Structure and function of the decomposer food webs of forests

Structure and function of the decomposer food webs of forests along a

European North-South-transect with special focus on Testate Amoebae (Protozoa)

Cover design: Rolf & Dagmar Schröter Current address: Dagmar Schröter, Department of Global Change & Natural Systems, Potsdam Institute for Climate Impact Research; P.O. Box 60 12 03, 14412 Potsdam, Germany; Phone: +49-331-288 2639, Fax: +49-331-288 2600, e-mail: [email protected] Citation of this thesis is recommended as follows: Schroeter, D. (2001). Structure and function of the decomposer food webs of forests along a European North-South-transect with special focus on Testate Amoebae (Protozoa). PhD-thesis, Department of Animal Ecology, University Giessen.

mailto:[email protected]

Structure and function of the decomposer food webs of forests along a

European North-South-transect with special focus on Testate Amoebae (Protozoa)

Dissertation zur Erlangung des Doktorgrades der Naturwissenschaftlichen Fakultät der Justus-Liebig-Universität Giessen

durchgeführt am Institut für Allgemeine und Spezielle Zoologie Bereich Tierökologie

vorgelegt von Dagmar Schröter

Giessen, März 2001

Dekan: Prof. Dr. Rainer Renkawitz 1. Berichterstatter: Prof. Dr. Volkmar Wolters, Universität Giessen

2. Berichterstatter: Prof. Dr. Peter C. De Ruiter, University of Utrecht, Niederlande

Dies ist keine leere Seite

Table of contents

Abbreviations............................................................................................................................................... 1 Introduction .............................................................................................................................1 1.1 Coniferous forests...................................................................................................................2 1.2 European North-South-transect ..............................................................................................3 1.3 The decomposer system.........................................................................................................4 1.3.1 Decomposition ........................................................................................................................4 1.3.2 The decomposer food web......................................................................................................5 1.3.3 Testate Amoebae (Rhizopoda, Protozoa) ...............................................................................6 1.3.4 Interactions within the decomposer food web .........................................................................8 1.3.5 Quantification of fluxes within the decomposer food web......................................................10 1.4 Structure and aims of this study............................................................................................11 1.4.1 Testate Amoebae community structure � Part 1...................................................................12 1.4.2 Decomposer food web function � Part 2...............................................................................13 1.4.3 Main hypotheses...................................................................................................................14 2 The study sites......................................................................................................................15 2.1 Site description .....................................................................................................................15 2.2 Sampling scheme and sample treatment ..............................................................................21 3 Material and methods............................................................................................................23 3.1 Functional groups of organisms ............................................................................................23 3.1.1 Microflora: fungi and bacteria................................................................................................23 3.1.1.1 Chloroform fumigation extraction method (CFE): microbial carbon (Cmic) .............................23 3.1.1.2 Ergosterol..............................................................................................................................25 3.1.1.3 Direct counting of bacteria ....................................................................................................25 3.1.1.4 Metabolic potential and metabolic quotient qCO2..................................................................25 3.1.2 Testate Amoebae..................................................................................................................26 3.1.2.1 Fixation and staining of substrate samples for quantitative analyses....................................27 3.1.2.2 Direct counting of Testate Amoebae.....................................................................................27 3.1.2.3 Flotation method: extraction of empty shells .........................................................................29 3.1.2.4 Batch cultures .......................................................................................................................29 3.1.2.5 Live observations ..................................................................................................................29 3.1.2.6 Taxonomic determination......................................................................................................29 3.1.2.6.1 Distinguishing Centropyxis aerophila sphagnicola and C. sylvatica ......................................32 3.1.2.6.2 Distinguishing Cyclopyxis eurystoma and Phryganella acropodia.........................................32 3.1.2.6.3 The taxon Euglypha cf. strigosa ............................................................................................32 3.1.2.6.4 The taxa Nebela parvula/tincta and N. tincta major/bohemica/collaris ..................................32 3.1.2.6.5 Distinguishing Trinema enchelys and T. lineare....................................................................33 3.1.2.7 Slide preparation with Euparal ..............................................................................................33 3.1.2.8 Slide preparation with Naphrax .............................................................................................34 3.1.2.9 Slide preparation with glycerine ............................................................................................34 3.1.2.10 Photography..........................................................................................................................34

Table of Contents

3.1.3 Nematoda .............................................................................................................................34 3.1.4 Microarthropoda (Collembola and Acari)...............................................................................35 3.1.4.1 Collembola ............................................................................................................................35 3.1.4.2 Acari .................................................................................................................................36 3.1.5 Enchytraeidae .......................................................................................................................36 3.2 Abiotic parameters ................................................................................................................37 3.2.1 Water content (WC) ..............................................................................................................37 3.2.2 pH of organic layer................................................................................................................37 3.2.3 Thickness of organic layer, mass-to-area-ratio and bulk density...........................................37 3.2.4 C:N-ratio of organic layer ......................................................................................................37 3.3 Statistical analyses ...............................................................................................................38 3.3.1 Analysis of variance ..............................................................................................................38 3.3.2 PEARSON product-moment correlation...................................................................................39 3.3.3 Canonical correspondence analysis (CCA)...........................................................................39 4 The food web model..............................................................................................................41 4.1 Detritus pool and resource quality.........................................................................................45 4.2 Model parameters .................................................................................................................46 4.2.1 Death rates ...........................................................................................................................46 4.2.2 Assimilation and production efficiencies................................................................................46 4.2.3 Feeding preferences .............................................................................................................47 4.3 Adapting the food web model to specific climatic conditions.................................................47 4.4 Biomass estimates................................................................................................................51 5 Results ..................................................................................................................................53 5.1 Testate Amoebae community structure.................................................................................53 5.1.1 Size structure and biomass...................................................................................................53 5.1.2 Species pattern .....................................................................................................................55 5.1.2.1 Species richness, diversity and evenness.............................................................................55 5.1.2.2 Species rank plots.................................................................................................................57 5.1.2.3 Relative biomass structure....................................................................................................58 5.1.3 Testate Amoebae communities in their environment ............................................................62 5.1.3.1 Other decomposer biota........................................................................................................62 5.1.3.2 Abiotic environment...............................................................................................................66 5.1.3.3 Multivariate analysis relating Testate Amoebae communities and environment ...................68 5.1.4 Total abundance, biocoenosis and necrocoenosis ...............................................................73 5.2 Testate Amoebae and the functioning of the food web .........................................................78 5.2.1 Schematic view of the decomposer food web .......................................................................78 5.2.2 Literature survey on the trophic relationships of Testate Amoebae.......................................79 5.2.2.1 Food sources of Testate Amoebae .......................................................................................79 5.2.2.2 Testate Amoebae as food source .........................................................................................81 5.2.3 Quantifying trophic interactions.............................................................................................81 5.2.3.1 Physiological parameters ......................................................................................................81 5.2.3.2 Feeding preferences .............................................................................................................82 5.2.3.3 Biomasses of functional groups ............................................................................................83 5.2.3.4 Simulated estimates of total C and N mineralisation.............................................................86

Table of Contents

5.2.3.5 Contribution of functional groups to C mineralisation............................................................88 5.2.3.6 Contribution of functional groups to N mineralisation............................................................90 6 Discussion.............................................................................................................................93 6.1 Testate Amoebae community structure.................................................................................93 6.1.1 Total number of Testate Amoebae species...........................................................................93 6.1.2 Similarity between Testate Amoebae communities...............................................................94 6.1.3 Comparison of species biomass pattern ...............................................................................95 6.1.4 Environmental factors explaining community structure .........................................................96 6.1.5 Size structure and biomass...................................................................................................98 6.1.6 Abundance of Testate Amoebae...........................................................................................99 6.1.7 Comparison of bio- and necrocoenosis...............................................................................100 6.2 Testate Amoebae and the decomposer food web...............................................................101 6.2.1 Total food web biomass and mineralisation along the transect...........................................101 6.2.2 Site specific characteristics of the food webs along the transect.........................................102 6.2.3 Common characteristics of the food webs along the transect .............................................105 6.2.4 Evaluation of model estimates ............................................................................................106 6.3 Research needs..................................................................................................................107 7 Conclusions ........................................................................................................................109 8 Summary.............................................................................................................................111 9 Zusammenfassung..............................................................................................................115 10 Appendix .............................................................................................................................119 10.1 Parameters for calculation of decomposer fauna biomass..................................................119 10.2 Importance of food web biomass structure in the model .....................................................120 11 References..........................................................................................................................121 List of figures.........................................................................................................................................131 List of tables..........................................................................................................................................133 Acknowledgements...............................................................................................................................135 curriculum vitae.....................................................................................................................................137

Abbreviations

Only the variables and constants that occur in a wider context are included in this glossary. In other cases see explanation beneath formula.

* significant (p-level of significance � 0.05) ** significant (p-level of significance < 0.01) A assimilation a assimilation efficiency (resp. when used as unit of a variable a = year) AB-DLO Research Institute for Agrobiology and Soil Fertility – Agricultural Research

Department, Netherlands ANOVA analysis of variance asl above sea level B biomass BML Bundesministerium für Ernährung,Landwirtschaft und Forsten C consumption CANIF Carbon and Nitrogen Cycling in Forest Ecosystems (EU funded research project) CCA canonical correspondence analysis cf. confer CFE chloroform fumigation extraction method Cmic microbial carbon cum(�A) cumulative explanatory power (variance explained) d death rate DE study site in Germany (Waldstein) DIC differential interference contrast DW dry weight E excretion EU European Union F feeding rate f. forma FR study site in France (Aubure) GCTE Global Change and Terrestrial Ecosystems GLOBIS Global Change and Biodiversity in Soils (EU funded research project) HSD honest significant difference IBP International Biological Programme IGBP International Geosphere-Biosphere Programme IPCC Intergovernmental Panel on Climate Change i index for prey j index for predator K deaths through predation ("kill") max maximum MC carbon mineralisation (organic C � CO2) min minimum

Abbreviations

MN nitrogen mineralisation (organic N � NOx-, NH4+, N2) n number of potential prey groups (model equations), number of replicates (ANOVA tables) n.s. not significant (p-level of significance > 0.05) NIPHYS EU funded project: Nitrogen Physiology of Forest Plants and Soils NPP net primary production N-SE study site in Northern Sweden (Åheden) mic microbial P production p productions efficiency resp. level of significance q C:N-ratio qCO2 metabolic quotient rC canonical correlation coefficient rpm rounds per minute SOM soil organic matter S-SE study site in Southern Sweden (Skogaby) TERI Terrestrial Ecosystem Research Initiative w feeding preference w/w weight-to-weight-ratio WHC water holding capacity

Chapter 1

Introduction

1 Introduction

"Woods are more than a group of trees." Emily Carr (1871-1945),

Canadian painter

Trees are a component of forest ecosystems that is rarely overlooked. These primary producers use solar energy, carbon dioxide, water and nutrients to build organic matter. The belowground counterpart of the primary producers, the decomposer system, remains largely unseen. All the same decomposition is a process equivalent to photosynthesis in its importance for the biogeochemical cycling of energy and nutrients (Heal et al. 1997). The share of net primary production (NPP) that is not consumed alive by herbivores enters the decomposer system directly as dead organic matter. In forests more then 90 % of NPP ends up on the ground, forming the organic layer (Swift et al. 1979, Ellenberg et al. 1986, Vedrova 1995). The amount of energy and matter that persists in the time lag between primary production and decomposition is the resource that all heterotrophic organisms expend their lives on. If the time lag between primary production and decomposition is especially long, dead organic matter is stored in what we refer to as ‘fossil fuels’. Decomposition serves two key ecological functions (sensu Likens 1992): the mineralisation of essential nutrient elements and the formation of soil organic matter (SOM). These functions link the decomposer system with the primary producers in a way that both systems determine each other (Wardle 1999). Without decomposition, primary producers could not assimilate carbon dioxide due to shortage of essential nutrients in available form. Without formation of SOM, the vegetation would lack the physical substrate to take root in and water would run off before it could be taken up. Conversely, without primary production, the decomposer system would lack its energy and nutrient resource. Decomposition is a consequence of the trophic activities of the decomposer biota (Swift et al. 1979, Verhoef and Brussaard 1990). The decomposer biota form a complex web of interacting organisms: the decomposer food web. The ecological understanding of the decomposer system and its contributions to biogeochemical cycling is essential to environmental management purposes and questions of global change (Currie 1999). For example, the latest report of the IPCC (Third Report of the Intergovernmental Panel on Climate Change, IPCC, 2001) mentions for the first time the effect of rising temperatures on

1

1 Introduction

the activity of decomposer microorganisms and the possibility of increased decomposition leading to increased release of the greenhouse gas CO2, a self-stimulating process. The soil has been described as our most precious non-renewable resource (Marshall et al. 1982). Today the need to protect this resource just in the same way as e.g. the atmosphere or the hydrosphere is understood by scientists and most policy makers (Hågvar 1998). This milestone in environmental protection concepts is the consequence of a great effort of ecologists throughout the world that have agreed on a common research goal within an initiative that was started in the 1960s, called the International Biological Programme (IBP). Aim of this programme was to investigate the biological basis of productivity and human welfare. Since then, human activity has been recognised to effect ecosystems on a global scale through anthropogenic induced land-use and climate change. A new initiative, the International Geosphere-Biosphere Programme (IGBP) has established a core project on Global Change and Terrestrial Ecosystems (GCTE) with the prime objective of predicting the effects of such global change on terrestrial agricultural and forest ecosystems. The European contribution to GCTE is the Terrestrial Ecosystem Research Initiative (TERI). Within the framework of TERI a number of projects have been started to provide the basic understanding of ecological processes essential to any attempt to predict effects of global change. The study reported here was part of such a project on Carbon and Nitrogen Cycling in Forest Ecosystems funded by the European Union (CANIF, Contract N° ENV4-CT95-0053, see Schulze 2000).

1.1 Coniferous forests It is estimated that globally 1580 Gt (= 1580 1015 g) of carbon is stored in soils and detritus, and 610 Gt C is fixed in the vegetation (Schimel 1995). The largest part of terrestrial primary production, namely 40 Gt C a-1 of carbon is annually fixed in forests. The flux to grasslands and agricultural fields is considerably smaller (15 Gt C a-1; Hobbie and Melillo 1984). Since the amounts of anthropogenic CO2 and nutrients that enter the ecosystems through industrial emissions have become a major concern, the net carbon storage of the terrestrial biosphere is of great interest. The Kyoto Protocol demands strategies to balance industrial emissions by biological fixation (IGBP Terrestrial Carbon Working Group 1998, WGBU 1998). Net carbon storage in ecosystems is determined by the balance between carbon gain (photosynthesis) and carbon loss (auto- and heterotrophic respiration) (Schimel 1995). Of the processes involved, decomposition is certainly the least understood (Hobbie and Melillo 1984). Biogeochemistry research indicates that the organic layer of forests is a key component in the responses of forests to aspects of global change (Currie 1999). Recent studies suggest that European

2

1 Introduction

forests act as a carbon sink on a global scale (Valentini et al. 2000), however, budgets of the exchange between biosphere and atmosphere are estimates involving many dynamic variables that need to be evaluated and refined in the long term (Schimel 1995). One of the Earth's largest terrestrial C pools is the boreal coniferous forest system (Bolger et al. 2000). During the last century coniferous plantations have replaced deciduous forests in large areas of Central-Europe because of their industrial profitability. In Germany, one of the most forested areas of the European Union, 66 % of the forest is coniferous (BML 1999). Today coniferous forests in the boreal and temperate zone play an important role in the world's biogeochemical cycles (Bolger et al. 2000). In this context the sites of this study were chosen to be coniferous forests along a European North-South-transect. Apart from their relevance to biogeochemical cycling, forest ecosystems are important reservoirs of biodiversity. The decomposer systems of forests tend to be especially species rich and may represent biodiversity hot spots in the landscape (Hågvar 1998).

1.2 European North-South-transect To discover general patterns within the functioning of ecosystems, studies on large geographical scale are needed (Menaut and Struwe 1994, Lawton 1999). When comparing forests along a large geographical gradient major shifts within the decomposer food webs and the mineralisation rates can be expected because of systematic changes in the physico-chemical environment. Environmental factors such as temperature and moisture have been shown to alter the impact of community structure of the decomposer food web on mineralisation processes (Sulkava et al. 1996). The sites investigated within this study lie on a transect that extends from close to the polar circle in Northern Sweden over ca. 2000 km to the North-East of France. A wide range of latitudes and climatic conditions is covered. The input of biologically reactive N compounds to forests has increased rapidly over much of Europe and Eastern North America, relative to pre-industrial levels. Nitrogen Emissions to the atmosphere remain elevated in industrialised countries and are accelerated in many developing regions (Galloway 1995). Atmospheric deposition of nitrogen is expected to alter C and N fluxes in forest systems because such systems are considered to be mostly N limited (Currie 1999). In this context the sites for this study were chosen to be subject to different levels of atmospheric N deposition, ranging from virtually non-polluted to heavy loads of N input.

3

1 Introduction

Intensive site-level experiments can be questioned as to their generality or applicability over a larger region, whereas regional surveys can be questioned due to their simultaneous change in several environmental parameters (e.g., climate, geology, soils) along any spatial transect (Aber, et al. 1998). In this study the second strategy is applied, being aware of its limitations. Furthermore, on such scale the number of sites that can be investigated is limited due to practical reasons. Generalisations from this transect study have to be evaluated in the light of other large scale studies within Europe.

1.3 The decomposer system

1.3.1 Decomposition The decomposer system is located in the interface between atmosphere and geosphere, i.e. in the organic layer of the soil. The process of decomposition is understood as the main link between the two largest terrestrial C pools: plant biomass (primary production) and soil organic matter (SOM) (Sollins et al. 1996). The process of decomposition has already been studied approx. 160 years ago by scientists dealing with the pressing task to increase agricultural soil fertility and to end the famine threatening most people of that time (Heal, et al. 1997; and Liebig 1840, Lawes 1861 and Müller 1887 cited therein). Since then the interest and concern for the decomposer system has increased and the first formalised paradigm of decomposition was published in 1979 (Swift et al. 1979). Today the decomposer system is recognised as an important component of global C and N cycles. Our understanding of decomposition was recently reviewed by Heal et al. (1997). Decomposition is a process of continuous breakdown and re-synthesis, resulting in a heterogeneous mixture of products (Andrén et al. 1990). It includes secondary production of microbial and animal biomass and organic metabolites, which in turn become resources. Decomposition of any resource is the result of three processes: (i) catabolism, i.e. chemical changes such as mineralisation of organic matter to inorganic forms (largely CO2, H2O, NH4+, NO3-, SO4-), and the synthesis of decomposer biomass and humus, (ii) comminution, i.e. physical reduction in particle size and selective redistribution of litter, and (iii) leaching, i.e. the abiotic transport of labile resources down the soil profile (Heal et al. 1997). The rate of decomposition, is controlled by the physico-chemical environment and the quality of the resource (Swift et al. 1979, Heal et al. 1997, Lavelle 1997). Since the process of decomposition is

4

1 Introduction

almost entirely mediated biologically, both groups of factors act by regulating the organisms of the decomposer food web (Heal et al. 1997). The main pathways of carbon into the decomposer system are the shedding of litter by the trees and the input of soluble organic matter via abiotic leaching or biotic exudation by roots (Swift et al. 1979, Elliott et al. 1984). The mineralisation of essential nutrient elements from these resources influences the energy flux within the ecosystem in two ways. First it controls the influx of energy by regulating primary production through controlling the availability of limiting nutrients. Second, it regulates the efflux of energy as nutrient availability interacts with substrate quality, and thereby controls decomposition rates (Elliott et al. 1984).

1.3.2 The decomposer food web Areas of especially large species richness have been called 'biotic frontiers', a term signifying the need to explore these systems. Two major biotic frontiers are the tropical rainforest canopy (Erwin 1983 as cited in Hågvar 1998) and the deep-sea benthos (Grassle & Maciolek 1992 as cited in Hågvar 1998). By now the soil is acknowledged as the third biotic frontier (André et al. 1994, Lawton et al. 1996, Hågvar 1998). However, the biodiversity within these decomposer systems is as poorly known as that of remote environments like the ocean floor (e.g. Elliott et al. 1984, Hobbie and Melillo 1984, Copley 2000). The functional diversity of soil biota is considered to be essential for the process of decomposition (Wolters 1996, Huhta et al. 1998, Wardle 1999). Decomposition is a process in which a web of heterotrophic organisms from almost the complete range of life forms is involved. With the exception of Echinodermata, every major phylum and many minor phyla of invertebrates are represented by species in the decomposer community (Swift et al. 1979, Anderson et al. 1981). Such are the microflora (bacteria, fungi, actinomycetes and yeasts), mircofauna (Protozoa, Nematoda, Rotatoria and Tardigrada), mesofauna (Enchytraeidae, Acari, Collembola) and macrofauna (e.g. earthworms, Diptera larvae, millipedes, woodlice, insects, slugs and snails). Thus, in contrast to the ‘one-plant-business’ photosynthesis, the process of decomposition is a real team effort. The decomposer biota range in size across five orders of magnitude (10-7 to 10-2 m) (Brussaard and Juma 1996). Because of this body size range the living space important to the various organisms is very different, ranging from a structural scale of millimeters for microarthropods down to a scale of macro-molecular level, on which e.g. microflora can define their resources (Swift et al. 1979). Over this wide spatial scale the organisms interact with each other. For example, even though their size is very

5

1 Introduction

different, competition between microflora and detritivorous fauna for high-quality resources appears to be intense (Seastedt 2000). Within this study the most important groups of microflora, micro- and mesofauna in coniferous forest floors were investigated: bacteria and fungi, Testate Amoebae (Protozoa), Nematoda, Microarthropoda (Collembola and Acari), and Enchytraeidae (Persson et al. 1980, Petersen and Luxton 1982). To reduce the complexity of the system the functional group concept was applied (sensu Moore et al. 1988). Special emphasis was put on the Testate Amoebae, the most important group of Protozoa in coniferous forest systems (Schönborn 1992c).

1.3.3 Testate Amoebae (Rhizopoda, Protozoa) Together with Nematoda, the Protozoa form an important component of the decomposer system: the microfauna. However, due to their small size and resulting methodological difficulties, Protozoa are often neglected in studies of the decomposer food web. Even if included, they are mostly considered with only coarse taxonomic resolution. Information on species distribution and diversity of such small organisms is rare (Foissner 1987). Almost all the relationships dealt with in macroecology (e.g. large-scale gradients in diversity) are formulated for macroscopic organisms. It has been shown that patterns found in such macroscopic groups of organisms may be different or absent for microscopic organisms like Protozoa (Hillebrand et al. 2001). The Testate Amoebae are a polyphyletic group of Protozoa that belong to the Rhizopoda (Meisterfeld 2001a, b). They distinguish from naked Amoebae in their ability to form an outer shell. This shell has an opening through which the pseudopodia extend, referred to as pseudostome. The shell length of the

Testate Amoebae ranges between 15 and 170 �m. The more common species tend to be smaller than

45 �m (Stout et al. 1982). They live in water filled soil pores and within the thin water-film around

detritus or soil particles. The Testate Amoebae are a comparably well-known group of soil Protozoa and so far approximately 200 species have been reported in terrestrial ecosystems (Foissner 1996). They are accompanied by around 400 species of ciliates, 260 species of flagellates and 60 species of Naked Amoebae (Foissner 1996). Among the Protozoa the study of Testate Amoebae is facilitated because taxonomy is based on their well-defined shell structure which allows simultaneous identification and counting (Coûteaux and Darbyshire 1998).

Important environmental factors determining Protozoan communities are moisture, food availability, pH, temperature, atmosphere, organic matter content, geometry of soil pores, root exudates, litter and humus

6

1 Introduction

type, soil type, vegetation cover, and site history (Stout 1980, Stout et al. 1982, Stout 1984, Foissner 1987, Bonnet 1988b, Cowling 1994, Ekelund and Ronn 1994, Meisterfeld 1995, Bamforth 1997). The heterogeneous soil environment imposes some general restrictions on the inhabiting organisms. Testate Amoebae meet these challenges with various physiological and morphological adaptations, such as protecting the cytoplasm with a shell (Schönborn 1962, Schönborn 1966). Further adaptations, like reduced size to reach small soil pores and inhabit thin water films around organic particles, and invagination of the pseudostome to protect against desiccation, have been described in detail (Schönborn 1968, Bonnet 1975). Testate Amoebae have been shown to be less sensible to decreasing moisture tension than other Protozoa (Stout and Heal 1967). Like most other Protozoa, they outwear periods of unfavourable conditions by encystment to dormant forms. Some species can secrete an internal protective membrane (epiphragm, Figure 1.1) and are able to survive several months without food in this temporal precystic form (Bonnet 1964, Coûteaux and Ogden 1988). Those species that possess resistant cysts, combined with the ability for rapid encystment and excystment, tend to dominate soil protozoan populations (Cowling 1994).

20 µm

epiphragmprecyst

Pseu

shell20 µm

epiphragmprecyst

Pseu

shell

Figure 1.1 Nebela lageniformis. 400x, DIC, in Euparal. Pseu = pseudostome.

7

1 Introduction

1.3.4 Interactions within the decomposer food web The organisms within the decomposer food web interact on a multiplicity of spatial, temporal and organisational scales within a heterogeneous habitat (Lee 1994). In the following a short overview of the major trophic strategies within the decomposer food web is given focussing on groups of major importance in coniferous forest systems. Saprotrophy and detritivory

The feeding on dead organic matter is termed saprotrophy (microflora) or detritivory (fauna). Saprotrophic microflora, i.e. bacteria and fungi, account for the largest share of the carbon dioxide efflux from the forest floor (Persson et al. 1980). Microflora generally immobilise nutrients (having e.g. C:N-ratios below that of their resources) and thus compete with plants for potentially limiting elements (Anderson et al. 1981). Ectomycorrhizal fungi, as symbiotic partners of trees, aid nutrient and water uptake in exchange of plant derived carbon (Read 1991, Leake and Read 1997, Lindahl et al. 2001). Ectomycorrhizal fungi obtain considerable amounts of energy from the symbiotic tree through allocation of assimilates to the roots. They expend this energy to form extracellular enzymes that enable saprotrophic feeding via the extraradical mycelium. Detritivorous fauna ingest dead organic material usually after it has been conditioned by microflora. To what extent the microflora colonizing the organic matter is used as food source during detritivorous feeding is difficult to quantify. It is assumed that most groups feeding on detritus are also microbivorous. For this feeding strategy of mites Luxton (1972) coined the term 'panphytophagous'. In this study the term is used generally for functional groups that are both, detritivorous and microbivorous. Many species of Testate Amoebae, Microarthropoda and Enchytraeidae are considered to be panphytophagous. Detritivorous and panphytophagous decomposer mesofauna significantly enhance decomposition through so-called 'indirect effects' of their feeding activity (Anderson et al. 1981, Petersen and Luxton 1982, Anderson 1995). Such effects are litter fragmentation or comminution, translocation and mixing of litter material, improvement of soil structure, re-concentration of limiting nutrients, and stimulation, transport and inoculation of microbes (Anderson and Ineson 1984, Lussenhop 1992). The 'direct' (i.e. metabolic) contribution of these fauna groups to carbon flux is usually very small compared with microbial activity (only a few percent of the total C mineralised) (Reichle 1977). Only Protozoan respiration may directly contribute a significant amount to the C flux from the organic layer (Foissner 1987). Significant enhancing effects of panphytophagous Microarthropoda and Enchytraeidae on both, C and N mineralisation, have been observed experimentally (Petersen and Luxton 1982,

8

1 Introduction

Woods et al. 1982, Coleman et al. 1983, Seastedt 1984, Verhoef and Brussaard 1990, Beare et al. 1992, Huhta et al. 1998). This enhancement of fluxes is assumed to be a result of the described indirect effects that stimulate microbial activity by reducing growth limiting factors. The detritivorous and panphytophagous fauna may be seen as catalysts for nutrient circulation. Microbivory

The direct contribution of microbivorous fauna to nitrogen mineralisation via excretion can be considerable, because they usually have higher C:N-ratios than their food (Anderson et al. 1985, De Ruiter et al. 1993b, Bolger et al. 2000). Especially the Protozoa increase microbial turnover rates and release nutrients immobilised in microbial tissue by grazing on microflora (Stout 1980, Clarholm 1981, Elliott et al. 1984, Kajak 1995, Berg 1997, Coûteaux and Darbyshire 1998). Interactions between Protozoa and microflora are not only concerned with ingestion of microbial biomass. Protozoa are believed to secrete metabolites that stimulate bacterial metabolism (Darbyshire 1994). Another important group of microbial grazers are microbivorous Nematodes, belonging also to the microfauna. Nematodes occur with low biomass compared to other biota. Nevertheless their indirect impact, e.g microbial inoculation and stimulation via grazing, is considered to be large (Yeates 1979, Anderson et al. 1981, Bardgett et al. 1999). Predaceous feeding

Predaceous groups kill and consume animals. Besides fascinating examples like fungi feeding on Protozoa (Darbyshire 1994) or Testate Amoebae feeding on Nematoda (Yeates and Foissner 1995) the most important predaceous interactions occur among e.g. Testate Amoebae feeding on other Testate Amoebae, Nematoda feeding on Nematoda, and predaceous Microarthropoda feeding on Microarthropoda and Nematoda. Although panphytophagous fauna may regularly ingest Protozoa, hence other animals, this is usually not referred to as predaceous feeding, but traditionally seems to be included in the term 'microbivorous'. The direct impact of predaceous groups on decomposition is small. Nevertheless, functionally important top-down control by predators on lower trophic levels may occur even though decomposer food webs are generally considered to be donor-controlled (Kajak 1995, Salminen et al. 1997, Mikola and Setälä 1998, Laakso and Setälä 1999a).

9

1 Introduction

1.3.5 Quantification of fluxes within the decomposer food web Studies manipulating key functional components of the decomposer food web have identified important effects on ecosystem function (Anderson et al. 1981, Ingham et al. 1985, Setälä and Huhta 1991, Bengtsson et al. 1995, Alphei et al. 1996, Wardle 1999). Such studies, like e.g. microcosm experiments, provide details of how soil organisms and processes interact under different environmental conditions (Moore et al. 1996). However, to specify the contribution of decomposer groups to C and N mineralisation and to assign particular flux rates to these groups is a difficult task. Direct measurements are impossible due to the multitude of organisms, their small size, and their invisibility and inaccessibility within the substrate they inhabit. Measurements of carbon efflux, nitrogen leaching or tree seedling growth in experimental systems or in situ treat the biota mediating these fluxes and effects as ‘black box’. Stable isotopes are useful tools to disentangle the complex structure of food webs (Eggers and Jones 2000). But even with those means the quantification of fluxes within the decomposer system is difficult. Stable isotope techniques rely on models calculating flux rates from differences in isotope signatures between food web components. There is uncertainty about the applicability of these models to the decomposer food web. The omnivory of the majority of functional groups within such webs might cause a need for refined descriptions of the fractionation process (Wolters, pers. com.). Nevertheless such techniques will undoubtedly enhance the understanding of the structure and function of decomposer food webs and the role of particular functional groups (Eggers and Jones 2000). Estimates of the contribution of functional groups to C and N mineralisation have been obtained from calculations based on the relationship between biomass and O2 consumption (resp. CO2 production) (Huhta and Koskenniemi 1975, Persson and Lohm 1977, Persson et al. 1980, Andrén et al. 1990). For Testate Amoebae a special method was developed making use of the fact that within this group necro- and biocoenosis can be assessed due to empty shells remaining after cell death (Lousier 1974a, Schönborn 1975). However, both approaches fail to consider synergistic effects due to feeding relationships within the food web. Such effects are considered in the food web model approach used by De Ruiter et al. (1993b, based on O'Neill 1969, and Hunt et al. 1987). The model includes the interactions of groups feeding on each other, e.g. the indirect effect that consumers have by stimulating the turnover of the organisms they feed upon. Food web modelling is recognised as a promising approach to quantify the flux of C and N across and within food webs (Coûteaux et al. 1988, Heal et al. 1997, Smith et al. 1998). The food web model applied in this study was originally developed for the short grass prairie (Hunt et al. 1987) and has since

10

1 Introduction

then been applied to agricultural systems and grasslands (De Ruiter et al. 1993a, De Ruiter et al. 1994, 1998) and to a forest ecosystem (Berg 1997).

1.4 Structure and aims of this study In this study the decomposer food webs of four coniferous forest sites along a North-South-transect across Europe were investigated. The sites cover a broad latitudinal and climatic range (Persson et al. 2000a). Moreover, they were chosen to be subject to different levels of atmospheric N deposition (Persson et al. 2000a). The study is divided into two parts. Part one is a descriptive approach on population ecological level. Within this part special attention is paid to an often neglected group of Protozoa, the Testate Amoebae. The structure of their communities at the different sites is described on species level. The diversity of the Testate Amoebae communities is investigated and their similarity is analysed. The Testate Amoebae communities are set in the context of the other major taxa of the decomposer food web of coniferous forests. A multivariate statistical approach is used to ordinate the dominance pattern of the communities and to test the correlation of Testate Amoebae species with environmental variables. The population size is estimated and active organisms, dormant cells and the necrocoenosis (empty shells) are monitored. In the second part population data are related to ecosystem processes (Figure 1.2). This second step intents to go beyond description towards a functional understanding, scaling-up from population to ecosystem science, a necessity that has been emphasised (e.g. Lawton 1994, Bengtsson et al. 1995). For this, descriptions of the decomposer food webs at the sites are used in a modelling approach to estimate the mineralisation of carbon and nitrogen by each particular functional group and by the whole food web. With this approach the structure of the food webs is linked to their function (Figure 1.2). General patterns in the functioning of ecosystems are sought by studying sites on a large geographical scale (Lawton 1999). In the following the aims of the two parts of this study are formulated and a short summary of the foundation underlying the hypotheses is given (section 1.4.1 and 1.4.2). A summary of the main hypotheses investigated in this study can be found in section 1.4.3.

11

1 Introduction

The Testate Amoebae community as part of the decomposer food web.

The community structure of Testate Amoebae.

The decomposer food web as part of the forest ecosystem, mediating the flux of energy and matter.

Figure 1.2 The ecological scales investigated within this study: linking the soil biota to ecosystem function (C and N flux).

1.4.1 Testate Amoebae community structure � Part 1 The Testate Amoebae communities inhabiting the transect sites are described on species level. The abundance and biomass structure is investigated, dormant forms and the necrocoenosis are considered separately. Trends in species, abundance and biomass patterns as well as similarity between the communities along the transect are sought. The Testate Amoebae communities are regarded in the context of other decomposer biota and the abiotic environment. The data set is analysed to identify those parameters that best explain the structure of Testate Amoebae communities found at the different sites. Like larger metazoan taxa, protozoan abundance and species richness are believed to increase with decreasing latitude and altitude (Foissner 1987, Smith 1996, Coûteaux and Darbyshire 1998, Chown and Gaston 2000). Moreover, increased N supply has been shown to enhance abundance and species richness of Protozoa (Chardez et al. 1972, Berger et al. 1986). Based on the equilibrium theory that a species community is determined by a balance between immigration and extinction (MacArthur and Wilson 1967) and accepting that a greater distance is a larger migration barrier, it is generally assumed that two communities are less similar the further they are apart. Among the local factors determining

12

1 Introduction

protozoan communities moisture and food availability have been proposed to be most important (Stout 1984, Cowling 1994). Due to the expected latitudinal trend in turnover rates (see Part 2 below) a larger necrocoenosis of Testate Amoebae is expected at the boreal site as a consequence of low decomposition rates (Meisterfeld 1980).

1.4.2 Decomposer food web function � Part 2 The structure of the decomposer food webs along the transect is described by calculating the biomasses of the functional groups involved and creating a schematic illustration of the trophic interactions within the webs. A food web model approach considering site specific climate and resource quality is applied to estimate the mineralisation of carbon and nitrogen by each particular functional group at each site. The total food web biomass and the simulated total C and N fluxes mediated by the decomposer biota at the different sites are compared. The estimated mineralisation rates are evaluated with experimentally obtained data from the project databank. The contribution of functional groups to biomass and C and N mineralisation rates is compared. Patterns in the importance of particular groups along the transect are sought. The estimated fluxes are related to the environmental conditions of each site. Total decomposer food web biomass tends to decrease towards North due to adverse climate and limited nutrient availability (Swift et al. 1979). Hence it is hypothesised that total C and N mineralisation will be smaller towards the North. Considering the general views of Parmelee (1995) and Tietema (1998) a gradual shift between two extreme types of decomposer systems is expected: a fungal-based food web with slow turnover rates in the N-limited North and a bacterial-based food web with fast turnover rates in the strongly N-polluted South. Considering that Testate Amoebae are mainly bacterivorous (Bonnet 1964) and based on the review on Protozoan diversity by Coûteaux et al. (1998) the relative importance of Testate Amoebae for total fluxes is hypothesised to increase towards South. Due to their climatic sensitivity the Microarthropoda are expected to be of less importance for the food web functions in the North (e.g. limited feeding capabilities during ice formation; Seastedt 2000) and to increase in importance towards South. Following Huhta et al. (1998) Enchytraeidae are expected to be of larger importance to mineralisation at the Northern sites than in the South.

