Abstract. Forty two vegetation samples (0.1 ha nested field plots), taken from lowland riparian forest, located on the right floodplain of Tundzha river (Southeastern Bulgaria), were analyzed. Coarsest picture of vegetation pattern was outlined, using classification and ordination techniques. Detailed analysis of alien and native species distribution was carried out by correlation and regression methods. Described variables were tested for difference with parametric and nonparametric tests. Classification resulted in four plant community types, dominated by different native and exotic trees, shrubs and herbaceous species. Ordination showed that main environmental gradient responsible for the vegetation pattern explanation was habitat moisture. Correlation and regression analyses revealed that most abundant alien and invasive species, such as Acer negundo and Amorpha fruticosa, preferred the most humid part of the moisture gradient, but other alien species had complex response surfaces. Logging is considered as unreliable predictor of alien species distribution at 0.1 ha scale, but as a promoter for xerophylous alien herbs establishment at the 1 m 2 scale on the studied territory. Flooding disturbance certainly has played essential role in this context too. Species diversity demonstrated patterns in accordance with Intermediate Disturbance Hypothesis. We recommend using different sampling scales with greater area covered.

Abstract. Detailed analysis of the relationship between vegetation and environment requires a thorough understanding of the environmental processes that influence vegetation. There is an intimate relationship between vegetation concepts' development, mathematical methods of analysis and knowledge of environmental processes. Traditionally most vegetation/environment studies have been done at a single scale of observation. It has been considered important to assess whether the processes that determine community structure and function are similar at different scales or whether the most important processes change with the scale of observation. This study aims to describe in a quantitative manner the vegetation pattern in a localized area and on that basis to test/verify statistically some previously proposed hypothesis and assumption concerning vegetation organization. Describing the coarsest vegetation pattern, ordination and classification analyses were used. The detailed relationships between environmental gradients and vegetation were examined using regression and correlation methods. Obtained results were tested with different statistical tests for their reliability. We found that the basic environmental gradients influencing the local vegetation are elevation and habitat moisture. The correlation between environmental gradients and vegetation increased with the increasing scale of observation. Environmental gradients influencing vegetation most strongly were the same at the different scales. With the increase of sampled area, correlation strength between environment and vegetation also increased. In the future, if possible, all studies of vegetation/environment relationships should be done at different spatial scales. Gathered quantitative information and verified hypothesis in the current study could be very successfully used in the management and conservation of the local and regional vegetation.

No abstract

Abstract. At the current stage of vegetation ecology development as a science it has not been reached consensus between the competing theories trying to explain vegetation pattern emerging under the influence of underlying environmental gradients. It is still not clear what is the response of different plant species towards the influence of complex gradients? This study tries to revise some classical and temporary hypotheses concerning species response surface/curve shape along environmental gradientswhether it is symmetric (Gaussian) or with other shape. It also attempts to check if species modes along complex gradients are distributed randomly, uniformly or they are clumped, as well as to inspect the species mode distribution among species importance value octaves -whether it is lognormal or lograndom. Field samples were gathered using gradsect method for laying down sampling plots. Obtained data were analyzed in the context of four complex gradients -elevation, habitat dryness, slope inclination and slope convexity. CCA ordination, GAM and LOEES regression and nonparametric correlation were used as analyzing tools. We have found that species response surface/curves do not have symmetric shape but rather they show asymmetric and complex forms. Mode distribution of dominant tree and shrub species was random but that of herbs followed clumped pattern. When species mode distribution was divided into octaves, trees and shrubs showed lograndom distribution but all species together had lognormal one. Herbs alone do not conform to neither lognormal nor lograndom distributional patter. It seems that each species has its own response shape towards the environment, determined by its physiology, interaction with other species and historical events. It is hoped that the current study will add a little drop to the vast ocean of vegetation ecology knowledge helping with the clarification of the basic understanding of the vegetation pattern.

No abstract