Search

The overall goal of this study was to describe a novel area-based semiempirical model for estimating growing stock from small-footprint light detection and ranging (lidar) data. The model assumes a linear relationship between growing stock and lidar-derived canopy volume that is stratified according to several canopy height classes to account for height dependent differences in canopy structure and nonlinear tree size-shape relationships. It was applied over a 128 [km.sup.2] alpine area in Austria where operational forest inventory data and lidar data acquired in winter and summer were available. The analysis showed that the semiempirical model was quite robust against changes in laser point density and acquisition time. Further, it was found that the model performed as well as a widely used iterative regression method based on a multiplicative model. Both models reached a high coefficient of determination ([R.sup.2] = 0.76-0.86) and a standard deviation of the residuals in the order of 20.4%-29.1%. Although it is less flexible than the multiplicative model, the advantages of the semiempirical model are its simplicity and the fact that its coefficients can be physically interpreted. These traits can be expected to enhance the applicability of the model in regions where high-quality inventory data are lacking. L'objectif global de cette etude consistait a decrire un nouveau modele semi-empirique base sur la superficie pour estimer le materiel ligneux sur pied a partir des donnees du Lidar a petite empreinte. Le modele suppose que la relation entre le materiel ligneux sur pied et le volume du couvert forestier derive du Lidar est lineaire. Le volume du couvert forestier est stratifie en plusieurs classes de hauteur pour tenir compte des differences dans la structure du couvert forestier qui sont fonction de la hauteur et des relations non lineaires entre la taille et la forme des arbres. Le modele a ete applique dans une zone alpine de 128 [km.sup.2] en Autriche pour laquelle des donnees d'inventaire d'exploitation et des donnees du Lidar acquises pendant l'hiver et l'ete etaient disponibles. L'analyse montre que le modele semi-empirique est tres robuste vis-a-vis des changements de densite des points laser et du moment d'acquisition. En outre, le modele performe aussi bien qu'une methode iterative de regression largement utilisee et basee sur un modele multiplicatif. Les deux modeles presentent un coefficient de determination eleve ([R.sup.2] = 0,76-0,86) et un ecart-type des residus de l'ordre de 20,4 a 29,1 %. Bien qu'il soit moins souple que le modele multiplicatif, les avantages du modele semi-empirique resident dans sa simplicite et le fait que ses coefficients peuvent etre interpretes physiquement. Ces caracteristiques devraient favoriser l'applicabilite du modele dans les regions ou il n'y a pas de donnees d'inventaire de bonne qualite. [Traduit par la Redaction], Introduction Light detection and ranging (lidar) is a remote sensing technology that was introduced in the 1990s as a tool for topographic mapping (Flood 2001). It uses pulsed lasers to [...]

Using 30 years of climate records from 20 weather stations, we investigate the magnitude of temperature and precipitation change, and change in the length of the growing season between 1961 and 1990. Special attention is paid to the period between 1981 and 1990, because recent research suggests that, during this time span, forest productivitymay have increased in the northern latitudes. In order to understandthe importance of changes in climate on forest growth, we use the ecosystem model FOREST BGC as a diagnostic tool to predict the annual net primary production (NPP). The results of our study indicate: no change in precipitation between 1961 and 1990; a significant (alpha =0.05) increase in mean annual temperature of 0.72 deg. C, mean annual minimum temperature (0.80 deg. C), winter temperature (2.36 deg. C) aswell as an increase in the length of the temperature-controlled growing season by 11 days, resulting in a significant increase in diameter increment obtained from 1179 cores of Norway spruce across Austria.The trends in NPP are consistent with observed increment rates validating the use of biogeochemical modeling as a diagnostic tool to search for possible causes on changing environmental conditions. [ABSTRACT FROM AUTHOR]