Question: Species optima or indicator values are frequently used to predict environmental variables from species composition. The present study focuses on the question whether predictions can be improved by using species environmental amplitudes instead of single values representing species optima.

Location: Semi-natural, deciduous hardwood forests of northwestern Germany.

Methods: Based on a data set of 558 relevés, species responses (presence/absence) to pH were modelled with Huisman-Olff-Fresco (HOF) regression models. Species amplitudes were derived from response curves using three different methods. To predict the pH from vegetation, a maximum amplitude overlap method was applied. For comparison, predictions resulting from several established methods, i. e. maximum likelihood/present and absent species, maximum likelihood/present species only, mean weighted averages and mean Ellenberg indicator values were calculated. The predictive success (squared Pearson's r and root mean square error of prediction) was evaluated using an independent data set of 151 relevés.

Results: Predictions based upon amplitudes defined by maximum Cohen's κ probability threshold yield the best results of all amplitude definitions (R² = 0.75, RMSEP = 0.52). Provided there is an even distribution of the environmental variable, amplitudes defined by predicted probability exceeding prevalence are also suitable (R² = 0.76, RMSEP = 0.55). The prediction success is comparable to maximum likelihood (present species only) and – after rescaling – to mean weighted averages. Predicted values show a good linearity to observed pH values as opposed to a curvilinear relationship of mean Ellenberg indicator values. Transformation or rescaling of the predicted values is not required.

Conclusions: Species amplitudes given by a minimum and maximum boundary for each species can be used to efficiently predict environmental variables from species composition. The predictive success is superior to mean Ellenberg indicator values and comparable to mean indicator values based on species weighted averages.

Nomenclature: Wisskirchen & Haeupler (1998); Koperski et al. (2000).

Question: How do different regeneration scenarios shape species composition at two stages of plant community establishment (emergence and net recruitment) in an early succession?

Location: Northern Spain.

Methods: In a recently ploughed field, we created eight regeneration scenarios with light, water and nitrogen availability (five replicates each). Seedlings of all species were monitored from emergence to death during one year. Abiotic and biotic variables were measured per quadrat, i.e. soil texture, nutrient contents, seed bank densities and composition, neighbour plant species densitiy and cover. We used partial ordination methods in order to separate the effect of each environmental variable on species composition during emergence and adult net recruitment.

Results: Light treatment determined annual plant density at time of emergence and recruitment, while water addition controlled the recruitment of perennials. Resource levels explained the emerged species composition; this effect was not translated into the recruited species composition. N-addition and N water addition were strongly associated to species abundances at the time of emergence. Seedling composition in summer was correlated with seed abundance of Cerastium spp. Neighbour species density and cover (mainly Arrhenatherum bulbosum, Agropyron repens and Picris echioides) explained significant fractions of species composition in the emergence and recruitment of the different cohorts. Interactions between species seem to vary in intensity among cohorts and in the key plant species that determined species abundance along succession.

Conclusions: Our scenarios exerted contrasting and multilevel effects on the development of our early succession community. Resource availability differently affected plant density and species composition at different life stages. It is relevant to consider different life stages in plant community studies. However, regeneration conditions and other abiotic factors are not enough to explain how community composition varies.

Nomenclature: Aizpuru et al. (1999).

Question: This study analysed the effect of severe soil erosion on species composition of plant communities by favouring species showing certain growth forms, root-sprouting and clonal growth abilities.

Location: The study area was located between the middle Ebro Valley and the Pre-Pyrenees (northeastern Spain).

Methods: Root-sprouting and shoot-rooting abilities, clonal reproduction and growth form were assessed for the 123 most common plant species from eroded lands in the study area. We obtained 260 vegetation relevés in three different substrata (gypsum outcrops, Miocene clays and Eocene marls) on areas with different degrees of soil erosion. The frequency of every plant trait in each relevé was estimated according to species presence. The effect of soil erosion on the frequency of plant attributes was assessed by correlation analyses.

