Ceplac – Central European Plant Communities

Topics

Introduction
Contents and Purpose
Structure
Current State
Syntaxonomic Concept
The Tasks
The Braun-Blanquet Approach, Character-species and the Core System Principle
The Problems
Twin and Layer Communities
The Double Species Concept
Presence/Absence versus Dominance
Vitality
Rankless Communities
Conclusions: the Ceplac Concept as Process and Synthesis
References

 

Introduction

This text contains an outline of the Ceplac concept conceived for the website only. An extended English version with more references to the literature is in preparation.
The actual state of the publishing and the download link to the demo version can be found at ‘Ceplac aktuell’.

Contents and Purpose

The Central European Plant Communities (Ceplac) are a project comprising the first overview of the phanerogamic and macrophytic algae vegetation of Central Europe that is based on comprehensive and detailed tables.
The actual version 2.0.1 is the German one; the English texts are in preparation.
The Ceplac are published as an easily operable electronic database, completed by several digital documents that are dealing extensively with the syntaxonomic concept and some other topics.
The Ceplac are not just a compilation of the existing knowledge but a thorough, critical reconstruction of the system, based on more than 15 years of research, teaching and practical experience. The specific combination of concept, tables and technology offers a help to identify and understand Central European vegetation as well as to build one’s own opinion on the concept and contents.
The research area covers Germany, Switzerland, Poland, the Czech Republic, Slovakia, Liechtenstein, Austria and Hungary. The alliances, orders and classes of Denmark, the Netherlands, Belgium, Luxembourg, Estonia, Latvia and Lithuania are supposed to be covered, too, as well as of greater parts of the other adjoining regions, especially Great Britain, Ireland and Scandinavia. In order to find optimally developed relevés/stands of units that in Central Europe assumingly only occur in an impoverished form, tables from many other countries have been processed.
Meant for professionals and interested amateurs, the Ceplac provide a unique inside into both the vegetation types of Central Europe and the sociology – and thus indirectly the ecology – of thousands of plant species. From his own work the author knows that the Ceplac are an everyday tool, helpful with all kinds of work in the field of vegetation ecology, whether it is pure scientific work, nature conservation assessment or the development of appropriate seed mixtures etc. It can be used as stand-alone implement and, through its references, as a door to the professional literature.

Structure

The focus of the Ceplac lies on the vegetation alliances, which due to the inflation of described associations, as already Heinz Ellenberg has stated (Ellenberg 1956), are the best characterised and most objective, meaningful and reliable units of the vegetation.
The database comprises synoptic tables at three levels of synthesis: at the lowest level each column as a rule contains the synthesised relevés of a source table that belong to either the same subassociation or, if no subassocations have been delimitated, to the same association. At this level the demarcations of probable well founded associations are already shown in most of the tables, and most of the associations are represented by several or many columns, which gives an insight into their subunits, too. There is a table for each of the 242 alliances and rankless groups that are included in the Ceplac so far. The other tables are synthesised at the level of alliance and class, respectively.
Thus not only the alliances but also an estimated 95% of the well defined associations of Central Europe are covered. So, regarding the tables the Ceplac provide both details and a more general overview of the vegetation units as well as of the sociological behaviour of the plant species.
The mentioned 242 tables are in fact nothing else but vertical stripes of a single, huge table containing all plant communities of Central Europe. One can switch from table to table or move to the table of a completely different alliance and will always find the same species in the same order (calling a table on the other hand means that the species that do not occur in this table are suppressed, what makes the tables clearer). This allows the comparison of arbitrary tables and guarantees the consistency of the species’ sociological assignment.
The tables have been sorted ‘manually’ similar as described by Ellenberg (1956), with the use of the computer as writing aid only. The method turned out to work with such a huge dataset without fundamental difficulties (the tables contain about 7,500 columns and 7,000 rows). Those who know about the problem and its current state will easily understand at the latest when reading the concept below that a software that is able to realize this or a similar advanced and adequate concept does not exist anyway.
The systematic part contains condensed texts based on the alliances or rankless groups and their higher units (altogether there are 427 units) that deal with the most important aspects of biology, ecology, human impact and history. These texts are focused on the new insights the project has generated. The level of the associations and relevant subordinate units will be discussed in later versions. The additional text files deal with the syntaxonomic concept, the process of data processing, nomenclatorial and taxonomical aspects etc.

