Effects of different thinning intensities on drought response in Norway spruce (Picea abies (L.) Karst.)

A thinning experiment has been conducted since 1971 on two 22-year-old stands of Norway spruce in the Belgian Ardenne. The stands have strongly contrasting water availability. In 1995, tree ring-width series were measured from increment cores sampled in control plots along with three different thinning intensity treatments. Relative growth indices were calculated by standardisation which involved fitting a curve to the measured series, and then dividing each measured value by the corresponding curve value. Based on reconstruction of the monthly values of soil water content (SWC), climatic response functions were calibrated on growth indices from the control treatment. In the humid stand, radial growth is mostly under lagged climatic control. In the dry stand, radial growth is instantaneously related to water supply during the early part of the vegetation period. The response functions were used to remove the climatic signal present in the growth indices calculated for the thinned treatments. This procedure does not allow the entire climatic signal in the growth indices of the dry stand to be removed. We show that reducing the stand density alters the classical relationship between climate and radial growth. At least in the short term, thinning improves the resistance of tree radial growth to drought stresses. This effect is more obvious on the humid stand than on the dry stand. We show also that observations and climate-based simulation of growth indices in the heavily thinned treatments have an inverse cycling behaviour. In the humid stand, trees from the heavily thinned plots showed growth reduction during the eighties, while simulation growth indices from the climatic response function were quite stable. While tree growth from the lightly thinned treatment remains coupled to climate because of high inter-tree competition for water supply, tree growth from the heavily thinned plots could be sensitive to other limiting factors that may involve atmospheric pollution. (C) 2003 Elsevier Science B.V. All rights reserved.