Results: Bipolar, non-clonal plants and annual species decreased their frequency with increasing soil erosion in the three substrata analyzed, whereas root-sprouters and woody plants (mostly sub-shrubs) increased their frequency in most of the substrata analysed.

Conclusions: Woody sub-shrubs, root-sprouters and clonal species are favoured in eroded lands in NE Spain. Bipolar species and annual plants might not be plastic enough to survive the high stress and frequent disturbances prevailing in such eroded areas.

Questions: How do climate conditions and the site's ecohydrological properties affect the age and size structure of natural Pinus sylvestris stands on pristine boreal mires? How do the long-term stand dynamics on mires proceed as stands age? Do the mire stands reach a balanced, old-growth stage?

Location: Boreal mire forests in southern and northern Finland.

Methods: Tree age and diameter distributions were analysed in 52 stands in two climate areas and in two mire site types with different ecohydrological properties. Temporal stand dynamics were examined by (1) comparing the graphs of the stands' mean tree ages by diameter at breast height (1.3 m) classes and (2) describing the changes in stand characteristics and stand age and size structures as a function of stand dominant age in a chronosequence.

Results: In the south, the DBH distributions were mostly unimodal and bell-shaped in both site type groups. Age distributions were multimodal and flat in fully-stocked sites but more uneven in sparsely forested composite sites. In the north, both the age and size distributions were clearly uneven in both site type groups. Tree age and size variation increased with stand age, but levelled out in the long term. Particularly in the south, the abundance of small trees decreased as stand age increased.

Conclusions: The pine stands on pristine boreal mires are more dynamic than anticipated and are generally not characterised by a balanced, self-perpetuating structure. Their dynamics reflect differences in climate and ecohydrology: on stocked sites in favourable boreal conditions, the stands showed structures typically resultant of inter-tree competition processes that control tree growth and regeneration, whereas in harsh boreal climates, the tree regeneration process is ongoing diversifying the stand structure.

Questions: What is the contribution of management continuity during the last 30–40 years to variation in species diversity and composition of a calcareous wooded meadow plant community? Is tree cover related to species diversity and composition of the herbaceous layer? What are the effects of local soil gradients on species diversity?

Location: Laelatu calcareous wooded meadow, Western Estonian coastal zone.

Methods: Plant community composition was assessed in 150 1 m × 1 m plots, located at 30 sites with known management history within Laelatu meadow (7 ha). Light and soil conditions and relative altitude were measured at each plot. DCA was used to analyse variation in species composition and general linear mixed models to analyse the effects of management and environmental parameters on diversity.

Results: Management continuity was the primary determinant of plant community composition, followed by light conditions and soil parameters. Species richness, diversity and evenness are positively dependent on management continuity. Spatial autocorrelation is important as well. Diversity started to decline under the tree canopy where 50% or less irradiation reached the level of the herbaceous layer. We did not find significant effects of soil conditions on small-scale diversity.

Conclusions: Management continuity, together with the cover of the tree layer, are the most important determinants of diversity. Despite grassland stands with different management history are located side by side, the regeneration of diversity and composition of plant communities after restoring regular management practices is a slow process.

Question: What is the influence of remnant trees on secondary forest structure and composition in tropical pastures many years after abandonment?

Location: Neotropical lowland wet forest, La Selva Biological Station, Costa Rica.

Methods: Tree and sapling density, basal area, and species richness were quantified at three distances from remnant trees, 0 – 10 m (inner), 20 - 30 m (intermediate), and ca. 50 m (distal) zones. A total of 15 remnant trees were sampled in pastures ~23 years after abandonment.

Results: Tree density decreased along a gradient from inner (1117 ± 377 individuals/ha) to distal (592 ± 282 individuals/ha) zones, and the number of large-seeded individuals (seeds > 1 cm diameter) was significantly greater in the inner zone. Basal area of tree individuals was greater in the inner (25.6 ± 12 m²/ha) and intermediate (28.3 ± 15.6 m²/ha) zones than the distal zone (14.7 ± 7.2 m²/ha), but there were no differences between inner and intermediate zones. Similar patterns are reported for species richness. Additionally, saplings (1 - 5 cm DBH) had higher density directly beneath and adjacent to remnants, suggesting that remnant trees can affect recruitment even many years after pasture abandonment and the formation of a surrounding secondary forest.

