Salt marshes in temperate regions are very productive natural vegetations. These vegetations frequently reach an above-ground production of more than 1 kg of dry weight per m 2per year. Herbivores consume only a small proportion of the annual plant production. Almost the entire amount of above ground plants dies after senescence. A small proportion may be washed away by the tides, but the major part remains at the salt marsh where it decomposes in the canopy or at the sediment surface.
Dead plant material is primarily decomposed by micro-organisms, such as fungi and bacteria. The chemical composition of the detritus to a large extent determines the rate of decomposition. A number of abiotic factors, such as temperature and humidity, also influence the decomposition process. In addition the process may be affected by fauna, present on the decomposing plant material.
In this thesis the role of nematodes in decomposition of Spartina anglica was studied. This plant species commonly occurs in salt marshes of Western Europe. In addition, one chapter is dedicated to the association between nematodes and decomposing seaweed in a completely different habitat: an Antarctic beach. In the first part of this thesis the relation between decomposition and naturally occurring nematode populations is studied. This part consists mainly of field studies. Nematodes, which are associated with the decomposition process are identified, and the population dynamics of one of these species is studied in detail. The second part of this thesis is dedicated to laboratory and model studies which were carried out to investigate the effects of nematodes on decomposition of S.anglica detritus and the possible mechanism underlying these effects.
Nematode populations on decomposing plant material
Nematodes are abundant on both S.anglica litter and on stranded Antarctic seaweed. We found that on standing dead Spartina anglica plant parts the nematode population frequently reached densities of 3000 individuals per g DW. When leaf material on the sediment surface was investigated even much higher nematode densities were found, up to 47,000 individuals per g DW. At Antarctic beaches nematode densities up to 26,000 individuals per g DW were found on seaweed wrack.
Although numerous, not all of these nematodes present on plant detritus are expected to influence the decomposition process. In chapter I an attempt was made to distinguish the nematode species which play a role in decomposition of S. anglica detritus from nematodes which do not have such a role. As decomposition is largely a microbial process, higher decomposition rates presumably coincide with a higher microbial production and, consequently, a higher availability of food for microbivorous nematodes. Amongst the microbivorous nematodes, those species were considered of possible importance to the decomposition process when their numbers increased with increasing decomposition rate. In the experiments, mesh containers, filled with Spartina anglica leaves, were placed on the sediment surface. Different decomposition rates were induced by using decaying leaf material of different ages and by repeating the experiments during four subsequent seasons. Mesh containers with inert material (plastic drinking straws) served as controls. Sixty nematode species were found in the mesh containers. Using a multivariate analysis (redundancy analysis) different nematode communities were found on plant material with different decomposition rates. These differences were caused by the changing abundance of only a few species. The majority of the species were found in equal numbers in treatments with decomposing Spartina leaves and in the control treatment. The numbers of individuals of those species which appeared closely correlated with the decomposition rate of Spartina anglica leaf-detritus were all bacterivorous nematodes. Numerically the most dominant were species of the family Monhysteridae ( Diplolaimelloides bruciei, Diplolaimella dievengatensis, Monhystera parva ). The highest numbers of these nematodes were found in treatments with the highest decomposition rates i.e. on decaying fresh leaves, during the warmer seasons. In the winter, when decomposition is slower, their numbers were lower.
The species diversity on standing dead plant parts of Spartina anglica is much lower than the species diversity on the sediment surface in mesh containers filled with S.anglica leaves. The dominant species on standing dead plants are the bacterivorous nematodes Diplolaimelloides bruciei, Monhystera disjuncta and Pellioditis marina. In chapter 11 a study is presented on the population dynamics of D. bruciei. This species was commonly found on above ground plant parts of Spartina. In a field study, population densities of this species were estimated on four classes of S.anglica plant material, representing the whole range of decomposition stages found in the canopy. D. bruciei was found throughout the year on all types of plant material, including living green plant parts. The population densities were highest on the older plant material, where densities of 1000-2000 individuals per g DW were reached. The highest densities were recorded in late summer and autumn.
