Insect pollinators provide a crucial ecosystem service, but are under threat. Urban areas could be important for pollinators, though their value relative to other habitats is poorly known. We compared pollinator communities using quantified flower-visitation networks in 36 sites (each 1 km2) in three landscapes: urban, farmland and nature reserves. Overall, flower-visitor abundance and species richness did not differ significantly between the three landscape types. Bee abundance did not differ between landscapes, but bee species richness was higher in urban areas than farmland. Hoverfly abundance was higher in farmland and nature reserves than urban sites, but species richness did not differ significantly. While urban pollinator assemblages were more homogeneous across space than those in farmland or nature reserves, there was no significant difference in the numbers of rarer species between the three landscapes. Network-level specialization was higher in farmland than urban sites. Relative to other habitats, urban visitors foraged from a greater number of plant species (higher generality) but also visited a lower proportion of available plant species (higher specialization), both possibly driven by higher urban plant richness. Urban areas are growing, and improving their value for pollinators should be part of any national strategy to conserve and restore pollinators.

Urban areas are often perceived to have lower biodiversity than the wider countryside, but a few small-scale studies suggest that some urban land uses can support substantial pollinator populations. We present a large-scale, well-replicated study of floral resources and pollinators in 360 sites incorporating all major land uses in four British cities. Using a systems approach, we developed Bayesian network models integrating pollinator dispersal and resource switching to estimate city-scale effects of management interventions on plant-pollinator community robustness to species loss. We show that residential gardens and allotments (community gardens) are pollinator ‘hotspots’: gardens due to their extensive area, and allotments due to their high pollinator diversity and leverage on city-scale plant-pollinator community robustness. Household income was positively associated with pollinator abundance in gardens, highlighting the influence of socio-economic factors. Our results underpin urban planning recommendations to enhance pollinator conservation, using increasing city-scale community robustness as our measure of success.

Summary1. Species distribution models could bring manifold benefits across ecology, but require careful testing to prove their reliability and guide users. Shortcomings in testing are often evident, failing to reflect recent methodological developments and changes in the way models are applied. We considered some of the fundamental issues. 2. Generalizability is a basic requirement for predictive models, describing their capacity to produce accurate predictions with new data, i.e. in real applications beyond model training. Tests of generalizability should be as rigorous as possible: ideally using a large number of independent test sites ( ≥ 200-300) that represent anticipated applications. Bootstrapping identifies the role of overfitting of the training data in limiting a model's generalizability. 3. Predictions from most distribution models are continuous variables. Their accuracy may be described by discrimination and calibration components. Discriminatory ability describes how well a model separates occupied from unoccupied sites. It is independent of species prevalence and is readily comparable between models. Rank correlation coefficients, such as the concordance index, are effective measures. 4. Calibration describes the numerical accuracy of predictions (e.g. whether 40% of sites with predicted probabilities of 0·40 are occupied) but is frequently overlooked in model testing. Poor calibration could mislead any conservation efforts utilizing models to estimate the 'value' of different sites for a given species. Effective assessments can be made using smoothed calibration plots. 5. The effects of species prevalence on nominal presence-absence predictions are well known. The currently preferred accuracy measure, Cohen's κ , has weaknesses. We argue that mutual information measures, based in information theory, may be more appropriate. 6. Synthesis and applications . Model evaluation must be informative and should ideally: (i) define generalizability in detail; (ii) separate the discrimination and calibration components of accuracy and test both; (iii) adopt assessment techniques that permit more valid intermodel comparisons; (iv) avoid nominal presence-absence evaluation where possible and consider information-theoretic measures; and (v) utilize the full range of techniques to help diagnose the causes of prediction problems. Few modellers in applied ecology and conservation biology satisfy these needs, making it difficult for others to evaluate models and identify potential misuses. The problems are real, and if uncorrected will damage conservation efforts through the inaccurate assessment of distribution and habitat preferences of important organisms.