Conclusions: Results indicate that remnant trees facilitate forest recovery over a broad temporal range, and appear to ‘nucleate’ forest regeneration by expanding their sphere of influence outward over time.

Questions: 1. Is there a primary role of disturbance at local scale and of environmental stress at regional scale? 2. Does disturbance increase or decrease environmental stress at local scale?

Location: The Atlantic coastal dune system of the Aquitaine Region (France).

Methods: Species biomass and 16 environmental variables were sampled in 128 quadrats along a local beach-inland gradient and a regional North-South gradient. Environmental data were analysed with ANOVAs and vegetation-environment relationships with Canonical Correspondence Analysis.

Results: At the local scale community composition was primarily driven by disturbance due to sand burial, whereas water and nutrient stress better explained regional differences. However, random biogeographical events are very likely to also affect community composition at the largest scale. The main interaction between environmental stress and disturbance was the mitigation of nutrient stress induced by disturbance at a local scale. This was due to a positive direct effect of sand burial and a positive indirect effect of wind (decrease in VPD by ocean spray). Although wind had also a significant effect on soil conductivity and pH, there was no evidence that these factors had any role in community composition.

Conclusions: Our results support the hypothesis that disturbance had a primary role at local scale and environmental stress at regional scale but further research is needed to separate the effect of stress from that of dispersal at regional scale. We also demonstrated that environmental stress in primary succession may not always decline with decreasing disturbance.

Nomenclature: Flora Europaea (Tutin et al. 1964–1980).

Questions: Studies of gap effects have been conducted mainly in forests. We studied gap ecology in a pyrogenic Ceratiola ericoides (Florida rosemary) dominated shrubland and asked: How do gap size and the frequency of large gaps change across the fire chronosequence? Do larger gaps differ from smaller gaps in vegetation structure or species diversity? Are effects of gaps independent of, or dependent upon, time-since-fire? Are larger gaps refugia for herbs and subshrubs?

Location: Archbold Biological Station, Lake Wales Ridge, south-central Florida, USA.

Methods: We investigated plant species occurrence and diversity in 805 gaps (areas free of shrubs taller than 50 cm) in 28 fire-dependent Florida rosemary scrub sites. We collected quantitative cover data in a subset of seven sites.

Results: Gap area distribution was lognormal. The largest gaps occurred throughout all but the longest time-since-fire intervals. Gaps were smallest in the longest unburned site but otherwise did not show strong patterns across the fire chronosequence. Species diversity measures increased with increasing gap area, with herbaceous diversity increasing with both gap area and bare sand. Herb diversity (H') decreased with time-since-fire. Larger gaps are refugia for some species. Of 14 species occurring in 25–75% of gaps, 13 had increased occupancy with increasing gap area, and gap area was the strongest predictor of occupancy for seven species of herbs and shrubs. Time-since-fire was the strongest predictor of occupancy for five species, including four ground lichens that increased with time-since-fire.

Conclusions: Community structure within Florida scrub gaps is influenced by gap size, which in turn is affected by fire, the dominant ecological disturbance. We present a conceptual model that considers both gap size and time-since-fire as drivers of community structure and herbaceous plant diversity in Florida scrub. Because gap properties (independently of fire) have strong influences on species assemblages in Florida rosemary scrub gaps, fire management should consider the number and size of gaps across the landscape.

Nomenclature: Wunderlin & Hansen (2003) for vascular plants; Brodo et al. (2001) for ground lichens unless otherwise noted.

