A major debate in the study of biodiversity concerns its influence on ecosystem functioning. We compared whether wood production in forests was associated with tree functional group identity (i.e. deciduous, conifer or sclerophylous), tree species richness (1-> or = 5) and tree functional group richness (1-3) by comparing more than 5000 permanent plots distributed across Catalonia (NE Spain). Deciduous forests were more productive than coniferous and sclerophylous forests. Wood production increased with tree species richness. However, functional group richness increased wood production only in sclerophylous forests. When other forest structure, environmental variables and management practices were included in the analysis, tree functional group identity and species richness still remained significant, while functional species richness did not. Our survey indicates that across a regional scale, and across a broad range of environmental conditions, a significant positive association exists between local tree species richness and wood production at least in typical early successional Mediterranean-type forests.

The Bitterlich relascope is a multiple use dendrometer widely used in forest inventory. Although it is most commonly used to estimate basal area, the relascope can also estimate other stand variables, including density and diameter distribution. However, forest stand inventories in Spain rarely use relascope plots to estimate these variables due to the belief that they lead to higher errors than fixed-radius plots due to the heterogeneity of many Mediterranean forests. This study compared the accuracy of the estimated averages of three main stand variables (basal area, stand density, and diameter class distribution) in forest stand inventories performed with relascope plots and with conventional fixed-radius circular plots, both measuring a similar number of trees (15)(16)(17)(18)(19)(20). A forest stand inventory simulator (DOMO) was used (1) to generate simulated forest stands corresponding to the nine most common types in the Mediterranean region of Catalonia (NE Spain), including even-aged and uneven-aged stands, and (2) to estimate and compare the average values of these variables at the forest stand level resulting from both plot types. In general, we did not find significant accuracy differences between the inventory systems for most of the stand variables and forest types studied, as expected by established angle-count sampling theory. However, the results show that for stands with multiple strata and open structures, the Bitterlich relascope provides a more accurate estimate for basal area than for density, while the reverse occurs for fixed-radius plots.