A MODEL DESCRIBING NATURAL REGENERATION RECRUITMENT OF NORWAY SPRUCE (PICEA-ABIES (L.) KARST.) IN AUSTRIA

In the absence of reliable and representative data on the frequency of seed years, seed amounts, germination and survival of seeds, the data of the observation period 1992-1996 of the permanent national forest inventory of Austria are used to develop a model describing the probab ility for the occurrence, density and height of Norway spruce (Picea a bies (L.) Karst) natural regeneration in the forests of Austria. The f orest inventory provided data on natural regeneration (saplings betwee n LO and 130 cm height) on clusters within a 3.89 km square sampling g rid. A logistic equation is used to predict the probability for natura l regeneration occurrence. Input parameters for this equation are the variables that describe the site, with slope and azimuth as continuous variables, and growth districts and vegetation types as discrete vari ables; the crown competition factor describes density and the quadrati c mean diameter describes the stage of development of the stand. The s ame equation type is used to predict the probability for the occurrenc e of Norway spruce, conditional on the occurrence of some regeneration . An additional variable in this species specific model is a dummy var iable which is set to 1 if Norway spruce occurs in the overstory and o therwise zero. Additional site variables entering this model are eleva tion and vegetation types characterizing soil fertility and moisture. Because the density and height of spruce regeneration depends on the s tand's susceptibility to browsing, the probability for browsing is als o modelled as a logistic equation depending on elevation, vegetation t ype, and stand density. Finally the probability distribution for heigh t and density of the regeneration is described by two bivariate Weibul l-distributions, each one describing browsed and unbrowsed Norway spru ce regeneration respectively. (C) 1997 Elsevier Science B.V.