No abstract

No abstract

No abstract

31Theory on plant succession predicts a temporal increase in the complexity of spatial 32 community structure and of competitive interactions: initially random occurrences of early 33 colonising species shift towards spatially and competitively structured plant associations in 34 late successional stages. Here we use long-term data on early plant succession in a German 35 post mining area to disentangle the importance of random colonisation, habitat filtering, and 36 competition on the temporal and spatial development of plant community structure. We used 37 species co-occurrence analysis and a recently developed method for assessing competitive 38 strength and hierarchies (transitive versus intransitive competitive orders) in multispecies 39 communities. We found that species turnover decreased through time within interaction 40 neighbourhoods, but increased through time outside interaction neighbourhoods. Successional 41 change did not lead to modular community structure. After accounting for species richness 42 effects, the strength of competitive interactions and the proportion of transitive competitive 43 hierarchies increased through time. Although effects of habitat filtering were weak, random 44 colonization and subsequent competitive interactions had strong effects on community 45 structure. Because competitive strength and transitivity were poorly correlated with soil 46 characteristics, there was little evidence for context dependent competitive strength associated 47 with intransitive competitive hierarchies. 48 49Running title: Plant community structure in early plant succession 50 51

Szwagrzyk, J. 2004. Spatio-temporal development of forests Á/ current trends in field methods and models. Á/ Oikos 107: 3 Á/15.We present a critical review of current trends in research of spatio-temporal development of forests. The paper addresses (1) field methods for the development of spatially-explicit models of forest dynamics and their integration in models of forest dynamics, (2) strengths and limitations of traditional patch models versus spatiallyexplicit, individual-based models, and (3) the potential for moment-based methods in the analysis of forest dynamics. These topics are discussed with reference to their potential for solving open questions in the studies of forest dynamics. The study of spatio-temporal processes provides a link between pattern and process in plant communities, and plays a crucial role in understanding ecosystem dynamics. In the last decade, the development of spatially-explicit, individual-based models shifted the focus of forest dynamics modelling from the dynamics of discrete patches to the interactions among individual organisms, thus encapsulating the theory of ''neighbourhood'' dynamics. In turn, the stochastic properties and the complexity of spatially-explicit, individual-based models gave rise to the development of a new suite of so-called moment-based models. These new models describe the dynamics of individuals and of pairs of individuals in terms of their densities, thus directly capturing second-order information on spatial structure. So far, this approach has not been applied to forests; we indicate extensions needed for such applications. Moment-based models may be an important complement to spatially explicit individual-based models in developing a general spatial theory of forest dynamics. However, both kinds of models currently focus on fine scales, whereas a critical issue in forest dynamics is to understand the interaction of fine-scale processes with coarser-scale disturbances. To obtain a more complete picture of forest dynamics, the relevant links and interactions between fine-, intermediate-, and coarse-scale processes ought to be identified. Intensive links between modelling work and field studies designed across different scales are a promising means to create a new perspective on forest dynamics.