Current State

The Ceplac are a very comprehensive work. So far about 53,000 vegetation relevés from 36 countries have been processed, and it would need more than 20,000 pages to print the entire matrix of the tables. The texts, condensed as they are, would fill about 260 pages, and the additional information in the database like header data and references would need more than another 1,000 pages. So, due to reasons of cost alone, the Ceplac would have been published in electronic form.
In addition, the electronic form provides several valuable options, like the abovementioned comparison of arbitrary tables, different sorting of the species or the use of search functions. Furthermore, actualisation and future development are much easier and cheaper.
However, the processing of relevés is the bottleneck of the Ceplac, because this step needs by far the most time. But since the availability of literary sources is an important factor, too, a certain degree of coincidence is ensured, which supports, alongside with several other measures, the objectivity of the work.

Syntaxonomic Concept

The Tasks

Systematic work has two tasks: to help to communicate about its object and to help to understand the object. These tasks often are competing: f.i. communication is facilitated by stability, whereas progress in understanding frequently leads to taxonomic changes. Communication is supported by simpleness of a system, whereas understanding often calls for more complex solutions. The author of this text starts from three convictions:

The following sections outline the author’s efforts in this field.

The Braun-Blanquet Approach, Character-species and the Core System Principle

The syntaxonomic concept of the Ceplac is grounded mainly on the Braun-Blanquet approach (Braun-Blanquet 1964, Westhoff et van der Maarel 1978). This basically means that the system is built on species fidelity, character-species and, as far as possible, the characteristic species combination.
According to this, in the Ceplac the influence of geographical borders and habitat features on the system has been reduced as much as possible, so that the inherent characteristics of the vegetation types themselves – the plant species, their abundance and vitality – are almost exclusively decisive for their distinction, acceptance or rejection, which is an important feature of up-to-date scientific systems.
Thus syntaxa that are not vegetation units but ‘context types’ like the Salicetea arenariae – shrubberies that only have in common that they grow in coastal dunes – or the Ulmenion minoris – forests that only have in common that they grow in older or higher regions of floodplains – have been dissolved as far as possible. It is, though, one of the outcomes of the work that a complete elimination of habitat features is practically impossible if an overseeable system is wanted.

In the Ceplac the relevancy of character-species is, compared to most other synopses, generally enhanced. There are no typological limitations of their validity and little geographical ones. Typological limitations have been made obsolete by the twin community and the double species concept (see below). As for geographical restrictions, the Ceplac tables clearly prove that in the region studied, which is much larger than Central Europe, there is little need for limitations of the character-species’ validity. This is very important, since the pessimism regarding this point was perhaps the most relevant reason for the inflation of associations, which more and more became nothing more than local phenomena.
The supra-regional necessity of geographical limitations still has to be researched with huge tables. Two aspects have to be considered: firstly, there is little to no need for such restrictions if the flora of the respective regions is very different, as it is for instance the case with Central Europe and the Mediterranean lowlands. That means among other things that most alliances and probably most associations, too, have a global validity.
Secondly the fact that a species in a certain region A occurs in other communities than in a region B does not necessarily prove a different sociological (and ecological) behaviour. It can be that this simply means that the occurrences in region A are transitions to the association of region B, which the species really is a character-species of.
On the other hand, the Ceplac research have shown that in huge regions with a similar flora like for instance the temperate zone of Eurasia there is a gradual decline of comparability of the plant communities. Although some of the associations are delimitable throughout the greater part of such a region, other species combinations are not comparable. Therefore it makes sense to include in original descriptions and synopses the probable geographical range of an association and the validity of its character-species. This range should be described in terms of great regions, if possible. Within Eurasia, for instance the Urals are a suitable boundary.
Finally, geographical limitation is certainly needed in dealing with communities dominated by neophytes originating from other continents. In these cases the restriction of character-species’ geographical validity to a single continent should usually be sufficient, whereas in Eurasia again the Urals may be used as a limitation.