Collembola comprise a major source of alternative prey to linyphiid spiders in arable fields, helping to sustain and retain these predators as aphid control agents within the crop. Polymerase chain reaction primers were developed for the amplification, from spider gut samples, of DNA from three of the most abundant species of Collembola in wheat crops in Europe, namely Isotoma anglicana, Lepidocyrtus cyaneus and Entomobrya multifasciata. The primers amplified fragments of the mitochondrial cytochrome oxidase subunit I (COI) gene and were designed following alignment of comparable sequences for a range of predator and prey species. Each of the primer pairs proved to be species-specific to a Collembola species, amplifying DNA fragments from 211 to 276 base pairs in length. Following consumption of a single collembolan, prey DNA was detectable in 100% of spiders after 24 h of digestion. We report the first use of DNA-based techniques to detect predation by arthropods on natural populations of prey in the field. All three species of Collembola were consumed by the spiders. By comparing the ratios of the Collembola species in the field with the numbers of spiders that gave positive results for each of those species, it was possible to demonstrate that the spiders were exercising prey choice. Overall, a single target species of Collembola was eaten by 48% of spiders while a further 16% of spiders contained DNA from two different species of Collembola. Preference was particularly evident for I. anglicana, the species most frequently found in spider guts yet the least numerous of the three target species in the field.

1. Floral resources are known to be important in regulating wild pollinator populations and are therefore an important component of agri-environment and restoration schemes which aim to support pollinators and their associated services.However, the phenology of floral resources is often overlooked in these schemes -a factor which may be limiting their success.2. Our study characterizes and quantifies the phenology of nectar resources at the whole-farm scale on replicate farms in Southwestern UK throughout the flowering season. We quantify the corresponding nectar demands of a subset of common wild pollinators (bumblebees) to compare nectar supply and pollinator demand throughout the year, thereby identifying periods of supply-demand deficit.3. We record strong seasonal fluctuations in farmland nectar supplies, with two main peaks of nectar production (May and July) and a considerable "June Gap" in between. March and August/September are also periods of low nectar availability.4. Comparing the phenology of nectar supply with the phenology of bumblebee nectar demand reveals "hunger gaps" during March and much of August/September when supply is unlikely to meet demand. 5. Permanent pasture and woodland produced the greatest share of farmland nectar because of their large area; however, linear features such as hedgerows and field margins provided the greatest nectar per unit area. Fifty percent of total nectar was supplied by just three species (Allium ursinum, Cirsium arvense and Trifolium repens), but some less productive species (e.g. Hedera helix and Taraxacum agg.) were important in ensuring phenological continuity of nectar supply. 6. Synthesis and applications. By comparing the phenology of farmland nectar supply with the phenology of pollinator demand, we demonstrate that the timing of nectar supply may be as important as total nectar production in limiting farmland pollinator populations. Considering phenology in the design of agri-environment or restoration schemes is therefore likely to improve their suitability for pollinators. Plant species which flower during periods of resource deficit (early spring and late summer) should be prioritized in schemes which aim to conserve or restore K E Y W O R D S agri-environment, bumblebees, floral resources, flowering phenology, nectar, pollination, pollinator conservation, restoration | 1587 Journal of Applied Ecology TIMBERLAKE ET AL.