13

1 Introduction

1.4.3 Main hypotheses

Part 1 �� Due to decreasing latitude and increasing N deposition the abundance and diversity of Testate

Amoebae increases from North to South. �� Similarity between Testate Amoebae communities increases with decreasing distance between the

sites. �� Among the environmental parameters explaining the Testate Amoebae community structure,

moisture and microbial parameters are the most important factors. �� The ratio of Testate Amoebae biocoenosis to necrocoenosis increases towards South, due to

enhanced decomposition (disappearance of empty shells).

Part 2 �� Total decomposer food web biomass, C and N mineralisation increase towards South. �� The expected structural changes within the food web along the transect determine ecosystem

function (e.g. C and N mineralisation). �� The importance of fungi to C and N fluxes within the food web decreases towards South, while that

of bacteria increases. �� The importance of decomposer fauna to C and N fluxes within the food web increases towards

South.

14

Chapter 2

The Study Sites

2 The study sites

2.1 Site description The study sites were selected to form a North-South-transect of European coniferous forests: N-SE (Northern Sweden, Åheden), S-SE (Southern Sweden, Skogaby), DE (Germany, Waldstein) and FR (France, Aubure) (Figure 2.1). They cover a range of latitudes and depositional loads and belong to a number of forest sites that were investigated within the European projects NIPHYS (Nitrogen Physiology of Forest Plants and Soils) and CANIF (Carbon and Nitrogen Cycling in Forest Ecosystems).

Figure 2.1 Schematic map of the study sites lying on a North-South transect within Europe. Northern latitude is given beneath site abbreviation. Total N deposition is indicated: 0 = very low; N = intermediate; NN = high. See Table 2.1 for details and site abbreviations.

15

2 The Study Sites

The climate ranges from boreal at N-SE, over humid oceanic at S-SE and FR, to humid continental at DE (Table 2.1). The altitudinal difference between the sites partly counteracts the latitudinal gradient of the transect due to increasing altitude at the sites of lower latitude. Both, mean annual temperature and mean annual precipitation are lowest at the boreal site N-SE (Table 2.1) and strong temperature extremes occur (Figure 2.2). Mean monthly temperatures at N-SE reach a maximum of 14.9°C in July, which is close to the maximum temperatures at the other sites. However, the minimum temperature (-11.9°C) lies far below and the time of the year when temperatures are below zero is considerably longer. In contrast to the strongly fluctuating temperatures the precipitation pattern at N-SE is more balanced than at the other site but lies on a considerably lower level (Figure 2.2). Taking the temperature curve into account, the availability of liquid water is bound to be relatively limited at N-SE. The other sites are quite similar concerning their temperature curves, DE being constantly a little colder and S-SE being the warmest site, due to its comparably low altitude (Figure 2.2, Table 2.1). Both, FR and S-SE are wet sites, with high fluctuations in precipitation pattern. DE is subject to much higher amounts of precipitation than N-SE, but receives less precipitation than FR and S-SE. The Northern Swedish site N-SE receives next to no nitrogen via the atmosphere, while the Southern Swedish and the French site are subject to intermediate levels of atmospheric N deposition. DE receives the highest amount of N. Atmospheric S deposition follows the same pattern. The litter at all sites consists almost entirely of spruce needles (Picea abies (L.) KARST.), except for the Northern Swedish forest site that is a mixed stand. The nutrient content of the litter reflects the characteristic history and environmental conditions of each site. Nutrient concentrations were usually higher in the needle litter of Central European sites, e.g. the concentration of N, P, S and K was higher in needles from FR and DE than in needles from the Swedish sites (Bauer et al. 2000).

16

2 The Study Sites

-15.0

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

25.0

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

air te

mpera

ture (

°C)

N-SES-SEDEFR

A

020406080

100120140160180200

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

precip

itatio

n (mm

)

N-SES-SEDEFR

B

Figure 2.2 Mean monthly temperature (A) and precipitation (B) at the sites.

17

2 The Study Sites

A detailed description of the sites and their history can be found in Persson et al. (2000c). In the following a short summary of the major characteristics of each particular site will be given (see also Table 2.1, and section 5.1.3.2., Table 5.9 for further abiotic parameters): N-SE (Northern Sweden, Åheden) N-SE, the site in Northern Sweden (Åheden), is a 180-year-old unmanaged pine forest (Pinus sylvestris

L.), mixed with spruce (Picea abies (L.) KARST.) and birch (Betula pubescens EHRHART). P. sylvestris is the dominant species. This site lies at 175 m asl and is characterised by a bottom layer of forest mosses. The field layer is dominated by the dwarf-shrubs Vaccinium myrtillus L. and Vaccinium vitis-

idaea L. The climate is boreal, with the bud break in early June and a mean annual temperature of 1.0°C (Figure 2.3). The soil type is a regolsol on sand. N deposition is very low and the site may be considered as virtually undisturbed.

S-SE (Southern Sweden, Skogaby) S-SE, the site in Southern Sweden (Skogaby) is a young homogeneous P. abies plantation situated about 20 km from the sea in the South-Western part of Sweden at around 105 m asl. The P. abies stand is the second-generation forest and was planted to replace a planted pine forest in 1966. Until 1913 the area was grazed Calluna heathland. S-SE is the site with the youngest and most productive tree stand (Scarascia-Mugnozza et al. 2000). About 50 % of the bottom layer is made up by mosses. The grass Deschampsia flexuosa (L.) TRIN. occurs only in glades or wider gaps between adjacent tree stands. The climate is humid oceanic, with the bud break in mid may and a mean annual temperature of 7.6°C. The soil type is a haplic podzol on sandy loam. The site is subject to an intermediate level of N-deposition. DE (Germany, Waldstein) The German site DE (Waldstein) is located at the North-Western border of the Fichtel Mountains (North-East Bavaria) at 700 m asl. Forest plantation in this area began as early as in the 16th century. The area consists mainly of planted P. abies forests. The P. abies stand at DE is a 146-year-old plantation with a dense field layer vegetation dominated by Vaccinium myrtillus, Calamagrostis villosa (CHAIX) J. F. GMEL., and Deschampsia flexuosa. The climate is humid continental with the bud break in late April and a mean annual temperature of 5.5°C. The soil type is a cambic podzol on loamy sand.

18

2 The Study Sites

FR (France, Aubure) The French site FR (Aubure) is located in the Strengbach catchment at the North-Eastern side of the Vosges Mountains at 1050 m asl. The P. abies stand is situated at a South-facing slope and is a 92-year-old forest planted after an old grazed declining fir forest. Since 1983 the canopy is partly defoliated (about 30 %) and some needles are yellow and deficient of magnesium. Patches of fern (Dryopteris filix-

mas (L.) SCHOTT) and grass (Deschampsia flexuosa) make up the field layer. The climate of the site is humid oceanic, with the bud break in late April and a mean annual temperature of 5.4°C. The soil type is a dystric cambisol on sandy loam.

Figure 2.3 Astrid Taylor & Anne Pflug during our autumn sampling at Åheden, N-SE.

19

Table 2.1 Characteristics of the four selected coniferous sites (from data given in Persson et al. 2000c). N-SE S-SE DE FR

dominant tree species Pinus sylvestris Picea abies

Betula pendula

Picea abies Picea abies Picea abies

understorey vegetation

dense layer of forest mosses, dwarf-shrubs

occasional mosses dense field layer of grasses and dwarf-shrubs

patches of grass and fern

latitude, longitude 64°13’ N, 19°30’ E 56°33’ N, 13°13’ E 50°12’ N, 11°53’ E 48°12’ N, 07°11’ E altitude asl (m) 175 95-115 700 1050 climate boreal humid oceanic humid continental

humid oceanic

mean annual air temperature (°C)

1.0

7.6 5.5 5.4

mean annual precipitation (mm)

488 1237 890 1192

bud break early June mid May late April late April stand age in 1995 (a)

180 33 142 92

type of stand natural planted planted planted total S deposition (kg S ha-1 y-1)

6 13 17 12

total N deposition (kg N ha-1 y-1)

2 16 20 15

P:N ratio of needlesa 0.21 0.08 0.11 0.16C:N-ratio organic layerb

39 29 22 26

organic layer C (10-3 kg C ha-1)

22.6 27.9 38.9 29.5

soil type regosol on sand haplic podsol on sandy loam

cambic podsol on loamy sand

dystric cambisol on sandy loam

site history umanaged, virtually undisturbed site

2nd generation, planted, former grazed Calluna heathland

planted planted, former grazeddeclining fir forest

a calculated from Bauer et al. (2000). b calculated from Persson et al. (2000c)

2 The Study Sites

2.2 Sampling scheme and sample treatment Samples were collected at four sampling times (Table 2.2): (i) October / November 1996, (ii) May / June 1997, (iii) September 1997, and (iv) March / April 1998. At each time, between 80 and 110 samples (soil corer: Ø 5 cm, length 12 cm) of the organic layer (litter, fermentation and humus layer, LFH) were taken at each site.

Table 2.2 Sampling times at the four sites. See Table 2.1 for site abbreviations. 1st sampling 2nd sampling 3rd sampling 4th sampling N-SE 02.11.1996 27.06.1997 07.09.1997 18.04.1998 S-SE 06.11.1996 21.06.1997 15.09.1997 16.04.1998 DE 08.10.1996 28.05.1997 29.09.1997 29.03.1998 FR 16.10.1996 27.05.1997 29.09.1997 29.03.1998

At the first sampling occasion 10 bulk samples were obtained at each site by merging 10 single soil cores per sample. An additional 10 soil cores were drawn to determine site specific organic layer thickness, bulk density and dry-mass-to-area ratio of the organic layer. After material had been taken out for the extraction of Microarthropoda, the 10 bulk samples were bulked again in pairs to deliver 5 bulk samples for the remaining measurements. This scheme results in 10 resp. 5 replicate samples of organic layer per site. For the subsequent sampling times (samplings 2-4) the sampling scheme was slightly altered. Eight

bulk samples were obtained by merging 7-10 single soil cores per sample. An additional 24 (3 � 8)

single soil cores were drawn and treated separately. Of these single soil cores 8 were used to extract Nematoda, another 8 to extract Enchytraeidae and the remaining 8 to extract Microarthropoda. The site specific organic layer thickness, bulk density and dry-mass-to-area ratios were also obtained from these 24 single soil cores. All other measurements were made with material from the bulk samples. This scheme results in 8 replicate samples of organic layer per site and sampling time. In the laboratory the single soil cores for faunal extractions were separated from living plants (mosses etc.). The bulk samples were mixed cautiously but thoroughly by hand, and bigger pieces of wood, twigs and living plant material were carefully removed. All fauna extractions and microbial measurements started within 3 days time after sampling. The samples were stored in the dark at 4° C in sealed polyethylene bags. Prior to microbial analyses and pH measurements the material was sieved using a 4 mm mesh.

21

Chapter 3

Material and Methods

3 Material and methods

3.1 Functional groups of organisms

3.1.1 Microflora: fungi and bacteria Soil microbial carbon (Cmic) was determined using the fumigation extraction method (section 3.1.1.1). To distinguish bacterial from fungal biomass additional measurements were undertaken with material collected at the 4th sampling time (ergosterol, section 3.1.1.2; direct counting of bacteria, section 3.1.1.3). A site specific bacterial-to-fungal-biomass-ratio was calculated and used to calculate the biomass pools of bacteria and fungi from the Cmic measurements. It was assumed that the measurements at the 4th sampling time appropriately estimate the average bacterial-to-fungal-biomass-ratio at each site. To estimate microbial activity the metabolic potential of the microflora was measured and a metabolic quotient was calculated (section 3.1.1.4). Prior to microbial analyses the material was sieved using a 4 mm mesh.

3.1.1.1 Chloroform fumigation extraction method (CFE): microbial carbon (Cmic) Chloroform fumigation causes cell lysis of microorganisms. The increase in extractable C following chloroform fumigation of soil was used to estimate the amounts of C held in the microbial biomass (Cmic) (Vance et al. 1987). From each sample (sieved, fresh organic material) two aliquots of an amount corresponding to 2 g DW each were weighed into two bottles. Half the aliquots were fumigated prior to extraction. For this, open bottles were put into an exsiccator. Two empty bottles served as controls. The bottom of the exsiccator was covered with moist paper towels and the exsiccator contained a beaker whose ground was covered with lime pellets. Among the sample bottles a beaker with approx. 50 mL ethanol-free chloroform (CHCl3) containing boiling stones was put. The exsiccator was evacuated until the chloroform boiled vigorously for approx. 2 min. The exsiccator was closed and the samples remained in the chloroform atmosphere in the dark at 20°C for 24 h. The exsiccator was then aerated and the paper towels as well as the beakers containing lime pellets and remaining chloroform were removed. The exsiccator was then repeatedly (at least ten times) evacuated and aerated to completely remove chloroform fumes (until no more smell of chloroform was detected from the samples). Fumigated as well as unfumigated samples were extracted using potassium sulfate solution. To each

23

3 Material and Methods

sample as well as to the controls 90 mL of K2SO4 solution was added (ratio sample to extractant: 1:45 w/w). The samples were shaken for 30 min (250 rpm) and percolated over filters. The extract was stored in polyethylene vials at –18°C. The extracts were analysed photometrically after defrosting using a continuous flow system (PERSTORP

ANALYTICAL GmbH, Perstorp, Sweden). For this the samples were acidified with sulphuric acid (1 N H2SO4) to convert mineral carbon to carbon dioxide (CO2), which was trapped in sodium hydroxide solution (1 N NaOH). The CO2-free samples were then merged with saturated persulfate solution (K2S2O8) and irradiated with UV radiation to ensure complete oxidation of organic carbon to CO2. The carbon dioxide was diffused through a silicone membrane and received by a weakly buffered phenolphthalein indicator solution. The decrease in the colour of the indicator is proportional to the carbon concentration in the extracts and was measured photometrically. The method was calibrated using potassium biphthalate solution (HO2CC6H5CO2K). Calculation The C content of the organic material was calculated from the C concentrations of the extracts as follows (equations 3.1 and 3.2):

� �DW1000

VV H2Oextr

�

��

�

nm

equation 3.1 m C content of organic material (mg g–1 DW) n C concentration of extract (mg L-1) Vextr volume of extractant (mL) VH2O volume of water in sample (mL) DW dry weight of sample (g) mbiomass = (mfumigated – munfumigated) k equation 3.2 mbiomass microbial C (mg g-1 DW) mfumigated C content in fumigated sample (mg g–1 DW) munfumigated C content in unfumigated sample (mg g–1 DW) k proportionality factor (k = 2.22, taken from Wu et al. 1990)

Equipment & reagents 80 mL bottles with screw caps; shaker; filter (SCHLEICHER & SCHÜLL 595 ½); glass funnels; 100 mL beakers; exsiccators; paper towels; boiling stones; lime pellets (NaCO3); polyethylene vials (ROTH,

24


article no. 0794.1); 0.5 M K2SO4 solution; chloroform (MERCK, article no. 2445, stabilised with 20 ppm 2-methyl-2-buten). 3.1.1.2 Ergosterol The ergosterol content of soil is an indicator of living fungal biomass (Ekblad et al. 1998). It may be used as a marker of the biomass of saprophytic and ectomycorrhizal fungi (Nylund and Wallander 1992). It was determined using HPLC analysis (Djajakirana et al. 1996) in the laboratory of Dr. Rainer Joergensen (Institut für Bodenwissenschaft, Georg-August-Universität, Göttingen). 3.1.1.3 Direct counting of bacteria The number of living bacteria cells and their cell volume was measured using automatic confocal laser-scanning microscopy picture analysis after fluorescent staining and bacterial biomass was calculated from these parameters (Bloem et al. 1997). The measurements were carried out in the laboratory of Dr. Jaap Bloem (Research Institute for Agrobiology and Soil Fertility (AB-DLO), Haren, The Netherlands). 3.1.1.4 Metabolic potential and metabolic quotient qCO2 The metabolic potential of the microflora within the organic material was estimated by measuring the metabolic release of carbon dioxide (respiration) at 10 °C and at an optimal water content (300 % DW). The amount of CO2 released per unit microbial biomass (metabolic quotient qCO2 (µg CO2-C g-1 DW h-

1)) is a measure of the microbial activity and was calculated by dividing the CO2 release by the microbial biomass in the same sample.For comparison with other studies it must be kept in mind that this particular metabolic quotient was calculated using the potential metabolic release of CO2 and thus represents a potential metabolic quotient. The material was sieved (4 mm mesh size) and an amount of fresh litter that equals 10 g DW (dry weight) is weighed into a microcosm (jar, volume 0.75 L) which was then sealed air-tight. For each sample two replicate microcosms were set up. Additionally four empty microcosms were treated identically to serve as control sets. The microcosms were pre-incubated in the dark at 10°C for five days to let the microbial community adapt to 'normal' activity after the sieving which usually causes a respiration peak due to release of substrate after tearing of fungal hyphae etc. After pre-incubation the water content was adjusted to 300 % DW by adding tab water if necessary. Then a small vessel filled with 4 mL 1.0 N NaOH was put into each microcosm and the microcosms were sealed air-tight. The microcosms were incubated for 6 days at 10°C in the dark.

25


During the incubation period carbon dioxide (CO2) that evolved from the organic material was trapped in the sodium hydroxide solution. After the incubation BaCl2 was added to the vessels in excess and reacted with dissolved CO2 to form BaCO3, an insoluble salt that precipitates and was thus removed from the reaction equilibrium, therefore shifting the reaction towards a complete consumption of the CO2. Simultaneously the lye is neutralised while NaCl is formed:

2 NaOH + BaCl2 + CO2 �� 2 NaCl + BaCO3 +H2O

After the incubation two aliquots of 0.5 mL were taken from each vessel and titrated with hydrochloric acid (0.1 N), using phenolphthalein as indicator. By means of substraction, the amount of metabolically formed CO2 that reacted with the sodium hydroxide was determined. Calculation of results The rate of respiration was calculated as follows:

� �

inc

SC

DWDE

tTNVVR

�

��

� equation 3.3

R respiration rate (mg CO2-C g-1 DW h-1) VC volume of acid needed to titrate the NaOH in control sets (mL) VS volume of acid needed to titrate the NaOH of sample (mL) E equivalent weight of CO2-C (= 6 mg mL-1) N normality of the acid (mol L-1) D factor of dilution of NaOH (= 8; because aliquots of 0.5 ml are taken from the vessels containing

a total of 4 mL) T titer of the acid tinc incubation time (h) DW dry weight of sample (g)

Equipment & reagents Microcosms (jars, volume 0.75 L) with air-tight lids; pipette; beakers; dark chamber with constant temperature of 10° C; burette or titration automat; small vessels for the alkali; glass-vials with air-tight lids for storage of the aliquots prior to titration; sodium hydroxide (NaOH) solution, 1.0 N; barium chloride (BaCl2) solution, 0.5 M; phenolphthalein indicator; hydrochloric acid (HCl) solution, 0.1 N.

3.1.2 Testate Amoebae Testate Amoebae were counted on species level. Most probable number culturing techniques as used for Naked Amoebae and Flagellata are insufficient to estimate the density of Testate Amoebae (Bunt and Tchan 1955, Foissner 1987, Ekelund and Ronn 1994). Thus a direct counting method using an

26


inversed microscope was used (Meisterfeld 1980, modified as described below). Testate Amoebae species were subsumed into five size classes (see section 5.1.1, Table 5.1 and 5.2) and biomass was calculated from the abundance of shells that were filled with cytoplasm using conversion factors from the literature (Volz 1951, Schönborn 1975, Schönborn 1977, 1981, 1982, Lousier and Parkinson 1984, Lousier 1985, Schönborn 1986, Wanner 1991). If useful measures from the literature were lacking the biomass of the species was calculated using measurements of cell length, width and height of at least 10 specimen and an ellipsoid formula (Heal 1965, Schönborn 1977). Specific gravity of the cytoplasm and C content of Protozoa were taken from the literature (specific gravity: 1.05 g mL-1 Schönborn 1981; C content 50 % DW, see Table 10.1, Berg 1997). Testate Amoebae species were assembled into two feeding groups: panphytophagous and predaceous species. The latter group comprises of the genera Nebela and Heleopera (Bonnet 1964 in Coûteaux 1976, Laminger 1980). See section and 5.2.2 for details. 3.1.2.1 Fixation and staining of substrate samples for quantitative analyses An aliquot of fresh material from bulk samples (1.00-16.00 g) was weighed into polyethylene vials and topped with alcoholic aniline blue within two days after soil sampling. The staining with aniline blue allows differentiation of three types of shells: empty shells, shells that were active at the moment of fixation, and cysts (Schönborn 1978, Aescht and Foissner 1992).

Equipment & reagents 1.5 g L-1 aniline blue ('waterblue') in 70 % alcohol (shake solution for 20 min); polyethylene vials (ROTH, article no. 0794.1). 3.1.2.2 Direct counting of Testate Amoebae The fixed sample was washed from the polyethylene vial into a 100 mL measuring cylinder with tab water. The volume was recorded and the suspension was transferred completely into a 250 mL or 500 mL round-bottomed flask, depending on density of material and expected amount of Protozoa in the sample. The polyethylene vial and the measuring cylinder were rinsed several times to transfer the entire material. When rinsing, the additional amount of water used was recorded each time. The suspension in the round-bottomed flask was filled up with tab water to a volume of 200 resp. 450 mL. The suspension was shaken for 20 min on a shaker with circular motions (600-700 rpm). Instantly after

the shaker stopped an aliquot of 100 �L or 500 �L (depending on the density of the suspension) was

taken from the middle of the round-bottomed flask using an EPPENDORF pipette. The very tip of the

27


pipette (approx. 1 mm) was clipped off to widen the opening and avoid the selective intake of smaller soil particles. The aliquot droplet was laid into a counting chamber that had been pre-filled with tab water and was already sat in the inversed microscope observation chamber holder, thus allowing the suspension to settle evenly. The round-bottomed flask with the remaining suspension was sealed with a rubber cap and stored in the dark at 4°C. The suspension in the counting chamber was allowed to settle

for at least 10 min. The entire chamber was then observed with an inversed microscope at 100�

magnification. Each specimen of Testate Amoebae found was recorded on species level (see section 3.1.2.6). Empty shells, active cells and cysts were differentiated. At least two aliquots of the aqueous suspension were counted. If the total of specimen found in one counting chamber deviated more than 10 % from the previous count, further aliquots were counted until

their deviation from the mean of the previous counts was � 10 %.

Larger species of Testate Amoebae occurred with much smaller frequency than small species. Therefore two different soil suspensions were counted for each soil sample. A light one, resulting in around 0.0005 g DW of substrate in the counting chamber, and a dense suspension, resulting in around 0.002 g DW of substrate in the counting chamber. In the light suspension larger shells occurred only rarely. But even small and hyaline species (e.g. Corythion dubium, Trinema lineare, Cryptodifflugia

oviformis) could easily be detected and the problem of losing smaller Testate Amoebae that may be masked by soil particles (Foissner 1987) was circumvented. In the dense suspension the smaller shells were hidden by soil particles, but the larger shells (length of 70 µm and more, e.g. Nebela spec.,

Trigonopyxis arcula, Centropyxis matthesi) could easily be seen and were recorded with a frequency that allowed extrapolation to the abundance in the sample. From each soil sample at least 2 aliquots of a light suspension and at least 2 more aliquots of a dense suspension were counted.

Equipment & reagents Inverse microscope (ZEISS Axiovert 135, magnifications: 100�, 250�, 400�, 1000�); counting chamber made of perspex with cover slip bottom (� 1.5 cm, depth 0.3 cm); 100 µL EPPENDORF pipette; 500 µL EPPENDORF pipette; 100 mL measuring cylinder; 250 and 500 mL round-bottomed flasks; funnel; spatula; 1 L water bottle; shaker; cat's whisker or human eyelash in collar holder (see sections 3.1.2.6 below); embedding fluids (Euparal, Naphrax, glycerine; see sections 3.1.2.7-9 below); object slides, cover-slips; micropipette (micro-capillary and flexible hose with mouth piece); conc. alcohol (approx. 96 %) (see section 3.1.2.6 below).

28


3.1.2.3 Flotation method: extraction of empty shells A simple method to extract empty shells from soil samples is the flotation-method (Chardez 1959). With this enrichment method empty Testate Amoebae shells can easily be accumulated for qualitative studies, e.g. to obtain enough specimen for exhaustive taxonomic evaluation and the compilation of a species list for a certain site. 250-500 mL soil material was spread out and left at room temperature until air-dry. The material was then passed subsequently through a coarse (mesh size 1 mm) and a fine (mesh size 0.250 mm) sieve. The sieving residue was discarded and the remaining, fine material was mixed with tab water in a large beaker using a glass rod for stirring. The mixture was then agitated using an aquarium's pump to bubble air through the sample for 1 min. The mixture was left to settle for another minute. While most soil particles sedimented the air-filled empty Testate Amoebae shells floated at the water surface and were collected at the wall of the beaker using a pipette. They were collected in a vial and put into an exsiccator. The exsiccator was evacuated so that the shells sedimented. The suspension was fixated with conc. formaldehyd (final concentration approx. 2 %) and stored in the dark at room temperature. 3.1.2.4 Batch cultures To be sure not to miss any species, batch cultures from each sampling occasion were kept for several weeks. To obtain the cultures 50-100 mL of organic soil material were put into jars and kept moist in the dark at 10°C for at least 6 weeks. Once a week liquid from the cultures was extracted by gently pressing the material to obtain run-off. The run-off was observed under the inverse microscope for live specimen

(magnification 100�).

3.1.2.5 Live observations For direct live observations 1-5 mL of organic soil material was suspended in 2-10 mL of tab water, and stirred or shaken for at least 1 min. An aliquot of this suspension was observed under the inverse

microscope for living specimen (magnification 100�).

3.1.2.6 Taxonomic determination In most cases the shell morphology allowed species determination. For this the shell often needed to be examined from all sides and the interior structures needed to be made visible. The shell was tossed around using a cat's whisker or a human eyelash fixed to a collar holder, manipulating the shell carefully

without disturbing the rest of the sample. Magnification was increased from 100� to 250�, 400�, 1000�

29


(oil) if necessary. Differential interference contrast (DIC) as well as phase contrast was used to increase visibility of low-contrast hyaline structures. In rare cases the shell was sucked out of the counting chamber using a micropipette and embedded in fixation fluid on an object slide to facilitate observation (e.g. glycerine, Euparal, Naphrax; see sections 3.1.2.7-9 below). Capped with a cover slip the slide served as a reference and may be stored for several years (edges were sealed with nail polish). No complete standard taxonomic works or determination keys can be found for Testate Amoebae. Therefore original species' descriptions, monographies of genera and compilation of biometric data as well as reviews of certain genera occurring in soil and freshwater were used. Tables 3.1 and 3.2 give an overview of the taxonomic literature.

Table 3.1 General taxonomic literature for the determination of Testate Amoebae. General determination literature Short description Meisterfeld (2001a, b) key to the genera, review of most important species Bonnet and Thomas (1960) review of species occurring in soil Grospietsch (1965) key to the genera Lüftenegger et al. (1988) review of biometric data Lüftenegger and Foissner (1991) review of biometric data Ogden and Hedley (1980) species descriptions (REM-illustrations) Rauenbusch (1987) soil species descriptions (REM-illustrations) Richter (1995) soil species descriptions Schroeter (1995) soil species descriptions

30

Table 3.2 Specialised taxonomic literature and monographies for the determination of Testate Amoebae. genus determination literature

Arcella EHRENBERG 1830 Deflandre (1928b), Decloitre (1976a, 1979b, 1982) Argynnia VUCETICH 1974 Assulina EHRENBERG 1871 Centropyxis STEIN 1957 Deflandre (1929), Decloitre (1978, 1979b, a, 1982), Rauenbusch (1987) Corythion TARANEK 1881 Decloitre (1960) Cryptodifflugia PENARD 1890 Grospietsch (1964), Page (1966) Cyclopyxis DEFLANDRE 1929 Decloitre (1977a, 1979b, 1982) Difflugia LECLERC 1815 Ogden (1983), Ogden and Zivkovic (1983) Edaphonobiotus SCHÖNBORN, FOISSNER & MEISTERFELD 1983 Euglypha DUJARDIN 1841 Coûteaux, et al. (1979), Chardez (1987b, a), Decloitre (1962, 1976b, 1979b, 1982) Geopyxella BONNET & THOMAS 1955 Bonnet (1974) Heleopera LEIDY 1879 Hyalosphenia STEIN 1857 Grospietsch (1965a) Microchlamys COCKERELL 1911 Microcorycia COCKERELL 1911 Nebela LEIDY 1874 Deflandre (1936), Decloitre (1977b, 1979b, 1982), Meisterfeld and Schüller (1982), Heal (1963) Phryganella PENARD 1902 Chardez (1969) Plagiopyxis PENARD 1910 Thomas (1958) Playfairina THOMAS 1961 Pseudawerintzewia BONNET 1959 Schoenbornia DECLOITRE 1964 Schönborn, et al. (1987) Schwabia JUNG 1942 Tracheleuglypha DEFLANDRE 1928 Deflandre (1928a), Coûteaux and Ogden (1988) Trachelocorythion BONNET 1979 Trigonopyxis PENARD 1912 Trinema DUJARDIN 1841 Decloitre (1981, 1982)


3.1.2.6.1 Distinguishing Centropyxis aerophila sphagnicola and C. sylvatica Although characteristic in their shell size the definite distinctive feature separating Centropyxis sylvatica (DEFLANDRE 1929) BONNET & THOMAS 1955 and C. aerophila sphagnicola DEFLANDRE 1929 is an inner diaphragm. This diaphragm is formed in C. sylvatica by the ventral lip of the pseudostome extending well into the shell and by a rim reaching from the dorsal shell wall toward the ventral lip (ventral = the side of the shell that opens in a pseudostome; dorsal = the side of the shell opposite to the pseudostome). C. aerophila sphagnicola lacks the dorsal rim. In this study the two species were differentiated by their length, because of the difficulty to judge the inner architecture of the shells under

the optical conditions when counting in watery suspension. Shells shorter than 68 �m were assigned to

C. aerophila sphagnicola, longer shells (� 68 �m) were assigned to C. sylvatica.

3.1.2.6.2 Distinguishing Cyclopyxis eurystoma and Phryganella acropodia The genera Phryganella is distinguished from the lobose genera Cyclopyxis and Difflugia only by its reticulolobose form of pseudopodia (Chardez 1969). Most observations within this study, however, were made with fixed material, rendering the observation of pseudopodia impossible. Therefore the species Cyclopyxis eurystoma DEFLANDRE 1929 and Phryganella acropodia (HERTWIG & LESSER 1874) HOPKINSON 1909, having a very similar shell architecture, were differentiated from each other according

to a size limit (see circumstantial discussion in Schroeter 1995). Shells with a diameter below 55 �m

were assigned to P. acropodia. 3.1.2.6.3 The taxon Euglypha cf. strigosa Some specimen of Euglypha could not without doubt be identified as Euglypha strigosa (EHRENBERG

1872) WAILES & PENARD 1911 because the mouth scales were obscured by debris. Furthermore specimen with and without spines and every degree of spine covering in between were found. It seems likely that E. strigosa sometimes loses its spines secondarily during the course of its lifetime. All these specimen were assigned to the taxon Euglypha cf. strigosa. Possibly this morphologic group also included the very similar species and forms Euglypha ciliata (EHRENBERG 1848) LEIDY 1878, E. ciliata f. glabra WAILES 1915, E. pseudociliata CHARDEZ 1962 and E. pseudociliata f. glabra CHARDEZ 1962. 3.1.2.6.4 The taxa Nebela parvula/tincta and N. tincta major/bohemica/collaris Nebela parva CASH & HOPKINSON 1909 and N. tincta (LEIDY 1879) AWERINTZEW 1906 differ from each other by the presence of two lateral pores above the pseudostome of the latter. Since this feature may

32


not be diagnosed unambiguously under the given optical conditions when counting, the two species were not distinguished but pooled as the taxon N. parvula/tincta.

Another group of shells has the appearance of N. parvula/tincta but exceeds a length of 95 �m. This

group consists of Nebela tincta major DEFLANDRE 1936, N. bohemica TARANEK 1881 and N. collaris (EHRENBERG 1848) LEIDY 1879. N. tincta major differs from the other two in having two lateral pores above the pseudostome but, as above, these cannot explicitly be judged in watery suspension. N.

collaris has a pseudostome with a curved rim in lateral view while the rim of N. bohemica appears as a straight line. However, both species cannot unequivocally be distinguished from each other (Heal 1963). Thus the three species were pooled together as the taxon N. tincta major/bohemica/collaris (cf. Schroeter 1995). 3.1.2.6.5 Distinguishing Trinema enchelys and T. lineare Following Lüftenegger et al. (1988) Trinema enchelys (EHRENBERG 1838) LEIDY 1878 may be

distinguished from T. lineare PENARD 1890 by its length being above 40 �m. Risk of misidentification is

considered to be very small, since the sizes overlap only in rare cases. Therefore, in this study length was used as distinctive feature. 3.1.2.7 Slide preparation with Euparal Embedding the Testate Amoebae in Euparal facilitates the examination of the interiors of the shell. This is due to the higher refraction index of Euparal (compared to water) making the shell and cell parts appear to be lighter. Under microscopic observation one or several specimen were transferred from watery suspension into the centre of a cover-slip lying on top of an object slide using a micropipette with a flexible hose and a mouth piece. A droplet of 96 % alcohol was used to supersede the water. When most of the alcohol had evaporated a small droplet of Euparal was put on top of the specimen and the cover-slip was turned and laid on the object slide. If the specimen were big, two short clips of human hair were laid alongside the specimen into the Euparal before turning the cover-slip to provide some footing and prevent the shell from being squashed when the Euparal dried. The slide preparation was kept upside down at room temperature, so that the specimen stayed close to the cover-slip and microscopic observation was not deterred by a thick layer of Euparal between the cover-slip and the specimen.

33


3.1.2.8 Slide preparation with Naphrax This slide preparation method was used for empty idiosome shells, especially of the genus Euglypha. The embedding fluid Naphrax (which is also used to embed diatoms) increases the contrast of the hyaline structures of silicious oxide shells that are characteristic of Testate Amoebae with ideosomes. The specimen were transferred from watery suspension into the centre of a cover-slip lying on top of an object slide using a micropipette with a flexible hose and a mouth piece. The shells were left to dry out completely and Naphrax was dropped on top of them. The cover-slip was turned and put on the slide. The slide preparation was carefully heated to about 50°C to liquefy the medium and expel bubbles of air. The slide was stored upside down at room temperature like the Euparal slide preparations. 3.1.2.9 Slide preparation with glycerine Like Euparal (section 3.1.2.7) glycerine lightens Testate Amoebae shells in the microscopic observation due to its optical density. Especially shells filled with cytoplasm were embedded in glycerine, since this medium imposes only a minor osmotic stress on the cytoplasm and nucleus and such structures are well-preserved. This method is especially well suited for observations using differential interference contrast (DIC). The handling was the same as for Euparal slide preparations, save that pure glycerine was taken instead of the Euparal. Glycerine slide preparations were not stored. 3.1.2.10 Photography The inverse microscope had an extra tubus equipped with a camera (CONTAX 167 MT). For microscopic photography the following films were used: Agfa Agfapan APX 25, Ilford b/w 125, Agfacolor 100.

3.1.3 Nematoda A modified O'CONNOR-wet-funnel-extraction followed by milk-filter-cleaning (s'Jakobs and van Bezooijen 1984) was used to extract Nematoda (1.5 d at 20°C + overnight temperature-increase to 60°C in wet funnels while cooling the sample below to 15°C, followed by 2 d milk-filter-cleaning at 20°C). The nematodes of the sampling occasions 1-3 were counted alive without further taxonomic determination. Nematodes of the 4th sampling occasion were killed by heat, stored in 4 % formaldehyde, counted on genus level (distinguishing juveniles from adults) and assembled into four feeding groups (see Table 3.3, extracted from the detailed Table of feeding habits of nematode genera in Yeates, et al. 1993). The taxonomic Nematode work was carried out by Dr. Ralf Lenz (Mainz, Germany). Nematode biomass was calculated from genus specific abundances using conversion factors from the literature (Berg 1997,

34


Ekschmitt et al. 1999) or calculated using the formula from Andrássy (1956) and length and width estimates from Bongers (1994). Body volume of juveniles was estimated to be on average 22 % of the adult body volume (Ilja Sonnemann, pers. com.). A list of the conversion factors used is given in the appendix (chapter 10, Table 10.1 and 10.2).

