ABSTRACT
We explored the semi-natural forests in south western Anatolia along a gradient between Mediterranean and continental climates to determine the site requirements of Juniperus excelsa in Turkey. We hypothesized that environmental variables and indicator species can be used to predict differences in occurrence and cover of J. excelsa and can therefore support decision making in reforestation management planning. Plant species composition and environmental variables were assessed in 153 plots. Association between J. excelsa and other plant species and environmental variables were analyzed using Fisher exact probability tests and stepwise discriminant analysis. High altitude (> 1000 m) as a proxy for an Oromediterranean climate, and high surface stoniness as a proxy for low competition by other tree species, are positive site properties for J. excelsa. The tree species avoids Eumediterranean and Supramediterranean plant communities. Twelve plant species, including the herbs Dianthus zonatus, Ajuga chamaepitys and Paronchia carica and the shrub Cotoneaster nummularia may be used as site indicators for J. excelsa restoration. Platanus orientalis, with similar site requirements but at present negatively associated to J. excelsa due to competitive effects, may be considered an additional indicator if stand conversion (harvesting and replacing P. orientalis) is part of the management plan.
ABSTRACT
The purpose of this study was to investigate the relationships between the plant distribution and the altitude-shape-size characteristics of sinkholes, and the landform characteristics inside sinkholes in the Mediterranean region of Turkey. Block kriging, Factor analysis, Cluster Analysis and Detrended Correspondence Analysis were performed. The sinkhole type and altitudinal zone were found to be the significant factors affecting the plant distribution. However, the sinkhole type was more important than the altitudinal zone. Hence, the sinkholes were first subdivided into groups according to types and then the groups were divided into subgroups according to the altitudinal zones. Consequently, 4 groups were defined; A-type sinkholes [1400-1550 m (A1), 1550-1700 m (A2)] and B-type sinkholes [1400-1550 (B1), 1550-1700 m (B2)]. The B-type was wider vertically and shorter horizontally than Atype sinkholes. Significant differences were found between the plant distribution and slope position inside the sinkholes. Plant distribution in the lower slopes was different from that in the flats and ridges in the B1 sub-type of B-type. Plant distribution in B2 subtype was different among the slope positions (ridge, middle slope, lower slope, and flat). Although distribution of plants is different in different parts (ridges, upper slope, middle slope, lower slope and basal flats) of A sinkhole, the differences between the parts of intermediate slope position are not significant. A high plant variability along short distances in the sinkholes was observed in the study area. That is why the site of sinkholes have a big potential for the distribution of many species. Hence, the area must be separated as strictly protected zone.
ABSTRACT
This study was conducted on Pinus nigra sub sp. pallasiana commonly known as crimean pine. The purpose of this study was to determine the important site factors affecting the development of these forests in Sutculer district. This district experiences a transitional climate between Mediterranean and Continental regions. There is a strong water deficit in summer. The data was collected from 37 sample plots. It was analyzed statistically by using upper stand height as a dependent variable and some site characteristics as independent variables. The results revealed that there was only one important negative linear relationship between upper stand height and Ah organic matter. This result was uncommon. The reasons for this depend indirectly other site factors (aspect, slope position, altitude, slope degree and humus forms). This result was explained at 69.53% of total variance using these site factors by means of factor analysis (FA). After FA, these site factors were transferred to stepwise disciriminant analysis (SDA) to determine eigenvalue ratios of the bonitet models. As a result of SDA, three bonitets and five variables (3B5V) were found as the best model with 71.5 % of variance and 0.007 % significance level.