Analysis of O-F-layer humus mass variation in a mixed stand of European beech and Norway spruce: An application of structural equation modelling

Abstract

A number of different research approaches have been used to investigate the influence of tree species on the particular ecological parameters in mixed forest stands. While it is widely recognized that trees have species-specific effects on particular resources like radiation, water and nutrients more interacting effects of mixtures are assumed to exist. In this paper we examine the spatial variation of humus mass in a mixed stand and discuss the influence of specific spatial factors on humus mass variation. Our study site is located in the Solling area, an intermediate mountain range in the south of Lower Saxony in Germany. It is located in a 130-140-year-old European beech (Fagus sylvatica L.) and Norway spruce (Picea abies (L). Karst.) mixed stand. To better understand processes which determine amount of humus mass variation within mixed stands of European beech and Norway spruce we measured humus mass of the O-F and ecological factors at sample points (litter fall of beech and spruce, radiation, and precipitation beneath the forest canopy). we used a structural equation model (SEM) to analyse data for the effect of species-specific litter input, radiation and throughfall on humus mass variation. Additional to classical regression approaches SEM enables us to consider latent, non-observed, factors or variables. In this paper we present a special SEM, a so-called linear structural relationship (LISREL) model, showing how self-formulated latent variables like "decomposition environment" and "decomposition inhibition" affect the humus mass of the O-F layer. Latent structures are of great importance in order to understand and analyse processes determining humus mass variation. In our model, "latent variables" are identified, which explain the contribution of measured variables to improvement or deterioration of humus decomposition, and hence more or less accumulation of organic material in the surface humus layers, in our case the OF layer. Spruce needle input impedes humus dynamics, while beech leaf litter input and higher throughfall amounts show supporting effects. These model results are in agreement with formerly reported findings, but yield further an interval-scaled combined measure which is better founded and more comprehensive than earlier descriptive approaches to the single ecological factors involved. Besides of the ecological interpretation of the results we succeed in determining a valid structural equation model taking into account all relevant ecological factors of humus decomposition. (C) 2008 Elsevier B.V. All rights reserved.