultural landscapes likely by a dual effect of reducing host crop area and increasing habitats with resources for parasitoids.

Habitat area and connectivity support cavity-nesting bees in vineyards more than organic management

The expansion and intensification of agriculture are the main causes of current insect declines. Pollinators like cavity-nesting bees can be limited by reduced nesting and feeding opportunities in farmland. As insects constitute the bulk of terrestrial biodiversity and fulfill important ecological functions, there is an urgent need to identify ways to combine agricultural land use and insect conservation. Perennial crops like grapevine can provide permanent habitats for numerous beneficial organisms including various pollinators. With their dominating character in viticultural areas and >7 million ha covered by vines globally, their potential to contribute to nature conservation should be more widely considered. We compared effects of organic management, inter-row vegetation characteristics and landscape parameters on the abundance and species richness of cavity-nesting bees in Central German vineyards. In a paired study design, we assessed cavity-nesting bees in 15 pairs of organically and conventionally managed vineyards along a gradient of landscape complexity. We found that organic management, even though it enhanced flower availability in the vineyards, was only partially beneficial for cavity-nesting bee abundance. Abundance and species richness were enhanced by either semi-natural habitat area or proximity of woody elements like hedges or forest remnants, most likely due to the nesting demands of this particular group of pollinators. We conclude that vineyards can help to sustain cavity-nesting bee abundance, given that landscapes are managed accordingly. We recommend maintaining or establishing woody elements between vineyards, which is likely to also benefit additional groups of organisms such as breeding birds in viticultural landscapes.

Interacting effects of landscape and management on plant–solitary bee networks in olive orchards

Abstract Understanding how multi‐scale environmental heterogeneity shapes the structure and functions of animal and plant communities is pivotal in agroecology. Our capacity to ensure the provision of ecosystem services (ES), the sustainability of agroecosystems and the efficiency of agri‐environmental schemes (AES) relies on this knowledge. There is growing interest in how biodiversity and ES are affected by the interplay between landscape characteristics and agricultural management (e.g., intermediate landscape complexity hypothesis; ILCH). However, studies have typically focused on classical measures of community structure (e.g., species abundance, richness and biodiversity), tending to neglect the effects on the structure and stability of ecological networks and the increased risk of biotic homogenization (i.e., loss of β‐diversity). In this work, we use bee trap nests to sample pollen–solitary bee mutualistic networks in 9 pairs of olive groves under different management regimes (conventional vs. organic) in a landscape complexity gradient in southern Spain. We analyse the mutualistic networks at farm level to test the ILCH and study how agricultural practices and landscape complexity interact to affect the properties of these networks. We also explore the effects on spatial biotic homogenization by performing multivariate analyses of the composition and abundance of bee–plant communities and their pairwise interactions. The results show that solitary bee–plant networks have greater complexity and stability on organic olive farms embedded in relatively heterogeneous landscapes. Differences from conventional management were higher in landscapes of intermediate complexity but were non‐significant on olive farms located in simpler landscapes. β‐diversity of bees, plants and their pairwise interactions was also greater on organic olive farms. In conclusion, human‐induced environmental heterogeneity interacts at different scales to shape plant–solitary bee networks in olive groves, which may have important implications for ES provision and the effectiveness of agri‐environmental measures. A free Plain Language Summary can be found within the Supporting Information of this article.

