Forest soil acidification assessment using principal component analysis and geostatistics |
| |
Authors: | Lubo Bor vka, Lenka Ml dkov , Ví t Pení ek, Ond ej Dr bek,Radim Va t |
| |
Affiliation: | aDepartment of Soil Science and Geology, Czech University of Agriculture in Prague, 165 21 Prague 6-Suchdol, Czech Republic |
| |
Abstract: | Soil acidification and consequent Al release is a problem particularly under forests in mountainous areas of the Czech Republic. It is controlled by a number of factors, like acid deposition, forest type, parent rock, altitude, etc. The Jizera Mountains region presents an area heavily influenced by acidification and forest decline. This paper focused on the effect of stand factors on spatial distribution of soil characteristics of the surface organic (O) and sub-surface (B) horizons from 98 sites using a combination of principal component analysis (PCA) and geostatistics. In the PCA, five principal components (PC) describing more than 70% of total variation were selected. The properties of the O and B horizons (pH, C, N, and S content, potentially dangerous Al forms) were in most cases separated, suggesting different processes and effects in each horizon. Spatial variation of PC scores was analysed using variograms, maps of their distribution were created using kriging. Spatial correlation with stand factors (altitude, slope aspect, forest type and age, soil unit, liming, and grass cover) was analysed using cross-variograms. The surface horizons are more sensitive to external influence (acid deposition, liming, grass expansion) and their spatial variation is stronger. The B horizons are more influenced by forest type (beech vs. spruce) and age, and by soil units (cambic vs. spodic horizons). The effect of stand factors is complex and often indirect. Nevertheless, used combination of pedometrical methods provided concise information about spatial variation and relationships between soil characteristics and the effect of stand factors. |
| |
Keywords: | Soil acidification Forest soils Principal component analysis Spatial distribution Geostatistics |
本文献已被 ScienceDirect 等数据库收录! |
|