| dc.description.abstracteng | Tree species influence soil chemical properties via the input of nutrients and protons through throughfall, stemflow, litterfall, and root respiration and/ or exudation. The effect of conifers versus hardwood trees on soil properties, such as beech (Fagus sylvatica L.), has often been investigated. More recent studies have focused on the influence of different broadleaved tree species on soil properties, and it was found that different broadleaved tree species may significantly influence soil properties, such as the C and N storage. However, most studies compared different mono-species stands or carried out common-garden experiments. Studies in an adult mixed forest are rare. Even fewer studies exist that compared the C and N partitioning during litter decomposition of different species. Species-specific or mixture related differences in the partitioning of C and N are therefore largely unknown. Identifying this gap in knowledge gave rise to the present work, which consists of the following three studies:
(1) The "cluster study" was conducted in a species-rich mixed forest stand of Hainich National Park. The soil type was a luvisol of loess over limestone. In a small area, three trees that were standing in a triangle to each other, so-called “clusters”, were selected. The clusters each consisted of one or two of the following tree species: beech, ash (Fraxinus excelsior L.) and lime (Tilia cordata Mill. or Tilia platyphyllos Scop.). The litterfall, the forest floor and topsoil (0-10 cm and 10-20 cm) were analyzed for their chemical composition.
(2) The "mesocosm study" was conducted in an acidified beech forest of Hainich National Park. The soil type was a luvisol of loess over limestone. The partitioning of C and N from 13C/15N-labelled beech and ash leaf litter was compared in pure and mixed variants. For this purpose, soil columns were transferred into PVC cylinders and returned to the place of sampling. The original fresh litter (L-horizon) was replaced by the respective isotopically labeled litter type or litter mixture to be examined. Via closed chambers, the total and litter derived CO2-respiration was measured biweekly over a period of twelve months. After five and ten months, the mass loss and the C- and N-loss of the original litter and the total and litter derived C and N contents in the O-horizon and mineral soil (0-4 cm) and in the microbial biomass (0-4 cm) were measured.
(3) The "microcosm study" was carried out at constant air temperature and soil moisture in a climatic chamber. Soil-litter mixtures were incubated for 206 days. The partitioning of litter C during decomposition of 13C-labeled leaf or root litter of beech and ash was compared in pure and mixed variants. The CO2-emission was recorded daily. At the beginning, the 13C of CO2 was measured every three days and later on every seven days. Total and litter derived contents of dissolved organic carbon (DOC) were analyzed on days 9, 29 and 206. Total and litter-derived C contents in the microbial biomass as well as the light and heavy density fractions were investigated on day 206.
The key findings from the three studies are presented below:
(1) Cluster study: Tree species influenced the chemical soil properties in the topsoil (0-10 cm) on a small spatial scale, while in 10-20 cm depth the clay content was more important. In 0-10 cm soil depth, the base saturation, the pH and the stock of exchangeable Mg2+ were highest under pure ash (98%, 5.1, 135 137 kg ha-1), and lowest under pure beech clusters (88%, 4.3, 70-76 kg ha-1). The proportion of exchangeable Al3+ to the cation exchange capacity (CEC) was lowest under pure ash (0-0.6%) and highest under pure beech clusters (3.7 7.8%). The soil properties under lime clusters were intermediate. Mixture effects were not detected. An important factor influencing chemical soil properties was the composition of leaf litter. Stocks of exchangeable Mg2+ and Ca2+ in the topsoil correlated positively with the annual inputs of the respective nutrient with the leaf litterfall. Since these were highest in the ash leaf litter, the stocks of exchangeable Mg2+ and Ca2+ in the topsoil also positively correlated with the proportion of ash leaf litter to total leaf litterfall. Ash leaf litter also had a positive effect on soil pH and the stocks of organic C and total N in the mineral soil, which was probably due to more rapid decomposition of ash leaf litter than of beech leaf litter, which in turn led to higher C stocks in the humus layer.
(2) Mesocosm study: Mass loss of ash leaf litter was faster than of beech leaf litter, which is reflected primarily in a more rapid mineralization of the ash leaf litter during the first 5 months (higher litter derived CO2-emissions compared to beech leaf litter). The mass loss of litter was positively correlated with the initial litter Ca-content and negatively with the initial litter lignin-content. The lignin:N ratio was not among the explaining variables, because both litter types contained high concentrations of N which differed only slightly. The mineralization of the ash leaf litter was accelerated in the mixture which contained beech leaf litter. No other mixture effects were detected. Differences in the distribution of litter derived C and N in the soil and the microbial biomass between the variants were not detected. In total, 7-20% of the litter derived C was found in the O-horizon and 1-5% was detected in the first 4 cm of mineral soil. Less than 1% of litter derived C was incorporated into the microbial biomass in the upper 4 cm of mineral soil. The partitioning of litter derived N to the O horizon (9-35%), the upper mineral soil (<8%) and the microbial biomass (<1%) was comparable with the partitioning of litter derived C.
(3) Microcosm study: Similar to the results of the mesocosm study, the mineralization (estimated by the litter derived CO2-emission) was higher for ash leaf litter (34% after 206 days) than beech leaf litter (24%). It was further accelerated when mixed with the latter (39%). Similarly, more C was mineralized from ash roots (29%) than from beech roots (23%). The amount of C mineralized was negatively correlated with the initial lignin:N ratio of the litter, and mineralization was the main path of litter decomposition. The release of DOC was negligible. Further, the DOC concentration was strongly declining with time, suggesting that most of it either mineralized, precipitated or associated to minerals. Four to twelve percent of litter derived C associated to minerals and there was no indication for a litter type or litter mixture effect. The microbial biomass incorporated less beech litter (0.2 0.4%) than ash litter derived C (0.7 1%), and did not differ between roots and leaves.
In summary, tree species can affect soil properties on a small spatial scale. An important control variable is the leaf litter. Thus, the nutrient stocks in the topsoil are linearly related to the return of nutrients via the litter. The differences in the topsoil C storage under beech and under ash could neither be related to different partitioning of leaf or root litter C into the mineral soil, nor to the minerals after 10 months of decomposition. This means that the positive influence of the ash leaf litter compared to the beech leaf litter on the C stocks in the topsoil is a long-term effect. In addition, differences in site properties, such as soil acidity and the composition and abundance of soil fauna, also cause different results. Finally, varying proportions of admixture of ash to beech dominated stands can cause a small-scale diversification of the soil habitat. | de |