Vulnerability of spruce ( Picea abies ) and beech ( Fagus sylvatica ) forest stands to storms and consequences for silviculture

Schütz, Jean-Philippe ; Götz, Michael ; Schmid, Willi ; Mandallaz, Daniel

In: European Journal of Forest Research, 2006, vol. 125, no. 3, p. 291-302

Ajouter à la liste personnelle
    Summary
    The form and magnitude of storm damage and stand disclosure patterns were assessed in 332 randomly chosen pure and regular stands of spruce (Picea abies L.) and beech (Fagus sylvatica L.) after storm lothar, within a region of the Swiss Midlands. This data was analysed in relation to maximal wind speed, measured with Doppler radar techniques and other influential factors such as relief, allometric characteristics, silvicultural history, and neighbourhood. In addition, storm damage, assessed from aerial photographs over an extended perimeter (about 70,000ha) was considered. A storm of the magnitude of lothar (December 26 1999), with an average maximal wind speed of 45ms−1 (160kmh−1) appears to have a highly chaotic wind field structure, with great spatial and temporal variation of wind gusts. Wind speeds were not a significant predictor for damage in spruce stands and only weakly influential for beech. The consequences of this high randomness were analysed to estimate the return time of such a storm at the stand level. It lies between 86 and 113years for spruce, 357 and 408 for beech. Only a few independent variables were significant and the overall explanatory strength of the model was unexpectedly low (R 2=0.07 for spruce and 0.30 for beech). Among the more reliable predisposing factors were mixture and aspect combined with gradient. An admixture of 10% or more broadleaved tree species or wind-firm conifers like Douglas fir [Pseudotsuga menziesii (Mirb.) Franco] significantly reduced the vulnerability of spruce stands (by a factor of more than three). On wind-exposed aspects, damage was more than twice the average. Steeper slopes caused a significant reduction in susceptibility (by a factor of six for slopes over 50%, in comparison to gentle slopes <20%). Other factors such as height to diameter ratio of trees or time since last thinning did not appear to be significant predictors