Forests are critical habitats for biodiversity and they are also essential for the provision of a wide range of ecosystem services that are important to human well-being. There is increasing evidence that biodiversity contributes to forest ecosystem functioning and the provision of ecosystem services. Here we provide a review of forest ecosystem services including biomass production, habitat provisioning services, pollination, seed dispersal, resistance to wind storms, fire regulation and mitigation, pest regulation of native and invading insects, carbon sequestration, and cultural ecosystem services, in relation to forest type, structure and diversity. We also consider relationships between forest biodiversity and multifunctionality, and trade-offs among ecosystem services. We compare the concepts of ecosystem processes, functions and services to clarify their definitions. Our review of published studies indicates a lack of empirical studies that establish quantitative and causal relationships between forest biodiversity and many important ecosystem services. The literature is highly skewed; studies on provisioning of nutrition and energy, and on cultural services, delivered by mixed-species forests are under-represented. Planted forests offer ample opportunity for optimising their composition and diversity because replanting after harvesting is a recurring process. Planting mixed-species forests should be given more consideration as they are likely to provide a wider range of ecosystem services within the forest and for adjacent land uses. This review also serves as the introduction to this special issue of Biodiversity and Conservation on various aspects of forest biodiversity and ecosystem services.

• This article synthesizes and reviews the available information on the effects of forestry practices on the occurrence of biotic and abiotic hazards, as well as on stand susceptibility to these...

Recent studies projecting future climate change impacts on forests mainly consider either the effects of climate change on productivity or on disturbances. However, productivity and disturbances are intrinsically linked because 1) disturbances directly affect forest productivity (e.g. via a reduction in leaf area, growing stock or resource-use efficiency), and 2) disturbance susceptibility is often coupled to a certain development phase of the forest with productivity determining the time a forest is in this specific phase of susceptibility. The objective of this paper is to provide an overview of forest productivity changes in different forest regions in Europe under climate change, and partition these changes into effects induced by climate change alone and by climate change and disturbances. We present projections of climate change impacts on forest productivity from state-of-the-art forest models that dynamically simulate forest productivity and the effects of the main European disturbance agents (fire, storm, insects), driven by the same climate scenario in seven forest case studies along a large climatic gradient throughout Europe. Our study shows that, in most cases, including disturbances in the simulations exaggerate ongoing productivity declines or cancel out productivity gains in response to climate change. In fewer cases, disturbances also increase productivity or buffer climate-change induced productivity losses, e.g. because low severity fires can alleviate resource competition and increase fertilization. Even though our results cannot simply be extrapolated to other types of forests and disturbances, we argue that it is necessary to interpret climate change-induced productivity and disturbance changes jointly to capture the full range of climate change impacts on forests and to plan adaptation measures.

The study presents a model for the probability of fire occurrence in forest stands of Catalonia. It was developed based on data from the Second Spanish National Forest Inventory (II IFN) and perimeters of the forest fires that occurred in Catalonia during a 12-year period that followed the II IFN measurement. The model was developed for forest planning, which means that the probability of fire was predicted from variables that are known in forest planning calculations. The model is suitable for both even-aged and uneven-aged stands, as well as pure and mixed stands. The variables included in the model represented easily measurable forest characteristics like elevation, tree size, stand structure and species composition. According to the model, the probability of a forest stand to become affected by forest fire increased with: lower altitudes, smaller diameters, larger basal areas, higher proportion of coniferous species and increasing variation in tree diameter. Although fire occurrence is a highly stochastic phenomenon, the model and all its parameters were significant, and the test results were consistent. The presented model can be used in stand and forest level optimizations that consider the effect of fire risk on optimal management. fire / model / forest stands / Catalonia Résumé-Modèle de probabilité de feux pour des peuplements forestiers de Catalogne (Nord-Est de l'Espagne). Cette étude présente un modèle pour les probabilités d'occurrence des feux dans des peuplements forestiers en Catalogne. Il a été développé sur la base de données du deuxième Inventaire Forestier National Espagnol (II IFN) et des périmètres des feux de forêts qui se sont produits en Catalogne pendant une période de 12 ans qui a suivi les mesures du II IFN. Le modèle a été développé pour la planification forestière ce qui veut dire que la probabilité de feux était prédite à partir de variables qui étaient connues dans les calculs de planification forestière. Le modèle est approprié pour les peuplements équiennes et inéquiennes, de même que pour les peuplements purs ou les peuplements mélangés. Les variables incluses dans le modèle sont des caractéristiques facilement mesurables comme l'altitude, la dimension des arbres, la structure des peuplements et les espèces constituant le peuplement. D'après le modèle, la probabilité qu'un peuplement forestier soit affecté par un feu augmente avec : une altitude basse, des petits diamètres, des surfaces terrières importantes, une forte proportion de conifères et une augmentation de la variation du diamètre des arbres. Bien que l'occurrence d'un feu soit un phénomène hautement stochastique, le modèle et tous ses paramètres étaient significatifs, et les résultats des tests étaient cohérents. Le présent modèle peut être utilisé à pour des optimisations, l'échelle du peuplement ou de la forêt, qui prennent en compte le risque de feu pour une gestion optimale. feu / modélisation / peuplements forestiers / Catalogne