Question: In the Northern Hemisphere, species with dispersal limitations are typically absent from secondary forests. In Australia, little is known about dispersal mechanisms and other traits that drive species composition within post-agricultural, secondary forest. We asked whether mode of seed dispersal, nutrient uptake strategy, fire response, and life form in extant vegetation differ according to land-use history. We also asked whether functional traits of Australian species that confer tolerance to grazing and re-colonisation potential differ from those in the Northern Hemisphere.

Location: Delatite Peninsula, NE Victoria, Australia.

Methods: The vegetation of primary and secondary forests was surveyed using a paired-plot design. Eight traits were measured for all species recorded. ANOSIM tests and Non-metric Multi-dimensional Scaling were used to test differences in the abundance of plant attributes between land-use types.

Results: Land-use history had a significant effect on vegetation composition. Specific leaf area (SLA) proved to be the best predictor of response to land-use change. Primary forest species were typically myrmecochorous phanerophytes with low SLA. In the secondary forest, species were typically therophytes with epizoochorous dispersal and high SLA.

Conclusions: The attributes of species in secondary forests provide tolerance to grazing suggesting that disturbance caused by past grazing activity determined the composition of these forests. Myrmecochores were rare in secondary forests, suggesting that species had failed to re-colonise due to dispersal limitations. Functional traits that resulted in species loss through disturbance and prevented re-colonisation were different to those in the Northern Hemisphere and were attributable to the sclerophyllous nature of the primary forest.

Nomenclature: Ross & Walsh (2003).

Question: How does a newly designed method of supervised clustering perform in the assignment of relevé (species composition) data to a previously established classification. How do the results compare to the assignment by experts and to the assignment using a completely different numerical method?

Material: Relevés analysed represent 4186 Czech grassland plots and 4990 plots from a wide variety of vegetation types (359 different associations or basal communities) in The Netherlands. For both data sets we had at our disposal an expert classification, and for the Czech data we also had available a numerical classification as well as a classification based on a neural network method (multi-layer perceptron).

Methods: Two distance indices, one qualitative and one quantitative, are combined into a single index by weighted multiplication. The composite index is a distance index for the dissimilarity between relevés and vegetation types. For both data sets the classifications by the new method were compared with the existing classifications.

Results: For the Czech grasslands we correctly classified 81% of the plots to the classes of an expert classification at the alliance level and 71% to the classes of the numerical classification. Correct classification rates for the Dutch relevés were 64, 78 and 83 % for the lowest (subassociation or association), association, and alliance level, respectively.

Conclusion: Our method performs well in assigning community composition records to previously established classes. Its performance is comparable to the performance of other methods of supervised clustering. Compared with a multi-layer perceptron (a type of artificial neural network), fewer parameters have to be estimated. Our method does not need the original relevé data for the types, but uses synoptic tables. Another practical advantage is the provision of directly interpretable information on the contributions of separate species to the result.

Question: Do shrubs influence the spatial pattern of soil seed banks in herbaceous vegetation and are these effects influenced by wind direction, sampling position (windward vs leeward sides of the shrub) and distance from the shrub?

Location: Horqin desert in eastern Inner Mongolia, China.

Methods: A pioneer shrub, Artemisia halodendron, occurring in a mobile sandy habitat was used as a case study. Species composition and abundance of the seed bank and established herbaceous vegetation around six target shrubs were sampled along transects aligned to the four main wind directions and at 0.5, 1, 2, 3, 4.5 and 6 m from the shrub base on both windward and leeward sides of a transect.

Results: The presence of shrubs significantly modified the spatial pattern of seed deposition, but effects varied with wind direction, sampling position and distance from the shrub. More seeds were deposited on the leeward side than on the windward sides in all four transects, especially on transects with the most prevailing wind directions. Shrubs also caused a marked variation in seed deposition across sampling locations; this effect was more pronounced on the leeward side of transects with the most prevailing wind directions, suggesting the mean range of the shrub's influence is within ca. 2 m.

Conclusions: The study shows clear evidence of shrubs as a source of spatial heterogeneity in seed availability in the herbaceous layer. Shrub presence effects were strongly influenced by complex interactions between wind direction, sampling position, and distance from the shrub.