Low‐intensity management benefits solitary bees in olive groves

Abstract One of the current challenges for applied ecologists is to understand how to manage/restore agroecosystems in a sustainable and cost‐effective way. The intermediate landscape complexity hypothesis (ILCH) predicts that the effectiveness of agri‐environmental measures (AES) on biodiversity and ecosystem services recovery is often largest in landscapes of intermediate complexity. This hypothesis has rarely been tested in savanna‐like permanent agroecosystems. Focusing on pollinators, we test the ILCH at the regional scale in Mediterranean olive orchards, one of the most important permanent agroecosystems in the world. We inferred abundance of cavity‐nesting pollinators in 40 paired olive orchards (extensively vs. intensively managed herbaceous cover) in 20 localities selected across a landscape complexity gradient. We also studied how different magnitudes in local management switches may affect pollinators by considering organic and intensive fields as management extremes in olive orchards. We used 208 trap nests for solitary bees to measure colonization rates. Additionally, we conducted pollinator surveys to ascertain that colonization rate was a representative proxy for pollinator activity. Our results showed that (a) changes in colonization rates due to local herb cover management peaked at intermediate landscape complexity, with extensively managed fields rendering higher colonization rates. (b) Organic fields had higher colonization rates than their control farms regardless of landscape complexity. (c) There was a highly significant correlation between nest colonization rates and density of pollinators foraging on flowers, which suggests that colonization rate is a good estimator of pollinator activity. Policy implications. The maintenance of ground herb cover (main agri‐environmental measure in olive orchards) is a cost‐effective investment allowing recuperation of pollinators when targeting olive farms located in landscapes of intermediate complexity. Additionally, fostering organic farming (still minority in olive groves) for the conservation of solitary bees should be a priority for policymakers since its effects are beneficial in any landscape.

Carabid functional diversity is enhanced by conventional flowering fields, organic winter cereals and edge habitats

The continued decline in farmland biodiversity in Europe despite substantial funding for agri-environment schemes (AES) has prompted calls for more effective biodiversity conservation measures. The current AES regime allows for both holistic measures, such as organic farming, that broadly target the agricultural environment and biodiversity-specific measures, such as flowering fields, but little is known of their relative efficacies. To address this gap, we studied carabids in 48 arable fields that presented four crop types under different management practices along a gradient of landscape complexity: (a) conventionally managed crop (winter wheat), (b) biodiversity-specific AES under conventional management (sown flowering field), (c) organically managed mono-crop (winter spelt) and (d) organically managed lentil-mixedcrop (lentil intercropped with cereal or camelina). For these crop-use types, we compared functional diversity of carabid assemblages at the edge and center of the fields. Using pitfall traps, we collected more than 55,000 carabids of 95 species over two years. We characterized diversity using community weighted means and functional divergence of three ecological traits – body size, feeding type, and flight ability. Conventional flowering fields and organic winter spelt, but not organic spring sown lentil-mixed-crop, increased the proportion of plant-feeding carabids; moreover, trait characteristics and their divergences were most affected by field edges, with smaller, less carnivorous and more flight-enabled carabid assemblages found there than in the center. Divergence of body size and feeding type but not of flight ability was larger at the field edges than centres. Surrounding landscape complexity did not affect carabid traits. We conclude that future AES policy should avoid strict decisions between biodiversity specific- and holistic measures. Instead, priority should be given to a diversity of different measures, targeting the enhancement of edge habitats as well as productive and non-productive measures.

Landscape-moderated biodiversity effects of ground herb cover in olive groves: Implications for regional biodiversity conservation

Studies assessing the effect of extensive versus intensive agricultural practices and addressing how biodiversity patterns and the effectiveness of agri-environmental practices (AES) to recover biodiversity are moderated by the landscape complexity (LMB framework), underlie large-scale biodiversity conservation programs and policies in anthropogenic landscapes. Such studies are numerous with annual crops and grasslands yet infrequent in more complex and structurally stable arboreal croplands, where high capacity to retain biodiversity is expected. Here, we explore to what extent landscape complexity and extensification of agricultural practices enhance biodiversity in olive groves of Andalusia (southern Spain). We monitored birds, ants and herbs in paired olive farms (extensive versus intensive ground herb cover management) from 20 localities spread over Andalusia and along a landscape complexity gradient. For each biological group, we obtained gamma diversity (diversity at the olive farm level), beta diversity (between the productive and unproductive areas located within the olive farm) and infield alpha diversity (in the productive area within the olive farm). We tested for multi-diversity, and for each group separately, three major hypotheses of the LMB: the intermediate-landscape complexity, the dominance of beta diversity, and the landscape species pool hypotheses. These hypotheses were corroborated with multi-diversity, which was affected by intensification of weed management and landscape simplification, suffering a combined impact of 26% of gamma biodiversity loss. The effectiveness of extensification to recover biodiversity depended on the landscape context and peaked at intermediate-complexity landscapes. Beta multi-diversity and infield alpha-diversity increased with landscape complexity. Birds, ants and weeds responded differently but were negatively affected either by landscape simplification or by management intensification. Birds mirrored better than other groups the multi-diversity pattern and were the best candidates for a rapid indicator of the impact of agriculture and land conversion on biodiversity. We provide recommendations for biodiversity conservation in olive groves-dominated landscapes and show that, if adequately managed, olive groves’ landscapes have potential for the conservation of biodiversity in the Mediterranean region. Our results illustrate the need to reformulate the future EU-Common Agricultural Policy and particularly, to adapt AES to each landscape.