In the Ceplac units up from alliance level must have character-species, and associations as a rule, too: only a few ‘central associations’ in the sense of Dierschke (1988) are accepted. These are communities that show a stable species combination and are very common in large regions, so that it would be inadequate to regard them as fragments.
At this place it might be added that the system of vegetation is, like most scientific systems, what can be called a core system. That means that the units are represented first and foremost by their cores, which consist from the most typical stands/relevés. These are the stands/relevés that show a maximum of community-owned species (mostly combined with a maximum abundance of these species) and a minimum of alien ones.
The boundaries of vegetations types cannot be defined; they have to be established by comparison with other communities: a stand/relevé always belongs to the community to which it shows the greatest similarity. If there is no other community, as it is for instance often the case at watersides, the stand belongs to the given association no matter how low the number of character-species and their abundance is. The task of a syntaxonomic concept is to define similarity.

The Problems

However, during the years it had turned out that the pure Braun-Blanquet approach is not sufficient to generate a logical, consistent and complete system of all vegetation units of a region. This has to be ascribed mainly to following reasons:

These problems often occur combined. For instance, the tall-herb communities of the Dauco-Melilotion have several character-species like Pastinaca sativa or Daucus carota that are important species of the Arrhenatherion meadows, too. Furthermore, the Dauco-Melilotion is the most important herbaceous layer type of nutrient poor ruderal pioneer groves that belong to the Sambuco-Salicion or the Populion albae.
As a result of these and other problems the existing regional or national systems are different from each other, in parts incompatible, inconsistent and incomplete. To give one example, combinations of a given tree layer type with herbaceous layer communities belonging to different classes are sometimes treated within a single association (f.i. Salicetum albae), sometimes as separate classes (e.g. Erico-Pinetea, Pulsatillo-Pinetea).

Twin and Layer Communities

In the Ceplac, after a long period of hesitation, a simple but fundamental decision has been made: to accept reality.
Consequently the mentioned problems have been solved among other things by making extensive use of the twin community concept, including the concept of layer communities, which dates back to the 19th century (Hult 1881). Twin communities consist from two or seldom more species groups that are sociological independent, meaning that they also occur together with other groups and as a rule alone, too. This is the case with forest and shrub communities and certain other types.
Therefore the systems of forest, shrub and herbaceous layer syntaxa are developed independently for each layer (e.g. ‘Fagion’ and ‘Luzulion luzuloidis’). In doing so it has turned out that as a rule the forest-owned types of herbaceous layer communities are similar enough to the respective fringe and open land units that they could be ascribed to the same associations. In the Ceplac tables of the alliances that include herbaceous layer columns the latter are depicted in cyan. Thus it is very easy to compare the similarity of open land and herbaceous layer forms.
The communities of the woody plants are called macro associations (‘Fagetum’). Fundamental combinations of the tree or shrub and the herbaceous layer are regarded as micro associations (‘Luzulo-Fagetum’). Thus well defined existing associations like the Luzulo-Fagetum can be adopted, and missing units of the ‘swarm’ of substitutional pioneer woods and bushes showing the same ‘Krautschichttyp’ (herbaceous layer community) can easily be added.
This process is connected with the segregation of community groups that had been both floristically and structurally inhomogeneous so far. For instance, the Alnetea glutinosae s.l. contain forests as well as tall and low growing shrub communities. Apart from the herbaceous layer, which can be the same within the mentioned structure types they have little if any in common, because the tall growing shrub species do not tolerate the shadow of dense forests, whereas the low growing scrub species for the same reason do not grow under the tall ones.
As a result of the mentioned segregation process the alliances and associations as a rule are physiognomically homogeneous. Thus the Ceplac have overcome the incompatibility of the floristic and the physiognomic approach in which Braun-Blanquet somewhat believed (though at other places he stated that typical stands of an association also show typical physiognomy).
This process, however, creates remarkably little need of new higher syntaxa except for about ten herbaceous layer types like the abovementioned Luzulion luzuloidis that are simply replacing existing higher forest units, and a few products of the segregation process. In this way a very clear and mostly familiar system is maintained.
The extensive use of the twin community concept also contains measures that are already accepted but not followed consequently: the splitting of units that are inhomogeneous in space (mosaics etc.), of communities with a different lifespan or phenology (short- and long-living communities, vernal, estival and autumnal communities) and of otherwise twinned units (e.g. the superposition of Secalietea stands by Isoëto-Nanojuncetea communities in temporarily wet fields).