Table 3.3 Nematode genera found on the sites and their feeding habits according to Yeates (1993). Following the simplified classification for the food web model further food sources or feeding modes that may occur are given in parentheses. Genus feeding habit Acrobeloides bacterivorous Acrolobus bacterivorous Alaimus bacterivorous Bunonema bacterivorous Geomonhystera bacterivorous (detritivorous) Heterocephalobus bacterivorous Metateratocephalus bacterivorous Panagrolaimus bacterivorous Plectus bacterivorous Prismatolaimus bacterivorous Pristionchus bacterivorous (predaceous) Rhabditis bacterivorous Teratocephalus bacterivorous Wilsonema bacterivorous Aphelenchoides fungivorous (epidermal cell and root hairs, migratory

endoparasite of plants, algae and lichens) Hexatylus fungivorous Laimaphelenchus fungivorous (predaceous, algae and lichens) Tylencholaimus fungivorous Prionchulus predaceous ("ingester") Seinura predaceous ("piercer") Eudorylaimus omnivorous (predaceous) Filenchus omnivorous (epidermal cell and root hairs) Malenchus omnivorous (epidermal cell and root hairs) Tylenchus omnivorous (algae and lichens, fungivorous)

3.1.4 Microarthropoda (Collembola and Acari) Microarthropods were extracted by means of the high-gradient-canister method (Macfadyen 1953, Kempson et al. 1963, Wolters 1983) into ethyleneglycol and transferred to ethanol (70 %) for storage. Organisms were sorted and counted using a dissecting microscope (Leica Wild M3C, magnification 40-

64 �). For more detailed information on the extraction procedure and determination see Pflug (2001).

3.1.4.1 Collembola All Collembola work was carried out by Anne Pflug (Department of Animal Ecology & Zoology, University Giessen, Germany). Collembola were counted on species level and aggregated into two

35


feeding groups. The genus Frisea is judged as predaceous, the remaining genera were judged as being panphytophagous. The body length of each species was calculated by taking the mean of all length measurements given in the determination keys (Gisin 1960, Palissa 1964, Fjellberg 1980, Zimdars and Dunger 1994, Jordana et al. 1997, Fjellberg 1998, Pomorski 1998). Collembola biomass was then calculated using the formula given by Persson & Lohm (1977) and the C content of Collembola given by Berg (1997, see Table 10.1 in the Appendix). Some regression coefficients for species were also taken from the calculation of Tanaka (1970) and Petersen (1975). For species not mentioned in these publications parameters of species of the same genus, family or suborder with similar body shape were used. Juveniles and adults were treated separately in the biomass calculations to account for the species specific smaller size of juvenile Collembola (for details see Pflug 2001). 3.1.4.2 Acari The Acari work was carried out by Astrid R. Taylor (Department of Animal Ecology & Zoology, University Giessen, Germany). Acari were counted on species level and sorted into two feeding groups (Luxton 1972, Walter and Proctor 1999): panphytophagous (juvenile and adult Cryptostigmata, Astigmata, Prostigmata and unidentified juveniles) and predaceous (juvenile and adult Mesostigmata) Acari. Length and width of a particular species were either taken from the literature that was used for species determination (Sellnick 1928, Willmann 1931, Giljarov and Krivolutsky 1975, Berg et al. 1990, Wunderle et al. 1990, Beck and Woas 1991) or measured for at least 3-10 specimen. Biomass was then calculated using the formulas and the dry-weight-to-live weight-ratio given by Persson & Lohm (1977) and the C content of Acari given by Berg (1997, see Table 10.1 in the Appendix). Nymphs and adults were treated separately to account for the species specific smaller size of juvenile Acari (for details see Taylor 2001).

3.1.5 Enchytraeidae Enchytraeidae were extracted using a modified O'CONNOR-wet-funnel method (O'Connor 1955, stepwise increase of temperature from 20-50°C in 5°C steps lasting 0.5 h each, while cooling the sample below to 15°C, entire procedure: 3.5 h). Enchytraeidae were counted alive under dissecting microscopes at

10� magnification without further taxonomic determination. Enchytraeid biomass was calculated from

abundances using a conversion factor from the literature (Heal 1967, Persson and Lohm 1977, Petersen and Luxton 1982, Dunger and Fiedler 1989, Górny and Grüm 1993, Berg 1997, see Table 10.1 in the Appendix).

36


3.2 Abiotic parameters

3.2.1 Water content (WC) From each sample about 3.00 g fresh weight (FW) of organic material was weighed into a beaker of known weight. The material was dried at 105°C for at least 24 h, cooled in an exsiccator over silicagel and weighed again (Alef 1991). The water content (WC) is expressed as percentage of dry weight (DW): WC (% DW).

3.2.2 pH of organic layer The organic material was sieved (mesh size 4 mm) before pH determination. Two sub-samples were drawn from each sample. For each sub-sample an amount of fresh litter equalling 2.00 g dry weight (DW) was weighed into a beaker. 20 mL distilled water (ratio litter-to-water 1:10) was added and the mixture was stirred with a glass rod until all needles were soaked. The samples were shaken for 60 min using a mechanical shaker. A calibrated pH-meter was used to measure the pH in the supernatant of each sample while gently moving the electrode.

3.2.3 Thickness of organic layer, mass-to-area-ratio and bulk density Thickness of organic layer (litter, fermentation and humus layer, LFH layer) was measured from single soil cores in the field (see sampling scheme section 2.2). Soil cores were then weighed and the mass-to-area-ratios and bulk density was calculated from the dry weight and the base area of the soil corer.

3.2.4 C:N-ratio of organic layer Total C and N content of the organic material was measured at the 4th sampling time only. The material was sieved (4 mm mesh size) and dried at 105°C. The material was powdered and the elements were measured gas chromatographically after dry combustion with oxygen using a Carlo Erba Elemental Analyser. The measurements were carried out in the laboratory of Dr. Rainer Joergensen (Institut für Bodenwissenschaft, Georg-August-Universität, Göttingen). The C:N-ratio was calculated from the C and N contents (%).

37


3.3 Statistical analyses Statistical analyses were carried out using STATISTICA 5.0 (Statsoft Inc., Tulsa, USA) and CANOCO for Windows 4.02 (Research Institute for Agrobiology and Soil Fertility (AB-DLO), Wageningen, The Netherlands; ter Braak and Smilauer 1998).

3.3.1 Analysis of variance Analyses of variance (ANOVAs) were performed to test for significant differences between means. Whenever possible two-way-ANOVAs were carried out including the factors 'site' and 'time'. All sites were sampled on four occasions covering the seasons of a whole year (see Table 2.2). However, this study was not designed to investigate population dynamics of the groups of organisms observed, nor to characterise fluctuations of the abiotic parameters assessed. Rather the aim of this study was to characterise the study sites with yearly averages reflecting the long term situation at each site. Nonetheless the factor 'time' is not neglected from the analyses, because it often explains an important part of the variance within the data sets. Strong fluctuations with time are mentioned whenever this is necessary to understand restrictions within the comparison of mean values per site or when the variance with time delivers some insight of general interest. Whenever necessary, data were adequately transformed prior to statistical analyses (arcsin for percentage values; logarithm for non-percentage values). The Tukey honest significant difference test (Tukey HSD test) was used to perform multiple comparisons between pairs of means of parameters. Homogeneity of variances was tested using the SEN & PURI-test. Lindman (1974) however showed that the F statistic is quite robust against violations of the homogeneity of variances assumption. Therefore, when this assumption was violated due to a number of zero values in the data set the ANOVA was occasionally calculated anyhow. Whenever this was the case it is pointed out clearly in the text. ANOVAs with model results The food web model described in detail in chapter 4 was run with the biomass input data from four separate sampling occasions per site, resulting in four sets of estimates for each site. Analyses of variance revealed no significant main effect of ‘time’ (sampling time) on the biomass and C and N mineralisation rates of any functional group nor the total biomass and total mineralisation of C and N. Therefore the four sets of modelling estimates per site were treated as replicate estimates of the mineralisation rates and one-way ANOVAs were calculated to compare the sites.

38


3.3.2 PEARSON product-moment correlation PEARSON product-moment correlations were used to test for significant relationships between two variables. The correlation coefficient (r) determines the extent to which values of two variables are linearly related to each other.

3.3.3 Canonical correspondence analysis (CCA) The canonical correspondence analysis (CCA) is a combination of ordination and multiple regression assuming unimodal (bell-shaped) response curves of species abundance to environmental variables (Jongman et al. 1987, ter Braak and Smilauer 1998). This assumption is an advantage over ordination techniques based on linear response models, such as principal component analysis (PCA), since a linear relationship between species abundance and environmental variables is rather unlikely within biological systems. The CCA is used to sort the sites according to their environmental conditions and to relate Testate Amoebae species abundance patterns to environmental information (faunal and microbial parameters, abiotic variables) and to test the data set for trends of ecological preferences of particular species. The analysis is performed in two steps (cf. ter Braak 1986). First the dominant pattern of variation in community composition is extracted from the species data by an ordination technique. Second this pattern is related to environmental variables. CANOCO uses an iterative ordination algorithm based on weighted averaging that is detailed in the appendix of ter Braak & Prentice (1988). Statistical significance of the regression with the ordination axes was determined at the 5 % significance level (p < 0.05) using a MONTE CARLO permutation test with 999 unrestricted permutations (ter Braak 1996). The sites are plotted on a two-dimensional factorial plane, representing a number of environmental variables, to understand the conditions at each site in relation to the other sites (biplot of sites). The Testate Amoebae species are plotted in the same way to test the data set for trends of ecological preferences of particular species (biplot of species).

39

Chapter 4

The Food Web Model

4 The food web model

The food web model approach of De Ruiter et al. (1993b, based on O'Neill 1969, and Hunt et al. 1987) was applied to the decomposer communities of the different sites. The principal output of the food web model are the carbon and nitrogen mineralisation by individual functional groups as well as the total mineralisation by the entire decomposer food web. Figure 4.1 gives a schematic view of the steps taken in this study to model the C and N mineralisation rates. The first necessity was to identify the organisms involved in the decomposer food web and to compile a list of the most important taxa. In the next step taxa were aggregated into functional groups on the grounds of their trophic function and energy processing rates (sensu Moore et al. 1988). Based on the comprehension of the trophic interconnectivity between the functional groups a diagram of the food web was constructed in form of a connectedness web (see Figure 5.13, section 5.2.1). The death rates were adapted to the climate of the specific study site (see section 4.2.3). Then energy flow descriptions of the food web were computed, in which the feeding rates were calculated from the observed population sizes (biomasses), the observed detritus pool and resource quality, and physiological parameters characterising the metabolism of the functional groups. Using the estimated feeding rates and the C:N-ratios of the interacting groups the C and N mineralisation were then calculated for each functional group and for the total food web.

41

4 The Food Web Model

compile participating organisms

aggregate into functional groups

comprehend trophic relationships

draw sketch of the food web:connectedness web

calculatefeeding rates

measure resource quality

calculate energy flow

calculatemineralisation rates (C + N)

compile physiologicalparameters

measurebiomasses(variables)

describe climate (forcing function)

Figure 4.1 Practical steps in applying the food web model approach to estimate C and N

mineralisation rates (schematic view).

The mathematic model is based on three principal assumptions: I. The decomposer system is in steady state, i.e. the annual average production of the organisms

balances the annual loss through natural deaths and predation. II. The top predators suffer from natural deaths only. Calculation of feeding rates starts from the top

predator and proceeds working backwards to the lowest trophic levels. III. If a predator feeds on more than one prey type, both, the preference of the predator for a given prey

and the relative population sizes of the prey types are taken into account.

42


The assumption that the annual growth rates of the populations balance the annual natural death rates (see section 4.2.1) and the death rates due to predation leads to the following equation (Hunt et al. 1987, De Ruiter et al. 1993b): production = natural deaths + deaths due to predation equation 4.1 The population biology equation states that production is the part of consumption (food incorporated) that is neither excreted nor respired but transformed into biomass (Figure 4.2).

consumption

assimilation excretion(faeces)

production(biomass)

mineralisation(CO2, NH4+, etc.)

assimilationefficiency

productionefficiency

Figure 4.2 Schematic illustration of the population biology equation, the pathway of energy from consumption to production resp. mineralisation.

The assimilation efficiency (a) is the rate with which the material consumed by a functional group is assimilated into the body (see section 4.2.2, equation 4.6). The production efficiency (p) is the rate with which the assimilated material is used for production. In other words (1-p) is the rate with which assimilated material is respired (see section 4.2.2, equation 4.7). Given that production depends on the feeding rate as well as on the efficiency of assimilation and production, equation 4.1 leads to equation 4.2 (Hunt et al. 1987, De Ruiter et al. 1993b).

43


jj

jjjj

jjjjjj

p ad

d p a

KBF

KBF

�

��

��

equation 4.2

aj assimilation efficiency pj production efficiency Fj feeding rate (kg C ha-1 a-1) dj death rate (a-1) Bj biomass (kg C ha-1) Kj loss through predation (kg C ha-1 a-1) j index of functional group

The top predator of the food web is considered not to be preyed upon. Thus the feeding rate of the top predator may be calculated from equation 4.2 by setting the deaths due to predation Kj = 0. Then the feeding rates of the trophic level below the top predator can be calculated and the calculation proceeds downwards working through the food web to the lowest trophic level, the primary decomposers. Most predators feed on more than one prey. If this is the case, both the preference of the predator for a given prey, and the relative population size (biomass) of the prey types are taken into account. To do so the relative feeding preference (wij, see section 4.2.3) of a predator j for a prey i is introduced (equation 4.3, Hunt et al. 1987, De Ruiter et al. 1993b). The total feeding rate of predator j (Fj) is split into prey specific feeding rates (Fij).

jij

jn

1kkj

iijij

j predatorby consumed biomass totalipreyfromconsumesjpredatorthatbiomass

w

w

FF

FB

BF

k

��

�

��

equation 4.3

wij feeding preference n number of prey types consumed by predator j index of predator i index of prey

44


The mineralisation rates are derived from the feeding rates (equation 4.4, Hunt et al. 1987, De Ruiter et al. 1993b). Carbon mineralisation is that part of the consumed biomass C that is assimilated (aj) and not incorporated into biomass but respired (1-pj):

� � jjjj p1a FMC �� equation 4.4

MC j carbon mineralisation (kg CO2-C ha-1 a-1) To calculate the nitrogen mineralisation both, the C:N-ratio of the consumed material (C:N-ratio of the prey) and that of the produced biomass (C:N-ratio of the predator) must be taken into account (equation 4.5, Hunt et al. 1987, De Ruiter et al. 1993b):

ijj

j

ijij q

pq1a FMN �

�

�

�

��

�

��

equation 4.5

MN ij nitrogen mineralisation (kg Nmin ha-1a-1) q C:N-ratio

4.1 Detritus pool and resource quality The size of the detritus pool (organic layer) at each site was calculated from the dry-mass-to-area ratios (Table 2.1). Since naturally occurring organic matter is a complicated mixture of degradable and recalcitrant substances (Andrén et al. 1990, Tezuka 1990, Sollins et al. 1996) it cannot be assumed that the entire C pool is equally available to primary decomposers. Instead for the food web model it was assumed that only 20 % of the total detritus pool is realisable for the decomposer organisms. Making use of the 14C-signature the mean residence time of carbon in the LF-layer at the sites was estimated to be 5-6 years (Harrison et al. 2000) which supports this assumption. The resource quality was characterised by the observed C:N-ratios of the organic layer (Table 2.1). Since the total C:N-ratio of the substrate in a heterogeneous environment like the organic layer may differ from the C:N-ratio of the material available to the primary consumers (Tezuka 1990, Hammel 1997) the site specific C:N-ratios of the substrate were reduced by 20 % for fungi and by 30 % for bacteria. This reflects the specific ability of fungi and bacteria to use recalcitrant substances (De Ruiter et al. 1993b, Dighton 1997) while the characteristic C:N-ratios, and thus the specific resource quality at

45


the particular site, are still taken into account.

4.2 Model parameters Parameters needed to run the food web model are the average C:N-ratio of the biomass of the functional groups, the death rate, the assimilation and production efficiencies and the feeding preferences. Estimates for these parameters were taken from the literature and are based on field budgets, laboratory measurements and laboratory cultures (see Table 5.14, section 5.2.3).

4.2.1 Death rates The nominal death rate (d) used in the food web model is defined by Hunt et al. (1987) as the inverse of the maximal life span observed under ideal laboratory conditions. This death rate determines the rate at which 'natural deaths' (non-predaceous losses) occur. The death rates used in this study are taken from the literature (see Table 5.14, section 5.2.3) (Hunt et al. 1987, De Ruiter et al. 1993a). They refer to a temperature of 10°C (De Ruiter et al. 1993a) and have to be adapted to the specific climatic conditions of a given site (see section 4.3).

4.2.2 Assimilation and production efficiencies The efficiency with which the consumed food is used for production depends on the efficiency with which the consumed food is assimilated (equation 4.6) and on the efficiency with which the assimilated food is used for production (equation 4.7).

� �CA

CEC

�

�

�a equation 4.6

a assimilation efficiency C consumption E excretion A assimilation

AP

ARA�

�

�p equation 4.7

46


p production efficiency R respiration P production

4.2.3 Feeding preferences The feeding preferences are entered in form of a matrix of wij values describing the relative preference

of predators on their prey (see Table 5.15, section 5.2.3). Within this matrix a value of � 1 indicates that

one functional group is predator of another functional group. Values > 1 put emphasis on a certain trophic link, with the weighting being dependent on the total prey biomass consumed by the predator. E.g. the feeding rate of panphytophagous Testate Amoebae Fpata is split to a feeding rate on three kinds of prey: fungi (Ffung,pata), bacteria (Fbact,pata) and detritus (Fdetr,pata). According to equation 4.3 the rate with which the panphytophagous Testate Amoebae feed on fungi, Ffung,pata is calculated as follows:

� �

� � patadetrpatadetr,bactpatabact,fungpatafung,

fungpatafung,patafung,

patatotal

fungpatafung,patafung,

wwww

w

FBBB

BF

FB

BF

�

��

��

��

equation 4.8

Btotal total biomass consumed by predator fung index of fungi pata index of panphytophagous Testate Amoebae bact index of bacteria detr index of detritus Hence, the amount of biomass that panphytophagous Testate Amoebae feed from fungi depends on their preference for fungi and the relative availability of all their food sources, including fungi.

4.3 Adapting the food web model to specific climatic conditions Besides resource quality, climate is one of the most important determinants for soil biota (Swift et al. 1979). Climatic influences on the decomposer food web enter the model in indirect and direct form. Indirectly, as the observed food webs (the connectedness of the functional groups and their biomasses) are a function of the long term site specific conditions and reflect the climatic and all other environmental influences. Directly, as temperature and moisture regime of a given site have an

47


influence on metabolic rates of the organisms of the food web. This is accounted for by making the death rate of the functional groups climate dependent. Climate has a massive influence on the death rate (natural deaths per year) of the functional groups in the field. This should be understood not so much in the terms of climatic extremes having a lethal effect but in terms of climatic influences on the metabolisms of the organism determining their number of generations per year. Climatic conditions that enhance the metabolism (e.g. increased temperature and moisture) will cause a higher turnover of the organisms, accounted for within the model by an increase in the death rates. Usually metabolic rates are adjusted to temperature using the Q10 rate constant. The Q10 value is the increase in activity for a 10°C rise in temperature. This connection is expressed in the so called 'power function':

� �10

Tb

10Q�

�

T

TE

equation 4.9 ET temperature factor Q10 rate constant T temperature Tb base temperature i.e. the temperature corresponding to ET=1 For decomposition modelling values of Q10 are usually within the range of 2-5 (Andrén et al. 1990), which means that with a temperature increase by 10°C metabolic rates at least double. The Q10-adaptation is not entirely satisfactory because it does not take soil moisture into account. Therefore, in this modelling study a method was used that considers both, temperature and soil moisture regime of the sites. Within the CANIF-project Persson et al. (2000a, 2000b) estimated respiration (CO2-evolution) and N mineralisation using a laboratory incubation method with material collected from the same study sites as studied here. C and N mineralisation rates were measured in the laboratory at a certain constant reference temperature (Tref) and a certain constant reference soil moisture (reference water potential

�ref). To extrapolate from these standard conditions to the soil temperature and moisture in the field

Seyferth (1998) within the working group of Tryggve Persson determined response functions in the laboratory, using material from the humus layer at a site close to Skogaby (S-SE). The range of temperatures and moistures studied was wide, taking account of both temperate and boreal systems: the studied incubation temperatures ranged from -4.0 to +25.0°C at water-holding capacities from

48


15 to 100 % (Seyferth 1998). Seyferth (1998) found that the temperature dependence of mineralisation rates in forest soils could well be described by a quadratic function (equation 4.10, Ratkowsky et al. 1982):

� �2min

2 Tb �� TM equation 4.10

M mineralisation rate at temperature T Tmin the minimum temperature at which activity starts (°C) T actual temperature (°C) b slope of temperature function (°C-1) Seyferth (1998) empirically determined Tmin, the minimum temperature at which activity started, to be -6.2°C and calculated a correction factor for the activity at a temperature T as follows (equation 4.11):

� �

� �2minref

2ref

2min

2

ref TTbTb�

�

�

TMM

equation 4.11

refMM correction factor

bref slope at the reference temperature Tref (°C-1) Tref reference temperature (°C) In further experiments she found the slope of this function to be moisture dependent according to a log-linear function (equation 4.12):

� � nlogm �� ψb equation 4.12

b slope of temperature function (equation 4.11) dependent on moisture m slope of the moisture function � water potential (Mpa) n constant For the calculation of M/Mref equations 4.11 and 4.12 can be combined:

49


� ��

� ��

� �

� �2minref

2min

2ref

2

ref TTT

nψlogmnlogm

�

�

�

�

�

�

TψMM

equation 4.13

�ref reference water potential (Mpa) The combination leads to equation 4.13, which Persson (2000a, 2000b) used to calculate monthly correction factors M/Mref in order to correct extrapolations from the laboratory to the field according to temperature and moisture regime of a specific site. Persson (2000a, 2000b) extrapolated from reference temperature of Tref = 15°C at optimal moisture

conditions (60 % water holding capacity (WHC), corresponding to a site specific water potential �ref that

was obtained from the CANIF database; e.g. for Skogaby, S-SE, �ref = 0.0035 MPa).

To be able to use Persson's correction factor M/Mref for the food web model the factors had to be modified to account for a difference in reference temperature. The death rates used for the model were obtained at Tref = 10°C (see De Ruiter et al. 1993a). All other variables and parameters stay constant, thus the term (Tref - Tmin)² is the only term that has to be recalculated (equation 4.14):

� �

� �2minref

2minref

refref TT'TT

''

�

�

��

MM

MM

equation 4.14

ref''

MM correction factor accounting for new reference temperature

refMM correction factor given by Persson

refT Persson's reference temperature (15°C)

refT' new reference temperature (10°C) For each month, Persson’s site specific monthly correction factor M/Mref was recalculated to obtain a corrected monthly M'/M'ref (equation 4.14). The recalculated correction factor M'/M'ref takes the reference temperature into account at which the death rates of the food web model were obtained. It makes use of site specific mean monthly temperature and moisture estimates from the CANIF databank. For each month and functional group a corrected death rate di' was obtained (equation 4.15):

50


refi '

'd'MMd ��

equation 4.15 di' climate corrected mean monthly death rate d nominal death rate (Hunt et al. 1987, De Ruiter et al. 1993a) From the monthly corrected di' an annual corrected death rate d'' is obtained for each functional group (equation 4.16):

��

�

12

1ii '12

1'' dd

equation 4.16 d'' mean annual death rate adapted to site specific climate

The site specific annual corrected death rates of each functional group are given in Table 5.14, section 5.2.3.

4.4 Biomass estimates The actual variables of the food web model are the observed biomasses of the functional groups at each site (section 5.2.3.3, Table 5.17). For each sites mean values of four sampling times covering the seasons of a year are used, representing an average biomass that takes part of the seasonal variation of the functional groups into account. The population dynamics of the functional groups were not studied. Rather the aim was to characterise the food web by yearly averages representing the standing crop biomass of the groups. To get realistic estimates of the standing crop biomass of the food webs at a certain site four samplings per year are sufficient (De Ruiter, pers. com.).

51

Chapter 5

Results

5 Results

The result chapter is divided into two parts. In part one (section 5.1) the Testate Amoebae communities are described and analysed. The Testate Amoebae are related to the abundance of other decomposer biota and to the abiotic environment at the sites along the transect. In part two (section 5.2) the results of the food web modelling approach described in chapter 4 are presented. The focus on Testate Amoebae is broadened towards the entire decomposer community and towards the forest ecosystem.

5.1 Testate Amoebae community structure

5.1.1 Size structure and biomass A total of 42 species were found on the study sites (Table 5.1). They were grouped into five size classes

I-V according to biomass, ranging from 0.0004 �g C ind-1 to 0.0210 �g C ind-1 (Table 5.2). The size

range was very large; the biomass per individual of the largest class being almost 53 fold that of the smallest class. Among the Testate Amoebae found, 14 species were assigned to the 1st (smallest), 8 to the 2nd resp. 3rd and 6 to the 4th resp. 5th size class (Table 5.2). The abundance of Testate Amoebae within a size class decreased gradually from the smallest to the largest class; small species occurred with higher abundance (Figure 5.1). A different picture emerged when the size classes were used to calculate the Testate Amoebae biomass per unit area. Larger species made up the larger share of the overall Testate Amoebae biomass (Figure 5.1). According to their size, Testate Amoebae species have been rated to be r-strategic (small species) or K-strategic (large, xenosome species) (Bamforth 1997). Both strategic groups similarly contribute to the total biomass of the community. Because body size varied so much between the different species of the Testate Amoebae community, abundance (number of individuals per unit area) cannot logically be linked to resource division by species (cf. Tokeshi 1993). Biomass measures serve this purpose better. On the other hand, abundance may more suitably reflect the activity of a species than does its biomass. In the comparison of the Testate Amoebae communities of the different sites emphasis will be put on biomass measures.

53

5 Results

Table 5.1 List of species that were found on the study sites and classification into size classes. See Table 5.2 for definition of size classes.

species size class

Arcella catinus PENARD 1890 IV Arcella arenaria compressa CHARDEZ 1957 III Assulina muscorum GREEFF 1889 I Assulina seminulum (EHRENBERG 1848) LEIDY 1879 II Bullinularia indica (PENARD 1907) DEFLANDRE 1953 V Centropyxis aerophila sphagnicola DEFLANDRE 1929 III Centropyxis gauthieri THOMAS 1959 III Centropyxis matthesi RAUENBUSCH 1987 V Centropyxis sylvatica (DEFLANDRE 1929) BONNET & THOMAS 1955 IV Corythion dubium TARANEK 1882 I Cryptodifflugia oviformis PENARD 1890 I Cyclopyxis eurystoma DEFLANDRE 1929 III Cyclopyxis kahli DEFLANDRE 1929 V Difflugia lucida PENARD 1890 II Difflugia minuta RAMPI 1950 I Edaphonobiotus campascoides SCHÖNBORN, FOISSNER & MEISTERFELD 1983 I Euglypha laevis (PERTY 1849) SCHÖNBORN 1992 I Euglypha rotunda minor WAILES 1915 I Euglypha cf. strigosa (EHRENBERG 1872) WAILES & PENARD 1911a III Heleopera sylvatica PENARD 1890 III Hyalosphenia subflava CASH & HOPKINSON 1909 III Microchlamys patella (CLAPAREDE & LACHMANN 1958) COCKERELL 1911 II Microcorycia flava (GREEFF 1866) PENARD 1902 II Nebela lageniformis PENARD 1890 IV Nebela militaris PENARD 1890 II Nebela parvula/tinctab IV Nebela tincta major/bohemica/collarisc IV Phryganella acropodia (HERTWIG & LESSER 1874) HOPKINSON 1909 II Phryganella paradoxa alta BONNET & THOMAS 1960 II Plagiopyxis declivis BONNET & THOMAS 1955 V Plagiopyxis intermedia BONNET 1959 IV Plagiopyxis labiata PENARD 1910 V Schoenbornia humicola (SCHÖNBORN 1964) DECLOITRE 1964 I Schoenbornia viscicula SCHÖNBORN 1964 I Tracheleuglypha dentata (PENARD 1890) DEFLANDRE 1928 I Trachelocorythion pulchellum (PENARD 1890) BONNET 1979 I Trigonopyxis arcula (LEIDY 1879) PENARD 1912 V Trigonopyxis minuta SCHÖNBORN & PESCHKE 1988 III Trinema complanatum PENARD 1890 I Trinema enchelys (EHRENBERG 1838) LEIDY 1878 I Trinema lineare PENARD 1890 I Trinema penardi THOMAS & CHARDEZ 1958 II

total number of species found 42 a Euglypha cf. strigosa is treated as a morphologic group. See section 3.1.2.6.3. b Nebela parva CASH & HOPKINSON 1909 and N. tincta (LEIDY 1879) AWERINTZEW 1906 were pooled

as the taxon N. parvula/tincta. See section 3.1.2.6.4. c Nebela tincta major DEFLANDRE 1936, N. bohemica TARANEK 1881 and N. collaris (EHRENBERG

1848) LEIDY 1879 were pooled as the taxon N. tinctamajor/bohemica/collaris. See section 3.1.2.6.4.

54

5 Results

Table 5.2 Size classes of Testate Amoebae and number of species found belonging to each size class.

size class conversion factor (µg C/individual)

number of species per size class

I 0.0004 14 II 0.0016 8 III 0.0035 8 IV 0.0077 6 V 0.0210 6

0

500

1000

1500

2000

I II III IV Vsize class

abun

danc

e (10

6 m-2

)

0

4

8

12

16

bioma

ss (k

g C ha

-1)

Figure 5.1 The abundance (columns) resp. biomass (dots) of living Testate Amoebae in the five size classes. See Table 5.1 and 5.2 for characterisation of size classes. Whiskers represent standard deviation.

5.1.2 Species pattern 5.1.2.1 Species richness, diversity and evenness Analysis of variance revealed significant effects of 'site' and 'time' on mean species number, diversity and evenness (Table 5.3). The F-value (inter-group variance divided by intra-group variance) indicates which of the tested effects accounts for the larger share of variance within the data set. The higher the F-value, the higher the amount of variance that is explained by the effect. The F-values in Table 5.3 show that the factor 'site' explains the largest part of variance within the data set of species number and diversity, while the variance of evenness is explained almost equally by both factors. The significant interaction between the two factors ‘site’ and ‘time’ reflect site specific fluctuations of the evenness with time resp. time specific relations of the sites in the equitability of the species abundances (Table 5.3).

55

5 Results

Table 5.3 Results of the two-way ANOVAs on the effect of 'site' and 'time' on number of species, diversity and evenness. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p = p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01, n = 50. site time site � time F p F p F p number of species 20.1 ** 6.3 ** 1.3 n.s. diversity 21.5 ** 6.6 ** 0.1 n.s. evenness 6.4 ** 8.0 ** 2.5 * There was little difference in the total number of Testate Amoebae species (species richness) found at the four sites, apart from a considerable larger value of 40 species at FR (Table 5.4). Besides the total number, the mean species number of each site was calculated in order to be able to perform statistical tests. While the total number of species is the cumulative number of species that were found at each site, the mean number of species is the average number of species encountered per sample. Concerning the mean species numbers, two groups distinguish from each other: the Swedish sites N-SE and S-SE with significantly lower values than the Southern sites DE and FR. To facilitate comparison with other studies, overall community diversity and evenness were estimated using the Shannon-Wiener diversity index (as reviewed in Krebs 1999). Parallel to the pattern of species richness, diversity was smallest at N-SE (3.2), increased towards South over 3.5 at S-SE, to a maximum of 3.7 at DE (Table 4.5). As a consequence of the evenness at S-SE being significantly higher than at N-SE, the only significant difference between the Shannon-Wiener diversities was that of N-SE having a lower diversity than all other sites (Table 4.5). The evenness estimates confirm a rather well-balanced species abundance pattern at FR, DE and S-SE, while at N-SE only 72 % of the maximum Shannon-Wiener diversity is reached (Table 5.4).

Table 5.4 Total number of species and mean number of species found at each site. For the calculation of diversity and evenness the relative abundance of living Testate Amoebae cells (active cells + cysts) was used; empty shells were not taken into consideration. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.3 for further details on ANOVA). N-SE S-SE DE FR number of species (total) 34 35 35 40 number of species (mean) 21A 23A 27B 27B diversity 3.2A 3.5B 3.7B 3.6B evenness 0.72A 0.78B 0.78B 0.76AB

56

5 Results

5.1.2.2 Species rank plots The most versatile approach of presenting species abundance data to facilitate comparison among different data sets is plotting logarithmic abundance resp. biomass against arithmetic rank order of species in a so-called rank plot (Tokeshi 1993). The rank plot was carried out with biomass data (see remarks on biomass and abundance measures in section 5.1.1). For such a plot the species that occur on each site are placed in rank order with the most dominant first. The total biomass is then log-transformed (base 10) and plotted against the ranks of the species. The rank curves of the four sites revealed strong similarities between community structures at N-SE, S-SE, DE and FR. All curves display a rather shallow non-continuous decrease towards the rarer species with more than one slope. This suggests that biomass is quite evenly distributed among species (Magurran 1988). The curve for S-SE exhibits a fast drop at the beginning of the curve, indicating that at this site a few species are of great relative importance.

1

10

100

1000

10000

100000

0 10spe

bioma

ss (g

C ha

-1)

log sc

ale

N-SES-SEDEFR

B

Figure 5.2 Species biomass rank plots. T

plotted against the ranks of th

20 30 40cies in rank order

he total Testate Amoebae biomass on a log scale are e species (Tokeshi 1993).

57

5 Results

5.1.2.3 Relative biomass structure The community structure of Testate Amoebae is shown in detail in Table 5.5. The relative biomass is compared to indicate the relative importance within resource division among a species community (see section 5.1.1). Species are subdivided into 6 groups (A-G) to identify patterns and to aid the comparison of community structures at the different sites (Table 5.5). Group A consists of species that were equally common at all sites. Among these, many species also occurred with relatively high biomasses, e.g. Trigonopyxis arcula, Nebela parvula/tincta, Trinema

lineare, Corythion dubium, Nebela tincta major/bohemica/collaris, Plagiopyxis declivis, Centropyxis

sylvatica, Schoenbornia humicola and Phryganella acropodia, Nebela militaris (Figure 5.3 C, D and E). Some species rank among the most important even though their individual biomass is small (size class I), e.g. Trinema lineare, Corythion dubium and Schoenbornia humicola. Consequently these

species occur in very high abundances. Group B comprises of species that exhibited relatively higher biomass towards North. Three of those four species, namely Arcella catinus, Cryptodifflugia oviformis and Difflugia lucida showed a clear biomass maximum at N-SE. Group C combines nine species that tended to occur with relatively higher biomass towards South, including five species that were not found at N-SE and/or S-SE, but occurred at the two Southern sites. Two species, Hyalosphenia subflava and Heleopera sylvatica (Figure 5.3 F), were absent from N-SE but quite common at all other sites. Trachelocorythion pulchellum and Trinema enchelys were absent from N-SE and S-SE but found at the two Southern sites. Group D consists of Tracheleuglypha dentata and Plagiopyxis intermedia that were found at all sites except S-SE. The latter species was very rare at N-SE. Group E unites two species (Bullinularia indica and Centropyxis gauthieri) that were lacking only at DE (Figure 5.3 A). Both species furthermore occurred with higher relative biomasses at N-SE than at S-SE and FR. Group F comprises of Plagiopyxis labiata and Cyclopyxis kahli, two species that were found at N-SE and S-SE but were lacking from the two Southern sites. The latter species was very rare at S-SE. Group G combines rare species that were found in low biomasses at one or two sites. Among those Edaphonobiotus campascoides (Figure 5.3 B) inhabited DE, but was absent from the other sites. Difflugia minuta and Nebela lageniformis (Figure 1.1) were restricted to, and very rare at FR.

58

5 Results

Table 5.5 The community structure of Testate Amoebae. Asterisks represent relative biomass (%). Species are arranged according to their occurrence or absence at certain sites. Dotted lines indicate groups labelled with capital letters (A-G). For explanation of the grouping see section 5.1.2.3. ****** > 31.9 %; ***** = 10.0 to 31.9 %; **** = 3.2 to 9.9 %; *** = 1.0 to 3.1 %; ** = 0.32 to 0.99 %; * < 0.32 %; empty space = 0 %; ° = found only once during counting or occurred only when analysing enriched material from flotations or batch cultures.

