One of the greatest concerns in community ecology is to find how species composition patterns are related to different environmental and spatial conditions. This approach is especially interesting when applied to high diversity heterogeneous forests such as the Brazilian Atlantic Rainforest sensu lato. The present study aims to verify the existence of indicator species in four southern Atlantic Rainforest formations and identify relationships among distributions of tree species with environmental and spatial variables. For that, tree species density data of 21 phytosociological surveys were collected from the literature. The data were analyzed using indicator species and partial canonical redundancy analysis (partial RDA). Sandy coastal formation contained the greatest number of indicator species (17), followed by Atlantic rainforest (10), cloud forest (4) and Araucaria forest (3). The partial RDA analysis explained 22% of total data variation, of which 11% was assigned to the environment, 5% to space, 6% to spatial component of environmental influence, and 78% remained undetermined. The forest formations present different sets of indicator species suggesting replacement of species along the forest formations. The largest and significant fraction of variation in the composition and abundance of tree species explained by environmental variables reflects the heterogeneity and complexity of habitats throughout the region of Atlantic Forest. The low spatial influence and the environmental results indicate a pattern of structured communities due to different requirements of niches by species (niche theory).
Authors:K. Mikulić, A. Radović, V. Kati, S. Jelaska, and N. Tepić
Land abandonment is a widespread phenomenon in agricultural systems, especially in former communist countries of Eastern and South-eastern Europe. Moreover, Croatia was affected by acts of war which enhanced the depopulation of marginal areas impelling further land abandonment. Agricultural landscapes in Croatia are highly parcelled with various proportions of forest habitats due to traditional smallholder farming systems. Secondary successions as a consequence of land abandonment affect farmland birds that are among the most endangered bird species in Europe. We examined bird communities along a habitat gradient in heterogeneous agricultural landscapes. We used the share of woody vegetation cover as a proxy measure for land abandonment that we classified in four classes. Our results showed no significant Shannon Wiener Index differences of bird communities along the land abandonment gradient. However, there were differences in abundances when we examined bird guilds such as farmland, forest and “other” birds separately. However, the conservation value of each of the four land abandonment classes did not show significant differences. We extracted single bird species such as the Yellowhammer (Emberiza citrinella), Red-backed Shrike (Lanius collurio), Song Thrush (Turdus philomelos) and European Robin (Erithacus rubecula) as potential indicator species for the four examined land abandonment levels. With these four species we successfully modelled the distribution of the recorded bird assemblages at the plot level along the four vegetation succession stages. We emphasized the need to develop new and integrative land use management concepts for areas affected by land abandonment in order to formulate sound conservation policy.
Ground beetles were studied among grassland — forest edge — beech forest habitats in Hungary, using pitfall traps. We hypothesised that the activity density and species richness of carabids were the highest in the forest edge, the activity density and the number of forest species were decreasing, while the activity density and the number of generalist species increasing from the forest towards the grassland. Carabid assemblage of the grassland was the most diverse and the forest was the least diverse if measured by Rényi diversity. The average species richness per trap was significantly higher in the grassland and in the edge than in the forest. The number of forest species was significantly higher in the edge than elsewhere. The number of generalist species was highest in the grassland and decreased towards the forest. The activity density of carabids was significantly higher in the forest and in the forest edge than in the grassland. The activity density of forest species was higher in the forest and in the edge than in the grassland. The activity density of the generalist species was higher in the grassland than in the forest edge and in the forest. There were seven species characteristic to the grassland as identified by IndVal; two species were characteristic to the edge, and two species were characteristic both to the forest and the edge. We found that humidity was the highest in the forest; Pterostichus oblongopunctatus and Molops piceus were associated with the forest habitat, while Abax ater and Pterostichus melanarius were associated with the forest edge according to the RDA. The the shrub cover was the most relevant factor in the edge; Abax ater and Pterostichus melanarius were associated with this habitat.
The paper analyzed the changes of beetle assemblages in the litter layer of eutrophic pine forests in the zone polluted by a nitrogen fertilizer plant Joint Stock Company “Achema” (Lithuania). We hypothesized that abundance, diversity and life traits of beetle assemblages depend on the distance from the pollution source. The samples of the litter layer were taken from pine stands at the distances of 3, 5, 10 and 20 km from the plant. The PCA and GLM analyses were used to reduce the number of variables to the main environmental gradient and assess the influence of environmental factors on beetle abundance, number of species, and life traits. The dependence of species number, abundance and the presence of forest and dendrophagous species on the distance from the plant was detected. A significant impact of organic carbon content, nitrogen emission and moss cover on other life traits of beetles was disclosed. The abundance of moss fraction in the litter layer was positively correlated with increasing distance from the plant. The ability to tolerate polluted sites by three species: Atheta fungi, Micrambe abietis and Brassicogethes aeneus, and intolerance of pollution by eight species: Bryaxis puncticollis, Quedius limbatus, Cyphon pubescens, Cephenium majus, Cyphon padi, Cyphon variabilis, Gabrius appendiculatus and Philonthus cognatus, were detected by IndVal analysis. The distribution of litter species was affected by the distance from the plant and by the richness of moss cover.
Authors:Z. Kemencei, R. Farkas, B. Páll-Gergely, F. Vilisics, A. Nagy, E. Hornung, and P. Sólymos
We determined microhabitat associations for 39 land snail species based on multimodel inference and generalized linear mixed models using a comprehensive and micro-scale data set from the Aggtelek Karst Area, Hungary. Patterns of microhabitat associations were highly nested among microhabitat types (litter, live trees, dead wood, rock) with high number of specialist species in dead wood and in rock microhabitats. Species composition was highly predictable in these microhabitats as opposed to live tree and litter faunas. Species richness was affected by microhabitat, topographic factors and local moisture conditions. Species richness in dead wood and rock microhabitats remained high irrespective of the topographic effects as opposed to litter and live tree microhabitats, where richness decreased with drier microhabitat conditions due to topography. Our results imply that consideration of topographic factors and microhabitat quality as part of coarse filter conservation measures could be beneficial to local land snail populations in the face of changing climate and disturbance regimes.
. Species assemblages and indicatorspecies: The need for a flexible asymmetrical approach. Ecol. Monog. 67: 345-366.
Species assemblages and indicatorspecies: The need for a flexible asymmetrical approach
For Lotka-Volterra population systems, a general model of state monitoring is presented. The model includes time-dependent environmental effects or direct human intervention (treatment) as control functions and, instead of the whole state vector, the densities of certain indicator species (distinguished or lumped together) are observed. Mathematical systems theory offers a sufficient condition for local observability in such systems. The latter means that, based on the above (dynamic) partial observation, the state of the population can be recovered, at least near equilibrium. The application of this sufficient condition is illustrated by three-species examples such as a one-predator two-prey system and a simple food chain.