The Double Species Concept

The Ceplac tables clearly show that species with a diffuse sociology are extremely rare. Strictly speaking companions do not exist, and the twin community concept enhances this effect. This is a major advance compared to the Braun-Blanquet approach, in which – strictly applied – for instance all of the very few Central European tree species would be companions, because all of them occur together with a great variety of herbaceous layer types, which are richer in species. This was one of the aspects of the Braun-Blanquet approach that wasn’t adopted by his supporters, and actually not even by himself.
Many authors have stressed that character-species are a special case of differential-species. The Ceplac tables suggest that it is more adequate to look at this inversely: although plant species as a rule occur in a quite a lot of vegetation units – forming transitions, or more or less incidentally –, the number of plant communities that they really are characteristic of is very small.

However, even if the analysis of ‘single vegetation’ types and the process of segregation of twin communities is completed, there remain species that are not character-species of a single vegetation unit of any given hierarchical rank but common character-species of two or more vegetation types that do not belong to the same next higher unit.
Sociologically most relevant of these are the species and species groups that are characteristic both of tall-herb and meadow communities, two basically distinct community types belonging to different main groups. Pastinaca sativa and Daucus carota for instance are common character-species of the tall-herb unit Dauco-Melilotion and the meadow alliance Arrhenatherion. This means that they really show a similar degree of fidelity within the different vegetation types and are not restricted to transitions. Such species often show somewhat enhanced vitality in tall-herb stands, but that is not sufficient to regard them as tall-herb character- and meadow differential-species.
Regarding this relationship, which can be described as incomplete twinning, the so called double species concept has been developed, which allows considering species with a complex sociology realistically without creating unnecessary confusion: the double species are treated – that means sorted and labelled – as tall-herb species in tall-herb community tables and as meadow species in meadow tables. The respective alternate ‘sociological diagnosis’ – the label – is shown in a quick info if the mouse is moved over the actual label, and the diagnoses can be switched with a menu command.
Thus it became possible to complete the so far somewhat gappy system of tall-herb communities consistently without basic changes in the system of meadows.

Presence/Absence versus Dominance

Having segregated twin communities and recognised double species, the next step is to examine what type of community the studied units belongs to. There are two very different types, with some communities holding an intermediate position: firstly there are communities with a stable species combination consisting from a few or many species that show a high constancy. These communities have a characteristic species combination. They are, so to speak, the Braun-Blanquet standard communities and therefore they can be treated with the methods described by Braun-Blanquet and many others. That means that for their acceptance and delimitation presence/absence are more important than abundance.
There are syntaxa belonging to this type that as a rule show dominance of one of their character-species, like the Caricetum firmae (Caricion firmae) or the Caricetum curvulae (Caricion curvulae), which stands usually are dominated by their name-giving species. It would be possible, however, to establish these associations on the basis of presence/absence alone. Abundance is just helpfully to set the exact borders to other communities, and to recognise the units in the field.

Luckily, with the standard Braun-Blanquet method most Central European communities can be treated. However, if the framework of these communities is established, there remain quite a few gaps, filled by species-poor vegetation types that are not compatible with that approach. They consist of species that show a high tendency towards dominance, building stands that have nothing in common but the regarding species and that can’t be ascribed to other communities without problems.
Species/communities like this occur especially on extreme habitats. There are a lot of them in amphibian and aquatic vegetation (f.i. Phragmito-Magnocaricetea, Thero-Salicornietea, Potametea, Charetea). But, regarded as layer communities, also several frequent tree and shrub units belong here (e.g. Fagetum sylvaticae, Pinetum sylvestris, Sambucetum nigrae). Most supporters of the Braun-Blanquet approach, including very experienced ones, seem to have no problem with treating one-layered communities of this type as associations, although strictly spoken they are sociations in the sense of many, mainly Nordic authors (for an overview see Trass et Malmer 1973).
Once again, in the Ceplac the decision has been made to accept reality: that there are vegetation types like the described. To treat them with the ‘strong’ Braun-Blanquet approach would mean to label such species as companions and to ascribe their stands with the help of the few further species growing in them to different associations that are better characterised.
But this would be very forced, since the further species in most cases only form rudimentary species combinations, and it would macerate the borders of the respective associations and deteriorate their ecological significance. Furthermore, it would still leave us with the problem of what to do with the stands that have no further species, which are not rare. So it seems to be best to adopt the handling performed by most vegetation scientists.
There is, though, a loophole in the Braun-Blanquet approach: Braun-Blanquet himself and some of his supporters have cited the fidelity scheme of Szafer and Pawlowski (Braun-Blanquet 1964). According to this, the analysis of the Ceplac tables reveals that for instance almost all tall sedges that are species of the Magnocaricion are ‘firm’ character-species of single-species associations, even though the table is built almost entirely on dominance.
The treatment of species-poor dominance vegetation types, however, leads us to the next problem of how to lump together such ‘dominance associations’ to alliances and higher units. Since they sometimes have little or nothing in common there are no intrinsic floristic features that would support any decision, and we would be left with huge, un-overseeable units. For instance, the Ceplac tables contain 52 Characeae species that are character-species of the Charetea fragilis. Most of them match the abovementioned criteria as character-species of their own association. Without the measures described below it would be impossible to group them into alliances. Using these measures five alliances can be distinguished, making the system much more comprehensible.
So what we have to do is to use other features than the species: firstly the fact that these ‘dominance associations’ don’t fit into other units, next using higher floristic ranks like genera and even families (all Central European Fagaceae are character-species of the Querco-Fagetea, and the class is built on this and other families and genera), physiognomy (tall sedges versus reeds) and, finally and worst, habitat features (the Phalaridion arundinaceae for instance is hardly justifiable without the habitat feature ‘usually on banks of medium fast flowing rivers’). With the Central European units mostly a combination of these features is needed. In using the mentioned procedure the Ceplac concept again follows the path experienced supporters of the Braun-Blanquet approach have gone without theorising.