N-SE S-SE DE FR A Trigonopyxis arcula **** ****** ***** ****** Nebela parvula/tincta ***** ***** ***** ***** Trinema lineare **** *** **** **** Corythion dubium **** **** **** *** Nebela tincta major/bohemica/collaris *** **** **** **** Plagiopyxis declivis **** *** **** **** Centropyxis sylvatica ***** *** **** *** Schoenbornia humicola **** *** *** **** Phryganella acropodia *** *** **** **** Microchlamys patella ***** *** *** *** Euglypha cf. strigosa *** *** **** *** Assulina muscorum *** *** *** ** Nebela militaris *** *** *** ** Trinema complanatum ** ** *** ** Cyclopyxis eurystoma ** ** *** ** Centropyxis aerophila sphagnicola ** * ** * Assulina seminulum ** * * * Euglypha rotunda minor * * * * Trinema penardi * * * * B Microcorycia flava ** *** * * Arcella catinus ***** * ** ** Cryptodifflugia oviformis **** * * ** Difflugia lucida *** * * ** C Euglypha laevis ** * *** ** Schoenbornia viscicula ** * ** *** Trigonopyxis minuta * * **** ** Centropyxis matthesi * * *** ** Hyalosphenia subflava **** **** **** Heleopera sylvatica ** ** *** Phryganella paradoxa alta * ° * Trachelocorythion pulchellum * * Trinema enchelys * * D Tracheleuglypha dentata * * * Plagiopyxis intermedia ° * * E Bullinularia indica ** * * Centropyxis gauthieri ** * * F Plagiopyxis labiata * ** Cyclopyxis kahli * ° G Edaphonobiotus campascoides * Arcella arenaria compressa ° ° Difflugia minuta ° Nebela lageniformis °

59

5 Results

ventral dorsal lip

50 µmA

20 µm

Ne

T

D 20 µm

Pseu cys

F

10 µm

PseuE10 µm

N

Pseu

C

Pseu

Nn

colla

P

shel

10 µmBFigure 5.3 Testate Amoebae species from the study sites. A. Bullinularia indica, focus on the dorsaland on the ventral lip. 200x, bright-field, in Euparal. B. Edaphonobiotus campascoides, lateral viewof the trumpet-like shell with cytoplasm and vesicular nucleus. 400x, DIC, in Euparal. C.Schoenbornia humicola, nucleus and cytoplasm stained with aniline blue. 400x, bright-field, inwatery suspension. D. Nematode and Trinema lineralis, 400x, bright-field, in watery suspension. E.Nebela militaris, cytoplasm stained with aniline blue. 400x, bright-field, in watery suspension. F.Heleopera sylvatica cyst, stained with aniline blue. 200x, bright-field, in watery suspension. Pseu =pseudostome; N = nucleus; n = nucleolus; P = cytoplasm; T = Trinema lineare; Ne = Nematode.

60

5 Results

The Testate Amoebae communities at the transect sites were analysed using two similarity indices; the number of unique species and the Bray & Curtis index (Bray and Curtis 1957, Southwood 1994). The number of unique species in comparison of a pair of sites is an important component of the complementarity of these sites (Colwell and Coddington 1994). The highest number of unique species was rendered by the comparison of site DE and N-SE (Table 5.6). In the comparison of FR and DE only 5 unique species were found.

Table 5.6 Total number of species, and matrices of the number of unique species (Colwell and Coddington 1994) and Bray & Curtis-similarities (Bray and Curtis 1957, Southwood 1994) in comparison of the four sites. N-SE S-SE DE FR total number of species 34 35 35 40 number of N-SE unique species S-SE 7 DE 11 8 FR 10 9 5 Bray & Curtis- N-SE similarity (%) S-SE 67 DE 34 43 FR 46 56 70

The Bray & Curtis-similarity index is based on species specific abundance values (Bray and Curtis 1957, Southwood 1994). Again the highest similarity was found when comparing the abundance patterns of DE and FR, followed by the similarity between N-SE and S-SE (Table 5.6). The number of unique species and the Bray & Curtis similarity were plotted against the geographical distance between the respective sites (Figure 5.4). With few exceptions the Bray & Curtis similarity increased with decreasing geographical distance between the sites, however this correlation is not significant (PEARSON correlation r = -0.68, p = 0.13; Figure 5.4). The number of unique species showed the reverse trend in a significant correlation (PEARSON correlation r = 0.89, p < 0.05; Figure 5.4). Both measures indicate an increasing dissimilarity of the Testate Amoebae communities with increasing geographical distance between the sites.

61

5 Results

0

10

20

30

40

50

60

70

80

0 400 800 1200 1600 2000

distance (km)

Bray

& C

urtis

simi

larity

(%)

0

2

4

6

8

10

12

numb

er of

uniqu

e spe

cies

Bray & Curtis similarity (%) number of unique species

D/F D/S S/N F/S D/N F/N

Figure 5.4 The Bray & Curtis similarity index and the number of unique species plotted against geographical distance between the sites. Bray & Curtis = dotted line (r = -0.68, p = 0.13); unique species = unbroken line (r = 0.89, p < 0.05). On the abscissa the distance between pair-wise compared sites is indicated using the following abbreviations: D = DE; F = FR; S = S-SE; N = N-SE.

5.1.3 Testate Amoebae communities in their environment The sites that were studied have been described in chapter 2 by general geographic and site historic variables (Table 2.1). It was part of the soil ecological survey carried out at each site to gather further information on the environmental conditions, in order to characterise the ecosystems in terms of microbial community, abiotic conditions, and faunal groups besides Testate Amoebae. 5.1.3.1 Other decomposer biota Besides Testate Amoebae other important biota of the decomposer food web were monitored: microorganisms (bacteria and fungi), Nematoda, Enchytraeidae and Microarthropoda (Collembola and Acari) and the main effect of 'site' and 'time' was analysed via two-way ANOVA. Both had a significant effect on all variables, except for the effect of ‘site’ on the abundance of Nematoda that was not significant (Table 5.7). According to the F-value, 'time' often explained the variance within the data set to

a larger extent than 'site'. Significant interactions of 'time � site' indicated that the dynamics of the biota

62

5 Results

were site specific and/or that the site-to-site relation of abundances varied with time.

Table 5.7 Results of the two-way ANOVAs on the effect of 'site' and 'time' on microbial and faunal parameters. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01. site time site � time F P F p F p n metabolic potential (mg C m-2 h-1) 20.4 ** 46.0 ** 4.8 ** 111 microbial biomass C (g m-2) 26.0 ** 53.7 ** 21.9 ** 113 metabolic quotient (mg C-CO2 g-1 Cmic) 22.2 ** 24.2 ** 11.5 ** 110 bacteria (% of total Cmic)a 19.9 ** 32 bacteria, frequency of dividing cells (%)a 3.8 * 32 Nematoda (103 m-2) 1.3 n.s. 22.8 ** 4.6 ** 116 Enchytraeidae (103 m-2) 100.5 ** 6.4 ** 2.1 * 116 Collembola (103 m-2) 14.7 ** 10.5 ** 4.1 ** 115 Acari (103 m-2) 3.1 * 7.3 ** 3.7 ** 55

a since these variables were monitored only at the last sampling occasion one-way ANOVAs were carried out. Microbial activity, expressed by the metabolic quotient that relates metabolic potential (CO2 evolution at 10°C and water content 300 % DW) to unit microbial biomass, was similar at N-SE, S-SE and DE, and significantly lowest at FR (Figure 5.5). Interestingly microbial biomass (Cmic) was highest at N-SE and FR, while the relative inactivity of the microflora at FR resulted in a comparably low metabolic quotient at this site (Figure 5.5). The high metabolic quotient at N-SE was a combination of relatively high microbial biomass and strong metabolic potential.

63

5 Results

0

10

20

30

40

50

60

70

80

N-SE S-SE DE FR

metab

olic p

otenti

al (m

g CO 2

-C m

-2 h-1

)

micro

bial b

iomas

s C (g

Cmi

c m-2

)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

metab

olic q

uotie

nt (m

g CO 2

-C g-1

Cmi

c)

metabolic potential microbial biomass C metabolic quotient

A

BC B C

A

B

C

A

A A

A

B

Figure 5.5 Microbial parameters (metabolic potential, microbial biomass C, metabolic quotient) along the transect. Symbols of the same shading labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.7 for further details on ANOVA). Whiskers represent standard deviation.

At all sites bacteria made up the smaller part of the total microbial biomass (4.8 – 12.6 %, Figure 5.6) and fungi accounted for around 90 %. The share of bacteria was significantly highest at the Southern-most site FR. The trend of the metabolic quotient (Figure 5.5) to some extent fitted the frequency of dividing bacterial cells which was significantly highest at N-SE (Figure 5.6).

64

5 Results

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

N-SE S-SE DE FR

%

bacteria frequency of dividing cells

AB B B

A

A

A

B

Figure 5.6 Percentage of bacterial from total microbial biomass C and frequency of dividing bacterial cells. Columns of the same shading labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.7 for further details on ANOVA). Whiskers represent standard deviation.

The sites also differed from each other with respect to other faunal groups (Figure 5.7) with the exception of Nematodes, which were found in similar abundances at all sites. Like the Testate Amoebae the Enchytraeidae showed a significant increase in abundance towards South, ranging from a minimum of 20 102 m-2 at N-SE over 50 102 m-2 at S-SE, to a maximum of 380 102 m-2 resp. 430 102 m-2 at DE resp. FR (Figure 5.7). The abundance of Collembola ranged from 59 103 m-2 to 154 103 m-2, with the maximum abundance found at FR and the minimum at S-SE. The abundance of Acari ranged from 135 103 m-2 to 255 103 m-2, and revealed the opposite pattern of Collembola: the maximum Acari abundance was found at S-SE and the minimum at FR.

65

5 Results

0

100

200

300

400

500

600

700

800

N-SE S-SE DE FR

Nema

toda,

Colle

mbola

, Aca

ri (10

3 m-2

)En

chytr

aeida

e (10

2 m-2

)

Nematoda Collembola Acari

A B

CC

A A

A

A

AB A B

CAB

AAB

B

Enchytraeidae

Figure 5.7 Abundance of the major faunal groups besides Testate Amoebae. Columns of the same shading labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.7 for further details on ANOVA). Whiskers represent standard deviation.

5.1.3.2 Abiotic environment Both 'site' and 'time' had a significant effect on the environmental variables and significant interactions between the two factors indicate site specific dynamics and/or time specific site patterns (Table 5.8). The F-values tell that for all variables except water content the effect 'site' explains more of the total variance within the data set than does 'time'. In Table 5.9 mean, minimum/maximum, and standard deviation (in parantheses) of the parameters are given in order to show the range of the abiotic environmental conditions at each site.

66

5 Results

Table 5.8 Results of the two-way ANOVAs on the effect of 'site' and 'time' on abiotic parameters. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01. site time site � time F p F p F p n thickness of organic layer (cm) 18.1 ** 15.8 ** 20.5 ** 116 bulk density (g DW mL-1) 38.7 ** 2.0 n.s. 11.1 ** 116 water content (% DW) 17.7 ** 38.2 ** 23.9 ** 115 pHH2O 136.6 ** 18.8 ** 18.2 ** 116 total C (% DW)a 24.0 ** 32 total N (% DW)a 66.8 ** 32 C:N-ratioa 218.1 ** 32

a since these variables were monitored only at the 4th sampling occasion one-way ANOVAs were carried out.

In the light of the precipitation patterns (Figure 2.2) the variance within the data set of the water content of the organic layer is easily understood (Table 5.9). The surprisingly high maximum water content at N-SE resulted from snow that was collected with the organic material on two sampling occasions. The moisture conditions at the sites is better reflected by the monthly precipitation patterns (Figure 2.2) than by the mean water contents. The structure of the organic layer, its thickness and density, leading to a certain amount of organic material per unit area, was characteristic at the sites. At S-SE and FR the organic layer was significantly thinner (4.7 resp. 5.1 cm) than at N-SE and DE (both 6.2 cm, Table 5.9). The density of material at N-SE was fairly small (0.09 g DW mL-1, Table 5.9). The bulk density at the other sites was higher (0.14 to 0.16 g DW mL-1). This was due to the nature of the material: at S-SE, DE and FR it comprised mainly of spruce needles, at N-SE it was a mixture of moss as well as lichen litter, spruce and pine needles and occasional birch leaves. The standard deviation of the layer thickness at DE and N-SE indicates that at these two sites, in contrast to the other two sites, thickness of organic layer varied considerably with time. At DE it was very much increased in September, probably due to increased needle fall. At N-SE it was increased during September and April, presumably due to increased litter fall in September and frost in April. Another group of environmental variables is strongly correlated with each other: pH and total N. These variables followed the patterning of total N and S deposition at the sites: minimum N content resp. acidity at N-SE, maxima at DE, followed by FR, and intermediate values at S-SE (Table 2.1 for

67

5 Results

deposition, and Table 5.9 for environmental parameters). This was accompanied by high C:N-ratios at N-SE, intermediate values at S-SE and lower values at DE and FR. The somewhat unexpected (regarding N content) low C:N-ratio at FR resulted from significantly lower total C content at this site (Table 5.9). The C:N-ratios given in Table 5.9 do not fully correspond to those in Table 2.1 (CANIF-databank). This is ascribed to natural variation and heterogeneity. They are however presented because the values in Table 5.9 are extracted from a full data set which allowed statistical testing. For the ratios given in Table 2.1 the raw data set was not available.

Table 5.9 Mean, minimum/maximum (2nd row) and standard deviation (2nd row, in parentheses) of the environmental parameters at the sites. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.8 for further details on ANOVA). N-SE S-SE DE FR n water content (% DW) 278A 183B 216C 250A 156/503(157) 143/227(35) 130/272(61) 176/338(66) 116 thickness of organic layer (cm) 6.2A 4.7B 6.2A 5.1B 4.3/9.4(2.4) 4.5/5.0(0.2) 5.1/9.1(1.9) 4.6/5.9(0.6) 116 bulk density (g DW mL-1) 0.09A 0.14B 0.14B 0.16C 0.05/0.13(0.04) 0.12/0.14(0.01) 0.12/0.15(0.01) 0.12/0.19(0.03) 116 pHH2O 3.9A 3.6B 3.4C 3.5D 3.8/4.1(0.1) 3.5/3.8(0.1) 3.1/3.6(0.2) 3.3/3.6(0.1) 116 total C (% DW)a 42A 44AB 46B 36C 39/46(2) 41/48(2) 42/49(2) 31/41(3) 32 total N (% DW)a 1.1A 1.5B 1.9C 1.5B 0.9/1.2(0.1) 1.4/1.6(0.1) 1.7/2.2(0.2) 1.3/1.8(0.2) 32 C:N-ratioa 39.6A 29.2B 23.5C 23.2C 35.8/42.9(2.5) 27.1/31.1(1.2) 22.4/24.7(0.8) 22.6/23.8(0.5) 32

a since these variables were monitored only at the last sampling occasion one-way ANOVAs were carried out. These values do not fully correspond to C:N-ratios given in Table 2.1 (from the CANIF-databank) due to natural variation and heterogeneity. They are presented because the full data set obtained from measurements within this study allowed statistical testing.

5.1.3.3 Multivariate analysis relating Testate Amoebae communities and environment The information on the study sites was explored focussing on its explanatory value for the Testate Amoebae species patterns found. A multivariate statistical method (canonical correspondence analysis, CCA) was used to relate the environmental data set to the Testate Amoebae species biomass pattern. In a first approach the entire environmental data set was included in the CCA and the most important variables were identified using 'forward selection'. The environmental variables judged as being most important were included into the final CCA. Those variable were 'total atmospheric pollution' (a combination of the parameters total atmospheric nitrogen and total atmospheric sulfur deposition), pH of organic layer, water content, mean annual precipitation mean annual temperature, and metabolic potential as well as microbial biomass C. Data on other components of the fauna did not deliver sufficient canonical correlation coefficients. Geographical information was not included into the analysis

68

5 Results

to avoid grouping of sites according to nominal data like ‘country’. In the resulting biplots sites are grouped together on the grounds of actual measurements of abiotic parameters in the field instead of pre-assumed classifications. The first canonical axis explains 13 % of the total variance within the data set, the second axis adds 8 % (eigenvalues, Table 5.10). The first two axes explain 27 % of the variance within the species data set and 73 % of the variance within the species-environment relation (cumulative percentage, Table 5.10).

Table 5.10 Summary of the CCA of species biomass pattern and environmental variables. axis 1 axis 2 eigenvalue 0.13 0.08 cumulative % variance of species data 17 27 cumulative % variance of species-environment relation 46 73

Including the explanatory variables one after the other using forward selection, total atmospheric

pollution is estimated to account for the biggest share of the total variance of the data set (cum(�A) in

Table 5.11). This variable shows a significant inter-set correlation coefficient (rC) with axis 1 (rC = –0.77, Table 5.11). To a lesser extent axis 1 represents substrate pH (rC = 0.60, Table 5.11), a factor that naturally has a strong negative correlation to atmospheric deposition of S and N. The second environmental variable with significant explanatory value is microbial biomass C. This variable is to a large extent represented by the second canonical axis (rC = -0.68, Table 5.11). Two further variables added significantly to the canonical regression model: mean annual temperature and precipitation. This is at least partly due to a correlation of the former with total atmospheric pollution (due to the fact that the more Southern and warmer sites are more polluted. According to their interrelation these explanatory parameters may be combined into three groups of factors: atmospheric pollution (S and N deposition, pH), microbial biomass, and climate (mean annual temperature and precipitation). In a two-dimensional diagram axes 1 and 2 are combinations of the environmental variables included in the CCA. On the factorial plane spanned by axes 1 and 2 the samples from each site and sampling time are plotted to understand the conditions at each site in relation to the other sites (biplot of sites, Figure 5.8). The environmental variables are depicted by arrows. The length of an arrow is a relative measure of its importance in influencing the data set. In a scatterplot of the samples the replicates from N-SE are grouped together at the far end of axis 1 (Figure 5.8). On the opposite side the samples from DE are found, sub-clustered among each other according to sampling time (number behind site abbreviation, Figure 5.8). The samples from S-SE and FR are scattered in between the clusters formed by samples from DE and N-SE, building sub-clusters

69

5 Results

according to sampling date. A cluster of samples from FR at sampling times 3 and 4 is separated along axis 2 towards the microbial measures. Samples from S-SE and those from FR that were collected at the 1st and 2nd sampling time are scattered alongside the environmental arrow representing water content. Judging from its relative length, this arrow is important in explaining the variance of the data set. It adds a further dimension, being neither correlated to atmospheric pollution nor to microbial measures and independent of, or almost contrary to, precipitation. This may reflect a methodological problem, since water content measurements were obscured by snow at two sampling times in N-SE (see section 5.1.3.2). In relation to each other and with respect to the factors of significant explanatory value the characteristics of the sites may be summarised as follows: N-SE is the site with the lowest pollution and acidification. This site is characterised by high microbial biomass and high metabolic potential and is subject to low amounts of precipitation and low temperatures. With respect to climate the conditions at S-SE, DE, FR are less severe and similar due to altitude counteracting latitudinal changes. The site DE is furthermore distinguished from the others by especially high loads of atmospheric pollution, resulting in low pH. The French site FR distinguishes from S-SE and DE in microbial features. This site had a high microbial biomass (likewise N-SE) but the significantly lowest metabolic quotient (in contrast to N-SE).

Table 5.11 Conditional effects of including the environmental variables into the CCA one after the other using forward selection. cum(�A) = cumulative explanatory power (variance explained) by including the environmental variable; rC = canonical correlation coefficient of the inter-set correlations of environmental variables with the axes. p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01.

cum(�A)

p

F rC with axis 1

rC with axis 2

atmospheric pollution 0.12 0.001 ** 8.6 -0.77 0.21 microbial biomass C 0.17 0.001 ** 4.5 0.01 -0.68 mean annual temperature 0.21 0.006 ** 3.3 -0.17 0.03 mean annual precipitation 0.24 0.033 * 2.0 -0.15 -0.38 pH 0.26 0.068 n.s. 1.8 0.60 -0.03 water content 0.27 0.155 n.s. 1.4 0.32 0.29 metabolic potential 0.28 0.538 n.s. 0.8 0.15 -0.35

70

5 Results

axis 1ax

is 2

-1.0 +1.0

-1.0

+1.0

pH

water content

temperature

metabolic potentialprecipitation

microbial biomass C

atmospheric pollution

2NS2

3NS21NS2

1NS32NS3

3NS3

3NS42NS4

1NS4

3NS12NS1

1NS1

2SS4

1DE1

2DE3

3DE3

2FR2

3SS41SS4

3DE2

1DE3

2DE21DE2

3DE4

1FR2

3DE1

4DE1

2DE1

3FR4

5DE1 4FR2

1SS22SS22DE4

1DE4

1FR3

2FR11FR1

2FR4

2FR3

3SS23FR1

3FR3

1FR4

1SS13SS1

3SS32SS3

2SS1

1SS3

Figure 5.8 Biplot of sites. The dots represent replicate samples from different sampling times at each site. Samples are identified by the replicate number followed by site abbreviation (NS = N-SE, SS = S-SE) and sampling time (1 = 1st sampling in Oct/Nov, 2 = 2nd sampling in May/Jun; 3 = 3rd sampling in Sept, 4 = 4th sampling in Mar/Apr).

71

5 Results

In the biplot of species (Figure 5.9) most species are grouped around the origin of the plot. The origin represents the overall mean of all explanatory variables. Thus the majority of Testate Amoebae species are characterised as being generalists with respect to the environmental variables and considering the breadth of the gradients studied. However, some interesting ‘outliers’ are found. Edaphonobiotus

campascoides is depicted along the atmospheric deposition arrow and seems thus to be associated with high deposition resp. low pH and higher temperature. Cyclopyxis kahli, Arcella catinus,

Cryptodifflugia oviformis and Centropyxis gauthieri are spread out in the opposite direction, along the pH arrow, seemingly preferring higher pH, less atmospheric pollution and lower temperatures. The species Bullinularia indica and Phryganella paradoxa alta seem to be strongly related to microbial parameters and are stretched out along the arrows metabolic potential and microbial biomass C (Figure 5.9). Furthermore, Schoenbornia humicola and S. viscicula are depicted in close proximity of the arrow representing metabolic potential. Trigonopyxis minuta is separated from the others in a direction indicating the association with high pollution, precipitation and microbial biomass.

72

5 Results

axis

2

axis 1

-1.0 +1.0

-1.0

+1.0

pH

water content

temperature

metabolic potentialprecipitation

microbial biomass C

atmospheric pollution

CkaAca

Cov

Cga

Tmi

Eca

Dlu

MpaMfl

Pla

PprBin

CeuCma

PacHsu

Tpe

Tde

Ela

CsyTco

Hsy

Est

Erm

SviTen

Pin

ShuPde

Tar

Csp

Tli

Nmb

Npt

Nmi

Tpu

Ase

Amu

Cdu

Figure 5.9 Biplot of species. Aca = Arcella catinus; Amu = Assulina muscorum; Ase = Assulina seminulum; Bin = Bullinularia indica; Cga = Centropyxis gauthieri; Cma = Centropyxis matthesi; Csp = Centropyxis sphagnicola; Csy = Centropyxis sylvatica; Cdu = Corythion dubium; Cov = Cryptodifflugia oviformis; Ceu = Cyclopyxis eurystoma; Cka = Cyclopyxis kahli; Dlu = Difflugia lucida; Dmi = Difflugia minuta; Eca = Edaphonobiotus campascoides; Ela = Euglypha laevis/rotunda; Erm = Euglypha rotunda minor; Est = Euglypha cf. strigosa; Hsy = Heleopera sylvatica; Hsu = Hyalosphenia subflava; Mpa = Microchlamys patella; Mfl = Microcorycia flava; Nla = Nebela lageniformis; Nmi = Nebela militaris; Npt = Nebela parvula/tincta; Nmb = Nebela tincta major/bohemica/collaris; Pac = Phryganella acropodia; Ppr = Phryganella paradoxa alta; Pde = Plagiopyxis declivis; Pin = Plagiopyxis intermedia; Pla = Plagiopyxis labiata; Shu = Schoenbornia humicola; Svi = Schoenbornia viscicula; Tde = Tracheleuglypha dentata; Tpu = Trachelocorythion pulchellum; Tar = Trigonopyxis arcula; Tmi = Trigonopyxis minuta; Tco = Trinema complanatum; Ten = Trinema enchelys; Tli = Trinema lineare; Tpe = Trinema penardi.

5.1.4 Total abundance, biocoenosis and necrocoenosis The direct counting method applied in this study allowed differentiation of living cells and empty shells of Testate Amoebae. In this way biocoenosis and necrocoenosis may be compared. Furthermore, within

73

5 Results

the biocoenosis, dormant specimen (cysts) were recorded separately from active cells. The effect of 'site' and 'time' on various parameters of the Testate Amoebae community was tested (Table 5.12). Both effects and the interaction were significant for all variables. Because of the apparent high variablity of the parameters, illustration of the results is restricted to some main features of the dat set.

Table 5.12 Results of the two-way ANOVAs on the effect of 'site' and 'time' on various markers of the Testate Amoebae community. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01. site time site � time F p F p F p n living cells (106 m-2) 28.3 ** 59.4 ** 3.1 ** 50 empty shells (106 m-2) 25.7 ** 46.6 ** 12.0 ** 50 living cells (% of total count) 8.0 ** 211.5 ** 7.0 ** 50 empty shells (% of total count) 10.3 ** 266.6 ** 8.5 ** 50 cysts (% of living cells) 6.9 ** 34.8 ** 3.9 ** 50 Testate Amoebae biomass (kg C ha-1) 62.2 ** 51.2 ** 10.8 ** 50

0

500

1000

1500

2000

2500

3000

3500

4000

4500

N-SE S-SE DE FR

abun

danc

e (10

6 m-2

)

0

10

20

30

40

50

60

70

80

90

100em

pty sh

ells (

% of

total

coun

t)

living cells empty shells empty shells (%)

AA

B

CAA

B

C

A

BB

B

Figure 5.10 The abundance of living cells (10-6 m-2) and empty shells (10-6 m-2 resp. %) along the transect. In comparison of dots and columns of the same shading those labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.12 for further details on ANOVA). Whiskers represent standard deviation.

74

5 Results

The number of living cells was similar at N-SE and S-SE and ranged around 700 106 m-2 (Figure 5.10). At both Southern sites the densities of Testate Amoebae were significantly higher. The highest density of living cells as well as empty shells was found at DE (2700 106 m-2 living cells resp. 1200 106 m-2 empty shells), further South, at FR, numbers were somewhat lower (1800 106 m-2 living cells resp. 800 106 m-2 empty shells). At all sites the average abundance of living cells was greater than that of empty shells. The relative size of the necrocoenosis (percentage of empty shells) was 24-37% of the total count, with the necrocoenosis at N-SE being significantly smaller (Figure 5.10). Large standard deviations indicate that the relationship of necro- to biocoenosis varied considerably with time. The F-values in Table 5.12 do indeed reveal that 'time' accounted for a larger share of the variance within the data set of those variables related to the proportion of bio- to necrocoenosis and the occurrence of dormant forms (percentage of living cells, empty shells and cysts). It seems thus reasonable to look at the dynamic of the overall averages of these variables (Figure 5.11).

%

0

10

20

30

40

50

60

70

80

90

100

Mar/Apr May/Jun Sept Oct/Nov

A

B

A A

AA

A

B

living cellscysts

Figure 5.11 Occurrence of living cells (% of all specimen found) and cysts (% of living cells) at the different sampling times (averages over all sites). In comparison of columns of the same shading those labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.12 for further details on ANOVA). Whiskers represent standard deviation.

75

5 Results

While the proportions of bio- to necrocoenosis remained fairly constant at three of the sampling times, the biocoenosis was significantly decreased at the summer sampling at all sites (May/June, Figure 5.11). At this sampling time the percentage of living cells of total count ranged from 43 % at N-SE over 25 % at S-SE to 12 resp. 11 % at DE resp FR. The percentage of dormant cells ranged from 0.1 and 2.3 % of the total living cells at three of the four sampling times. Corresponding to the decrease in living cells at the May/June sampling, however, the number of dormant cells (cysts) was significantly increased at this sampling time, ranging from 3.2 % at N-SE over 5.1 % at DE to 9.4 % at FR to a maximum of 13.7 at S-SE. For the modelling approach applied in this study the Testate Amoebae biomass is of particular interest (see section 5.2). The biomass of the total Testate Amoebae community at each site and its variation with sampling time is shown in Figure 5.12. Both, 'site' and 'time', had a significant effect on the total Testate Amoebae biomass, with the effect of 'site' explaining a larger amount of the total variance within

the data set (Table 5.12). The significant interaction ('site � time') indicates that the biomass dynamics

were unique at each site and/or that at each sampling time the biomass at the sites related differently to each other. Figure 5.12 delivers an argument for the latter explanation since it reveals a common trend in the biomass dynamics at all sites, displaying seemingly unfavourable conditions at the May/June sampling. In accordance with the decrease in living cells and increase in cysts (Figure 5.11) the overall Testate Amoebae biomass dropped at the May/June sampling at all sites, which is significant for DE and FR (Figure 5.12). The two Southern sites showed a greater dynamic of Testate Amoebae biomass than the two Swedish sites (Figure 5.12). At the two Swedish sites the biomass of Testate Amoebae remained rather constantly at low values. The relationship between bio- and necrocoenosis of the Testate Amoebae was similar at the four sites, apart from the surprisingly low relative size of the necrocoenosis at N-SE. All sites revealed a common dynamic in the parameters concerning the realtionship between bio- and necrocoenosis that hinted to the importance of moisture availability. This common dynamics of parameters, reflecting the activity of the communities, was paralled by Testate Amoebae total biomass. The biomass of Testate Amoebae at the two Southern sites was significantly higher and showed higher variation with time than at the two Swedish sites.

76

5 Results

Testa

te Am

oeba

e (kg

C ha

-1)

N-SE

0

20

40

60

80

100

120

Mar/Apr May/Jun Sept Oct/NovS-SE


DE

0

20

40

60

80

100

120

Mar/Apr May/Jun Sept Oct/NovFR


AA

A AA

AA A

A A A

B B

C

C

AC

Figure 5.12 Testate Amoebae biomass at the different sites and sampling times. Within a site specific sub-plot columns labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.12 for further details on ANOVA). Whiskers represent standard deviation.

77

5 Results

5.2 Testate Amoebae and the functioning of the food web

5.2.1 Schematic view of the decomposer food web For each site a list of the most important functional groups was compiled. The diets of these groups were identified following the works of Hunt (1987) and De Ruiter et al. (1993a) and other literature on specific taxa (see section 5.2.2 on the trophic relationships of Testate Amoebae and section 3.1.3 on feeding groups of Nematoda). The trophic relationships between the functional groups of the decomposer web are schematically illustrated by a connectedness web (Figure 5.13). Within this sketch of the food web feeding relationships are indicated by arrows pointing from prey to predator. Food web complexity as the number of trophic groups times connectance, where connectance is the fraction of trophic interactions of all possible interactions (sensu Paine 1988), was very similar at the sites. Merely the functional group of predaceous Nematoda was absent at S-SE and FR, reducing food web complexity, as defined above, from 3.5 (N-SE, DE) to 3.2 (S-SE, FR). Keeping this restriction in mind, Figure 5.13 applies for all sites. The primary decomposers of the food webs at the four study sites were bacteria and fungi (Figure 5.13). These microflora groups are saprotrophic and directly metabolise soil organic matter in form of detritus and humus. The microflora was grazed upon by fungivorous and bacterivorous Nematoda, as well as panphytophagous Collembola, Acari and Testate Amoebae, and Enchytraeidae. The term 'panphytophagous' refers to functional groups that are microbivorous as well as detritivorous (sensu Luxton 1972). Many of the functional groups were omnivorous, feeding on several trophic levels (e.g. omnivorous Nematoda feeding on detritus, microflora and grazers; Figure 5.13). Loss due to secondary consumers (predators preying on grazers) occurred through predaceous Testate Amoebae, omnivorous and predaceous Nematoda, predaceous Collembola and predaceous Acari. The top predators of the food webs were the predaceous mites.

78

5 Results

litterdetritushumus

Bacteria

Enchytraeidae

pred. Acari

panphyt. Acari

panphyt. CollembolaFungi

fung. Nematoda

panphyt. Testate Amoebae

bact. Nematoda pred. Nematoda

omni. Nematoda

pred. Testate Amoebae

pred. Collembola

Figure 5.13 Sketch of the decomposer food web (connectedness web). Feeding relationships are indicated by arrows pointing from prey to predator. fung. = fungivorous; bact. = bacterivorous; panphyt. = panphytophagous; omni. = omnivorous; pred. = predaceous.

5.2.2 Literature survey on the trophic relationships of Testate Amoebae 5.2.2.1 Food sources of Testate Amoebae Terrestrial Testate Amoebae are the main consumers of microbial biomass, exerting a selective grazing pressure on edible microorganisms (Meisterfeld 1987). They primarily feed on bacteria (Bonnet 1964, Coûteaux 1976, Stout & Heal 1967 in Lousier and Parkinson 1984) but also ingest fungal hyphae, spores and yeasts (Barron 1978 in Coûteaux and Devaux 1983, Foissner 1987, Meisterfeld 1987). Moreover Testate Amoebae feed on algae and on other Protozoa (Bonnet 1964, Laminger 1978, 1980, Lousier and Parkinson 1984, Laminger & Bucher 1984). Two larger Testate Amoebae species in a New Zealand forest soil have been proven to feed on micrometazoans: Nebela vas and Difflugia spec. preyed on 7 different Nematode species (Yeates and Foissner 1995). Schönborn (1965, 1982) verifies that they can live solely on particulate organic matter like detritus and humus particles. Bamforth (1997) even concluded that most Testate Amoebae feed on humus particles. Some authors claim that Testate Amoebae use dissolved organic matter via pinocytosis (Coûteaux 1976, Lousier and Parkinson 1984). Because of the apparent variety of food sources, Testate Amoebae have been called polyphagous

79

5 Results

(Schönborn 1965). Information on particular species or genera is rare and is summarised in Table 5.13, regarding genera common in soil.

Table 5.13 Feeding behaviour of some soil Testate Amoebae species resp. genera. Genus resp. species feeding behaviour, food Author Arcella arenaria var. sphagnicola

algae Laminger & Bucher (1984)

Argynnia dentistoma algae Laminger & Bucher (1984) Bullinularia wood detritus, humus particles Bonnet (1964) in Laminger (1980) Centropyxis wood detritus, humus particles Bonnet (1964) in Laminger (1980) Centropyxis aerophila algae Bonnet (1964) in Laminger (1980) Cyclopyxis wood detritus, humus particles Bonnet (1964) in Laminger (1980) Euglypha rotunda detritus particles, fungal spores, bacteria Schönborn (1978) Heleopera petricola humicola Testate Amoebae Bonnet (1964) in Laminger (1980) Heleopora petricola Protozoa Laminger & Bucher (1984) Nebela Testate Amoebae, humus particles, Sphagnum

detritus Schönborn et al. (1987)

Nebela collaris Protozoa Laminger & Bucher (1984) Nebela collaris detritus particles, fungal spores, bacteria Schönborn (1978) Phryganella acropodia fungi Coûteaux and Devaux (1983),

Ogden and Pitta (1990) Pseudawerintzewia wood detritus, humus particles Bonnet (1964) in Laminger (1980) Schoenbornia humicola humus particles, bacteria Schönborn et al. (1987) Schwabia wood detritus, humus particles Bonnet (1964) in Laminger (1980) Trigonopyxis wood detritus, humus particles Bonnet (1964) in Laminger (1980) Trigonopyxis arcula detritus, fungal spores Coûteaux (1976), Bonnet (1964) in

Laminger (1980) Trinema complanatum detritus particles, fungal spores, bacteria Schönborn (1978) Trinema enchelys detritus, bacteria, diatoms, algae, fungal spores,

small Testate Amoebae (Euglypha laevis minor, Trinema lineare), Naked Amoebae (Limax)

Laminger (1978, 1980)

Trinema lineare bacteria Laminger (1980) For the model the Testate Amoebae were divided into two functional groups: the panphytophagous Testate Amoebae and the predaceous Testate Amoebae (Figure 5.13). The former group is considered to be primarily bacteria feeding, to graze on fungi to a lesser extent and to use detritus as a minor food source. The latter group comprises of the genera Nebela and Heleopera, two genera that are reported as being predaceous (Bonnet 1964 in Coûteaux 1976, Laminger 1980) although some species may additionally feed on fungal spores, bacteria and detritus (e.g. Nebela collaris, see Table 5.13). Five predaceous species were found at the sites: Heleopera sylvatica, Nebela lageniformis, N. militaris, N.

parvula/tincta and N. tincta major/bohemica/collaris.