Predation-mediated ecosystem services and disservices in agricultural landscapes.

Ecological intensification may reduce environmental externalities of agriculture by harnessing biodiversity to benefit regulating ecosystem services. However, to propose management options for the production of such services, there is a need to understand the spatiotemporal dynamics of net effects between ecosystem services and disservices provided by wild organisms across taxonomic groups in relation to habitat and landscape management. We studied the contribution of predatory vertebrates and invertebrates (including both carnivores and seed herbivores) to regulating ecosystem services and disservices in 16 cereal fields in response to a local habitat contrast and a landscape complexity gradient. From May to November 2016, we provided weed (predation reflects an ecosystem service) and crop (predation reflects a disservice) seeds, as well as pest (predation reflects an ecosystem service) and beneficial (predation reflects a disservice) invertebrate prey to predators. Seed predation was dominated by vertebrates, while vertebrates and invertebrates contributed equally to predation of animal prey. Before harvest, predation steadily increased from very low levels in May to high levels in July independent of the resource type. After harvest, ecosystem services declined more rapidly than disservices. The presence of adjacent seminatural grasslands promoted crop seed predation, but reduced pest prey predation. Predation on beneficial prey decreased with increasing proportions of seminatural grassland in the landscape. Predatory vertebrates and invertebrates provide important ecosystem services due to the consumption of pests. However, beneficial invertebrates and crop seeds were often consumed to a similar or even higher extent than harmful invertebrates or weed seeds. Our results therefore raise concerns that management options aimed at enhancing service providers may simultaneously increase levels of disservices. By considering positive and negative effects simultaneously, this study addresses an important knowledge gap and highlights the importance of interactions between local management, landscape composition, and service and disservice provision across taxa and over time. Considering trade-offs between ecosystem services and disservices when evaluating the net effects of biodiversity conservation measures on ecosystem service provision is crucial. Future agri-environment schemes that offer payments for seminatural habitats may need to provide higher compensation for farmers in cases where net effects are likely to be negative.

Unpacking the push-pull system: Assessing the contribution of companion crops along a gradient of landscape complexity

The push-pull system, a stimulo-deterrent cropping strategy consisting of intercropping cereals with herbaceous legumes and surrounded by fodder grasses, is presented as a promising crop diversification strategy for smallholder farmers in Africa as it may contribute to maize stemborer Busseola fusca (Fuller) suppression, while improving soil fertility and providing feed for livestock. The push-pull system has often been assessed at plot level and as a package (e.g., Maize + Desmodium + Napier grass). However, it is unclear how the system performs in different landscape settings or when companion crops are changed to better meet household needs. Here we evaluate the potential of the push-pull system to suppress maize stemborer infestations in three landscapes in the Rift Valley region of Ethiopia along a gradient of landscape complexity. Within each landscape, experimental plots were established on four representative smallholder farms. At each farm we used a split-plot factorial design with main plots surrounded or not by Napier grass, and subplots consisting of sole maize, maize-bean or maize-Desmodium. We assessed stemborer infestation level and maize grain and stover yields during two years, as well as natural enemies abundance and egg predation at two maize development stages in the second year. In the simple landscape, which was dominated by maize, all treatments had high stemborer infestation levels, irrespective of within-field crop diversity; the presence of Napier grass was associated with higher predator abundance, while egg predation rates were the highest in the maize-bean intercrop. In the intermediate complexity landscape, subplots with sole maize had higher stemborer infestation levels compared to maize-bean or maize-Desmodium. In the complex landscape, infestation levels were low in all treatments. However, none of these effects led to significant differences in maize grain and stover yields among treatments in any of the landscapes. The benefits of the push-pull system accrued from the companion crops (bean, Desmodium and Napier), rather than from stemborer suppression per se. Our findings highlight the importance of the surrounding landscape in mediating the performance of the push-pull system, provide new insights on the contribution of the different components of push-pull system and can guide the design of ecologically intensive agroecosystems.

