Assessing European forest biodiversity: a multi-dimensional approach

Assessing forest diversity is still a critical task at local, national and global level. Monitor biodiversity nationwide is re- quired for countries participating to the Convention of Biological Diversity, however, measuring forest diversity is a challenge due to the complexity and the functionality of forest and to the difficult of harmonizing sampling design at continental scale. The present research is aimed to analyze a large forest di- versity data to assess forest diversity at the pan-continental level, to investigate how the tree layer diversity can be used as a proxy of the community diversity at different scale levels, and to demonstrate the importance of forest monitoring for forest management. We used a multi-dimensional pan-European data acquired in the framework of BioSoil project (Li-BioDiv database), as part of ICP forest Level I network. The database contains forest struc- ture and vegetation data collected in 19 European countries in the period 2005-2008. In addition, for one of the purpose we used also local-scale diversity data as well. The dissertation is divided into two main parts. The first one includes paper I and II that deepen data stored in the LI-BioDiv database, while the second one in- cludes paper III and IV investigating the relationship between tree and ground vegetation data. The paper I presents the LI-BioDiv database and com- pares plot level forest variable with data of National Forest Inven- tories and demonstrates that the database appears useful mainly for research purpose aimed at studying cross-relationships between multiple forest variables. Paper II is aimed of evaluating the uncertainty of ground vegetation data related to forest diversity descriptors using a diffusion-based cartogram approach that visually displays how data information change in function to different uncertainty degrees. The study highlights than an awareness of the negative relation- ship between the period of the survey and species richness can lead to a better data handling and analysis. In addition it also demon- strate that cartograms are efficient tools for evaluating and man- aging uncertainty and can strengthen the results of data analysis by providing alternative perspectives and interpretations of spatial phenomena. In Paper III, we explore the relationship between struc- tural diversity indicators and floristic diversity indices, across biore- gions, in different management conditions and in different forest types at pan-European scale. As expected, no relationship was found between the two group of indices. The results confirm that structural diversity indices are complementary to floristic biodiver- sity indices. A selection of groups of different types of indices is highly encouraged to better explain forest biodiversity. In Paper IV stand structure and plant diversity changes were tested at local scale. The effects of alternative managements were investigated by comparing the structure and plant diversity along a four-stages temporal gradient of stand development under similar site conditions. The study underlines species turnover along a temporal gradient in order to promote conservation-oriented man- agement. The results underline that the frequency of silvicultural interventions seems to influence the turnover in species composition and structural condition variation. In conclusion, understanding heterogeneity related to bi- ological data is a complex task that can lead to good data use and analysis. The relationship between tree layer and ground veg- etation layer change across scales and local conditions. National Forest inventories should be taken into consideration as a platform to include multiple forest variable in order to properly monitoring and assessing forest biodiversity status and trends, and to promote a targeted management.