80

5 Results

5.2.2.2 Testate Amoebae as food source A rarely regarded aspect of the trophic relationships of Protozoa is the fact that apart from grazing on others, they themselves serve as a food source. In previous production studies predation upon Testate Amoebae has not been considered (Lousier 1974a, 1985, Meisterfeld 1987, Schönborn 1992b). Testate Amoebae seem to be directly preyed upon by Nematoda (Varga 1959, 1960 and Foissner 1995 in Yeates and Foissner 1995). Lousier (1974a) considers detritivorous fauna, such as Enchytraeidae, to be the most important consumers of Testate Amoebae in soil. Testate Amoebae may be consumed rather indirectly along with large quantities of plant remains that panphytophagous fauna ingest while grazing on soil microflora. Rusek (1998) mentions Protozoa as food source of Collembola. In this study Testate Amoebae are considered to be consumed by Enchytraeidae, omnivorous and predaceous Nematoda, predaceous Collembola, predaceous mites and by predaceous Testate Amoebae (Figure 5.13).

5.2.3 Quantifying trophic interactions 5.2.3.1 Physiological parameters As described in chapter 4, a number of physiological parameters are needed to quantify the trophic interactions within the decomposer food webs. These parameters (assimilation and production efficiencies, C:N-ratios, death rates) and the literature they were obtained from are listed in Table 5.14. The same assimilation and production efficiencies as well as C:N-ratios of functional groups were taken for all sites. Climatic data (mean monthly temperatures and mean monthly precipitation, Figure 2.2) served as forcing functions and the model was adapted according to the climate of each specific study site. This was done by adapting basic death rates obtained from Hunt et al. (1987) and De Ruiter et al. (1993a) to temperature and moisture regime of the specific sites as described in section 4.3. The quality and quantity of the primary resource is taken into account via the site specific C:N-ratio of the organic material and the site specific quantity of the C pool (section 4.1 and Table 2.1).

81

5 Results

Table 5.14 Physiological parameters for each functional group. Assimilation and production efficiencies (a and p) taken from Andrén et al. (1990) and C:N-ratios (q) from Hunt et al. (1987) if not stated otherwise. Basic death rates (at 10°C) were obtained from Hunt et al. (1987) and De Ruiter et al. (1993a) and adapted according to temperature and moisture regime of the specific sites.

death rate (a-1) a p q N-SE S-SE DE FR

microflora Bacteria 1.00a 0.30a 4 0.4 0.7 0.6 0.6 Fungi 1.00a 0.30a 10 0.4 0.7 0.6 0.6

microfauna Testate Amoebae

panphytophagous 0.70 0.43 7 2.0 3.6 3.0 3.0 predaceous 0.70 0.43 7 2.0 3.6 3.0 3.0

Nematoda bacterivorous 0.30 0.40 5b 0.9 1.6 1.4 1.3 fungivorous 0.30 0.40 5b 0.6 1.1 1.0 1.0 omnivorous 0.60 0.33 7c 1.5 2.6 2.2 2.2 predaceous 0.60 0.33 5b 0.5 1.0 0.8 0.8

mesofauna Acari

panphytophagous 0.25 0.40 5.5b 0.6 1.1 0.9 0.9 predaceous 0.80 0.30 8 0.6 1.1 0.9 0.9

Collembola panphytophagous 0.25 0.40 8 0.6 1.1 0.9 0.9 predaceous 0.80 0.30 8 0.6 1.1 0.9 0.9

Enchytraeidae 0.25 0.40 6c 1.7 3.0 2.5 2.5 a Hunt et al. 1987. b Persson 1983. c Berg 1997.

5.2.3.2 Feeding preferences Since most functional groups fed on more than one food source their feeding preferences were considered via feeding preferences which enter the model as described in section 4.2.3 (Table 5.15). Most preferences were taken from Hunt (1987). The preferences concerning Testate Amoebae result

from literature studies as summarised above (section 5.2.2). Within Table 5.15 a value of � 1 means

that the functional group indicated in the header of a specific column feeds on the functional group in the specific row. Values > 1 put emphasis on a certain trophic link, with the weighting being dependent on the total food biomass consumed by the functional group. The same feeding preferences were used at all sites.

82

5 Results

Table 5.15 The feeding preferences wij. Explanation see section 4.2.3. prmi = predaceous Acari; prco = predaceous Collembola; prne = predaceous Nematoda; omne = omnivorous Nematoda; pami = panphytophagous Acari; paco = panphytophagous Collembola; prta = predaceous Testate Amoebae; pata = panphytophagous Testate Amoebae; fune = fungivorous Nematoda; bane = bacterivorous Nematoda; ench = Enchytraeidae; fung = fungi; bact = bacteria; detr = total detritus.

saprotrophs, grazers, predators prmi prco prne omne pami paco prta pata fune bane ench fung bact

prmi 0 0 0 0 0 0 0 0 0 0 0 0 0 prco 1 0 0 0 0 0 0 0 0 0 0 0 0 prne 1 1 0 0 0 0 0 0 0 0 0 0 0 omne 1 1 1000 0 0 0 0 0 0 0 0 0 0 pami 2 1 0 0 0 0 0 0 0 0 0 0 0 paco 2 1 0 0 0 0 0 0 0 0 0 0 0 prta 1 1 10 1 0 0 0 0 0 0 0 0 0 pata 1 1 10 1 0 0 1 0 0 0 1 0 0 fune 1 1 1000 1 0 0 0 0 0 0 0 0 0 bane 1 1 1000 1 0 0 0 0 0 0 0 0 0 ench 0 0 0 0 0 0 0 0 0 0 0 0 0 fung 0 0 0 1 10 100 0 10 1 0 10 0 0 bact 0 0 1 1 100 0 0 1000 0 1 10 0 0

food r

esp.

prey

detr 0 0 0 1 1 1 0 1 0 0 1 1 1 5.2.3.3 Biomasses of functional groups Analyses of variance on the effect of 'site' on absolute and relative contributions and on total food web biomass were performed (Table 5.16). A significant effect of 'site' not only on absolute but also on relative biomass indicates a structural difference independent of total food web biomass. To be able to compare the different sites concerning their biomass structure the relative contributions of functional groups to the total biomass at each site are given in Table 5.17. Such structural differences were found for bacteria, fungi, bacterivorous, fungivorous and predaceous nematodes, predaceous microarthropods and enchytraeids (Table 5.16 and 5.19).

83

5 Results

Table 5.16 Results of ANOVAs on the main effect of ‘site’ on biomass (% resp. kg C ha-1) and on simulated C and N mineralisation (% resp. kg ha-1 a-1) of individual functional groups within the decomposer food web (one-way ANOVA). Because of negative N mineralisation values (immobilisation by bacteria) no relative contributions were calculated for this variable. df-effect = 3; F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * ≤ 0.05; ** < 0.01.

biomass simulated mineralisation rates (kg C ha-1) (%) C (kg ha-1 a-1) C (%) N (kg ha-1 a-1) F p F p F p F p F p

microflora Bacteria 9.6 ** 575 ** 3.1 n.s. 1.7 n.s. 2.5 n.s. Fungi 0.8 n.s. 12.6 ** 4.2 * 2.0 n.s. 14.7 **


panphytophagous 2.5 n.s. 2.3 n.s. 3.5 * 2.4 n.s. 3.0 n.s. predaceous 3.7 * 3.4 n.s. 4.1 * 4.4 * 4.1 *

Nematoda bacterivorous 8.7 ** 4.6 * 4.3 * 22.8 ** 4.3 * fungivorous 7.6 ** 3.6 * 4.0 * 21.6 ** 4.0 * omnivorous 1.5 n.s. 1.2 n.s. 1.7 n.s. 1.2 n.s. 4.0 * predaceousa 92.1 ** 38.1 ** 97.7 ** 89.5 ** 92.6 **

mesofauna Acari

panphytophagous 5.0 * 2.1 n.s. 7.0 ** 0.4 n.s. 5.8 * predaceous 13.8 ** 3.5 * 16.2 ** 1.0 n.s. 16.1 **

Collembola panphytophagous 3.8 * 2.0 n.s. 4.1 * 0.7 n.s. 4.2 * predaceous 7.8 ** 12.7 ** 8.0 ** 21.2 ** 8.0 **

Enchytraeidae 14.8 ** 28.0 ** 31.1 ** 13.9 ** 11.4 **

total food web 0.9 n.s. 4.3 * 39.0 ** a due to a number of zero values in this particular data set the homogeneity of variances

assumption is violated according to the SEN & PURI-test. However; Lindman (1974) shows that the F statistic is quite robust against violations of this assumption.

The total biomass of the decomposer food web ranged from 360 to 540 kg C ha-1 without any statistical significant differences between sites (Table 5.16 and 5.19). The difference between the food webs thus principally lay in the pattern of functional group contributions to total food web biomass, in the following referred to as biomass structure. More specifically, fungi made up 79 to 91 % of the total food web biomass. Their relative contribution was significantly highest at N-SE, and gradually decreasing towards South (Table 5.17). Relative bacterial biomass was significantly different between all sites with a maximum at FR, but no gradual trend along the transect (Table 5.17). The biomass of fungi strongly dominated over that of bacteria. There was an overall trend of increasing faunal biomass towards South (Table 5.17), reaching a maximum of 14 % of the total food web at DE. Within the soil fauna the Testate Amoebae were the most important contributors, with a particular prominence at the Southern site DE (Table 5.17). Nematoda

84

5 Results

contributed little to total food web biomass and were the only faunal group that showed maximum biomass at N-SE, with significantly higher relative biomass contributions of predaceous, fungivorous and bacterivorous Nematodes. Relative microarthropod biomass increased towards South with a significant maximum for predaceous mites at DE. While the relative biomass of mites was surprisingly low at FR, predaceous Collembola revealed a significant biomass maximum at this site (Table 5.17). Enchytraeidae biomass increased significantly towards South.

Table 5.17. Biomasses of functional groups (%) and total (kg C ha-1) at each site. Mean values of four sampling occasions are shown, standard deviations are given in parentheses. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.16 for details on the ANOVA).

N-SE S-SE DE FR

microflora Bacteria 5.1A (0.1) 6.9B (0.4) 4.1C (0.2) 11.4D (0.3) Fungi 91A (1) 86AB (5) 82B (4) 79B (2)


panphytophagous 1.9A (1.2) 4.7A (3.7) 7.7A (4.0) 4.5A (2.6) predaceous 0.4A (0.3) 1.4A (0.6) 2.2A (1.3) 1.1A (0.6)

Nematoda bacterivorous 0.16A (0.05) 0.04B (0.03) 0.06AB (0.06) 0.07AB (0.06) fungivorous 0.04A (0.01) 0.02A (0.01) 0.02A (0.02) 0.02A (0.02) omnivorous 0.02A (0.01) 0.02A (0.01) 0.05A (0.05) 0.02A (0.01) predaceous 0.013A (0.004) 0.000B (0.000) 0.001B (0.001) 0.000B (0.000)

mesofauna Acari

panphytophagous 0.3A (0.2) 0.5A (0.4) 0.8A (0.4) 0.4A (0.1) predaceous 0.03A (0.01) 0.08AB (0.04) 0.15B (0.10) 0.08AB (0.01)

Collembola panphytophagous 0.5A (0.3) 0.4A (0.2) 1.6A (1.4) 1.3A (0.9) predaceous 0.003A (0.004) 0.002A (0.002) 0.017A (0.002) 0.091B (0.047)

Enchytraeidae 0.1A (0.2) 0.4A (0.2) 1.9B (0.5) 1.8B (0.4)

total food web 476A (105) 357A (122) 469A (197) 538A (195) Plotting correlation coefficients between the relative biomass structure at different sites (pair-wise PEARSON correlation) against their geographical distance led to the observation of a similarity trend. Similarity of biomass structure generally reflected geographical distance; the further apart two sites the less similar they were. An extreme outlier to this trend was the correlation r between DE and FR, which revealed that these sites were least similar even though they lie closest to each other (Figure 5.14). When this pair was omitted, the correlation between food web similarity and geographical distance became significant (PEARSON correlation r = -0.91, p < 0.05, point FR/DE excluded from regression; Figure 5.14)

85

5 Results

0.993

0.994

0.995

0.996

0.997

0.998

0.999

1.000

0 500 1000 1500 2000 2500

distance (km)

corre

lation

coeff

icien

t

( FR/DE)

DE/S-SE

S-SE/N-SE

FR/S-SE DE/N-SE

FR/N-SE

Figure 5.14 Correlation coefficients (r) between the biomass structure at different sites (relative contributions to total biomass, pair-wise correlation) against the geographical distance (km) between the sites (straight line: r = -0.91, p < 0.05, point FR/DE excluded from regression).

5.2.3.4 Simulated estimates of total C and N mineralisation Simulated total carbon and nitrogen mineralisation were significantly different between the sites (Table 5.16, Figure 5.15). The mineralisation rates did not simply follow the total food web biomasses: significant differences were found between total mineralisation rates although the total food web biomasses were similar at all sites (Table 5.16). The simulated rate of carbon mineralisation at N-SE (800 kg C ha-1 a-1, Figure 5.15 A) was significantly lower than the rates at the two Southern sites DE and FR (2600 resp. 2400 kg C ha-1 a-1). For S-SE, the site situated in between the boreal and the two Southern sites, an intermediate C mineralisation rate was estimated (1500 kg C ha-1 a-1). The pattern of simulated nitrogen mineralisation followed that of carbon (Figure 5.15 B). N mineralisation was estimated to be almost zero at the boreal site N-SE. At S-SE and FR nitrogen mineralisation was calculated to be intermediate with 30 resp. 60 kg N ha-1 a-1. The simulated N mineralisation rate of almost 100 kg N ha-1 a-1 at DE was significantly higher than the rates at the Swedish sites N-SE and S-SE.

86

5 Results

0

500

1000

1500

2000

2500

3000

N-SE S-SE DE FR

C mi

neral

isatio

n (kg

ha-1

a-1)

simulatedobserved

a

ab

b b

A

0

20

40

60

80

100

120

N-SE S-SE DE FR

N mi

neral

isatio

n (kg

ha-1

a-1)

simulatedobserved

a

b

c

bc

B

Figure 5.15 Estimates of C and N mineralisation rates at the different sites obtained using the food web model ("simulated") and laboratory incubations of soil cores ("observed"). Laboratory incubation data are taken from Persson et al. 2000ab. In comparison of columns of the same shading those labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.16 for further details on the ANOVA).

The food web model estimates of total carbon and nitrogen mineralisation were compared to observed mineralisation rates from extrapolated laboratory incubation experiments for the same forest sites and soil layers conducted by Persson et al. (Figure 5.15, Persson et al. 2000a, Persson et al. 2000b). At the two Swedish sites N-SE and S-SE the food web model approach delivered total C mineralisation rates very similar to the observed rates from the laboratory incubation approach. At DE and FR simulated C mineralisation rates exceeded the experimentally observed rates (Figure 5.15 A). The simulated N mineralisation rates were in good agreement with the observed rates at all sites (Figure 5.15 B).

87

5 Results

5.2.3.5 Contribution of functional groups to C mineralisation In order to see if energy pathways within the decomposer food webs shifted in their importance for total fluxes, absolute and relative contributions of functional groups to C mineralisation were compared (Table 5.16, Figure 5.16). When comparing relative contributions the difference in overall mineralisation totals does not mingle principle differences within the functioning of the food webs. Analyses of variance revealed significant effects of the factor ‘site’ on absolute and relative C mineralisation of a number of functional groups (Table 5.16). The microflora was responsible for around 90 % of the total C mineralisation and only around 10 % was performed by the soil fauna (Figure 5.16). At the boreal site N-SE carbon mineralisation was estimated to be largely performed by fungi (bacterial to fungal respiration ca. 30/70), while bacteria and fungi contributed similarly at the other sites (ca. 40/50). The relative contribution of bacteria and fungi was not significantly different at the sites (Figure 5.16), in contrast to the relative biomasses (Table 5.17). The Testate Amoebae as a whole contributed substantially to C mineralisation with a significant maximum at DE (from 6 % or 44 kg C ha-1 a-1 at N-SE up to 12 % or 343 kg C ha-1 a-1 at DE, Figure 5.16). Nematoda added little to overall carbon mineralisation (0.1-0.2 % or 1-2 kg C ha-1 a-1) reaching a maximum at N-SE. Only at this site did predaceous Nematoda contribute significantly. The contribution of total Acari reached a maximum of 0.3 % (or 6 kg C ha-1 a-1) at the German site DE, and ranged around 0.2 % (or 1-4 kg C ha-1 a-1) at the other sites. Predaceous mites made the largest contribution at DE. Collembola as a whole mineralised between 0.2-0.3 % (or 2-3 kg C ha-1 a-1) at the Swedish sites, and up to around 0.6 % (or 10-11 kg C ha-1 a-1) at the two Southern sites. Predaceous Collembola contributed significantly most at FR. The contribution of Enchytraeidae became increasingly important from North to South. It ranged from 0.2 % (or 1 kg C ha-1 a-1) at N-SE and 0.4 % (or 5 kg C ha-1 a-1) at S-SE to up to 1.4-1.6 % (or more than 30 kg C ha-1 a-1) at the two Southern sites (Figure 5.16).

88

0.0 0.5 1.0 1.5 2.0

Enchytraeidae

predaceous

panphytophagous

predaceous

panphytophagous

predaceousomnivorous

fungivorous

bacterivorous

predaceous

panphytophagous

N-SE

5 aa

aa

a

a

aa

aa

a

Testate Amoebae

Nematoda

Acari

Collembola

0.0 0.5 1.0 1.5 2.0

S-SE

10 a

a

a

a

aa

a

bb

b

ab

0.0 0.5 1.0 1.5 2.0

DE

10 a

b

b

b

b

b

a

a

a

a

a

0 10 20 30 40 50 60 70 80

FungiBact a

aBacteria

0 10 20 30 40 50 60 70 80

aa

0 10 20 30 40 50 60 70 80

aa

0 10 20 30 40 50 60 70 80

aa

0.0 0.5 1.0 1.5 2.0

FR

bb

7 a

ab

ba

a

aa

bb

0.0 0.5 1.0 1.5 2.0

Enchytraeidae

predaceous

panphytophagous

predaceous

panphytophagous

predaceousomnivorous

fungivorous

bacterivorous

predaceous

panphytophagous

N-SE

5 aa

aa

a

a

aa

aa

a

Testate Amoebae

Nematoda

Acari

Collembola

0.0 0.5 1.0 1.5 2.0

S-SE

10 a

a

a

a

aa

a

bb

b

ab

0.0 0.5 1.0 1.5 2.0

DE

10 a

b

b

b

b

b

a

a

a

a

a

0 10 20 30 40 50 60 70 80

FungiBact a

aBacteria

0 10 20 30 40 50 60 70 80

aa

0 10 20 30 40 50 60 70 80

aa

0 10 20 30 40 50 60 70 80

aa

0.0 0.5 1.0 1.5 2.0

FR

bb

7 a

ab

ba

a

aa

bb

0 10 20 30 40 50 60 70 80

aa

0.0 0.5 1.0 1.5 2.0

FR

bb

7 a

ab

ba

a

aa

bb

Figure 5.16 Relative contributions of the functional groups to C mineralisation (%) at the sites. In comparison of the mineralisation by a particular functional group at the sites (i.e. the horizontal comparison of bars on the same level of the ordinate axis) values labelled with identical letters are not significantly different from each other according to the Tukey HSD test of a one-way ANOVA on the main effect ‘site’ (see Table 5.16 for further details on the ANOVA). Whiskers on bars represent standard deviation.

5 Results

5.2.3.6 Contribution of functional groups to N mineralisation The contributions of most functional groups to N mineralisation were significantly different between the sites (Table 5.16). Negative contributions to N mineralisation represent N immobilisation. At all sites bacteria were estimated to immobilise N (Table 5.18). Because of the negative contribution of the bacteria to N mineralisation, relative contribution estimates in percentages are not given. At N-SE the absolute amount of N immobilisation was smaller than at the other sites (-11 kg N ha-1 a-1 at N-SE and -14 to –32 kg N ha-1 a-1 at the other sites). At all sites the largest contribution to N mineralisation is made by panphytophagous Testate Amoebae (9-65 kg N ha-1 a-1), followed by fungi (2-39 kg N ha-1 a-1), and predaceous Testate Amoebae (0.7-5.9 kg N ha-1 a-1). These three groups reveal the same pattern in between sites: they contribute most at DE, then FR, less at S-SE and a minimum at N-SE, but only some of these differences were significant (Table 5.18). N-SE was the only site at which the fungivorous and predaceous Nematoda contributed to N mineralisation, and the bacterivorous Nematodes showed a maximum contribution at N-SE (0.34 kg N ha-1 a-1). The mesofauna played a minor role at N-SE compared to the other sites. Its contribution increased from North to South, with maxima at the Southern-most site FR for panphytophagous mites, panphytophagous and predaceous Collembola, and Enchytraeidae (Table 5.18). The faunal groups are of greater importance for N than for C mineralisation at all sites. At N-SE fungi alone counterbalanced only about one fifth of bacterial N immobilisation (bacterial to fungal N mineralisation -11/2), a remaining 11 kg N ha-1 a-1 was mediated by fauna (Table 5.18). At S-SE more than half of the N immobilisation was counterbalanced by fungi (-24/13), at FR the relationship was nearly balanced (-32/26). At DE the fungal N mineralisation was estimated to exceed bacterial N immobilisation (DE –14/39).

90

5 Results

Table 5.18 Contributions of functional groups within the decomposer food web to N mineralisation (kg N ha-1 a-1). Mean values of four sampling occasions are shown, standard deviations are given in parentheses. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test of a one-way ANOVA on the main effect ‘site’ (see Table 5.16 for details on the ANOVA). N-SE S-SE DE FR

microflora Bacteria -11A (4) -24A (9) -14A (8) -32A (20) Fungi 2A (0) 13AC (4) 39B (14) 26BC (9)


panphytophagous 9A (4) 34A (14) 65A (40) 57A (38) predaceous 0.7A (0.5) 3.2AB (1.0) 5.9B (3.3) 3.7AB (2.4)

Nematoda bacterivorous 0.34A (0.07) 0.09B (0.02) 0.13AB (0.13) 0.19AB (0.16) fungivorous 0.01A (0.00) 0.00B (0.00) 0.00AB (0.00) 0.00AB (0.00) omnivorous 0.00A (0.00) 0.00AB (0.00) 0.02B (0.02) 0.01AB (0.01) predaceous 0.01A (0.00) 0.00B (0.00) 0.00B (0.00) 0.00B (0.00)

mesofauna Acari

panphytophagous 0.06A (0.03) 0.15AB (0.10) 0.20AB (0.07) 0.34B (0.15) predaceous 0.03A (0.01) 0.10B (0.02) 0.19C (0.05) 0.13BC (0.03)

Collembola panphytophagous 0.17A (0.11) 0.18A (0.11) 0.68A (0.43) 0.74A (0.04) predaceous 0.00A (0.00) 0.00A (0.00) 0.03A (0.01) 0.17B (0.11)

Enchytraeidae 0.0A (0.0) 0.0A (0.1) 0.5B (0.3) 0.9B (0.4)

91

Chapter 6

Discussion


6 Discussion

6.1 Testate Amoebae community structure Totalling the observations on Testate Amoebae community structure, it is to say that in spite of the considerable geographical distance covered by the North-South-transect (approx. 2000 km) the communities at the sites were quite similar. Differences in latitude, climate, atmospheric pollution and altitude seemed to have only a modulating effect. Within these small differences in the community structures some interesting trends emerged.

6.1.1 Total number of Testate Amoebae species The total number of Testate Amoebae species found was 42. This is in the range previously reported for spruce forests. In spruce forests in Germany Wanner (1991) found 32 species, Rauenbusch (1987) 16 species and Schroeter (1995) 42-61 species. Coûteaux (1976) reported 71 species from a spruce forest in France. All species found in this study have previously been found at temperate spruce sites (Coûteaux 1976, Rauenbusch 1987, Aescht and Foissner 1989, Wanner 1991). Like other faunal and floral taxa, numbers of Protozoan species are assumed to increase with decreasing latitude and altitude (Foissner 1987, Cowling 1994, Smith 1996, Coûteaux and Darbyshire 1998, Chown and Gaston 2000). Moreover, increased N supply has been shown to enhance species richness (Chardez et al. 1972, Berger et al. 1986). Along the transect positive effects of latitude and increased N availability counteract with negative effects of altitude. Nevertheless, the total number of Testate Amoebae species increased towards South. The mean number of species, diversity and evenness was significantly lowest at the boreal site. It has been argued that biogeographical rules of species richness distribution generally true for larger animal species do not apply on the microbial scale (Finlay and Fenchel 1996, Finlay and Fenchel 1999, Hillebrand et al. 2001). Because this study was part of a larger project comparisons with species numbers of other taxa from the same sites can be made. During the same investigation period Collembola and ectomycorrhizal fungi were studied (Taylor et al. 2000, Pflug 2001). At the four sites a range of 30-38 Collembola species were found, with the highest species number at N-SE and the lowest at S-SE (Pflug 2001). Further sampling may have retrieved further Collembola species at FR and N-SE (Pflug 2001). Such effort is not expected to render further species of Testate Amoebae due the general rule that relieable accounts of total Protozoa species richness can be retrieved from 'small samples'

93

6 Discussion

(Finlay and Fenchel 1996). The amount of material investigated in this study was well above the sample size to which Finlay and Fenchel (1996) refer. Taylor et al. (2000) studied the number of ectomycorrhizal morphotypes on root tips and found a range between 14-19 species, with the maximum at N-SE and the minimum at DE. Thus the pattern of species richness was different for the three taxa Testate Amoebae, Collembola and ectomycorrhizal fungi. The comparison of Testate Amoebae species numbers with other taxa must be interpreted considering that the species definition within this group, likewise the group of ectomycorrhizal fungi, is still obscure (Finlay and Fenchel 1996). Most Testate Amoebae species reproduce asexually by cell division (Grospietsch 1965b). The species described are classified by their shell morphology and represent morphospecies. Shell morphology is, however, known to be subject to modifications according to environmental conditions (Schönborn 1992a, Bobrov et al. 1995). To conclude, the hypothesis that diversity of Testate Amoebae increases with decreasing latitude is corroborated by the finding that species richness increased towards South. The same trend was revealed by comparison of the Shannon-Wiener diversities of the communities. The effect of geographical location on both parameters was significant as tested by ANOVA. However, overall differences were small and comparison of means by Tukey HSD test indicated high similarity among S-SE, DE and FR.

6.1.2 Similarity between Testate Amoebae communities The similarity between the Testate Amoebae communities was compared by the number of unique species (Colwell and Coddington 1994) and the Bray & Curtis-similarity index (Bray and Curtis 1957, Southwood 1994). The latter index takes the abundance pattern of the species into account. Both measures indicate that, on an overall high level, similarity decreased with increasing geographical distance, e.g. number of unique species significantly increased between sites that were further apart. It has been argued that Protozoan species communities are very similar over broad geographical ranges (Finlay and Fenchel 1996). The large abundances of Protozoa are believed to result in high migration rates, improbability of extinctions and rarity of allopatric speciation (Finlay and Fenchel 1999). The observed biogeographical trend in community similarities took place on a high level of similarity (Bray & Curtis Index 34-70 %). The similarity between Collembola species communities of the same sites was considerably lower and decreased with increasing geographical distance between the sites (Bray & Curtis-Index 11-39 %; Pflug, 2001). Thus, in contrast to patterns in species richness, Testate

94

6 Discussion

Amoebae and Collembola revealed the same pattern of community similarity as measured by the Bray & Curtis index. In summary, the similarity between Testate Amoebae communities was large compared to a metazoan taxon along the same transect. Nevertheless, the general rule that larger distances represent a barrier for migration may apply also for Testate Amoebae. It has to be considered that, besides migration, local conditions leading to extinctions of species determine the species community at a given site.

6.1.3 Comparison of species biomass pattern A comparison of the relative biomass patterns of species at the sites indicated that many species were 'cosmopolitan', i.e. they occurred at all sites studied. Ordination and multivariate regression with environmental parameters (CCA) further suggested that most species are generalists. It has to be kept in mind that this assignment concerns only the range of environmental variables taken into account and the breadth of gradients studied. The range of pH was, for example, quite small (pH 3.1-4.1). Many Testate Amoebae species occurred with high relative biomasses. A number of those have been described before as being generalists in acid forest litters, e.g. Nebela parvula/tincta, Trinema lineare,

Corythion dubium, Nebela tincta major/bohemica/collaris, Plagiopyxis declivis, Centropyxis sylvatica,

Trinema complanatum, and Cyclopyxis eurystoma (Bonnet 1988a, Aescht and Foissner 1989, Bonnet 1989b, a, 1990, 1991a, b). Few species were found that occurred only at a single site. The restriction of Edaphonobiotus

campascoides to DE indicates a possible preference of this species for high N supply and acidity, which is supported by the species biplot within the CCA. Conversely Bullinularia indica and Centropyxis

gauthieri were found at all sites but DE. The CCA indicates that Bullinularia indica may positively be linked to higher pH and microbial biomass. Characteristic species or absence of some species were expected especially for the boreal site N-SE because of its distinctive features. Three species had higher relative biomasses at N-SE than at all other sites (Arcella catinus, Cryptodifflugia oviformis and Difflugia lucida). Hyalosphenia subflava and Heleopera sylvatica were absent from N-SE but occurred with high relative biomass at all temperate sites. H. subflava has been described to rely on constant humidity and H. sylvatica is judged to be sensitive to extreme environmental conditions (Bonnet 1990), their absence from the boreal site which is low in precipitation and subject to climatic extremes might be explained by this sensitivity. However, there were also two species that were absent only from DE, and two species that were absent only from

95

6 Discussion

S-SE. Cyclopyxis kahli, a species that was found regularly only at N-SE (and very rarely at S-SE) was associated to the higher pH at this site according to CCA. At present there are divergent perspectives on the species distribution of Protozoa. Some believe that 'everything is (almost) everywhere', i.e. that most species are ubiquist cosmopolitans (Finlay and Fenchel 1996, Fenchel et al. 1997, Finlay and Fenchel 1999). Others argue that increased scientific effort will discover more and more specialist and endemic species and patterns in species distribution (Foissner 1987, Foissner 1997, Coûteaux and Darbyshire 1998, Foissner 1998). According to the definitions of Smith (1978) cosmopolitans occur in all regions of the world, but not necessarily in all types of habitats, while ubiquists occur in all types of habitats, but not necessarily in all regions of the world. This study cannot answer the question if Testate Amoebae are generally ubiquists, because only one type of habitat was studied. However, the findings reported provide some argument that many Testate Amoebae may be distributed over a wide geographical distance, considering that only a range of temperate to boreal sites over a transect of 2000 km were studied. Techniques like scanning electron microscopy were not applied. Such techniques might have resulted in finer species resolution within the taxonomic groups Euglypha cf. strigosa, Nebela tincta major/bohemica/collaris and N tincta/parvula. It has to be taken into consideration that very rare specialists species may falsely have been recorded within these groups. Apart from this taxonomic restriction there is evidence that the species richness estimates were almost exhaustive. The species richness at the sites was studied not only from quantitative samples. Additionally, enrichment techniques (flotations, batch cultures) were used in order to find species that were very rare. However, these techniques did render only one species that had otherwise been overlooked. It is concluded that the sample size of the quantitative samples were sufficient to discover a next to complete range of the species present at the sites. Recapitulating, the hypothesis that the boreal site N-SE will distiguish from the other sites by a number of species was not confirmed. Only two species were common at all other sites were absent from N-SE. However, such cases occurred also at other sites. No species was found only at the boreal site but one species was absent or very rare at all sites except N-SE. One species was restricted to DE and seemed to be associated to higher N-supply according to CCA.

6.1.4 Environmental factors explaining community structure The difference between the Testate Amoebae communities was small and lay largely in the pattern of

96

6 Discussion

relative biomasses of the species, referred to as community structure. When exploring the data set for factors modulating the Testate Amoebae community structure, the canonical correspondence analysis (CCA) emphasised the importance of parameters that may be united into three groups of factors, according to their interrelation. These factors were atmospheric pollution (S and N deposition, pH of organic layer), microbial biomass, and climate (mean annual temperature and precipitation). Moisture and food availability (i.e. availability of microbes) have been proposed to be the most important factors determining Protozoan communities (Stout 1984, Cowling 1994). Positive effects of irrigation have been observed in forest soils (Lousier 1974a, b). The correlation of Testate Amoebae and moisture content in mires is, in fact, used in palaeohydrology to reconstruct mire surface wetness based on qualitative estimates of the hydrological preferences of individual Testate Amoebae species (Tolonen 1986). 'Rhizopod analyses' in palaeoecological studies have established transfer functions for Testate Amoebae species assemblages to water table and percentage moisture in mires (Woodland et al. 1998). Some studies in spruce forest did not find significant correlation between soil moisture and the abundance of Testate Amoebae (Petz and Foissner 1989, Wanner and Funke 1989, Wanner 1991), pointing to overriding effects of other factors or a lack of correlation above a certain threshold value. Although the aim of this study was not to investigate the abundance and biomass dynamics within the Testate Amoebae communities, a common trend was observed that seems worth mentioning in the context of the importance of moisture availability for Testate Amoebae. At the summer sampling relative number of cysts and empty shells was increased and total biomass decreased. Such decreases in population size and shifts towards increased number of dormant cells and empty shells have previously been reported for the dry summer months (Coûteaux 1976, Lousier 1984b, a, Lousier 1984c, 1985). These observations support the view that moisture availability is a limiting factor for the Testate Amoebae communities. The factor microbial biomass probably is important because it indicates food availability for the Testate Amoebae. Peaks of Protozoan biomass have been observed to quickly follow rain induced peaks of bacterial biomass, demonstrating the dependence of Protozoa on food availability (Clarholm 1981). Increased fungal biomass in microcosms has been shown to enhance numbers of the Testate Amoebae species Phryganella acropodia (Coûteaux and Devaux 1983, Coûteaux 1985). It has been pointed out that moisture may effect Protozoa indirectly through affecting the availability of microbial food (Foissner 1987). Literature on the effect of atmospheric deposition on soil Protozoa is quite sparce (Coûteaux et al. 1998). However, positive effects of N fertilisation on the species richness of Testate Amoebae have

97

6 Discussion

been reported (Chardez et al. 1972, Berger et al. 1986). This study confirms the positive effect of increased N supply on a larger geographical scale. In a biplot of sites the CCA was used to aid ordination of the sites according to environmental parameters. With respect to the three main groups of factors N-SE is clearly differentiated from the other sites by low atmospheric pollution, comparatively higher pH, severe climate (low temperatures and precipitation) and high microbial biomass. The other sites are more similar to each other, however, remarkable features are extremely high loads of atmospheric pollution at DE, leading to low pH. To conclude, the hypothesis that moisture and microbial parameters are the most important factors modulating the Testate Amoebae community structure was corroborated. Furthermore, evidence was found that a group of parameters concerning atmospheric pollution has high explanatory value. Increased N supply, decreased pH and/or related factors may exert a positive influence on Testate Amoebae communities. Increased N-supply, like moisture, may act indirectly through enhancing food supply by stimulating bacterial growth.

6.1.5 Size structure and biomass Testate Amoebae species within a large biomass range were found. The biggest species had a biomass of more than 50 times that of the smallest. Because of this range, abundance (number of individuals per unit area) could not logically be linked to resource division (cf. Tokeshi 1993). Biomass measures serve this purpose better, while abundance measures may better reflect community activity. A comparison of both measures revealed that small species occurring with high frequencies and large species occurring in low frequencies both contribute substantially to the biomass of the total community. In a study of German spruce forests similar relationships between body size and frequency of Testate Amoebae were found (Wanner 1991). In fact, a general rule has been stated that numerical density of individual animal taxa (e.g. numbers per unit area) increases with the inverse of body mass (Peters 1983). Small species like Trinema lineare, Cryptodifflugia oviformis and Euglypha laevis are rated to be r-selected, while larger species with xenosome shells, e.g. Centropyxis sylvatica, Heleopera sylvatica and Bullinularia indica, are rated to be K-selected (Wodarz et al. 1992, Bamforth 1997). Among both groups some species occurred with high relative biomasses, suggesting that even within the group of Testate Amoebae, strategies along a wide range of the r/K-continuum are successfully realised. The general rule that rates of growth and metabolism are inversely proportional to body size applies

98

6 Discussion

also on microscopic scale (Fenchel 1988). Therefore, smaller Testate Amoebae may play a larger role for energy and nutrient fluxes than bigger species, even though their total biomass is equal or less. See section 6.2 for further discussion of Testate Amoebae biomass along the transect and within the context of the decomposer food web.

