Geoderma, 2009, Vol.152(3), pp.218-230
Few data are available on the relative importance of different site-specific factors (e.g. topographic parameters, vegetation, soil-specific factors) for soil organic carbon (SOC) distribution in northern forests. Moreover, an increasing frequency of large-scale forest dieback in these forests may change the SOC stock and its spatial distribution. Univariate and multivariate geostatistics were applied to elucidate the spatial variability of SOC stocks of different landscape units under a high-elevation Norway spruce ( L.) forest in central Europe and its relation to topographic parameters, soil-specific factors, and the impact of recent and past stand composition. We studied four high-elevation sites: Three are stocked with unmanaged Norway spruce ( L.) forest, but differ in their positions on hillslope, topographic parameters, and soil type. At an adjacent site the spruce forest had been destroyed by bark-beetle ( ) infestation 25 years ago (dieback site). The fallen logs remained at the site and the site was not reforested. Soil samples were taken from the forest floor and the mineral soil. Fine root biomass and SOC stocks were significantly negatively spatially correlated at the spruce sites, but spatially independent at the dieback site due to the homogeneous distribution of grass roots at this site. For none of the sites with healthy Norway spruce forest, forest floor thickness or the magnitude of the SOC stocks at a given sampling point was related to its distance to the nearest tree. This is attributed to an overwhelming impact of small-scale topography (≤ 5 m) on the SOC stock distribution. In contrast, at the dieback site the SOC stocks in the forest floor and the topsoil horizon showed a significant negative spatial correlation with the distance to the nearest fallen log. The spatial variability of the SOC stocks in the mineral soil increased with increasing stone content and curvature heterogeneity. Clay content and SOC stocks were spatially positively correlated at each site, demonstrating the importance of soil texture on SOC distribution. Our results indicate that topographic parameters dominate the distribution of topsoil and subsoil SOC stocks at stand scale at the investigated sites. In contrast, tree distribution is only of minor importance at these sites with a distinctive patchy topographic structure. Particularly in sloped terrain, fallen logs have the potential to change the spatial distribution of SOC stocks in the forest floor, but also in the mineral topsoil considerably, and within a period of a few decades.
Multivariate Geostatistics ; Topography ; Soil Mapping ; Tree Distribution ; Som Pool ; Agriculture
View record in ScienceDirect (Access to full text may be restricted)