Questions: How similar are solutions of eight commonly used vegetation classification methods? Which classification methods are most effective according to classification validity evaluators? How do evaluators with different optimality criteria differ in their assessments of classification efficacy? In particular, do evaluators which use geometric criteria (e.g. cluster compactness) and non-geometric evaluators (which rely on diagnostic species) offer similar classification evaluations?

Methods: We analysed classifications of two vegetation datasets produced by eight classification methods. Classification solutions were assessed with five geometric and four non-geometric internal evaluators. We formally introduce three new evaluators: PARTANA, an intuitive variation on evaluators which use the ratio of within/between cluster dissimilarity as the optimality criterion, an adaptation of Morisita's index of niche overlap, and ISAMIC, an algorithm which measures the degree to which species are either always present or always absent within clusters.

Results and Conclusions: 1. With the exception of single linkage hierarchical clustering, classifications resulting from the eight methods were often similar. 2. Although evaluators varied in their assessment of best overall classification method, they generally favored three hierarchical agglomerative clustering strategies: flexible beta (β = − 0.25), average linkage, and Ward's linkage. 3. Among introduced evaluators PARTANA appears to be an effective geometric strategy which provides assessments similar to C-index and Gamma evaluators. Non-geometric evaluators ISAMIC and Morisita's index demonstrate a strong bias for single linkage solutions. 4. Because non-geometric criteria are of interest to phytosociologists there is a strong need for their continued development for use with vegetation classifications.

Questions: Is seedling emergence limited by the set of viable seeds, by incompatibility between the phenology of seed shedding and timing of mowing, or by dry weather in germination periods? Does seedling mortality fluctuate with season and weather?

Location: Negrentino, southern Alps, Switzerland.

Methods: Fecundity estimates of the dominant grass Bromus erectus; highly frequent counts of spontaneous seedlings by species and calculation of a community-level average mortality rate across 5 years; species-level records of seed shedding date and measurements of seed mass; measurement of soil moisture.

Results: B. erectus produced 143.9 viable seeds/m²/year while the density of its seedlings was a 55 times smaller fraction. Grasses had fewer seedlings than forbs and their phenology of seed shedding was less compatible with mowing date. Soil moisture was a strong determinant of seedling emergence in spring and less so in autumn. Average seedling mortality declined with age of the populations and reached a maximum in an extremely dry summer. In relatively wet summers establishment success was positively related to seed mass.

Conclusion: Community structure is susceptible to drought through mechanisms that selectively reduce recruits of coexisting plant functional groups. We propose that (1) more frequent intense droughts tend to reduce species that depend on frequent recruitment from seed, hence favour long-lived clonally spreading species, (2) drought timing selects between species with different germination phenology and drought resistance, and (3) drought impacts can be mitigated by changing management regimes that affect seed shedding.

Nomenclature: Lauber & Wagner (1996).

Question: Is it possible to model the germinative and resprouting behaviour of plant species in Atlantic shrublands and woodlands in relation to fire intensity? Is it possible to recognise different functional regenerative types in these plant species?

Location: Galicia, NW Iberian Peninsula.

Methods: We explored the patterns of germination and resprouting plant responses in relation to different intensities of fire using data from 37 trees, shrubs and herbaceous species growing in Atlantic shrublands and woodlands.

Results: Synthesizing their germinative and resprouting behaviour, we created two graphical models: the Functional Germinative Model (FGM) and the Functional Sprouting Model (FSM). Integrating the germinative and resprouting responses, and taking into account fire intensity, we created the Functional Regenerative Model (FRM), which predicts the post-fire recuperation of the populations of each species. The FRM has been validated with data from four Atlantic communities. We identified four plant functional regenerative types (PFRT) for Atlantic forest vegetation and we propose three intensities of response.

Conclusions: The extracted models (FGM, FSM and FRM) and the grouping of species in four PFRTs could be applicable to more Atlantic species, to disturbance ecology in general and to population, community and landscape management.

Nomenclature: Tutin et al. (1964–1983).