6.1.6 Abundance of Testate Amoebae The total abundance of living Testate Amoebae increased towards South, with maximum values found at DE followed by FR. Counteracting influences seem to take effect. On one hand, both, decreasing latitude and increasing N supply have a positive effect on Testate Amoebae abundance (Chardez et al. 1972, Berger et al. 1986, Foissner 1987, Foissner 1994). On the other hand, increasing altitude has a negative effect on abundance (Foissner 1987). Therefore it is suggested that the non-continuous abundance trend along decreasing latitude may be explained by the counteracting altitudinal gradient (high altitudes at DE resp. FR: 700 m asl resp. 1050 m asl). In addition to the maximum altitude of FR, N deposition at this site was lower than at DE. The total abundance of living cells found (700-2700 106 m-2) were in the upper range or above values reported from other coniferous forests. In the review by Petersen and Luxton (1982) a range of 190-1050 106 m-2 is given for woodland moder. Wanner (1991) reported 20-370 106 m-2 in German spruce forests, using an extraction method based on the flotation principle (Chardez 1959). Rauenbusch (1987) found up to 900 106 m-2 Testate Amoebae in a beech forest, using a direct method involving several dilution steps. Estimates of the abundance of soil Protozoa are generally obtained with much variation due to spatial and temporal hererogeneity of their substrate and methodological difficulties related to the small size of these organisms (Foissner 1987, Ekelund and Ronn 1994). Today direct counting methods from soil suspensions are preferred for some groups of Protozoa (Foissner 1987, Ekelund and Ronn 1994). In an extensive review of the different methods Foissner (1987) resumes that the most reliable abundance values are obtained with direct counting techniques. When directly counting Testate Ameobae from soil solutions a compromise has to be obtained concerning the concentration of the solution. On the one hand, in a dense suspension the risk of overlooking specimen that are hidden by soil particles is high. On the other hand, in a more diluted suspension bigger species that occur with lower frequency are not found in sufficient numbers for extrapolation. Furthermore a certain inaccuracy is unavoidable in diluting the samples and such errors accumulate with each dilution step. In this study each sample was investigated in a direct counting approach involving two steps: (i) counting a sufficient number of sub-

99

6 Discussion

samples of a dense suspension to obtain estimates of the larger specimen, and (ii) counting a sufficient number of sub-samples of a thin suspension to obtain estimates of the smaller specimen and to reduce risk of overlooking. The hypothesis that the abundance of Testate Amoebae increases from North to South is corroborated. Considering the results discussed in 6.1.4 this increase could be explained by increasing moisture and food availability at the sites with milder climate and high N deposition.

6.1.7 Comparison of bio- and necrocoenosis The number of living cells and cysts (biocoenosis) was on average two to three times the number of empty shells (necrocoenosis). The relative size of the biocoenosis (63-76 %) was unusually high at all sites compared to values from the literature (Meisterfeld 1980, Wanner 1991). Meisterfeld (1980) found 5-18 % living Testate Amoebae in a spruce forest and up to 46% in the upper 2 cm of a meadow soil. At the boreal site N-SE the relative size of the necrocoenosis was significantly smaller than at the other sites: relative to the total community less empty shells occurred. A small necrocoenosis is a consequence of high decomposition rates and/or low mortality rates (Meisterfeld 1980). The hypothesis that the relative size of the necrocoenosis of Testate Amoebae would be larger at the boreal site due to decreased decomposition is not corroborated. Quite contrary the reverse was found. The low relative number of empty shells occurring at N-SE cannot be explained by high decomposition rates, since N-SE revealed smaller rates than all other sites (see section 6.2.1). A possible explanation is that the relative amount of empty shells was small due to low metabolic rates of the Testate Amoebae in consequence of the severe climatic conditions. This is in accordance with the climatic adjustment used in the modelling approach (chapter 4).

100

6 Discussion

6.2 Testate Amoebae and the decomposer food web

"It is true that mathematical models simplify very much. But this is not characteristic of mathematical models

– it is characteristic of any attempt to comprehend the world." Peter Yodzis 1989

6.2.1 Total food web biomass and mineralisation along the transect A dynamic and interactive set of environmental variables like climate and litter quality determine the structure and function of the decomposer food web through their long-term influence on the ecosystem. The latitudinal transect from Northern to Southern Europe formed by the coniferous sites covers a range of climate and N depositional loads. The conditions at N-SE are rather extreme (boreal, proximately no N deposition) compared to the three Southern sites (humid-oceanic resp. humid continental, N deposition ranging from 15-20 kg N ha-1 a-1). In consequence, N-SE clearly differentiates from the other three sites (see section 6.1.4). The complexity of the food webs as referring to the number of functional groups times connectance was similar at all sites (sensu Paine 1988). Results show that opposed to the initial hypothesis total food web biomasses were also alike. However, the biomass structures of the functional groups were different. A tendency of increasing dissimilarity between the biomass structure of the food webs with increasing geographical distance between the sites was observed. However, the correlation beween dissimilarity and distance was only significant when an outlier was omitted from the analysis: The similarity beween FR and DE did not fit the trend but was unexpectedly low. Total C and N mineralisation rates of the decomposer food webs in the organic layer were site specific. The lowest rates of N and C mineralisation were found at the Northern-most site N-SE, intermediate rates at the Southern-Swedish site S-SE and maximum rates at the two Southern-most sites DE and FR. Since the food web complexity and the total food web biomass at the sites was similar, the pattern of total mineralisation rates could be a consequence of three remaining aspects: (i) direct climatic effects on the process rates of the organisms, (ii) differences in resource quality, and (iii) different biomass structures of the food webs. To test this, the food webs were modelled again without taking climatic differences and differences in resource quality (C:N-ratio of the organic material) of the sites into

101

6 Discussion

account. While the absolute estimates changed drastically, the pattern of mineralisation at the different sites remained (Table 10.3, section 10.2). Hence, the differences in mineralisation rates were to some extent attributed to the characteristic pattern of functional group contributions to total biomass. This corroborates the view that the structure and not only the total biomass of the decomposer food web determines C and N fluxes (Moore et al. 1993, Setälä et al. 1996, Setälä et al. 1998).

6.2.2 Site specific characteristics of the food webs along the transect At the boreal site N-SE the fungal pathway was of enhanced importance. The bacteria to fungi ratio of C mineralisation was estimated to be around 30/70. Total N mineralisation at N-SE was estimated to be negligible. The role of fungi has been stated to be pivotal for boreal forest ecosystems in general (Näsholm et al. 1998, Lindahl et al. 2001). This can be attributed to their ability to translocate carbon and nutrients and the capacity of ectomycorrhizal fungi to utilise organic nutrients. These abilities are a competitive advantage under conditions of low N supply (Leake and Read 1997). On a fertility gradient of a number of boreal sites, increasing fertility decreased the relative abundance of fungi while total microbial biomass remained unchanged (Pennanen et al. 1999). The view that fungi find especially favourable conditions at N-SE is supported by Taylor et al. (2000) who compared the ectomycorrhizal community of three of the studied sites (namely N-SE, DE and FR) and found the highest species richness, highest diversity and highest abundance of mycorrhizal root tips at the Northern-most site N-SE. Regarding the decomposer fauna at N-SE the food web structure is shifted in favour of Nematoda and to the disadvantage of Testate Amoebae in comparison to the other three sites. Nevertheless the contribution of Testate Amoebae to mineralisation remained higher than that of Nematoda. In comparison to the other sites, the fauna at N-SE was of relatively less importance for C mineralisation (7 % compared to 10-14 % at the other sites). Without fauna, however, the microbial N balance would have been negative at N-SE (bacterial to fungal N mineralisation -11/2). The positive influence of decomposer fauna, especially Protozoa, on N mineralisation is well established (Ekelund and Ronn 1994, De Ruiter et al. 1993b, Beare et al. 1995). The contribution of Enchytraeidae to biomass and mineralisation was estimated to be of minor importance at N-SE and S-SE and increased towards South. At DE and FR the Enchytraeidae were the second important faunal group for C mineralisation after the Testate Amoebae. The estimates for the contribution of Enchytraeidae to C mineralisation at N-SE and S-SE were 0.2 resp 0.4 % which corresponds to 1.4 resp. 5.2 kg C ha-1 a-1. This is in agreement with findings from field investigations

102

6 Discussion

(Huhta and Koskenniemi 1975, Huhta 1976, Berg 1997). For example Huhta and Koskenniemi (1975) found the C mineralisation by Enchytraeidae to be ca. 3.3 kg C ha-1 a-1 at a boreal site comparable to N-SE and ca. 8.7 kg C ha-1 a-1 at a eutrophic spruce site comparable to S-SE. Many microcosm studies support the view that Enchytraeidae are a keystone group in boreal forest soils (Huhta et al. 1998, Laakso and Setälä 1999b). The direct relative contributions of Enchytraeidae as estimated in this study suggest that the importance of this group is largely due to indirect influences which are included in results from microcosm studies but remain largely unconsidered in the presented approach. However, Sulkava et al. (1996) point out that microcosm studies may overestimate the effect of Enchytraeidae due to artificial conditions favouring the growth of semi-aquatic animals which may then reproduce to densities far exceeding those in the field. In a microcosm experiment where Enchytraeidae were kept at field density levels their presence had no measurable effect on soil respiration (Hedlund and Augustsson 1995). Compared to the fungal-based food web at the boreal site N-SE with slow turnover rates the three Southern sites may be characterised as being bacterial-based with rather fast turnover rates. With respect to C mineralisation bacteria and fungi contributed similarily at these sites, the bacteria to fungi ratio was estimated to be ca. 40/50. A ratio very similar to that has been found in a long term study of a scots pine stand in central Sweden (i.e. 55/45, Jädraås, Persson et al. 1980). Apart from this common feature there are some differences among the bacterial-based food webs and in the nutrient cycling rates at S-SE, DE and FR. In spite of rather mild humid-oceanic climate and considerable N deposition at S-SE mineralisation rates at this site were lower than at DE and FR. Collembola and Enchytraeidae were of relatively less importance compared to the food webs at DE and FR. Similar to N-SE, but to a lesser extent, decomposer fauna was estimated to be essential in order to counterbalance bacterial N immobilisation. The forest at S-SE revealed a shortage in nutrients according to markers for N and P availability (C:N-ratio, P:N-ratio, Table 2.1). Firstly, the C:N-ratio of the organic layer was considerably higher at S-SE than at DE and FR. Secondly, the P:N-ratio of the tree foliage of 0.08 indicated a P deficiency at S-SE. The ratio between phosphorus and nitrogen in the tree foliage is an indicator to interpret nutrient status of the soil. P is considered to be deficient at ratios between 0.10-0.12 (Ingestad 1979 and Nihlgård 1990 cited in Setälä et al. 1997). As a comparably young tree plantation with the highest net primary production (Scarascia-Mugnozza et al. 2000) N and P defficiency at S-SE might therefore have effected the plant-soil interaction in a negative way, leading to decreased activity in the organic layer. The likewise low P:N-ratio (0.11) at DE resulted from increased

103

6 Discussion

N concentration of the needles and this site does not suffer from shortage of P (Bauer et al. 2000). The soil nutrient status despite high N deposition at S-SE acknowledges the need to consider site history (past land use management practices, see e.g. Watson and Mills 1998, Aber et al. 1998). At FR and DE, likewise two sites with favourable climate and high N deposition, mineralisation rates were high. The contribution of Testate Amoebae, Microarthropoda and Enchytraeidae to biomass and C mineralisation was increased at these Southern-most sites. This is principally attributed to less adverse conditions towards South (Lavelle et al. 1995, Seastedt 2000). The relative contribution of Enchytraeidae to C mineralisation was considerable and exceeded that of the total microarthropods. The importance of Enchytraeidae for N and C mineralisation has been reported to be greater than that of Microarthropoda in microcosm studies of coniferous raw humus (Setälä et al. 1991, Sulkava et al. 1996). Enchytraeids were found to out-compete Microarthropoda under moist conditions (Sulkava et al. 1996). DE is characterised by an especially high N mineralisation rate. At DE estimates of fungal N mineralisation exceeded bacterial immobilisation more than 3-fold. For decomposer microflora the availability of degradable C as energy source is prerequisite to successfully exploit available nutrients (Paul and Clark 1989, Mikola et al. 1998). Enhanced C input increases microbial activity and consequently the N flow through the ecosystem (Andrén et al. 1990). Bioavailability of C is considered a key determinant of soil N cycling (Currie 1999). The increased N mineralisation rate at DE might thus be explained by increased energy availability originating from different sources. One source may be the nitrophilous dense understorey vegetation at DE that is likely to supply easily degradable root exudates to the decomposer system. Another source of energy for the microflora may be the trees, whose N nutrition was especially good at DE (Bauer et al. 2000, Scarascia-Mugnozza et al. 2000). It has been suggested that increased nitrogen availability leads to increased carbon allocation to tree roots (Burton et al. 2001). Well nourished trees might supply especially large amounts of energy to their mycorrhizal fungi. The ectomycorrhizal fungi could expend this supplementary energy to produce more degradative enzymes, thus increasing the availability of nutrients within the decomposer system and enhancing overall turnover rates (Read 1991, Leake and Read 1997, Lindahl et al. 2001).

104

6 Discussion

6.2.3 Common characteristics of the food webs along the transect Some general characteristics of the food webs can be summarised. A common feature at all sites was the prominence of the microflora. The contribution of fauna to total C mineralisation was 7-14%. This estimate is considerably higher than others reported from studies in coniferous forest soils (i.e. 1-5% as reviewed by Persson 1989). This divergence is suggested to be due to the fact that in this study the Testate Amoebae, as the major Protozoan group in coniferous forest soils (Schönborn 1992c), were quantified using a direct counting method instead of a culturing technique. Culturing techniques like the most probable number (MPN) method that have been used in previous studies are unsuitable for this group of Protozoa (Foissner 1987, Aescht and Foissner 1992, Ekelund and Ronn 1994). In general the most important module within the decomposer food web besides fungi seemed to be the Protozoa feeding mainly on bacteria. The contribution of Testate Amoebae to the C flux ranged from 6-12 % (or 44 to 343 kg C ha-1 a-1) and the absolute values were similar to values previously reported for other forest sites (Lousier and Parkinson 1984, Meisterfeld 1986, Schönborn 1992b). The pronounced positive effect of Protozoan grazing on N mineralisation has been demonstrated in microcosm studies (Clarholm 1985, Vreeken-Buijs et al. 1997). The need for including Protozoa in studies of the energy and nutrient cycle in soils that has been stressed by various authors is once again underlined (as reviewed by Petersen and Luxton 1982, Foissner 1987). In agricultural and grassland soils bacteria are reported to contribute substantially to N mineralisation (Hunt et al. 1987, De Ruiter et al. 1993a). In the organic layers of a pine forest, however, they were estimated to immobilise N (Berg 1997). The results from coniferous forests also suggest that N is immobilised by the bacteria and released by organisms grazing on them. Possibly this indicates a principle difference beween agricultural and forest decomposer systems. Because the resources of secondary consumers are relatively N rich, compared to detritus, the soil fauna was of much more importance for N than for C mineralisation, a fact that has been recognised in various studies (Anderson et al. 1981, Hunt et al. 1987, Persson 1989, Setälä et al. 1990, Andrén et al. 1990). Regarding the estimates of the contributions of predaceous Microarthropoda and Nematoda to mineralisation, the view that the functional importance of soil fauna is inversely related to the trophic position of the group is supported (Laakso and Setälä 1999).

105

6 Discussion

6.2.4 Evaluation of model estimates In the following the estimates of total C and N mineralisation rates along the transect shall be evaluated in general. The food web model estimates were compared to extrapolated mineralisation rates from laboratory incubation experiments for the same forest sites and soil layers conducted by Persson et al. ( Persson et al. 2000a, Persson et al. 2000b). The food web model approach delivered values very similar to the experimentally observed rates except for two cases: total C mineralisation at DE and FR . It is hard to draw conclusion with respect to what causes these differences. They might be due to the model formulations, the assumptions underlying the equations and the uncertainties with respect to particular values of the input parameters. Especially the population sizes of the microflora and their specific death rates, as well as the C:N ratios of the microbial substrate are difficult to establish and may have a great impact on the outcome of the model (see e.g. sensitivity analyses carried out by De Ruiter et al., 1993a). It is however also possible that the discrepancy between model results and observations are due to uncertainties with respect to the observed rates. This is underlined by the fact that further mineralisation estimates based on other techniques delivered estimates of C mineralisation that lay inbetween the estimates reported here (14C technique Harrison et al. 2000, and NUCOM model simulation by van Oene et al. 2000). The impression is reinforced that there is no flawless method for estimating mineralisation rates of the decomposer system. A considerable part of the variation within all approaches is due to deviating estimates of the total C pool of the forest soils, site variability and differences in sampling frequency (e.g. Persson et al. collected soil cores at a single sampling time while the food web model runs with estimates from 4 sampling times). Nevertheless, the food web model calculated mineralisation rates that satisfactorily matched the observed rates for N mineralisation at all sites and for C mineralisation at two sites. Therefore, the model calculations were used to estimate the contribution of the various functional groups to the C and N mineralisation rates, being aware that the estimates of contributions to C mineralisation at DE and FR should be treated with care. In this study the food web model was applied to explore the data sets of soil biota and environmental variables, and not as a predictive tool. The model delivered an impression of how environmental variables may effect the contribution of the various functional groups to nutrient cycling.

106

6 Discussion

6.3 Research needs Biogeographical monitoring of Protozoan species has just started. Further studies on a wider range of habitats and biomes are needed. Moreover, autecological studies are called for to refine the trophic and physiological classification of Testate Amoebae and the other taxa involved in the decomposer food web of coniferous forest sites. Much remains to be studied about the very base of the decomposer food web. The organic matter as basic resource of the primary decomposers needs to be characterised in more detail. Life strategies like cycles of activity and inactivity of soil bacteria as well as fungi are poorly understood. Their capacity to alter their C:N-ratio according to the availability of N needs to be further explored (Tezuka 1990), especially since the food web model is quite sensitive to the microbial C:N-ratio (De Ruiter et al. 1993a). Trees as symbiotic organisms and their interchange of energy and nutrients with ectomycorrhizal fungi are of great importance to the decomposer system (Lindahl et al. 2001). Including direct C and N fluxes between primary producers and primary decomposer into the food web model would improve the approach. Discrepancies between results from microcosm studies and food web model estimates for a functional group (e.g. Enchytraeidae) point to the indirect effect of mesofauna within decomposition, which is known to be of great importance (Anderson 1995). A demanding task is to better quantify such indirect effects and to further include them into the food web model. Finally, validation of the simulated estimates on functional group level is called for but remains hard to achieve. Recent studies using stable isotopes have been the most advanced attempts to disentangle the complex structure of food webs (as reviewed in Eggers and Jones 2000). The combination of stable isotope techniques and food web modelling may allow further insight into the functioning of belowground food webs.

107

Chapter 7

Conclusions

7 Conclusions

Part 1 � The Testate Amoebae community structure at four spruce forests sites along a large geographical

transect was remarkably similar. High abundances of this Protozoan taxon may have resulted in high migration rates and improbability of extinctions. Furthermore a strong influence of the common primary resource (spruce litter) may have had an equalising effect on community structure.

� Within the tight range of differences found, the hypothesis that the diversity of Testate Amoebae

increases with decreasing latitude was corroborated.

� As hypothesised, the similarity between Testate Amoebae communities increased with decreasing

geographical distance between the sites, even though the overall level of similarity was quite high. The high similarities found suggest that biogeographical rules of species richness distribution generally true for larger animal species cannot without restrictions be applied to Protozoan taxa.

� Among the small taxonomic group of Testate Amoebae a wide range of strategies within the r/K-

continuum was realised. Both, r-strategic species and K-strategic species, contributed considerably to Testate Amoebae biomass.

� The hypothesis that moisture and microbial parameters are the most important factors modulating

Testate Amoebae community structure was corroborated. Moreover, evidence was found that atmospheric pollution is a relevant factor.

� The hypothesis that the abundance of Testate Amoebae increases from North to South was

confirmed. This increase is suggested to be due to increasing moisture and food availability at the sites with milder climate and high N deposition.

� The hypothesis that the relative size of the necrocoenosis of Testate Amoebae is larger at the boreal

than at the temperate sites due to decreased decomposition is not corroborated. Quite contrary the reverse was found. A possible explanation is that the relative amount of empty shells was small due to low metabolic rates of the Testate Amoebae in consequence of the severe climatic conditions.

109

7 Conclusions

Part 2 � Contrary to expectations the total food web biomass was similar at the sites along the transect,

suggesting a strong influence of the common primary resource (spruce litter). However, a trend of increasing dissimilarity of the food web biomass structures with increasing geographical distance was observed.

� The hypothesis that total C and N mineralisation rates of the decomposer food web increase towards

South was corroborated.

� The differences in mineralisation rates were, besides climate and resource quality, to some extent

attributed to the characteristic pattern of functional group contributions to total biomass. This corroborates the view that the structure of the decomposer food web determines C and N fluxes.

� The hypothesis that the decomposer system at the boreal site N-SE is dominated by fungal pathways

and exhibits 'slow cycles' was corroborated. Severe climate and low N availability seem to favour such systems.

� Increasing N input and mild climate resulted in gradual realisations of high input systems favouring

bacterial pathways and 'fast cycles'. Especially the two Southern-most sites were nutrient rich with high turnover rates.

� The importance of fauna to absolute and relative C mineralisation increased from North to South. The

absolute contribution of fauna to N mineralisation showed the same trend. However, relative to the estimated amount of N immobilisation by the microflora the decomposer fauna, as grazers recycling microbial N, was of increasing importance towards the North.

� The Testate Amoebae played a larger direct role for C and N fluxes than the total mesofauna. The

impact of this microfauna group on C and N fluxes was estimated to be significant.

� The rating of Enchytraeidae as a keystone group in boreal systems was not corroborated by the

estimates of their direct contribution to the fluxes of C and N. Such differences between microcosm studies and modelling results may hint to indirect effects having key influences which were only partly taken into account by the food web model.

� Within the decomposer fauna the highest trophic level (predaceous Microarthropoda and Nematodes)

played a minor direct role for mineralisation, which supports the view that the functional importance of soil fauna is inversely related to the trophic position of the group.

110

Chapter 8

Summary

Corythion dubium 10 µm

8 Summary

The decomposer systems of four coniferous forest sites on a European latitudinal transect (North Sweden to North-East France) were studied. The sites were subject to different climate as well as to different levels of atmospheric N deposition. The decomposer communities and their abiotic environment were monitored on four sampling dates. The study links different ecological scales and attempts to go beyond description towards a functional understanding. It is divided into two parts. In part one the community structure of Testate Amoebae, the dominant Protozoan group of coniferous forest soils, was studied on species level. Diversity, abundance of active cells, cysts and empty shells as well as species biomass pattern at the sites were compared and related to environmental parameters. In part two the Testate Amoebae community was set into the context of the decomposer food web. The food web structure and biomass of the functional groups were assessed. A numerical model was applied to each site to link the structure of the food webs to their function. The C and N flux across and within the food web was simulated. The contribution of the Testate Amoebae and the other functional groups to C and N mineralisation at the different sites was compared. A total of 42 Testate Amoebae species was found. The species number of Testate Amoebae ranged between 34 to 40 species and increased with decreasing latitude. Diversity and evenness of the Testate Amoebae communities were lowest at the boreal site. The species biomass pattern at the sites was quite similar. Many species occurred at all sites and with high relative biomass. The boreal site was not clearly distinguished from the temperate sites by specialist species. Nevertheless, the similarity between Testate Amoebae community structure increased with decreasing geographical distance between the sites. The size range among the Testate Amoebae community was very large, the biggest species found had an individual biomass of more than 50 times that of the smallest. A wide range of strategies within the r/K-continuum was realised. Both, r-strategic species and K-strategic species contributed considerably to Testate Amoebae biomass. A multivariate approach (CCA) suggested that three factors were most important in explaining the Testate Amoebae community structures: atmospheric pollution, microbial biomass and climate. Increased N-supply through deposition may impose an indirect positive effect by stimulating bacterial growth and thus enhancing food supply. The abundance of Testate Amoebae ranged from 700 106 m-2 at the boreal site to 2700 106 m-2 at the temperate site with the highest load of atmospheric pollution. The view that abundances increase towards South is supported. This increase is suggested to be due to increasing moisture and food availability at the sites with milder climate and high N deposition. The relative size of the necrocoenosis of Testate Amoebae was smaller

111

8 Summary

at the boreal site than at the other sites. This could not be explained by high decompostion rates. A possible explanation is that the relative amount of empty shells was small due to low metabolic rates of the Testate Amoebae in consequence of severe climatic conditions. The food web modelling approach revealed that mineralisation rates were not a simple function of food web biomass. Besides climate and resource quality, food web structure determined C and N fluxes. Total food web biomass was similar at all sites, ranging from 360 to 540 kg C ha-1. Total mineralisation rates ranged from 800 to 2600 kg C ha-1 a-1 and from 0-100 kg N ha-1 a-1. Mineralisation rates were lowest at the low N-input boreal site with a food web dominated by fungal pathways (bacterial to fungal respiration ca. 30/70). Further South, as N availability increased, bacterial pathways became more important (bacterial to fungal respiration ca. 40/50) and the cycling of C and N was accelerated. Including estimates from direct counts of Testate Amoebae resulted in considerably higher estimates of faunal mineralisation than have previously been reported (7-14% of total C mineralisation). The estimated contribution of Testate Amoebae to C mineralisation ranged from 6 % or 44 kg C ha-1 a-1 at the boreal site to 12 % or 343 kg C ha-1 a-1 at the temperate site with the highest load of atmospheric pollution. With a similar patterning between sites the estimated contribution of Testate Amoebae to N mineralisation ranged from 10 to 71 kg N ha-1 a-1. Testate Amoebae were the most important contributers to N cycling, and were essential to balance bacterial N immobilisation especially in the North. Within the decomposer fauna the highest trophic level (predaceous Microarthropoda and Nematodes) played a minor direct role for mineralisation which supports the view that the functional importance of soil fauna is inversely related to the trophic position of the group. The contribution of Nematoda to C and N fluxes was estimated to be quite low in general. In contrast to the other faunal groups, their importance did not increase towards South but was relatively higher at the boreal site. The microfauna played a larger direct role for C and N fluxes than the mesofauna. The rating of Enchytraeidae as a keystone group in boreal systems was not corroborated by the estimates of their direct contribution to C and N fluxes. Such differences between microcosm studies and modelling results hint to indirect effects having key influences which were only partly taken into account by the food web model. In general, the importance of decomposer fauna to absolute (kg ha-1 a-1) and relative (%) C mineralisation increased from North to South. The absolute (kg ha-1 a-1) contribution of fauna to N mineralisation showed the same trend. However, relative to the estimated amount of N immobilisation by the microflora, the decomposer fauna was of increasing importance towards the North. Along a European North-South-transect, changes within the biomass structure of four decomposer food webs resulted in specific total C and N mineralisation rates. Besides resource quality and climate, the

112

8 Summary

structure of the decomposer food web was essential to C and N fluxes.

113

Chapter 9

Zusammenfassung

Corythion dubium 10 µm

9 Zusammenfassung

Die Zersetzergemeinschaften von vier europäischen Nadelwäldern wurden untersucht. Die Flächen lagen auf einem Europäischen Transekt von Nord-Schweden bis Nord-Ost–Frankreich, der sich über eine Länge von ca. 2000 km erstreckte. Sie unterschieden sich hinsichtlich der klimatischen Bedingungen und des atmosphärischen Stickstoffeintrags. Die Zersetzergemeinschaften und ihre abiotische Umwelt wurden an 4 Probeterminen erfasst. Die vorliegende Studie untersucht die Zersetzergemeinschaft auf verschiedenen ökologischen Skalenebenen. Es wurde angestrebt, den deskriptiven Ansatz zu erweitern und ein funktionelles Verständnis der Lebensgemeinschaft im Zusammenhang mit ihrer Umgebung zu erlangen. Die Studie ist in zwei Teile untergliedert. Im ersten Teil wird die Gemeinschaftsstruktur der beschalten Amöben (Rhizopoda, Protozoa) auf Artebene untersucht. Beschalte Amöben stellen die bedeutenste Protozoengruppe in Fichtenwaldböden dar. Diversität, Abundanz belebter Zellen, encystierter Tiere und leerer Schalen sowie die Biomasse der beschalten Amöben an den einzelnen Standorten wurden verglichen und in Bezug zu Umweltparametern gesetzt. Im zweiten Teil wird die Gemeinschaft der beschalten Amöben in den Zusammenhang des gesamten Zersetzernahrungsnetzes gesetzt. Die Nahrungsnetzstruktur und die Biomasse der funktionellen Gruppen wurden erfasst. Ein mathematisches Modell wurde angewendet, um von der Struktur des Nahrungsnetzes auf seine Funktion zu schließen. Der Kohlenstoff- und Stickstofffluss durch das gesamte Netz und innerhalb des Netzes wurde simuliert. Der Beitrag der beschalten Amöben sowie anderer funktioneller Gruppen zur Kohlenstoff- und Stickstoffmineralisation an den einzelnen Standorten wurde verglichen. Insgesamt wurden 42 Arten beschalter Amöben gefunden. Der Artenreichtum an den einzelnen Standorten betrug 34 bis 40 Arten und nahm mit abnehmendem Breitengrad zu. Diversität und Evenness der Amöbengemeinschaft waren am borealen Standort am niedrigsten. Die Biomassestruktur der Gemeinschaften an den Standorten war ähnlich. Viele Arten kamen an allen Standorten vor und wiesen hohe relative Biomassen auf. Die Gemeinschaft der beschalten Amöben am borealen Standort wurde nicht durch spezialisierte Arten klar von den gemäßigten Standorten abgegrenzt. Dennoch nahm die Ähnlichkeit zwischen den Amöbengemeinschaften mit zunehmender geographischer Distanz zwischen den Standorten ab. Die Größenspanne innerhalb der Amöbengemeinschaften war sehr weit. Die individuelle Biomasse der größten Arten war mehr als fünfzig mal größer als die der kleinsten Arten. Sowohl r- als auch K-Strategen trugen wesentlich zur Gesamtbiomasse der Amöbengemeinschaft bei. Innerhalb der Gruppe der beschalten Amöben scheinen eine Reihe von Strategien entlang des r/K-

115

9 Zusammenfassung

Kontinuums erfolgreich umgesetzt zu werden. Anhand multivariater Analyse (kanonische Korrespondenzanalyse) wurden drei Umweltfaktorengruppen ermittelt, die für die Biomassestruktur der Amöbengemeinschaft von besonderer Bedeutung waren: Atmosphärischer Schadstoffeintrag, mikrobielle Biomasse und Klima. Der positive Effekt erhöhter Stickstoffeintrags war möglicherweise indirekt. Erhöhtes Stickstoffangebot könnte, ähnlich wie erhöhte Bodenfeuchte, bakterielles Wachstum anregen und somit das Nahrungsangebot für die beschalten Amöben verbessern. Die Abundanz der beschalten Amöben lag zwischen 700 106 m-2 am borealen Standort und 2700 106 m-2 an dem gemäßigten Standort mit der höchsten Last an atmosphärischem N-Eintrag. Die Abundanz war im Süden höher als im Norden. Dies wird auf erhöhte Standortfeuchte und verbessertes Nahrungsangebot an den Standorten mit milderem Klima und hohem N-Eintrag zurückgeführt. Die relative Größe der Nekrozönose (leere Schalen beschalter Amöben) war am borealen Standort kleiner als an den anderen Standorten. Da dies nicht auf erhöhte Abbauraten zurückgeführt werden konnte, wird vermutet, dass die metabolischen Raten der Amöbengemeinschaft durch widrige klimatische Bedingungen herabgesetzt waren. Dies könnte die Nekrozönose im Verhältnis zur Biozönose verkleinern. Ergebnisse des Zersetzernahrungsnetzmodells zeigten, dass die Mineralisationsraten des Nahrungsnetzes keine simple Funktion der Gesamtbiomasse sind. Zusammen mit Klima und Ressourcenqualität bestimmte die Nahrungsnetzstruktur die Kohlenstoff- und Stickstoffflüsse. Die Gesamtbiomasse des Nahrungsnetzes war an allen Standorten ähnlich und betrug zwischen 360 und 540 kg C ha-1. Die simulierten Gesamtmineralisationsraten waren 800 bis 2600 kg C ha-1 a-1 und 0 bis 100 kg N ha-1 a-1. Die Mineralisationsraten am borealen Standort waren am niedrigsten. Die höchsten Raten wurden an dem gemäßigten Standort mit dem höchsten Eintrag an atmosphärischem N-Eintrag ermittelt. Am borealen Standort wurde das Nahrungsnetz durch die Pilze dominiert (Verhältnis bakterieller zu pilzlicher Respiration ca. 30/70). Mit zunehmender Stickstoffverfügbarkeit in Richtung Süden nahm der bakterielle Abbauweg an Bedeutung zu (Verhältnis bakterieller zu pilzlicher Respiration ca. 40/50) und die Kreisläufe von C und N waren schneller. Der Beitrag der Fauna an der gesamten C Mineralisation wurde mit 7-14% vergleichsweise höher eingeschätzt als in anderen Studien. Dies wird hauptsächlich auf die erhöhte Einschätzung des Protozoenbeitrags zurückgeführt, da die beschalten Amöben in dieser Studie nicht mittels Kulturmethoden sondern durch direkte Zählverfahren erfasst wurden. Mit Hilfe des Nahrungsnetzmodells ergab sich ein Beitrag der beschalten Amöben zur Kohlenstoffmineralisation zwischen 6 % oder 44 kg C ha-1 a-1 am borealen Standort und bis zu 12 % oder 343 kg C ha-1 a-1 an dem gemäßigten Standort mit der höchsten Last an atmosphärischem N-Eintrag. Der Beitrag der beschalten Amöben zur Stickstoffmineralisation betrug 10

116

9 Zusammenfassung

bis 71 kg N ha-1 a-1 und zeigte den gleichen Verlauf im Standortvergleich. Die beschalten Amöben mineralisierten die vergleichsweise größte Menge Stickstoff innerhalb des Nahrungsnetzes. Besonders im Norden war ihr Beitrag essentiell für den Ausgleich bakterieller N Immobilisation. Innerhalb des Nahrungsnetzes war der Beitrag der höheren trophischen Ebenen (räuberische Mikroarthropoden und Nematoden) gering. Somit wurde die Auffassung bestätigt, dass die Bedeutung einer Gruppe für die C und N Flüsse umgekehrt proportional zu ihrer trophischen Stellung ist. Der Beitrag der Nematoden zum C und N Fluss wurde insgesamt niedrig eingeschätzt. Im Gegensatz zu den anderen Tiergruppen nahm ihr Beitrag nicht gen Süden zu, sondern war am borealen Standort vergleichsweise höher. Der Beitrag der Mikrofauna zum C und N Fluss war insgesamt höher als der der Mesofauna. Die Einschätzung, dass Enchytraeiden eine Schlüsselgruppe in borealen Wäldern sind, wurde durch die ermittelten direkten Beiträge zur C und N Mineralisation nicht bestätigt. Solche Differenzen zwischen Ergebnissen aus Mikrokosmos-Studien und modellierten Werten weisen auf wichtige indirekte Effekte im Zersetzernahrungsnetz hin. Solche indirekten Effekte werden nur zum Teil durch das Nahrungsnetzmodell erfasst. Im Allgemeinen nahmen die absoluten (kg ha-1 a-1) und relativen (%) Beiträge der Zersetzerfauna zur Kohlenstoffmineralisation von Norden nach Süden zu. Die absoluten (kg ha-1 a-1) Beiträge der Fauna zur Stickstoffmineralisation zeigten den selben Trend. Relativ zur N-Immobilisation der Bakterien hingegen wurde der Beitrag der Fauna für die N-Mineralisation des gesamten Netzes nach Norden hin wichtiger. Entlang eines Europäischen Nord-Süd-Transekts ergaben standortspezifische Zersetzernahrungsnetze spezifische C und N Mineralisationsraten. Neben Ressourcenqualität und Klima war die Struktur der Nahrungsnetze für die C und N Umsätze entscheidend.

117

Chapter 10

Appendix

10 Appendix

10.1 Parameters for calculation of decomposer fauna biomass

Table 10.1 C contents (from Berg 1997) of body dry weight and biomass conversion factors of decomposer fauna groups (for details see sections 3.1.2-5).

