Search

Abstract. A study of the forest lines, tree lines and the structures of the sub-alpine forest was performed in Vallone Vallanta and in Alevé forest in the Varaita Valley (Cottian Alps, Piedmont, Italy). Forest- and tree lines were analysed over 1728 ha while forest structures were studied on six 3000-m 2 plots located at the tree line (2), at the forest line (2) and inside the sub-alpine forest (2). Dendro-ecological analysis of living plants and stumps showed that Larix decidua was more abundant in the past than today and that Pinus cembra has expanded, both upwards and within sub-alpine forests. Age structure analysis revealed that the current sub-alpine forest stands were established 200 - 220 yr ago, probably following a clearcut. At the forest lines the tree density decreases, and some trees are more than 500 yr old, whereas at the tree lines most of the trees (almost exclusively Pinus cembra) are younger than 100 yr. Growth dynamics were investigated both by observing Basal Area Increment (BAI) in the old and dominant trees, and by comparing the BAIs of classes of trees with a given cambial age range in different time periods. The results showed that the growth rates of mature Pinus cembra and Larix decidua had increased. These increments are more substantial for Pinus than for Larix. The growth rate of young trees (100 yr) of both species has decreased over recent decades. This could be due to competition caused by increased tree densities that have resulted from a decrease in grazing. [ABSTRACT FROM AUTHOR]

Interest in coppices is growing due to the need to replace fossil fuels with renewable energy sources. In Italy, beech covers one million hectares, half of which originated by coppicing. This study tested which factors drive the presence and growth of beech resprouts, with a focus on fertility, cutting intensity, age, and size at time of coppicing. We analyzed 509 stools in 24 stands coppiced between 1 and 26 years before sampling. We fitted Generalized Linear Mixed Models of the probability of sprouting and height of the tallest resprout for each stool as a function of elevation, slope, aspect, bedrock, precipitation, temperature, age at coppicing, time since coppicing, residual shoot density, the sum, average and coefficient of variation of the diameter of cut shoots, and type of stool treatment. Of all harvested stools, 249 (49%) had sprouted with an average of 7.6 resprouts per stool. Height of the tallest resprout on each stool ranged from 3 to 800 cm, mainly as a function of time since coppicing. Resprout mortality was on average 1.4% per plot. Sprouting decreased with decreasing site fertility, increasing precipitation, and increasing size of cut stems. Leaving one or more shoots on the stool after felling produced a high proportion of sprouting stools (82%). Although based on a limited sample, our quantitative analysis of the driving factors of sprouting in beech can be used to support silvicultural decisions in over-mature beech coppices, and to optimize trade-offs between ecosystem services such as biomass production, biodiversity, and hydro-geologic protection. [ABSTRACT FROM AUTHOR]

Aerial LiDAR (Light Detection and Ranging) derived data are widely adopted for the study and characterization of forests. In particular, LiDAR derived-CHM (Canopy Height Model) has proved essential in identifying tree height variability and estimating many forest features such as biomass and wood volume. However, CHM quality may be affected by internal limits and anomalies caused by raw data (point cloud) processing (i.e., vertical errors), which are quite often disregarded by users, thus generating potentially erroneous results in their applications. In this work, an auto-consistent procedure for the fast evaluation of CHM accuracy has been developed based on the assessment of internal anomalies affecting CHM data obtained by differencing gridded DSM (Digital Surface Model) and DTM (Digital Terrain Model). To this purpose, a CHM was generated using the gridded DTMs and DSMs provided by the Cartographic Office of the Piemonte Region (north-western Italy). We estimated the local potential CHM error over the whole region, and demonstrated its strictly dependence on the terrain morphometry, particularly slope. The relationship between potential CHM error and slope was modeled separately for mountain, hill and flat terrain contexts, and used to produce a potential error map over the whole region. Our results showed that approximately 20% of the regional territory suffers from CHM uncertainty (in particular high elevation areas, including the treeline), though the majority of regional forest categories was affected by negligible CHM error. The potential consequences of CHM error in forest applications were evaluated, concluding that the tested LiDAR dataset provide a reliable basis for forest applications in most of the regional territory. [ABSTRACT FROM AUTHOR]