Landscape complexity promotes hoverflies across different types of semi‐natural habitats in farmland

Abstract Semi‐natural habitats (SNH) provide essential resources for many organisms in agricultural landscapes and can increase biodiversity at the local and landscape scale. For the management of ecosystem services, it is crucial to understand how local characteristics of SNH and the surrounding landscape complexity affect beneficial species. We investigated this for hoverflies (Diptera: Syrphidae)—an important functional group providing both pest control and pollination services—in a total of 138 SNH within 35 agricultural landscapes in Switzerland and Germany. SNH differed in type (woody, herbaceous), shape (areal, linear) and availability of food resources (floral resources and aphids). They were located along a gradient of landscape complexity (1%–75% SNH in a 1 km radius). In total, 9,030 hoverflies belonging to 89 species were collected. In both countries, hoverfly community composition was mainly driven by SNH type and SNH shape, in addition to landscape complexity in Switzerland. Species richness of hoverflies increased with additional SNH at the landscape scale. However, the ubiquitous and dominant aphidophagous species Episyrphus balteatus was indifferent to landscape composition. Species richness of total and aphidophagous hoverflies was higher in woody than in herbaceous SNH. Hoverfly species richness was similar in linear and areal SNH, but non‐aphidophagous hoverflies were more vulnerable to landscape simplification in linear than areal SNH. Effects of the SNH type partly differed between the two countries. In Germany, the dominant aphidophagous species E. balteatus preferred woody over herbaceous SNH, while no such difference was found in Switzerland. Overall, local richness and abundance of floral resources were poor predictors for hoverfly abundance, richness or community composition. Synthesis and applications. Hoverfly species richness and community composition in semi‐natural habitats (SNH) were mainly driven by the SNH type and landscape complexity. Conservation and restoration of complex agricultural landscapes with a high proportion of different SNH types is therefore key for the conservation of hoverfly diversity, and thus likely to promote pest control and pollination services provided by them. Thereby, local improvement of SNH to promote hoverflies has to consider regional differences in habitat characteristics.

Relationships between multiple biodiversity components and ecosystem services along a landscape complexity gradient

The assessment of effects of anthropogenic disturbance on biodiversity (BD) and ecosystem services (ES) and their relationships are key priorities of the Intergovernmental Panel for Biodiversity and Ecosystem Services. Agricultural landscapes and their associated BD provide multiple ES and it is crucial to understand how relationships between ES and BD components change along gradients of landscape complexity. In this study, we related eight ES potentials to the species richness of five invertebrate, vertebrate and plant taxonomic groups in cereal farming systems. The landscape complexity gradient ranged from areas dominated by annually tilled arable land to areas with high proportions of unfertilized, non-rotational pastures and uncultivated field borders. We show that after accounting for landscape complexity relationships between yield and bird richness or biological control became more positive, but relationships between bird richness and biological control became less positive. The relationship between bird and plant richness turned from positive to negative. Multidiversity (overall biodiversity), was positively related to landscape complexity, whereas multifunctionality (overall ES provision), was not significantly related to either one of these. Our results suggest that multidiversity can be promoted by increasing landscape complexity; however; we found no support for a simultaneous increase of several individual ES, BD components or multifunctionality. These results challenge the assumption that biodiversity-friendly landscape management will always simultaneously promote multiple ES in agricultural landscapes. Future studies need to verify this pattern by using multi-year data, larger sets of ES and BD components and a study design that is appropriate to address larger spatial scales and relationships in several regions.