C content of dry weight (%) conversion factor (µg C individual-1)

Testate Amoebae 50.0 size class specific, see Tables 5.1 and 5.2 Nematoda 50.0 genus specific, see Table 10.2 Acari 47.7 genus specific, see Taylor (2001) Collembola 47.5 species specific, see Pflug (2001) Enchytraeidae 50.0 22.2

Table 10.2 Nematode biomass was calculated from genus specific abundances using conversion factors from the literature (Ekschmitt et al. 1999) or from calculations using the formula from Andrássy (1956) and size estimates from Bongers (1994). Body volume of juveniles was estimated to be on average 22 % of the adult body volume (Ilja Sonnemann, pers. com.).

family genus

conversion factor for adult specimen

(µg C individual-1) Cephalobidae Acrobeloides 0.021 Acrolobus 0.013 Alaimidae Alaimus 0.051 Bunonematidae Bunonema 0.006 Monhysteridae Geomonhystera 0.030 Heterocephalobus 0.041 Teratocephalidae Metateratocephalus 0.009 Panagrolaimidae Panagrolaimus 0.089 Plectidae Plectus 0.076 Prismatolaimidae Prismatolaimus 0.034 Neodiplogasteridae Pristionchus 0.598 Rhabditidae Rhabditis 0.598 Teratocephalus 0.009 Wilsonema 0.006 Aphelenchoididae Aphelenchoides 0.018 Neotylenchidae Hexatylus 0.248 Laimaphelenchus 0.012 Leptonchidae Tylencholaimus 0.050 Mononchidae Prionchulus 0.987 Seinura 0.019 Qudsianematidae Eudorylaimus 0.416 Tylenchidae Filenchus 0.007 Malenchus 0.006 Tylenchus 0.055

119

10 Appendix

10.2 Importance of food web biomass structure in the model To test the importance of food web biomass structure for the estimated mineralisation rates, the food web model was run without site specific adjustment of climate and resource quality (Scenario "equal"). Within this scenario equal climate (10°C, optimal moisture availability) and resource quality (C:N-ratio 22) was assumed for all sites. Differences in mineralisation estimates between sites from scenario "equal" are due only to the site specific food web structure. The estimates from the scenario are compared to those obtained after adequately adjusting the model to site specific conditions (Scenario "site specific", Table 10.3). The estimates obtained with scenario "equal" are much higher than those of scenario "site specific". However, a similar pattern of mineralisation rates in comparison of the sites is observed. Food web biomass structure thus accounts for a considerable part of the differences in mineralisation rates.

Table 10.3 Simulated N and C mineralisation rates (kg ha-1 a-1) at the different sites from a scenario assuming equal climate and resource quality at all sites (Scenario "equal") compared to the estimates obtained after adjusting the model to site specific conditions (Scenario "site specific").

Scenario

"equal" Scenario

"site specific" mineralisation (kg ha-1 a-1) N C N C N-SE 80 2245 1 764 S-SE 94 2550 28 1514 DE 192 5130 97 2600 FR 176 4913 57 2428

120

Chapter 11

References

11 References

Aescht, E., and W. Foissner. 1989. Catalogus Faunae Austriae - Teil I a: Stamm: Rhizopoda (U.-Kl. Testacealobosia, Testaceafilosia). Verlag der Österreichischen Akademie der Wissenschaften, Wien.

Aescht, E., and W. Foissner. 1992. Enumerating soil testate amoebae by direct counting. Pages B-6.1-B-6.4, Society of Protozoologists, Conference Proceedings.

Alef, K. 1991. Methodenhandbuch Bodenmikrobiologie. ecomed, Landsberg/Lech. Alphei, J., M. Bonkowski, and S. Scheu. 1996. Protozoa, Nematoda and Lumbricidae in the rhizosphere of Hordelymus

europaeus (Poaceae): faunal interactions, response of microorganisms and effect on plant growth. Oecologia 106: 111-126.

Anderson, J. M. 1995. Soil organisms as engineers: microsite modulation of macroscale processes. Pages 94-106 in C. G. Jones and J. H. Lawton, editors. Linking Species and Ecosystems. Chapman & Hall, London.

Anderson, J. M., S. A. Huish, P. Ineson, M. A. Leonard, and P. R. Splatt, editors. 1985. Interactions of invertebrates, micro-organisms and tree roots in nitrogen and mineral element fluxes in decidiuous woodland soils. Blackwell Scientific, Oxford.

Anderson, J. M., and P. Ineson. 1984. Interactions between microorganisms and soil invertebrates in nutrient flux pathways of forest ecosystems. Pages 89-132 in J. M. Anderson, A. D. M. Rayner, and D. W. S. Watson, editors. Invertebrate-microbial interactions. Cambridge University Press, Cambridge.

Anderson, R. V., D. C. Coleman, and C. V. Cole. 1981. Effects of saprophytic grazing on net mineralization. Ecol. Bull. 33: 201-216.

André, H. M., M.-I. Noti, and P. Lebrun. 1994. The soil fauna: the other last biotic frontier. Biodiversity & Conservation 3: 45-56.

Andrén, O., T. Lindberg, U. Boström, M. Clarholm, A.-C. Hansson, G. Johansson, J. Lagerlöf, K. Paustian, J. Persson, R. Pettersson, J. Schnürer, B. Sohlenius, and M. Wivstad. 1990. Organic carbon and nitrogen flows. Pages 85-125 in O. Andrén, T. Lindberg, K. Paustian, and T. Rosswall, editors. Ecology of Arable Land - Organisms, Carbon and Nitrogen Cycling. Munksgaard, Copenhagen.

Bamforth, S. S. 1997. Protozoa: Recyclers and Indicators of Agroecosystem Quality. Pages 63-84 in G. Benckiser, editor. Fauna in Soil Ecosystems. Marcel Dekker, Inc., New York.

Bardgett, R. D., R. Cook, G. W. Yeates, and C. S. Denton. 1999. The influence of nematodes on below-ground processes in grassland ecosystems. Plant and Soil 212: 23-33.

Bauer, G., H. Persson, T. Persson, M. Mund, M. Hein, E. Kummetz, G. Matteucci, H. van Oene, G. Scarascia-Mugnozza, and E.-D. Schulze. 2000. Linking plant nutrition and ecosystem processes. Pages 63-98 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Beare, H. M., R. W. Parmelee, P. F. Hendrix, and W. Cheng. 1992. Microbial and faunal interactions and effects on litter nitrogen and decomposition in agroecosystems. Ecological Monographs 62: 569-591.

Beck, L., and S. Woas. 1991. Die Oribatiden-Arten (Acari) eines südwestdeutschen Buchenwaldes I. carolinea 49: 37-82. Bengtsson, J., H. Setälä, and D. W. Zheng. 1995. Food Webs and Nutrient Cycling in Soils: Interactions and Positive

Feedbacks. Pages 30-38 in G. A. Polis and K. O. Winemiller, editors. Food Webs: integration of patterns and dynamics. Chapman & Hall, New York.

Berg, J., S. Woas, and L. Beck. 1990. Zur Taxonomie der Phthiracarus-Arten (Acari, Oribatei) eines südwestdeutschen Buchenwaldes. andrias 7: 61-90.

Berger, H., W. Foissner, and H. Adam. 1986. Field experiments on the effects of fertilizers and lime on the soil microfauna of an alpine pasture. Pedobiologia 29: 261-272.

Bloem, J., M. Veninga, and J. Shepherd. 1997. Vollautomatische Messung von Bodenbakterien mit Hilfe der konfokalen Laser-Scanningmikroskopie und der Bildanalyse. Mitteilungen für Wissenschaft und Technik XI: 143-148.

BML. 1999. Unser Wald � Natur und Wirtschaftsfaktor zugleich. Bundesministerium für Ernährung, Landwirtschaft und Forsten (BML).

Bobrov, A. A., S. B. Yazvenko, and B. G. Warner. 1995. Taxonomic and ecological implications of shell morphology of three Testaceans (Protozoa: Rhizopoda) in Russia and Canada. Archiv für Protistenkunde 145: 119-126.

Bolger, T. M., L. J. Heneghan, and P. Neville. 2000. Invertebrates and nutrient cycling in coniferous forest ecosystems:

Aber, J., W. McDowell, K. Nadelhoffer, A. Magill, G. Berntson, M. Kamakea, S. McNulty, W. Currie, L. Rustad, and I. Fernandez. 1998. Nitrogen saturation in temperate forest ecosystems. BioScience 48: 921-934.

Andrássy, I. 1956. Die Rauminhalts- und Gewichtsbestimmung der Fadenwürmer (Nematoden). Acta Zoologica Hung. Acad. Sci. 2: 1-15.

Berg, M. 1997. Decomposition, nutrient flow and food web dynamics in a stratified pine forest soil. PhD-thesis. Vrije Universiteit, Amsterdam.

121

11 References

spatial heterogeneity and conditionality. Pages 255-269 in D. C. Coleman and P. F. Hendrix, editors. Invertebrates as Webmasters in Ecosystems. CAB International, Wallingford.

Bongers, T. 1994. De Nematoden van Nederland � Een identificatietabel voor de in nederland aangetroffen zoetwater- en bodembewonende nematoden, 2nd edition. Uitgeverij Pirola, Schoorl, Utrecht.

Bonnet, L. 1964. Le peuplement thécamoebien des sols. Revue d'Écologie et de Biologie du Sol 1: 123-408. Bonnet, L. 1974. A propos de Geopyxella sylvicola et de Pseudawerintzewia calcicola (Rhizopodes Thécamoebiens

édaphiques). Revue d'Écologie et de Biologie du Sol 10: 509-522. Bonnet, L. 1975. Types morphologique, écologie et évolution de la thèque chez les thécamoebiens. Protistologica XI: 363-

378. Bonnet, L. 1988a. Écologie du genre Plagiopyxis. (Thécamoebiens des sols). Extrait du Bulletin de la Société d'Histoire

Naturelle de Toulouse 124: 13-21. Bonnet, L. 1988b. Le signalement écologique des Thécamoebiens des sols 3ème partie: Caractéristiques des faciès. Extrait

du Bulletin de la Société d'Histoire Naturelle de Toulouse 124: 7-12. Bonnet, L. 1989a. Données écologiques sur quelques Centropyxidae (Thécamoebiens) des sols. Extrait du Bulletin de la

Société d'Histoire Naturelle de Toulouse 125: 7-16. Bonnet, L. 1989b. Données écologiques sur quelques Hyalospheniidae et Paraquadrulidae (Thécamoebiens) des sols

(Première partie). Extrait du Bulletin de la Société d'Histoire Naturelle de Toulouse 125: 17-22. Bonnet, L. 1990. Données écologiques sur quelques Hyalospheniidae et Paraquadrulidae (Thécamoebiens) des sol

(Deuxième partie: genre Nebela). Extrait du Bulletin de la Société d'Histoire Naturelle de Toulouse 126: 9-17. Bonnet, L. 1991a. Écologie de quelques Euglyphidae (Thécamoebiens, Filosea) des milieux édaphiques et paraédaphiques

(Deuxième partie). Extrait du Bulletin de la Société d'Histoire Naturelle de Toulouse 127: 15-20. Bonnet, L. 1991b. Écologie de quelques Euglyphidae (Thecamoebiens, Filosea) des milieux édaphiques et paraédaphiques

(Première partie: genres Corythion et Trinema). Extrait du Bulletin de la Société d'Histoire Naturelle de Toulouse 127: 7-13.

Bonnet, L., and R. Thomas. 1960. Thécamoebiens du Sol. in Faune terrestre et d'eau douce des Pyrénées-Orientales, 5, 103 pp., Vie et Mileu, Paris.

Bray, J. R., and C. T. Curtis. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecological Monographs 27: 325-349.

Brussaard, L., and N. G. Juma. 1996. Organisms and Humus in Soil. Pages 329-359 in A. Piccolo, editor. Humic Substances in Terrestrial Ecosystems. Elsevier Science B.V., Amsterdam.

Bunt, J. S., and Y. T. Tchan. 1955. Estimation of protozoan populations in soils by direct microscopy. Proc. Linn. Soc. N. S. 80: 148-153.

Burton, A. J., K. S. Pregitzer, and R. L. Hendrick. 2001. Relationships between fine root dynamics and nitrogen availability in Michigan northern hardwood forests. Oecologia 125: 389-399.

Chardez, D. 1959. Thécamoebiens des terres de Belgique I. Hydrobiologia 14: 72-78. Chardez, D. 1969. Le genre Phryganella Pénard. Bull. Rech. Agr. Gembloux 4: 315-322. Chardez, D. 1987a. Notes protozoologiques au sujet de quelques especes des genres Nebela, Paraquadrula et Euglypha

(Protozoa, Rhizopoda, Testacea). Trav. Lab. Zool. Gen. Faun. Gembloux 9: 1-6. Chardez, D. 1987b. Sur les epines des Euglypha et Placocista (Protozoa Rhizopoda Testacea. Euglyphidae). Trav. Lab.

Zool. Gen. Faun. Fac. Sc. Agr. Gembloux 8: 1-5. Chardez, D., F. Delecour, and F. Weissen. 1972. Évolution des populations thécamoebiennes de sols forestiers sous

l'influence de fumures artificielles. Rev. Ecol Biol Sol 9: 185-196. Chown, S. L., and K. J. Gaston. 2000. Areas, cradles and museums: the latitudinal gradient in species richness. Trends in

Ecology & Evolution 15: 311-315. Clarholm, M. 1981. Protozoan grazing of bacteria in soil � impact and importance. Microbial Ecology 7: 343-350. Coleman, D. C., C. P. P. Reid, and V. V. Cole. 1983. Biological strategies of nutrient cycling in soil systems. Advances in

Ecological Research 13: 1-55. Colwell, R. K., and J. Coddington. 1994. Estimating terrestrial biodiversity through extrapolation. Phil. Trans. R. Soc. Lond. B

345: 101-118. Copley, J. 2000. Ecology goes underground. Nature 406: 452-454. Coûteaux, M.-M. 1976. Dynamisme de l'Équilibre des Thécamoebiens Dans Quelques Sols Climacique. Mémoires du

Muséum National D'Histoire Naturelle, Série A, Zoologie XCVI, 183 pp. Imprimerie Nationale, Paris. Coûteaux, M.-M. 1985. Relationships between testate amoebae and fungi in humus microcosms. Soil Biology &

Biochemistry 17: 339-345. Coûteaux, M.-M., and J. F. Darbyshire. 1998. Functional diversity amongst soil protozoa. Applied Soil Ecology 10: 229-237. Coûteaux, M.-M., and J. Devaux. 1983. Effect d'un enrichissement en champignons sur la dynamique d'un peuplement

thecamoebiens d'un humus. Revue d'Écologie et de Biologie du Sol 20: 519-545. Coûteaux, M.-M., G. Faurie, L. Palka, and C. Steinberg. 1988. La relation prédateur-proie (protozoaires-bactéries) dans les

122

11 References

sol: rôle dans la régulation des populations et conséquences sur les cycles du carbone et de l'azote. Revue d'Écologie et de Biologie du Sol 25: 1-31.

Coûteaux, M.-M., A. Munsch, and J.-F. Ponge. 1979. Le genre Euglypha: Essai de taxinomie numérique. Protistologica XV: 565-579.

Coûteaux, M.-M., and C. G. Ogden. 1988. The Growth of Tracheleuglypha dentata (Rhizopoda: Testacea) in Clonal Cultures under Different Trophic Conditions. Microbial Ecology 15: 81-93.

Coûteaux, M.-M., M. Raubuch, and M. Berg. 1998. Response of protozoan and microbial communities in various coniferous forest soils after transfer to forests with different levels of atmospheric pollution. Biology And Fertility Of Soils 27: 179-188.

Cowling, A. J. 1994. Protozoan distribution and adaptation. Pages 5-42 in J. F. Darbyshire, editor. Soil Protozoa. CAB International, Wallingford.

Currie, W. S. 1999. The responsive C and N biogeochemistry of the temperate forest floor. Trends in Ecology & Evolution 14: 316-320.

Darbyshire, J. F., editor. 1994. Soil Protozoa. CAB International, Wallingford. De Ruiter, P. C., J. C. Moore, K. B. Zwart, L. A. Bouwman, J. Hassink, J. Bloem, J. A. De Vos, J. C. Y. Marinissen, W. A. M.

Didden, G. Lebbink, and L. Brussaard. 1993a. Simulation of nitrogen mineralization in the below-ground food webs of two winter wheat fields. Journal of Applied Ecology 30: 95-106.

De Ruiter, P. C., A.-M. Neutel, and J. C. Moore. 1994. Modelling food webs and nutrient cycling in agro-ecosystems. Trends in Ecology & Evolution 9: 378-383.

De Ruiter, P. C., A.-M. Neutel, and J. C. Moore. 1998. Biodiversity in soil ecosystems: the role of energy flow and community stability. Applied Soil Ecology 10: 217-228.

De Ruiter, P. C., J. A. van Veen, J. C. Moore, L. Brussaard, and H. W. Hunt. 1993b. Calculation of nitrogen mineralization in soil food webs. Plant and Soil 157: 263-273.

Decloitre, L. 1960. Structure de la theque du genre Corythion Taranek. Hydrobiologia 16: 215-217. Decloitre, L. 1962. Le genre Euglypha DUJARDIN. Arch.Protistenkd. 106: 51-100. Decloitre, L. 1976a. Le genre Arcella EHRENBERG. Complements a jour au 31. decembre 1974 de la monographie du

genre parue en 1928. Arch.Protistenk. 118 a: 291-309. Decloitre, L. 1976b. Le genre Euglypha. Complements a jour au 31. decembre 1974 de la monographie du genre parue en

1962. Arch.Protistenk. 118 b: 18-33. Decloitre, L. 1977a. Le genre Cyclopyxis. Complements a jour au 31. decembre 1974 de la monographie du genre parue en

1929. Arch.Protistenk. 119 a: 31-53. Decloitre, L. 1977b. Le genre Nebela. Complements a jour au 31. decembre 1974 du genre parue en 1936. Arch.Protistenk.

119 b: 325-352. Decloitre, L. 1978. Le genre Centropyxis I. Complements a jour au 31.12.1974 de la monographie du genre parue en 1929.

Arch.Protistenk. 120: 63-85. Decloitre, L. 1979a. Le genre Centropyxis II. Complements a jour au 31. decembre 1974 de la monographie du genre parue

en 1929. Arch.Protistenk. 121: 162-192. Decloitre, L. 1979b. Mises a jour au 31.12.1978 des mises a jour au 31.12.1974 concernant les genre Arcella, Centropyxis,

Cyclopyxis, Euglypha et Nebela. Arch.Protistenk. 122 b: 387-397. Decloitre, L. 1981. Le genre Trinema DUJARDIN, 1841. Revision a jour au 31.XII.1979. Arch.Protistenk. 124: 193-218. Decloitre, L. 1982. Complements aux publication precedentes. Mise a jour au 31.XII.1981 des genre Arcella, Centropyxis,

Euglypha, Nebela et Trinema. Arch.Protistenk. 126: 393-407. Deflandre, G. 1928a. Deux genres nouveaux de rhizopodes testacés. 1. Wailesella gen. nov. 2. Tracheleuglypha gen. nov.

Ann. Protistol. 1: 37-43. Deflandre, G. 1928b. Le genre Arcella EHRENBERG, Morphologie-Biologie. Archiv Protistenkd. 64: 201-287. Deflandre, G. 1929. Le genre Centropyxis Stein. Archiv für Protistenkunde 67: 322-375. Deflandre, G. 1936. Etude monographique sur genre Nebela Leidy (Rhizopoda - Testacea). Annales Protistol. 5: 201-286. Dighton, J. 1997. Nutrient Cycling by Saprophytic Fungi in Terrestrial Habitats. Pages 271-279 in Wicklow and Söderström,

editors. Environmental and Microbial Relationships. Springer, Heidelberg. Djajakirana, G., R. G. Joergensen, and B. Meyer. 1996. Ergosterol and microbial biomass relationship in soil. Biology And

Fertility Of Soils 22: 299-304. Dunger, W., and H. J. Fiedler. 1989. Methoden der Bodenbiologie. Gustav Fischer, Stuttgart. Eggers, T., and T. H. Jones. 2000. You are what you eat...or are you? Trends in Ecology & Evolution 15: 265-266. Ekblad, A., H. Wallander, and T. Näsholm. 1998. Chitin and ergosterol combined to measure total and living fungal biomass

in ectomycorrhizas. The New Phytologist 138: 143-149. Ekelund, F., and R. Ronn. 1994. Notes on protozoa in agricultural soil with emphasis on heterotrophic flagellates and naked

amoebae and their ecology. Fems Microbiology Reviews 15: 321-353. Ekschmitt, K., G. Bakonyi, M. Bongers, T. Bongers, S. Boström, H. Dogan, A. Harrison, A. Kallimanis, P. Nagy, A. G.

123

11 References

O'Donnel, B. Sohlenius, G. P. Stamou, and V. Wolters. 1999. Effects of the nematofauna on microbial energy and matter transformation rates in European grassland soils. Plant and Soil 212: 45-61.

Ellenberg, H., R. Mayer, and J. Schauermann. 1986. Ökosystemforschung � Ergebnisse des Sollingprojektes 1966 - 1986. Ulmer, Stuttgart.

Elliott, E. T., D. C. Coleman, R. E. Ingham, and J. A. Trofymow. 1984. Carbon and energy flow through microflora and microfauna in the soil subsystem of terrestrial ecosystems. Pages 424-433 in M. J. Klug and C. A. Reddy, editors. Current perspectives in microbial ecology. American Society of Microbiology, Washington.

Fenchel, T. 1988. Marine Plankton Food Chains. Ann. Rev. Ecol. Syst. 19: 19-38. Fenchel, T., G. F. Esteban, and B. J. Finlay. 1997. Local versus global diversity of microorganisms: cryptic diversity of

ciliated protozoa. Oikos 80: 220-225. Finlay, B. J., and T. Fenchel. 1996. Global diversity and body size. Nature 383: 132-133. Finlay, B. J., and T. Fenchel. 1999. Divergent perspectives on protist species richness. Protist 150: 229-233. Fjellberg, A. 1980. Identification keys to Norwegian Collembola. Norsk entomologisk Forening, Ås. Fjellberg, A. 1998. The Collembola of Fennoscandia and Denmark. Part I. Poduromorpha. Brill Academic Publishers. Foissner, W. 1987. Soil protozoa: fundamental problems, ecological significance, adaptations in ciliates and testaceans,

bioindicators, and guide to the literature. Progr. Protistol. 2: 69-212. Foissner, W. 1994. Soil protozoa as bioindicators in ecosystems under human influence. Pages 147-193 in J. F. Darbyshire,

editor. Soil Protozoa. CAB International, Wallingford. Foissner, W. 1996. Soil protozoan diversity: the state of the art. Mitt. Deut. Bodenk. Gesell. 81: 219-220. Foissner, W. 1997. Global soil ciliate (Protozoa, ciliophora) diversity: a probability-based approach using large sample

collections from Africa, Australia and Antarctica. Biodiversity and Conservation 6: 1627-1638. Foissner, W. 1998. An updated compilation of world soil ciliates (Protozoa, Ciliophora), with ecological notes, new records,

and descriptions of new species. European Journal of Protistology 34: 195-235. Galloway, J. N. 1995. Acid deposition: Perspectives in time and space. Water, Air and Soil Pollution 85: 15-24. Giljarov, M. S., and D. A. Krivolutsky. 1975. Bestimmungsbuch der bodenbewohnenden Milben - Sarctoptiformes (russian).

Akad. Nauka SSSR, Moscow. Gisin, H. 1960. Collembolenfauna Europas. Mus. Hist. Nat., Genève. Górny, M., and L. Grüm, editors. 1993. Methods in Soil Zoology. Polish Scientific Publishers, Warszawa. Grospietsch, T. 1964. Die Gattungen Cryptodifflugia und Difflugiella (Rhizopoda, Testacea). Zool. Anz. 172: 243-257. Grospietsch, T. 1965a. Monographische Studie der Gattung Hyalosphenia STEIN (Rhizopoda, Testacea). Hydrobiologia 26:

211-241. Grospietsch, T. 1965b. Wechseltierchen (Rhizopoden), 3 edition. Franckh'sche Verlagshandlung, Stuttgart. Hågvar, S. 1998. The relevance of the Rio-Convention on biodiversity to conserving the biodiversity of soils. Applied Soil

Ecology 9: 1-7. Hammel, K. E. 1997. Fungal degradation of lignin. CAB International, Oxon. Harrison, A. F., D. D. Harkness, A. P. Rowland, J. S. Garnett, and P. J. Bacon. 2000. Annual carbon and nitrogen fluxes in

soils along the European forest transect, determined using 14C-bomb. Pages 237-256 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Heal, O. W. 1963. Morphological variation in certain Testacea (Protozoa: Rhizopoda). Archiv für Protistenkunde 106: 351-368.

Heal, O. W. 1965. Observations on testate amoebae (Protozoa: Rhizopoda) from Signy Island, South Orkney Islands. Brit. Antarct. Surv. Bull. 6: 43-47.

Heal, O. W. 1967. Quantitative feeding studies on soil amoeba. Pages 120-125 in O. Graff and J. E. Satchell, editors. Progress in Soil Biology. North Holland Publ. Co., Amsterdam.

Heal, O. W., J. M. Anderson, and M. J. Swift. 1997. Plant litter quality and decomposition: an historical overview. CAB International, Oxon.

Hedlund, K., and A. Augustsson. 1995. Effects of Enchytraeid grazing on fungal growth and respiration. Soil Biology & Biochemistry 27: 905-909.

Hillebrand, H., F. Watermann, R. Karez, and U.-G. Berninger. 2001. Differences in species richness patterns between unicellular and multicellular organisms. Oecologia 126: 114-124.

Hobbie, J. E., and J. M. Melillo. 1984. Role of microbes in global carbon cycling. Pages 389-393 in M. J. Klug and C. A. Reddy, editors. Current perspectives in microbial ecology. American Society of Microbiology, Washington.

Huhta, V. 1976. Effects of clear-cutting on numbers, biomass and community respiration of soil invertebrates. Ann. Zool. Fennici 13: 63-80.

Huhta, V., and A. Koskenniemi. 1975. Numbers, biomass and community respiration of soil invertebrates in spruce forests at two latitudes in Finland. Ann. Zool. Fennici 12: 164-182.

Huhta, V., T. Persson, and H. Setälä. 1998. Functional implications of soil fauna diversity in boreal forests. Applied Soil Ecology 10: 277-288.

124

11 References

Hunt, H. W., D. C. Coleman, E. R. Ingham, R. E. Ingham, E. T. Elliott, J. C. Moore, S. L. Rose, C. P. P. Reid, and C. R. Morley. 1987. The detrital food web in a shortgrass prairie. Biology And Fertility Of Soils 3: 57-68.

IGBP Terrestrial Carbon Working Group. 1998. The terrestrial carbon cycle: implications for the Kyoto protocol. Science 280: 1393-1394.

Ingham, E. R., J. A. Trofymow, E. R. Ingham, and D. C. Coleman. 1985. Interactions of bacteria, fungi and their nematode grazers: Effects on nutrient cycling and plant growth. Ecological Monographs 55: 119-140.

Jongman, R. H. G., C. J. F. ter Braak, and O. F. R. van Tongeren, editors. 1987. Data analysis in community and landscape ecology. Pudoc, Wageningen.

Jordana, R., J. I. Arbea, C. Simón, and M. J. Luciáñez. 1997. Collembola - Poduromorpha. Pages 808 in M. A. Ramos and et al., editors. Fauna Ibérica. Museo Nacional de Ciencias Naturales, Madrid.

Kajak, A. 1995. The role of soil predators in decomposition processes. European Journal of Entomology 92: 573-580. Kandeler, E., C. Kampichler, R. G. Joergensen, and K. Mölter. 1999. Effects of mesofauna in a spruce forest on soil

microbial communities and N cycling in field mesocosms. Soil Biology & Biochemistry 31: 1783-1792. Kempson, D., M. Lloyd, and R. Ghelardi. 1963. A new extractor for woodland litter. Pedobiologia 3: 1-21. Krebs, C. J. 1999. Ecological Methodology, 2nd edition. Harper & Row Publishers, New York. Laakso, J., and H. Setälä. 1999a. Population- and ecosystem-level effects of predation on microbial-feeding nematodes.

Oecologia 120: 279-286. Laakso, J., and H. Setälä. 1999b. Sensitivity of primary production to changes in the architecture of belowground food webs.

Oikos 87: 57-64. Laminger, H. 1978. The Effects of Soil Moisture Fluctuations on the Testacean Species Trinema enchelys (Ehrenberg) Leidy

in a High Mountain Brown-earths-podsol and its Feeding Behaviour. Archiv für Protistenkunde 120: 446-454. Laminger, H. 1980. Bodenprotozoologie. Mikrobios 1: 1-142. Lavelle, P. 1997. Faunal Activities and Soil Processes: Adaptive Strategies That Determine Ecosystem Function. Advances

in Ecological Research 27: 93-132. Lavelle, P., C. Lattaud, D. Trigo, and I. Barois. 1995. Mutualism and biodiversity in soils. Plant and Soil 170: 23-33. Lawton, J. H. 1994. What do species do in ecosystems? Oikos 71: 367-374. Lawton, J. H. 1999. Are there general laws in ecology? Oikos 84: 177-192. Lawton, J. H., D. E. Bignell, G. F. Bloemers, P. Eggleton, and M. E. Hodda. 1996. Carbon flux and diversity of nematodes

and termites in Cameroon forest soils. Biodiversity and Conservation 5: 261-273. Leake, J. R., and D. J. Read. 1997. Mycorrhizal Fungi in Terrestrial Habitats. Pages 281-301 in Wicklow and Söderström,

editors. Environmental and Microbial Relationships. Springer, Heidelberg. Lee, K. E. 1994. The biodiversity of soil organisms. Applied Soil Ecology 1: 251-254. Likens, G. E. 1992. The ecosystem approach: Its use and abuse. pages 1-166 in Kinne, O. ,editor. Excellence in Ecology.

Book 3, Ecology Institute, Oldendorf/Luhe. Lindahl, B., A. F. S. Taylor, and R. D. Finlay. 2001. In press. Defining nutritional constraints on carbon cycling - towards a

less "phytocentric" perspective. Plant and Soil: 17. Lindman, H. R. 1974. Analysis of variance in complex experimental designs. W. H. Freeman & Co., San Francisco, USA. Lousier, J. D. 1974a. Effects of experimental soil moisture fluctuations on turnover rates of Testacea. Soil Biology &

Biochemistry 6: 19-26. Lousier, J. D. 1974b. Response of Soil Testacea to Soil Moisture Fluctuations. Soil Biology & Biochemistry 6: 235-239. Lousier, J. D. 1984a. Population dynamics and production of Phryganella acropodia and Difflugiella oviformis (Testacea,

Rhizopoda, Protozoa) in an aspen woodland soil. Pedobiologia 26: 331-347. Lousier, J. D. 1984b. Population dynamics and production studies of species of Euglyphidae (Testacea, Rhizopoda,

Protozoa) in aspen woodland soil. Pedobiologia 26: 309-330. Lousier, J. D. 1984c. Population Dynamics and Production Studies of Species of Nebelidae (Testacea, Rhizopoda) in an

Aspen Woodland Soil. Acta Protozoologica 23: 145-159. Lousier, J. D. 1985. Population dynamics and production studies of species of Centropyxidae (Testacea, Rhizopoda) in an

aspen woodland soil. Archiv für Protistenkunde 130: 165-178. Lousier, J. D., and D. Parkinson. 1984. Annual population dynamics and production ecology of Testacea ( Protozoa,

Rhizopoda) in an aspen woodland soil. Soil Biology & Biochemistry 16: 103-114. Lüftenegger, G., and W. Foissner. 1991. Morphology and biometry of twelve soil testate amoebae (Protozoa, Rhizopoda)

from Australia, Africa, and Austria. Bull. Br. Mus. nat. Hist. (Zool.) 57: 1-16. Lüftenegger, G., W. Petz, H. Berger, W. Foissner, and H. Adam. 1988. Morphologic and Biometric Characterization of

Twenty-four Soil Testate Amoebae (Protozoa, Rhizopoda). Archiv für Protistenkunde 136: 153-189. Lussenhop, J. 1992. Mechanisms of microarthropod-microbial interactions in soil. Advances in Ecological Research 23: 1-33. Luxton, M. 1972. Studies on the oribatid mites of a Danish beech wood Soil - I. Nutritional Biology. Pedobiologia 12: 434-

463. MacArthur, R. H., and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, NJ.

125

11 References

Macfadyen, A. 1953. Notes on methods for the extraction of small soil arthropods. Journal of Animal Ecology 22: 65-78. Magurran, A. E. 1988. Ecological diversity and its measurement. Croom Helm, London. Marshall, V. G., D. K. M. Kevan, J. V. Matthews, and A. D. Tomlin. 1982. Status and research needs of Canadian soil

arthropods. Bull. Soc. Can. Supplement 14: 5. Meisterfeld, R. 1980. Die Struktur von Testaceenzönosen (Rhizopoda, Testacea) in Böden des Sollings. Verhandlungen der

Gesellschaft für Ökologie VIII: 435-447. Meisterfeld, R. 1986. The importance of protozoa in a beech forest ecosystem. Adv. in Protozool. Res. (Symp. Biol. Hung.)

33: 291-299. Meisterfeld, R. 1987. Die Bedeutung der Protozoen im Kohlenstoffhaushalt eines Kalkbuchenwaldes (Zur Funktion der

Fauna in einem Mullbuchenwald 3). Verhandlungen der Gesellschaft für Ökologie XVII: 221-227. Meisterfeld, R. 1995. Thekamöben - ihr Potential für Ökosystemforschung und Bioindikation. Pages 87-95 in W. Dunger and

K. Voigtländer, editors. Bedeutung, Stand und aktuelle Entwicklung der Systematik von Bodentieren. Naturkundemuseum Görlitz, Staatliches Museum für Naturkunde, Görlitz.

Meisterfeld, R. 2001a. in press. Order Arcellinida. In J. J. Lee et al., editors. The Illustrated Guide to the Protozoa. Allen Press, Lawrence.

Meisterfeld, R. 2001b. in press. Testate amoebae with filopoda. In J. J. Lee et al., editors. The Illustrated Guide to the Protozoa. Allen Press, Lawrence.

Meisterfeld, R., and R. Schüller. 1982. Le genre Nebela. Une étude biométrique (Rhizopoda, Testacea). J Protozool 29: 520A.

Menaut, J.-C., and S. Struwe. 1994. Terrestrial Ecosystem Research Initiative (TERI) Science Plan. Ecosystems Research Report 17, European Commission.

Mikola, J., and H. Setälä. 1998. No evidence of trophic cascades in an experimental microbial-based soil food web. Ecology 79: 153-164.

Moore, J. C., P. C. De Ruiter, H. W. Hunt, D. C. Coleman, and D. W. Freckman. 1996. Microcosms and soil ecology: critical linkages between field studies and modelling food webs. Ecology 77: 694-705.

Moore, J. C., P. C. De Ruiter, and W. H. Hunt. 1993. Influence of Productivity on the Stability of Real and Model Ecosystems. Science 261: 906-908.

Moore, J. C., D. E. Walter, and H. W. Hunt. 1988. Arthropod regulation of micro- and mesobiota in belowground food webs. Annual Review of Entomology 33: 419-439.

Näsholm, T., A. Ekblad, A. Nordin, R. Giesler, M. Högberg, and P. Högberg. 1998. Boreal forest plants take up organic nitrogen. Nature 392: 914-916.

Nylund, J.-E., and H. Wallander. 1992. Ergosterol analysis as means of quantifying mycorrhizal biomass. Methods in Microbiology 24: 77-88.

O'Connor, F. B. 1955. Extraction of Enchytraeid Worms from a Coniferous Forest Soil. Nature 175: 815-816. Ogden, C. G. 1983. Observations on the systematics of the genus Difflugia in Britain (Rhizopoda, Protozoa). Bulletin of the

British Museum (Natural History) 44: 1-73. Ogden, C. G., and R. H. Hedley. 1980. An Atlas of Freshwater Testate Amoebae. Oxford University Press, London. Ogden, C. G., and P. Pitta. 1990. Biology and ultrastructure of the mycophagous soil testate amoeba, Phryganella acropodia

(Rhizopoda, Protozoa). Biology & Fertility of Soils 9: 101-109. Ogden, C. G., and A. Zivkovic. 1983. Morphological studies on some Difflugiidae from Yugoslavia (Rhizopoda, Protozoa).

Bulletin of the British Museum (Natural History) 44: 341-375. O'Neill, R. V. 1969. Indirect Estimation of Energy Fluxes in Animal Food Webs. J. Theoret. Biol. 22: 284-290. Page, C. F. 1966. Cryptodifflugia operculata n. sp. (Rhizopodea: Arcellinida, Cryptodifflugiidae) and the status of the genus

Cryptodifflugia. Trans. Amer. Microsc. Soc. 85: 506-515. Paine, R. T. 1988. Food webs: road maps of interactions or grist for theoretical development? Ecology 69: 1648-1654. Palissa, A. 1964. Insekten - I. Teil Apterygota. Pages 299 in P. Brohmer, P. Ehrmann, and G. Ulmer, editors. Die Tierwelt

Mitteleuropas. Quelle & Meyer, Heidelberg. Parmelee, R. W. 1995. Soil fauna: linking different levels of the ecological hierarchy. Pages 107-116 in C. G. Jones and J.

