This is the author's manuscript for a work that has been accepted for publication. Changes resulting from the publishing process, such as copyediting, final layout, and pagination, may not be reflected in this document. The publisher takes permanent responsibility for the work. Content and layout follow publisher's submission requirements.

Recent years have seen a rapid expansion in the ability of earth system models to describe and predict the physical state of the ocean. Skilful forecasts ranging from seasonal (3 months) to decadal (5-10 years) time scales are now a reality. With the advance of these forecasts of ocean physics, the first generation of marine ecological forecasts has started to emerge. Such forecasts are potentially of great value in the management of living marine resources and for all of those who are dependent on the ocean for both nutrition and their livelihood; however, this is still a field in its infancy. We review the state of the art in this emerging field and identify the lessons that can be learnt and carried forward from these pioneering efforts. The majority of this first wave of products are forecasts of spatial distributions, possibly reflecting the inherent suitability of this response variable to the task of forecasting. Promising developments are also seen in forecasting fish-stock recruitment where, despite well-recognized challenges in understanding and predicting this response, new process knowledge and model approaches that could form a basis for forecasting are becoming available. Forecasts of phenology and coral-bleaching events are also being applied to monitoring and industry decisions. Moving marine ecological forecasting forward will require striking a balance between what is feasible and what is useful. We propose here a set of criteria to quickly identify "low-hanging fruit" that can potentially be predicted; however, ensuring the usefulness of forecast products also requires close collaboration with actively engaged end-users. Realizing the full potential of marine ecological forecasting will require bridging the gaps between marine ecology and climatology on the one-hand, and between science and end-users on the other. Nevertheless, the successes seen thus far and the potential to develop further products suggest that the field of marine ecological forecasting can be expected to flourish in the coming years.

We assessed bee diversity and abundance in urban areas of Vancouver, British Columbia, Canada, to determine how urban environments can support bees. Habitats examined were community and botanical gardens, urban wild areas, Naturescape flower beds and backyards, and traditional flower beds and backyards. A total of 56 bee species (Hymenoptera), including species of the genera Andrena Fabr. (Andrenidae), Bombus Latr. (Apidae), Osmia Panzer and Megachile Latr. (Megachilidae), and Halictus Latr. and Dialictus Pauly (Halictidae), were collected. Abundance exhibited strong seasonal variation. Wild bees were most abundant during late spring, whereas honey bees peaked at the end of the summer. The most abundant species seen was the managed honey bee Apis mellifera L. (Apidae), followed by wild Bombus flavifrons Cresson. Community and botanical gardens, and plants such as cotoneaster (Cotoneaster Medik. sp.) and blackberry (Rubus discolor Weihe & Nees) (Rosaceae), centaurea (Centaurea L. sp.; Asteraceae), buttercup (Ranunculus L. sp.; Ranunculaceae), and foxglove (Digitalis L. sp.; Scrophulariaceae), had the highest abundance of bees, while bee populations in wild areas were the most diverse. Weeds such as dandelions (Taraxacum officinale G.H. Weber ex Wiggers; Asteraceae) dominated these wild areas and had one of the highest diversities of bee visitors. Traditional flower beds with tulips (Tulipa L. sp.; Liliaceae) and petunias (Petunia Juss. sp.; Solanaceae) had relatively poor diversity and abundance of bees throughout the year. Our study suggests that urban areas have the potential to be important pollinator reservoirs, especially if both bloom and habitat heterogeneity are maintained and enhanced through sustainable urban planning. Wiggers (Asteraceae), dominaient ces habitats et présentaient la diversité d'abeilles visiteuses las plus élevée. Les parterres traditionnels, avec fleures telles que les Tulipa L. sp. (Liliaceae) et les Petunia Juss. sp. (Solanaceae), présentaient une diversité d'abeilles relativement pauvre tout au cour de l'an. Notre étude suggère que les zones urbaines ont du potentiel comme important réservoir de pollinisation, surtout quand et l'hétérogénéité d'habitat et hétérogénéité de fleures sont maintenues et accrues à travers le développement urbain durable. 852 THE CANADIAN ENTOMOLOGIST November/December 2004 Volume 136 THE CANADIAN ENTOMOLOGIST 853 FIGURE 1. Map of the study area in metropolitan Vancouver, British Columbia, Canada. Numbers represent study sites, which were grouped into four categories: sites 1-6 are city gardens; sites 7-12 are Naturescape sites; sites 13-20 are wild areas; and sites 21-25 are traditional sites (see Table 1). Shaded grey areas correspond to protected green spaces and watersheds.

No abstract