RENECOFOR - Facteurs explicatifs de la variabilité et évolution temporelle de la nutrition foliaire dans les réseaux de suivi à long terme des écosystemès forestiers de France (RENECOFOR), de Belgique (Wallonie) et du Grand-Duché de Luxembourg (period 1993-2005) : Période 1993-2005

Fontainebleau

2008

67 S.

19 Abb., 14 Tab., 43 Lit. Ang.

Bandaufführung

978-2-84207-324-4

153051

Due to the growing public awareness of possible adverse effects of air pollution on forests, an intensive monitoring program of forest ecosystems (level II) was launched in the early 90ties in most European countries and was coordinated by ICP Forests. Fifteen years later, this monitoring program provides a unique data set which enables us to analyse the emporal Variation in the nutritional Status of numerous species growing on very diverse sites (soil, climate). This study focuses on foliar concentrations from 1993 to 2005 on French (RENECOFOR), Walloon and Luxemburger plots. The aim is to reveal long term trends while taking interannual variability into account. In a second part, we propose hypotheses to explain the observed variations over time. The data set available for this study includes different monitoring periods and sampling years according to country. Concerning France, foliar analyses were done in 1993, 1994, 1995, 1996, 1997, 1999, 2001 and 2005; in Wallonia, similar foliar analyses were carried out in 1996, 1997, 1999, 2001, 2003 and 2005 while the two plots in Luxemburg were sampled in 1994, 1995, 1996, 1997, 1998, 1999, 2001, 2003 and 2005. The sampling method, the sample treatments and the chemical analyses were comparable in each country. For the Statistical analyses, the following variables were taken into account: dry weight of 100 leavesor 1000 needles, macro-nutrient concentrations (N, S, P, Ca, Mg, KL), micro-nutrient concentrations (Na, Cl, Mn, Fe, AI, Cu, Zn), and nutrient ratios (S/N, N/P, N/Ca, N/Mg, N/K, K/Ca, K/Mg, Ca/Mg, Ca/Al). In order to explain the temporal variations in the foliar nutrient status, other variables measured on the same plots were also used (chemical and physical soil properties, atmospheric deposition, litterfall, and meteorological data). Two kinds of Statistical treatments were used to analyse the data: principal component analysis (PCAs) and linear mixed models. PCAs reveal the main factors explaining foliar concentration variability as well as inter-annual variations. Linear mixed models were used lo evaluate the temporal trends (linear fixed effect) while taking into account the inter-plot and inter-annual variability (random factors). These statistical analyses show a generalised decrease in P foliar concentrations on the broadleaved plots and on the Scots pine and maritime pine plots located on acid soils. Generally, concentrations are rather decreasing for the macro-nutrients, except for K whose foliar concentration is increasing with time on the coniferous plots. Concerning micro-nutrients, the trends are less clear; however, we found decreasing concentrations of Fe, AI, Zn, Mn and Cu under common oak located on acid soils. The inter-annual variability in the foliar concentrations seems to be associated with climatic conditions, concentrations being higher during wet years than during dry ones. Various hypotheses may explain these results; however, additional information is needed to verify them. We suggest using a mechanistic modelling approach to do this.