Vitality

What hasn’t been mentioned so far is the role of vitality. In accordance with the Braun-Blanquet approach in the Ceplac vitality is used during all steps of the described process as an additional feature that helps us to establish and delimitate associations. A species should not be regarded as a character-species of an association in which it shows reduced vitality, even if it occurs with high frequency and abundance. In these cases it is very likely that the respective species is a relic of another community. This is an example of that it is unwise to build a system on atypical cases.
The concept of vitality includes the aspect of maturity. Species that occur in juvenile state only should not be treated as character-species of a community, too. However, if juvenile specimens of species dominate a stand – not a community –, it may be sensible to regard this stand as a juvenile one, f.i. a juvenile forest.

Rankless Communities

With the procedure described above most vegetation stands can be ascribed to associations. However, there are stands that have no character-species of any association. They are to be described as rankless fragment communities.

Conclusions: the Ceplac Concept as Process and Synthesis

As a result of the depicted processes and concepts, the units of the Ceplac are reconstructed or sometimes newly described as maximally well characterised and regarding to their ecological interpretation reliable syntaxa. It is an indication of the quality of the scientific community’s work and of the Braun-Blanquet approach that by far the most Ceplac units have been described beforehand, most of them long ago. However, in many if not most cases the author had to decide between competing existing solutions, and sometimes a new solution had to be found.
There are two lessons we can learn from that process of conceiving an adequate, realistic and informative system for all vegetation types of a larger area:

During the last eight years the Ceplac concept has been tested extensively both in the field and with the tables. It’s the author’s opinion that it has achieved a high degree of maturity, creating a clearer, simpler, more consistent and complete system as well as it provides a deeper insight in the ecological context of many syntaxa.

Rolf Diran, 29 November 2016

References

Braun-Blanquet, Josias (1964): Pflanzensoziologie – Grundzüge der Vegetationskunde (3. ed.). Springer, Berlin Wien New York: 864 pp.
Dierschke, Hartmut (1988): Zur Benennung zentraler Syntaxa ohne eigene Kenn- und Trennarten. – Tuexenia 8: 381–382.
Ellenberg, Heinz (1956): Grundlagen der Vegetationsgliederung – Einführung in die Vegetationskunde für Studierende der Hochschulen. – In: Einführung in die Phytologie 4/1 Aufgaben und Methoden der Vegetationskunde Part 1 Eugen Ulmer, Stuttgart: 156 pp.
Hult, Ragnar (1881): Försök till analytisk behandling af växtformationerna. – Medd. Soc. Fauna et Flora fennica 8: 1–155.
Trass, Hans, Malmer, Nils (1973): North European approaches to classification. – In: Whittaker, R. H. (ed): Handb. Veg. Sci. 5 Ordination and classification of communities: 529–574.
Westhoff, Victor, van der Maarel, Eddy (1973): The Braun-Blanquet approach. – In: Whittaker, R. H. (ed): Handb. Veg. Sci. 5 Ordination and classification of communities: 617–726. Den Haag.