Lawton, editors. Linking Species and Ecosystems. Chapman and Hall, London. Paul, E. A., and F. E. Clark. 1989. Soil microbiology and biochemistry. Academic Press, San Diego. Paul, E. A., and N. G. Juma. 1981. Mineralization and immobilization of soil nitrogen by microorganisms. Pages 179-195 in

F. E. Clark and T. Rosswall, editors. Terrestrial Nitrogen Cycles. Ecol. Bull., Stockholm. Pennanen, T., J. Liski, E. Bååth, V. Kitunen, J. Uotila, C. J. Westman, and H. Fritze. 1999. Structure of the microbial

communities in coniferous forest soils in relation to site fertility and stand development stage. Microbial Ecology 38: 168-179.

Persson, T. 1983. Influence of soil animals on nitrogen mineralisation in a northern Scots pine forest. Pages 117-126 in P. Lebrun, H. M. André, A. De Medts, C. Grégorie-Wibo, and G. Wauthy, editors. New trends in soil biology. Dieu-Brichart, Ottignies-Louvain-la-Neuve.

126

11 References

Persson, T. 1989. Role of soil animals in C and N mineralisation. Pages 185-189 in M. Clarholm and L. Bergström, editors. Ecology of arable land. Kluwer Academic Publishers, Dordrecht.

Persson, T., E. Bååth, M. Clarholm, H. Lundkvist, B. E. Söderström, and B. Sohlenius. 1980. Trophic structure, biomass dynamics and carbon metabolism of soil organisms in a scots pine forest. Pages 419-459 in T. Persson, editor. Structure and Function of Northern Coniferous Forests � An Ecosystem Study. Swedish Natural Science Research Council (NFR), Stockholm.

Persson, T., P. S. Karlsson, U. Seyferth, R. M. Sjöberg, and A. Rudebeck. 2000a. Carbon mineralisation in European forest soils. Pages 257-275 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Persson, T., and U. Lohm. 1977. Energetical significance of the Annelids and Arthropods in a Swedish grassland soil. Ecological Bulletins 23: 218.

Persson, T., A. Rudebeck, J. H. Jussy, M. Colin-Belgrand, A. Priemé, E. Dambrine, P. S. Karlsson, and R. M. Sjöberg. 2000b. Soil nitrogen turnover - mineralisation, nitrification and denitrification in European forest soils. Pages 297-331 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Persson, T., H. van Oene, A. F. Harrison, P. Karlsson, G. Bauer, J. Cerny, M.-M. Coûteaux, E. Dambrine, P. Högberg, A. Kjøller, G. Matteucci, A. Rudebeck, E.-D. Schulze, and T. Paces. 2000c. Experimental sites in the NIPHYS/CANIF project. Pages 14-46 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Peters, R. H. 1983. The ecological implications of body size. Cambridge University Press, Cambridge, UK. Petersen, H. 1975. Estimation of dry weight, fresh weight, and calorific content of various Collembolan species. Pedobiologia

15: 222-243. Petersen, H., and M. Luxton. 1982. A comparative analysis of soil fauna populations and their role in decomposition

processes. Oikos 39: 286-388. Petz, W., and W. Foissner. 1989. Effects of irrigation on the protozoan fauna of a spruce forest. Verhandlungen der

Gesellschaft für Ökologie (Göttingen 1987) 17: 397-399. Pflug, A. 2001. Determinants of soil community structure and function in european coniferous forests with particular

emphasis on Collembola. PhD-thesis. Justus-Liebig-University, Giessen. Pomorski, R. J. 1998. Onychiurinae of Poland (Collembola: Onychiuridae). Genus International Journal of Invertebrate

Taxonomy (Supplement 9): 201. Ratkowsky, D. A., J. Olley, T. A. McMeekin, and A. Ball. 1982. Relationship between temperature and growth rate of

bacterial cultures. Journal of Bacteriology 149: 1-5. Rauenbusch, K. 1987. Biologie und Feinstruktur (REM-Untersuchungen) terrestrischer Testaceen in Waldböden (Rhizopoda,

Protozoa). Archiv für Protistenkunde 134: 191-294. Read, D. J. 1991. Mycorrhizas in ecosystems. Experientia 47: 376-391. Reichle, D. E. 1977. The role of soil invertebrates in nutrient cycling. Pages 145-156 in Soil Organisms as Components of

Ecosystems, Stockholm. Richter, E. 1995. Untersuchungen zur Ökologie bodenbewohnender beschalter Amöben (Protozoa) in naturnahen Wäldern

der Eifel. Diploma-thesis. Rheinisch-Westfälische Technische Hochschule Aachen, Aachen. Rusek, J. 1998. Biodiversity of Collembola and their functional role in the ecosystem. Biodiversity and Conservation 7: 1207-

1219. Salminen, J., H. Setälä, and J. Haimi. 1997. Regulation of decomposer community structure and decomposition processes in

herbicide stressed humus soil. Applied Soil Ecology 6: 265-274. Scarascia-Mugnozza, G., G. A. Bauer, H. Persson, G. Matteucci, and A. Masci. 2000. Tree biomass, growth and nutrient

pools. Pages 49-62 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Schimel, D. S. 1995. Terrestrial ecosystems and the carbon cycle. Global Change Biology 1: 77-91. Schönborn, W. 1962. Zur Ökologie der sphagnikolen, bryokolen und terrikolen Testaceen. Limnologica 1: 231-254. Schönborn, W. 1965. Untersuchungen über die Ernährung bodenbewohneneder Testaceen. Pedobiologia 5: 205-210. Schönborn, W. 1966. Beschalte Amöben (Testacea). A. Ziemsen-Verlag, Wittenberg-Lutherstadt. Schönborn, W. 1968. Vergleich der zönotischen Größen, der Verteilungsmuster und der Anpassungsstandards der

Testaceen-Taxocönosen in der Biotopreihe vom Aufwuchs bis zum Erdboden. Limnologica 6: 1-22. Schönborn, W. 1975. Ermittlung der Jahresproduktion von Boden-Protozoen. I. Euglyphidae (Rhizopoda, Testacea).

Pedobiologia 15: 415-424. Schönborn, W. 1977. Production Studies on Protozoa. Oecologia 27: 171-184. Schönborn, W. 1978. Untersuchungen zur Produktion der Boden-Testaceen. Pedobiologia 18: 373-377. Schönborn, W. 1981. Populationsdynamik und Produktion der Testaceen (Protozoa: Rhizopoda) in der Saale. Zool. Jb. Syst.

108: 301-313. Schönborn, W. 1982. Estimation of annual production of Testacea (Protozoa) in mull and moder (II). Pedobiologia 23: 383-

127

11 References

393. Schönborn, W. 1986. Population dynamics and production biology of testate Amoeba (Rhizopoda, Testacea) in raw humus

of two coniferous forest soils. Archiv für Protistenkunde 132: 325-342. Schönborn, W. 1992a. Adaptive Polymorphism in Soil-Inhabiting Testate Amoebae (Rhizopoda): Its Importance for

Delimitation and Evolution of Asexual Species. Archiv für Protistenkunde 142: 139-155. Schönborn, W. 1992b. Comparative Studies on the Production Biology of Protozoan Communities in Freshwater and Soil

Ecosystems. Archiv für Protistenkunde 141: 187-214. Schönborn, W. 1992c. The Role of Protozoan Communities in Freshwater and Soil Ecosystems. Acta Protozoologica 31: 11-

18. Schönborn, W., W. Petz, M. Wanner, and W. Foissner. 1987. Observations on the Morphology and Ecology of the Soil-

Inhabiting Testate Amoeba Schoenbornia humicola (SCHÖNBORN, 1964) DECLOITRE, 1964 (Protozoa, Rhizopoda). Archiv für Protistenkunde 134: 315-330.

Schroeter, D. 1995. Charakterisierung von Lebensgemeinschaften beschalter Amöben (Protozoa) in Böden von Fichtenforsten. Diploma-thesis. Rheinisch-Westfälische Technische Hochschule Aachen, Aachen.

Seastedt, T. R. 1984. The role of microarthropods in decomposition and mineralization processes. Annual Review of Entomology 29: 25-46.

Seastedt, T. R. 2000. Soil fauna and control of carbon dynamics: Comparison of rangelands and forests across latitudinal gradients. Pages 293-312 in D. C. Coleman and P. F. Hendrix, editors. Invertebrates as Webmasters in Ecosystems. CAB International, Wallingford.

Sellnick, M. 1928. Formenkreis: Hormilben, Oribatei. Pages 1-42 in P. Brohmer, P. Ehrmann, and G. Ulmer, editors. Die Tierwelt Mitteleuropas. Quelle & Meyer, Leipzig.

Setälä, H., and V. Huhta. 1991. Soil fauna increase Betula pendula growth: laboratory experiments with coniferous forest floor. Ecology 72: 665-671.

Setälä, H., J. Laakso, J. Mikola, and V. Huhta. 1998. Functional diversity of decomposer organisms in relation to primary production. Applied Soil Ecology 9: 25-31.

Setälä, H., V. G. Marshall, and J. A. Trofymow. 1996. Influence of body size of soil fauna on litter decomposition and N-15 uptake by poplar in a pot trial. Soil Biology and Biochemistry 28: 1661-1675.

Setälä, H., E. Martikainen, M. Tyynismaa, and V. Huhta. 1990. Effects of soil fauna on leaching of nitrogen and phosphorus from experimental systems simulating coniferous forest floor. Biology And Fertility Of Soils 10: 170-177.

Setälä, H., J. Rissanen, and A. M. Markkola. 1997. Conditional outcomes in the relationship between pine and ectomycorrhizal fungi in relation to biotic and abiotic environment. Oikos 80: 112-122.

Seyferth, U. 1998. Effects of soil temperature and moisture on carbon and nitrogen mineralisation in coniferous forests. licentiate thesis. University of Agricultural Science, Uppsala.

s'Jakobs, J. J., and J. van Bezooijen. 1984. A manual for practical work in nematology. Nematology Department, Wageningen.

Smith, H. G. 1978. The distribution and ecology of terrestrial protozoa of sub-antarctic and maritime antarctic islands. Br. Antarct. Surv. Sci. Rep. 95: 1-104.

Smith, H. G. 1996. Diversity of Antarctic terrestrial protozoa. Biodiversity and Conservation 5: 1379-1394. Smith, P., O. Andrén, L. Brussaard, M. Dangerfield, K. Ekschmitt, P. Lavelle, and K. Tate. 1998. Soil biota and global change

at the ecosystem level: describing soil biota in mathematical models. Global Change Biology 4: 773-784. Sollins, P., P. Homann, and B. A. Caldwell. 1996. Stabilization and destabilization of soil organic matter: mechanisms and

controls. Geoderma 74: 65-105. Southwood, T. R. E. 1994. Ecological Methods � with particular reference to the study of insect populations. Chapman &

Hall, London. Stout, J. D. 1980. The Role of Protozoa in Nutrient Cycling and Energy Flow. Adv. Microb. Ecol. 4: 1-50. Stout, J. D. 1984. The protozoan fauna of seasonally inundated soil under grassland. Soil Biology & Biochemistry 16: 121-

125. Stout, J. D., S. S. Bamforth, and J. D. Lousier. 1982. Protozoa. Pages 1103-1120 in Methods of Soil Analysis Part 2 -

Chemical and Microbiological Properties. Stout, J. D., and O. W. Heal. 1967. Protozoa. Pages 149-195 in N. Burges and F. Raw, editors. Soil Biology. Academic

Press, London. Sulkava, P., V. Huhta, and J. Laakso. 1996. Impact of soil faunal structure on decomposition and N-mineralisation in relation

to temperature and moisture in forest soil. Pedobiologia 40: 505-513. Swift, M. J., O. W. Heal, and J. M. Anderson. 1979. Decomposition in terrestrial ecosystems. Blackwell Scientific

Publications, Oxford. Tanaka, M. 1970. Ecological studies on communities of soil Collembola in Mt. Sobo, southwest Japan. Jap. J. Ecol. 20: 102-

110. Taylor, A. F. S., F. Martin, and D. J. Read. 2000. Fungal diversity in ectomycorrhizal communities of Norway spruce (Picea

128

11 References

abies [L.] Karst.) and beech (Fagus sylvatica L.) along north-south transects in Europe. Pages 343-365 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Taylor, A. R. 2001. in prep. Acari in forest soils. PhD-thesis. Justus-Liebig-University, Giessen. ter Braak, C. J. F. 1986. Canonical correspondence analysis: a new eigenvector method for multivariate direct gradient

analysis. Ecology 67: 1167-1179. ter Braak, C. J. F. 1996. Unimodal models to relate species to environment. DLO-Agricultural Mathemetics Group,

Wageningen. ter Braak, C. J. F., and I. C. Prentice. 1988. A theory of gradient analysis. Advances in Ecological Research 18: 271-317. ter Braak, C. J. F., and P. Smilauer. 1998. CANOCO reference manual and user's guide to Canoco for Windows � Software

for Canonical Community Ordination (version 4). Centre for Biometry, Wageningen. Tezuka, Y. 1990. Bacterial regeneration of ammonium and phosphate as affected by carbon:nitrogen:phosphorus ratio of

organic substrates. Microbial Ecology 19: 227-238. Thomas, R. 1958. Le genre Plagiopyxis PENARD. Hydrobiologia 10: 198-214. Tokeshi, M. 1993. Species abundance patterns and community structure. Advances in Ecological Research 24: 112-186. Tolonen, K. 1986. Rhizopod analysis. Pages 645-666 in B. E. Berglund, editor. Handbook of Holocene Palaeoecology and

Palaeohydrology. John Wiley & Sons Ltd, New York. Valentini, R., G. Matteucci, A. J. Dolman, E.-D. Schulze, C. Rebmann, E. J. Moors, A. Granier, P. Gross, N. O. Jensen, K.

Pilegaard, A. Lindroth, A. Grelle, C. Bernhofer, T. Grünwald, M. Aubinet, R. Ceulemans, A. S. Kowalski, T. Vesala, Ü. Rannik, P. Berbigier, D. Loustau, J. Gudmundsson, H. Thorgeirsson, A. Ibrom, K. Morgenstern, R. Clement, J. Moncrieff, L. Montagnani, S. Minerbi, and P. G. Jarvis. 2000. Respiration as the main determinant of carbon balance in European forests. Nature 404: 861-865.

van Oene, H., F. Berendse, T. Persson, A. F. Harrison, E.-D. Schulze, B. R. Andersen, G. A. Bauer, E. Dambrine, P. Högberg, G. Matteucci, and T. Paces. 2000. Model analysis of carbon and nitrogen cycling in Picea and Fagus forests. Pages 419-467 in E.-D. Schulze, editor. Carbon and Nitrogen Cycling in Forest Ecosystems. Springer, Heidelberg.

Vance, E. D., P. C. Brookes, and D. S. Jenkinson. 1987. An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19: 703-707.

Vedrova, E. F. 1995. Carbon pools and fluxes of 25-year old coniferous and deciduous stands in middle Siberia. Water Air and Soil Pollution 82: 239-246.

Verhoef, H. A., and L. Brussaard. 1990. Decomposition and nitrogen mineralization in natural and agroecosystems – the contribution of soil animals. Biogeochemistry 11: 175-211.

Volz, P. 1951. Untersuchungen über die Mikrofauna des Waldbodens. Zool. Jb. Syst. 79: 514-566. Walter, D. E., and H. C. Proctor. 1999. Mites: Ecology, Evolution and Behaviour. CABI Publishing, Wallingford. Wanner, M. 1991. Zur Ökologie von Thekamöben (Protozoa: Rhizopoda) in süddeutschen Wäldern. Archiv für

Protistenkunde 140: 237-288. Wanner, M., and W. Funke. 1989. Zur Mikrofauna von Waldböden: I. Testacea (Protozoa: Rhizopoda)Auswirkungen

anthropogener Einflüsse. Verhandlungen der Gesellschaft für Ökologie 17: 379-384. Wardle, D. A. 1999. How soil food webs make plants grow. Trends in Ecology & Evolution 14: 418-420. WGBU. 1998. The accounting of biological sinks and sources under the Kyoto protocol: a step foreward or backward for

global environmental protection? Special report German Advisory Council on Global Change (WGBU), Bremerhaven.

Willmann, C. 1931. Moosmilben oder Oribatiden (Oribatei). Pages 79-200 in F. Dahl, editor. Die Tierwelt Deutschlands. V. G. Fischer, Jena.

Wodarz, D., E. Aescht, and W. Foissner. 1992. A Weighted Coenotic Index (WCI): Description and application to soil animal assemblages. Biology And Fertility Of Soils 14: 5-13.

Wolters, V. 1983. Oekologische Untersuchungen an Collembolen eines Buchenwaldes auf Kalk. Pedobiologia 25: 73-85. Wolters, V. 1996. Functional implications of biodiversity in soil. Pages 133 in Functional implications of biodiversity in soil.

European Communities, Schloß Rauischholzhausen, Germany. Wolters, V., and R. G. Joergensen. 1992. Die mikrobielle Biomasse in Böden der Sukzessionsreihe Acker, Brache, Wald.

Pages 883-886 in VDLUFA-Schriftenreihe, editor. Woodland, W. A., D. J. Charman, and P. C. Sims. 1998. Quantitative estimates of water tables and soil moisture in Holocene

peatlands from testate amoebae. Holocene 8: 261-273. Woods, L. E., C. V. Cole, E. T. Elliott, R. V. Anderson, and D. C. Coleman. 1982. Nitrogen transformations in soil as affected

by bacterial-microfaunal interactions. Soil Biology & Biochemistry 14: 93-98. Wu, J., R. G. Joergensen, B. Pommerening, R. Chaussod, and P. C. Brookes. 1990. Measurement of soil microbial biomass

C by fumigation-extraction - an automated procedure. Soil Biology & Biochemistry 22: 1167-1169. Wunderle, I., L. Beck, and S. Woas. 1990. Zur Taxonomie und Ökologie der Oribatulidae und Scheloribatulidae (Acari,

Oribatei) in Südwestdeutschland. Andrias 7: 15-60.

129

11 References

Yeates, G. W. 1979. Soil nematodes in terrestrial ecosystems. J. Nematol. 11: 213-229. Yeates, G. W., T. Bongers, R. G. M. de Goede, D. W. Freckman, and S. S. Georgieva. 1993. Feeding habits in soil

nematode families and genera - an outline for soil ecologists. Journal of Nematology 25: 315-331. Yeates, G. W., and W. Foissner. 1995. Testate amoebae as predators of nematodes. Biology & Fertility of Soils 20: 1-7. Yeates, G. W., S. Saggar, and B. K. Daly. 1997. Soil microbial C, N, and P, and microfaunal populations under Pinus radiata

and grazed pasture land-use systems. Pedobiologia 41: 549-565. Zimdars, B., and W. Dunger. 1994. Synopses on Parlaearctic Collembola, Part 1. Tullbergiinae Bagnall, 1935. Abh. Ber.

Naturkundemus. Görlitz 68: 1-71.

130

List of figures

Figure 1.1 Nebela lageniformis. 400x, DIC, in Euparal. Pseu = pseudostome...........................................................................7 Figure 1.2 The ecological scales investigated within this study: linking the soil biota to ecosystem function (C and N flux). ..12 Figure 2.1 Schematic map of the study sites lying on a North-South transect within Europe. Northern latitude is given

beneath site abbreviation. Total N deposition is indicated: 0 = very low; N = intermediate; NN = high. See Table 2.1 for details and site abbreviations. ........................................................................................................................................15

Figure 2.2 Mean monthly temperature (A) and precipitation (B) at the sites. ...........................................................................17 Figure 2.3 Astrid Taylor and Anne Pflug during our autumn sampling at Åheden, N-SE. ........................................................19 Figure 4.1 Practical steps in applying the food web model approach to estimate C and N mineralisation rates (schematic

view)................................................................................................................................................................................42 Figure 4.2 Schematic illustration of the population biology equation, the pathway of energy from consumption to production

resp. mineralisation.........................................................................................................................................................43 Figure 5.1 The abundance (columns) resp. biomass (dots) of living Testate Amoebae in the five size classes. See Table 5.1

and 5.2 for characterisation of size classes. Whiskers represent standard deviation.....................................................55 Figure 5.2 Species biomass rank plots. The total Testate Amoebae biomass on a log scale are plotted against the ranks of

the species (Tokeshi 1993).............................................................................................................................................57 Figure 5.3 Testate Amoebae species from the study sites. A. Bullinularia indica, focus on the dorsal and on the ventral lip.

200x, bright-field, in Euparal. B. Edaphonobiotus campascoides, lateral view of the trumpet-like shell with cytoplasm and vesicular nucleus. 400x, DIC, in Euparal. C. Schoenbornia humicola, nucleus and cytoplasm stained with aniline blue. 400x, bright-field, in watery suspension. D. Nematode and Trinema lineralis, 400x, bright-field, in watery suspension. E. Nebela militaris, cytoplasm stained with aniline blue. 400x, bright-field, in watery suspension. F. Heleopera sylvatica cyst, stained with aniline blue. 200x, bright-field, in watery suspension. Pseu = pseudostome; N = nucleus; n = nucleolus; P = cytoplasm; T = Trinema lineare; Ne = Nematode.......................................................................................................................................................................58

Figure 5.4 The Bray & Curtis similarity index and the number of unique species plotted against geographical distance between the sites. Bray & Curtis = dotted line (r = -0.68, p = 0.13); unique species = unbroken line (r = 0.89, p < 0.05). On the abscissa the distance between pair-wise compared sites is indicated using the following abbreviations: D = DE; F = FR; S = S-SE; N = N-SE...........................................................................................................................................62

Figure 5.5 Microbial parameters (metabolic potential, microbial biomass C, metabolic quotient) along the transect. Symbols of the same shading labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.7 for further details on ANOVA). Whiskers represent standard deviation. ...................................64

Figure 5.6 Percentage of bacterial from total microbial biomass C and frequency of dividing bacterial cells. Columns of the same shading labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.7 for further details on ANOVA). Whiskers represent standard deviation. ............................................65

Figure 5.7 Abundance of the major faunal groups besides Testate Amoebae. Columns of the same shading labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.7 for further details on ANOVA). Whiskers represent standard deviation. ................................................................................66

Figure 5.8 Biplot of sites. The dots represent replicate samples from different sampling times at each site. Samples are identified by the replicate number followed by site abbreviation (NS = N-SE, SS = S-SE) and sampling time (1 = 1st sampling in Oct/Nov, 2 = 2nd sampling in May/Jun; 3 = 3rd sampling in Sept, 4 = 4th sampling in Mar/Apr). ..................71

Figure 5.9 Biplot of species. Aca = Arcella catinus; Amu = Assulina muscorum; Ase = Assulina seminulum; Bin = Bullinularia indica; Cga = Centropyxis gauthieri; Cma = Centropyxis matthesi; Csp = Centropyxis sphagnicola; Csy = Centropyxis

131

List of figures

sylvatica; Cdu = Corythion dubium; Cov = Cryptodifflugia oviformis; Ceu = Cyclopyxis eurystoma; Cka = Cyclopyxis kahli; Dlu = Difflugia lucida; Dmi = Difflugia minuta; Eca = Edaphonobiotus campascoides; Ela = Euglypha laevis/rotunda; Erm = Euglypha rotunda minor; Est = Euglypha cf. strigosa; Hsy = Heleopera sylvatica; Hsu = Hyalosphenia subflava; Mpa = Microchlamys patella; Mfl = Microcorycia flava; Nla = Nebela lageniformis; Nmi = Nebela militaris; Npt = Nebela parvula/tincta; Nmb = Nebela tincta major/bohemica/collaris; Pac = Phryganella acropodia; Ppr = Phryganella paradoxa alta; Pde = Plagiopyxis declivis; Pin = Plagiopyxis intermedia; Pla = Plagiopyxis labiata; Shu = Schoenbornia humicola; Svi = Schoenbornia viscicula; Tde = Tracheleuglypha dentata; Tpu = Trachelocorythion pulchellum; Tar = Trigonopyxis arcula; Tmi = Trigonopyxis minuta; Tco = Trinema complanatum; Ten = Trinema enchelys; Tli = Trinema lineare; Tpe = Trinema penardi. .......................................................................73

Figure 5.10 The abundance of living cells (10-6 m-2) and empty shells (10-6 m-2 resp. %) along the transect. In comparison of dots and columns of the same shading those labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.12 for further details on ANOVA). Whiskers represent standard deviation. ........................................................................................................................................................................74

Figure 5.11 Occurrence of living cells (% of all specimen found) and cysts (% of living cells) at the different sampling times (averages over all sites). In comparison of columns of the same shading those labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.12 for further details on ANOVA). Whiskers represent standard deviation...........................................................................................................................75

Figure 5.12 Testate Amoebae biomass at the different sites and sampling times. Within a site specific sub-plot columns labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.12 for further details on ANOVA). Whiskers represent standard deviation....................................................................77

Figure 5.13 Sketch of the decomposer food web (connectedness web). Feeding relationships are indicated by arrows pointing from prey to predator. fung. = fungivorous; bact. = bacterivorous; panphyt. = panphytophagous; omni. = omnivorous; pred. = predaceous. ...................................................................................................................................79

Figure 5.14 Correlation coefficients (r) between the biomass structure at different sites (relative contributions to total biomass, pair-wise correlation) against the geographical distance (km) between the sites (straight line: r = -0.91, p < 0.05, point FR/DE excluded from regression).................................................................................................................86

Figure 5.15 Estimates of C and N mineralisation rates at the different sites obtained using the food web model ("simulated") and laboratory incubations of soil cores ("observed"). Laboratory incubation data are taken from Persson et al. 2000ab. In comparison of columns of the same shading those labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.16 for further details on the ANOVA). .............87

Figure 5.16 Relative contributions of the functional groups to C mineralisation (%) at the sites. In comparison of the mineralisation by a particular functional group at the sites (i.e. the horizontal comparison of bars on the same level of the ordinate axis) values labelled with identical letters are not significantly different from each other according to the Tukey HSD test of a one-way ANOVA on the main effect ‘site’ (see Table 5.16 for further details on the ANOVA). Whiskers on bars represent standard deviation..............................................................................................................89

132

List of tables

Table 2.1 Characteristics of the four selected coniferous sites (from data given in Persson et al. 2000c)...............................20 Table 2.2 Sampling times at the four sites. See Table 2.1 for site abbreviations. ....................................................................21 Table 3.1 General taxonomic literature for the determination of Testate Amoebae. ................................................................30 Table 3.2 Specialised taxonomic literature and monographies for the determination of Testate Amoebae. ............................31 Table 3.3 Nematode genera found on the sites and their feeding habits according to Yeates (1993). Following the simplified

classification for the food web model further food sources or feeding modes that may occur are given in parentheses.35 Table 5.1 List of species that were found on the study sites and classification into size classes. See Table 5.2 for definition of

size classes.....................................................................................................................................................................54 Table 5.2 Size classes of Testate Amoebae and number of species found belonging to each size class. ..............................55 Table 5.3 Results of the two-way ANOVAs on the effect of 'site' and 'time' on number of species, diversity and evenness. df-

Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p = p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01, n = 50. ..........................................................56

Table 5.4 Total number of species and mean number of species found at each site. For the calculation of diversity and evenness the relative abundance of living Testate Amoebae cells (active cells + cysts) was used; empty shells were not taken into consideration. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.3 for further details on ANOVA)...56

Table 5.5 The community structure of Testate Amoebae. Asterisks represent relative biomass (%). Species are arranged according to their occurrence or absence at certain sites. Dotted lines indicate groups labelled with capital letters (A-G). For explanation of the grouping see section 5.1.2.3. ****** > 31.9 %; ***** = 10.0 to 31.9 %; **** = 3.2 to 9.9 %; *** = 1.0 to 3.1 %; ** = 0.32 to 0.99 %; * < 0.32 %; empty space = 0 %; ° = found only once during counting or occurred only when analysing enriched material from flotations or batch cultures........................................................................59

Table 5.6 Total number of species, and matrices of the number of unique species (Colwell and Coddington 1994) and Bray & Curtis-similarities (Bray and Curtis 1957, Southwood 1994) in comparison of the four sites.............................. 61

Table 5.7 Results of the two-way ANOVAs on the effect of 'site' and 'time' on microbial and faunal parameters. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01. .........................................................................................63

Table 5.8 Results of the two-way ANOVAs on the effect of 'site' and 'time' on abiotic parameters. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01......................................................................................................................67

Table 5.9 Mean, minimum/maximum (2nd row) and standard deviation (2nd row, in parentheses) of the environmental parameters at the sites. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.8 for further details on ANOVA). .....................68

Table 5.10 Summary of the CCA of species biomass pattern and environmental variables. ...................................................69 Table 5.11 Conditional effects of including the environmental variables into the CCA one after the other using forward

selection. cum(�A) = cumulative explanatory power (variance explained) by including the environmental variable; rC = canonical correlation coefficient of the inter-set correlations of environmental variables with the axes. p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01. .........................................................................................70

Table 5.12 Results of the two-way ANOVAs on the effect of 'site' and 'time' on various markers of the Testate Amoebae community. df-Effect = 3 (site, time); df-Effect = 9 (interaction site � time); F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * � 0.05; ** < 0.01.................................................74

Table 5.13 Feeding behaviour of some soil Testate Amoebae species resp. genera. .............................................................80

133

List of tables

Table 5.14 Physiological parameters for each functional group. Assimilation and production efficiencies (a and p) taken from Andrén et al. (1990) and C:N-ratios (q) from Hunt et al. (1987) if not stated otherwise. Basic death rates (at 10°C) were obtained from Hunt et al. (1987) and De Ruiter et al. (1993a) and adapted according to temperature and moisture regime of the specific sites...............................................................................................................................82

Table 5.15 The feeding preferences wij. Explanation see section 4.2.3. prmi = predaceous Acari; prco = predaceous Collembola; prne = predaceous Nematoda; omne = omnivorous Nematoda; pami = panphytophagous Acari; paco = panphytophagous Collembola; prta = predaceous Testate Amoebae; pata = panphytophagous Testate Amoebae; fune = fungivorous Nematoda; bane = bacterivorous Nematoda; ench = Enchytraeidae; fung = fungi; bact = bacteria; detr = total detritus. ...................................................................................................................................................................83

Table 5.16 Results of ANOVAs on the main effect of ‘site’ on biomass (% resp. kg C ha-1) and on simulated C and N mineralisation (% resp. kg ha-1 a-1) of individual functional groups within the decomposer food web (one-way ANOVA). df-effect = 3; F = inter-group variance divided by intra-group variance; p-level of significance: n.s. > 0.05 (not significant); * ≤ 0.05; ** < 0.01. .......................................................................................................................................84

Table 5.17. Biomasses of functional groups (%) and total (kg C ha-1) at each site. Mean values of four sampling occasions are shown, standard deviations are given in parentheses. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test (see Table 5.16 for details on the ANOVA)......................................................................................................................................................85

Table 5.18 Contributions of functional groups within the decomposer food web to N mineralisation (kg N ha-1 a-1). Mean values of four sampling occasions are shown, standard deviations are given in parentheses. In comparison within a specific row values labelled with identical letters are not significantly different from each other according to the Tukey HSD test of a one-way ANOVA on the main effect ‘site’ (see Table 5.16 for details on the ANOVA). ...............................91

Table 10.1 C contents (from Berg 1997) of body dry weight and biomass conversion factors of decomposer fauna groups (for details see sections 3.1.2-5). ........................................................................................................................................119

Table 10.2 Nematode biomass was calculated from genus specific abundances using conversion factors from the literature (Ekschmitt et al. 1999) or from calculations using the formula from Andrássy (1956) and size estimates from Bongers (1994). Body volume of juveniles was estimated to be on average 22 % of the adult body volume (Ilja Sonnemann, pers. com.). ...................................................................................................................................................................119

Table 10.3 Simulated N and C mineralisation rates (kg ha-1 a-1) at the different sites from a scenario assuming equal climate and resource quality at all sites (Scenario "equal") compared to the estimates obtained after adjusting the model to site specific conditions (Scenario "site specific"). .........................................................................................................120

134

Acknowledgements

Acknowledgements

"Because we do most things relying only on our own sagacity we become self-interested, turn our backs on reason, and things do

not turn out well. As seen by other people this is sordid, weak, narrow and inefficient. When one is not capable of true intelligence, it is good to consult

with someone of good sense. [...] It is, for example, like a large tree with many roots.

One man's intelligence is like a tree that has been simply stuck in the ground."

(by Yamamoto Tsunetomo 1659-1713, from 'Hagakure' alias 'The Way of the Samurai',

the inspiration to the film 'Ghost Dog' by Jim Jarmusch) Prof. Dr. Volkmar Wolters is gratefully acknowledged for supporting this thesis and for giving me the opportunity to participate within two international research projects (CANIF and GLOBIS). I thank him for stimulating discussions and for his confidence. Thanks to his support I was able to meet, interact with and learn from many fellow researchers. I am very grateful to Prof. Dr. Peter C. De Ruiter, University of Utrecht, whom I first met on one of the TERI workshops that were hosted in Giessen. He taught me to use the food web model. I greatly appreciate his help and his friendly and fruitful ways of discussion. I am grateful that he accepted to be referee of this thesis. Dr. Ralf Meisterfeld was the supervisor of the Testate Amoebae work within this thesis and I dearly thank him for his support. Thank you, Anne Pflug, Astrid Taylor and Jens Dauber, for continuous support and encouragement. Discussions, many journeys and evenings with these three made my PhD-years worth while. Special thanks to Anne Pflug for providing the Collembola data, to Astrid Taylor for providing the Acari data, to Jens for sharing his views and discussing any aspect of ecology, back and forth and back again. Thanks to all three of them for corrections and helpful comments on many chapters of this thesis. For statistical advice I thank Prof. Dr. Wolfgang Köhler, Dr. Klemens Ekschmitt and Dr. Gabriel Schachtel. I thank Monika Leonardt, Christine Tandler, Birgit Wasmus, Susanne Vesper, Barbara Beier and Martin Kröckel for their help in the laboratory. I very much enjoyed the co-operation with the CANIF and GLOBIS partners. � I am particularly grateful to Niklas Lindberg, Janne Bengtsson and Tryggve Persson for stimulating discussions, introduction to Bob Hund, support and company, especially on our sampling trips and visits to Uppsala, and during our twitching tour around Halkidiki. I dearly thank Tony Harrison and Harmke van Oene for openly sharing information on their modelling

135

Acknowledgements

approaches and the exchange of early drafts of publications. I thank Andy Taylor for answering many questions on ectomycorrhiza and life (in soil) in general. Discussing N mineralisation (HA! here it is, the N-word...!) with him was always a source of motivation. I am very grateful to Karl Reiter and Georg Grabherr for their help with multivariate statistics and an enjoyable time in Vienna. For creating a joyous working atmosphere and for their helpful ways I am grateful to Johannes Frisch, Ilja Sonnemann, Maria Robeck, Anna Kohler, Stephanie Holzhauer and Tobias Purtauf. Special thanks to the last three for adding a great deal of enthusiasm and new ideas to our working group; to Anna for correcting the introduction; to Ilja for her help with the Nematoda; to Johannes for help on taxonomic nomenclature and countless candy...! Thank you, Heide and Uwe Schröter, for your support, patience and friendship especially during the last two years. Thank you, Hanna Tigges, for your heartiness and love of life, that I find worth pursuing. � "The way to hell is paved with good will" was one of her favorite sayings... Thank you, Ralph Hückelhoven, Christiane Heineck, Reinhild Biermann, Rolf Schröter, Susann Beetz, Jürgen Geerlings and Jutta Herzogenrath, friends indeed, for their support. Thanks especially to Ralph for proof-reading many parts of this thesis and for his trustiness in emergency; to Rolf for his advice about lay-out and help with the cover, and to Christiane for making our home sweet. Thank you, Eva Gessner, Jörg Espelta and Mike Fechner for adding essential joy. "Echte Fründe ston zesamme, su wie eine Jott un Pott...."

Sampling time in Åheden, N-SE.

136

curriculum vitae

name Dagmar Schröter date of birth 27.05.1970

nationality German address Gutenbergstr. 90, 14467 Potsdam, Germany

Professional development since May 2001 Scientific co-ordinator at the Potsdam Institute for Climate Impact Research

Apr 1996 – Dec 00 Research assistant at the Institute of Animal Ecology & Zoology, Justus-Liebig-University Giessen

Dec 1995 – Jan 96 Research assistant at the Institute of General Biology, Technical University Aachen

Education Nov 1995 Biology Diploma

Nov 1994 – Nov 95 Diploma thesis at the Institute of Ecology and Zoology, Technical University Aachen: "Characterisation of communities of Testate Amoebae (Protozoa) in spruce forest soils". Supervisors: Prof. Dr. P. Schmidt and Dr. R. Meisterfeld

Oct 1989 – Nov 94 Biology studies at the Technical University Aachen May 1989 Graduation (Abitur): German high school, Grevenbroich

Aug 1986 – Jul 87 Graduation: American high school, Little Valley, New York, USA Aug 1987 – May 89

Aug 1980 – Jul 86 German high school, Grevenbroich (grades 5 – 10, grades 12 – 13)

137

Structure and function of the decomposer food webs of forests

Documents

Transcript of Structure and function of the decomposer food webs of forests