ABSTRACT1. The assessment of links between ecology and physical habitat has become a major issue in river research and management. Key drivers include concerns about the conservation implications of human modifications (e.g. abstraction, climate change) and the explicit need to understand the ecological importance of hydromorphology as prescribed by the EU's Water Framework Directive. Efforts are focusing on the need to develop 'ecohydromorphology' at the interface between ecology, hydrology and fluvial geomorphology. Here, the scope of this emerging field is defined, some research and development issues are suggested, and a path for development is sketched out.2. In the short term, major research priorities are to use existing literature or data better to identify patterns among organisms, ecological functions and river hydromorphological character. Another early priority is to identify model systems or organisms to act as research foci. In the medium term, the investigation of patternprocesses linkages, spatial structuring, scaling relationships and system dynamics will advance mechanistic understanding. The effects of climate change, abstraction and river regulation, eco-hydromorphic resistance/ resilience, and responses to environmental disturbances are likely to be management priorities. Large-scale catchment projects, in both rural and urban locations, should be promoted to concentrate collaborative efforts, to attract financial support and to raise the profile of eco-hydromorphology.3. Eco-hydromorphological expertise is currently fragmented across the main contributory disciplines (ecology, hydrology, geomorphology, flood risk management, civil engineering), potentially restricting research and development. This is paradoxical given the shared vision across these fields for effective river management based on good science with social impact. A range of approaches is advocated to build sufficient, integrated capacity that will deliver science of real management value over the coming decades.

Bees, hoverflies and butterflies are taxa frequently studied as pollinators in agricultural and conservation contexts. Although there are many records of non-syrphid Diptera visiting flowers, they are generally not regarded as important pollinators. We use data from 30 pollen-transport networks and 71 pollinator-visitation networks to compare the importance of various flower-visiting taxa as pollen-vectors. We specifically compare non-syrphid Diptera and Syrphidae to determine whether neglect of the former in the literature is justified. We found no significant difference in pollen-loads between the syrphid and non-syrphid Diptera. Moreover, there was no significant difference in the level of specialization between the two groups in the pollen-transport networks, though the Syrphidae had significantly greater visitation evenness. Flower visitation data from 33 farms showed that non-syrphid Diptera made up the majority of the flower-visiting Diptera in the agricultural studies (on average 82% abundance and 73% species richness), and we estimate that non-syrphid Diptera carry 84% of total pollen carried by farmland Diptera. As important pollinators, such as bees, have suffered serious declines, it would be prudent to improve our understanding of the role of non-syrphid Diptera as pollinators.

Summary 1. Anthropogenic activities can increase fine sediment supply to streams over multiple spatial and temporal extents. Identifying the processes responsible, and the scale at which any effects on stream organisms become evident, are key management needs, but appropriately scaled surveys are surprisingly few. 2. We surveyed macroinvertebrates and superficial fine sediments at two spatial resolutions (reach‐ and patch‐scale) in tributaries of the River Usk, a temperate, montane catchment in rural Wales (U.K.). Land use, habitat and geomorphological character were measured on‐site or derived from an existing database (=Fluvial Audit). We aimed to identify: (i) how in‐stream sediments varied with land use and associated geomorphology; (ii) likely consequences for macroinvertebrates and (iii) any scale‐dependence in relationships between macroinvertebrates and sediment character. 3. At both the reach‐ and patch‐scales, bed cover by fine sediment was related directly to the extent of eroding banks 500 m upstream. In turn, sedimentation and bank erosion were negatively correlated with catchment or riparian woodland extent. 4. At the reach scale, macroinvertebrate composition varied with catchment land use and stream chemistry, with richness declining as rough grazing or woodland was replaced by improved grassland. There was no response to deposited sediment except for weak increase in the relative abundance of oligochaetes. 5. By contrast, at the patch scale, fine sediments were accompanied by pronounced changes in invertebrate composition, and we ranked the 27 most common taxa according to their apparent sediment tolerance. General estimating equations showed that total and Ephemeroptera, Plecoptera and Trichoptera richness decreased significantly by 20% and 25% at the most sediment impacted sites (30% cover) by comparison with sediment‐free sites. 6. We conclude that sediment deposition in the upper Usk system mostly reflects local bank erosion, with riparian woodland likely to mediate this process through bank stability. Fine sediment release had marked ecological effects, but these were detectable only at patch‐to‐patch scales. We suggest that investigation of localized sediment release in streams will benefit from scale‐dependent or scale‐specific sampling, and some effects could go undetected unless sample resolution is selected carefully.