We compared results derived using three different approaches to assess the suitability of common tree species on the Franconian Plateau in southern Germany under projected warmer and drier climate conditions in the period 2061-2080. The study area is currently a relatively warm and dry region of Germany. We calculated species distribution models (SDMs) using information on species’ climate envelopes to predict regional species spectra under 63 different climate change scenarios. We complemented this with fine-scale ecological niche analysis using data from 51 vegetation surveys in seven forest reserves in the study area, and tree-ring analysis (TRA) from local populations of five tree species to quantify their sensitivity to climatic extreme years. The SDMs showed that predicted future climate change in the region remains within the climate envelope of certain species (e.g. Quercus petraea), whilst for e.g. Fagus sylvatica, future climate conditions in one third of the scenarios are too warm and dry. This was confirmed by the TRA: sensitivity to drought periods is lower for Q. petraea than for F. sylvatica. The niche analysis shows that the local ecological niches of Quercus robur and Fraxinus excelsior are mainly characterized by soils providing favorable water supply than by climate, and Pinus sylvestris (planted) is strongly influenced by light availability. The best adapted species for a warmer and potentially drier climate in the study region are Acer campestre, Sorbus torminalis, S. aria, Ulmus minor, and Tilia platyphyllos, which should therefore play a more prominent role in future climate-resilient mixed forest ecosystems.

Grid-based inventories of 1,924 deciduous forests plots in Germany and 4,775 in Romania were used to investigate tree species composition as affected by browsing and grazing under different forest management (rotation forestry, selectively cut forest, protected forest). At regional scale, the loss of tree species in the dominant layer was between 52 to 67% in Germany and of 10 to 30% in Romania, with largest effects in protected nature reserves in Germany. At plot level, only 50% (Germany) to 54% (Romania) of canopy species were found in the regeneration layer with a height of 1.5 m. Browsing was influenced by the proportion of Fagus in the regenerating trees in Germany, and by stand density, basal area, and management in both regions. Structural equation modeling explained 11 to 26% of the variance in species loss based on the fresh loss of the terminal bud in the winter prior to the inventory work (one season browsing). Browsing (and grazing in Romania) is shown to be a significant cause of species loss across both countries and all management types. Potential cascading effects on other organisms of deciduous forest ecosystems are discussed. We conclude that the present hunting practices that support overabundant ungulate populations constitute a major threat to the biodiversity of deciduous forests in Germany and Romania and to other places with similar ungulate management, and that changes my only be possible by modernizing the legal framework of hunting.

The implementation of standardised methods for the monitoring of NATURA 2000 sites in Europe is still a key topic in environmental research. Effective, economically priced and, as far as possible, automated applications are required. Rapidly developing sensor technology together with advanced image processing methods offer new possibilities for application of remote sensing data to NATURA 2000 monitoring. The studies presented here combine commonly available GIS data, such as Biotope Type Maps or Forestry Site Maps with remote sensing classifications of the very high spatial resolution (VHSR) QuickBird sensor. Two knowledge-based approaches under inclusion of a priori object-based information are utilised to detect the extent of habitats as well as their quality according to the German NATURA 2000 mapping guidelines. While one method used a segmentation of forested sites in Bavaria (southern Germany), the second technique applied available objects to classify heathland habitats in the Brandenburg Region (northern Germany). The results were subsequently compared, in close cooperation with local environmental authorities, with habitats mapped terrestrially for NATURA 2000 management plans. These results indicate that different remote sensing methods can be a valuable support for terrestrial mapping. Woodland habitats can be detected and specific NATURA 2000 quality parameters (e.g., percentage of natural forest types) are recognisable. In the case of heath-dominated sites, terrestrial mapping can even be replaced by remote sensing of certain habitat types for which it is also possible to obtain adequate measures of quality. Having evaluated the quality of forest and heathland NATURA 2000 habitats, two general challenges when implementing the guideline regionally could be indicated. Firstly, the very general scope of the Habitats Directive contradicts to specific local protection purposes. Secondly, the protection aims given for NATURA 2000 areas are very static. The Directive could be improved by adapting existing management and conservation strategies to pro-actively respond on likely anthropogenic influences

Introduction: Forests are a crucial part of the global carbon cycle and their proper management is of high relevance for mitigating climate change. There is an urgent need to compile for each region reference data on the carbon (C) stock density and C sequestration rate of its principal forest types to support evidence-based conservation and management decisions in terms of climate change mitigation and adaptation. In the Andean Mountains of northern Patagonia, extensive areas of temperate forest have developed after massive anthropogenic fires since the beginning of the last century.Methods: We used a plot design along belt transects to determine reference values of carbon storage and annual C sequestration in total live (above- and belowground biomass) and deadwood mass, as well as in the soil organic layer and mineral soil (to 20 cm depth) in different forest types dominated by Nothofagus spp. and Austrocedrus chilensis.Results: Average total carbon stock densities and C sequestration rates range from a minimum of 187 Mg.ha⁻¹ and 0.7 Mg.ha⁻¹.year⁻¹ in pure and mixed N. antarctica shrublands through pure and mixed A. chilensis forests taller than 7 m and pure N. pumilio forests to a maximum in pure N. dombeyi forests with 339 Mg.ha⁻¹ and 2.2 Mg.ha⁻¹.year⁻¹, respectively. Deadwood C represents between 20 and 33% of total wood mass C and is related to the amount of live biomass, especially for the coarse woody debris component. The topsoil contains between 33 and 57% of the total estimated ecosystem carbon in the tall forests and more than 65% in the shrublands, equaling C stocks of around 100–130 Mg.ha⁻¹ in the different forest types.Conclusion: We conclude that the northern Patagonian temperate forests actually store fairly high carbon stocks, which must be interpreted in relation to their natural post-fire development and relatively low management intensity. However, the current high stand densities of these forests may well affect their future carbon storage capacity in a warming climate, and they represent a growing threat of high-intensity fires with the risk of a further extension of burned areas in the future.