S. anglica vegetations are regularly flooded at high tide, which potentially reduces the nematode population density on the plant material, as nematodes may be flushed from the plants. Since in situ measurements of the flooding effect are not possible, the population dynamics of D. bruciei was studied in the laboratory under a controlled flooding regime. The population densities of D.bruciei indeed seemed to be highly influenced by flooding. A considerable part of the population disappeared during flooding, but on younger, yellow, decomposing leaves the rate of removal by flushing was much lower than on older, brown, leaves. This is probably caused by the change of the leaf structure during decomposition. Nematodes may become less well attached to the leaf surface when the groove structure of the leaves disappears with progressive decay; consequently, a higher proportion is flushed away. The growth rate of the population, however, was equal on both leaf types. The growth rate of the nematode population, as estimated in the laboratory, was used to calculate the total production of nematodes in the field. It was shown that the total biomass production of D. bruciei equalled 114 mg C per m 2per year. If 30% of the detritus was decomposed by bacteria and a trophic efficiency of 10 % is assumed, the total amount of bacteria] carbon ingested by D.bruciei accounted for 7.5 % of the total bacterial biomass produced. It was estimated that the dominant bacterivorous nematodes together may consume over 20% of the total bacterial biomass production.
In chapter III a study of nematodes found in stranded seaweed at an Antarctic beach is presented. Large amounts of seaweed are deposited along the coast of Admiralty Bay, King George Island, Antarctica. The stranded seaweed partly decomposes on the beach and supports populations of various meiofauna species, mostly nematodes. The factors determining the number of nematodes found in the seaweed packages were studied. The densities of nematodes appeared to be correlated primarily with salinity, height and C:N ratio of the detritus. Salinity and height were most likely related to the flooding regime in conjunction with the off-stream of melt water. Decomposition rate appeared mainly determined by the water content and the sediment composition. Melt water run-off or the impact of the surf probably increased seaweed weight losses in these situations.
The effect of nematodes on decomposition of S.anglica
Experiments with D. bruciei, a species numerously present on standing dead S.anglica plants (see chapter II), were set up to study the effect of this nematode on decomposition (chapter IV). Green and yellow leaves were placed on agar in petri dishes and inoculated with D. bruciei. CO 2 production was determined regularly after inoculation. Weight, carbon and nitrogen losses were determined at the end of the experiment, 30 days after inoculation. In the presence of nematodes, CO 2 -production on green, decaying leaves increased by 20 - 25 %. Losses of dry weight, carbon and nitrogen during decomposition increased with at least 30 %. On yellow, more senescent leaves no effect on CO 2 -production was found, but losses of dry weight, carbon and nitrogen tended to be higher in the presence of nematodes. The results of this study show that D.bruciei may enhance the decomposition rate of S.anglica -leaves; the extent of the stimulatory effect, however, depends on leaf condition and the population density of the nematode. The minimal nematode population density for a measurable stimulatory effect was estimated to be 4000 individuals per g DW of S.anglica leaves. As described in chapter II, field population densities are often of the same order of magnitude.
A part of the senescent S.anglica leaves and stems decompose at the sediment surface, where the material is covered with sediment. In chapter I a clear correlation was found between the number of the bacterivorous nematode Diplolaimella dievengatensis and the decomposition rate of S.anglica detritus present on the sediment surface. The effect of the D.dievengatensis on the carbon mineralization of S.anglica detritus was examined in a laboratory experiment (chapter V). Detritus mixed with sediment appeared to decompose at higher rates in the presence of the nematodes. CO 2 production per hour was 74 % higher in the presence of the nematode than in its absence; O 2 consumption per hour increased to a similar extent. Diffusion coefficients were calculated from measurements of both O 2 consumption, using gas chromatography, and O 2 micro-gradients, using micro-electrodes. The apparent diffusion coefficient of O 2 in the sediment in the presence of nematodes was 40% to 70 % higher than the bulk sediment diffusion coefficient. Since the increase of the CO 2 production and of the diffusion of oxygen in the presence of nematodes was of the same magnitude, we concluded that the enhanced turnover time of Spartina detritus presumably was largely caused by the bioturbation activity of the nematodes.
A simulation model was constructed to quantify the relations between decomposing S.anglica detritus, bacteria and their grazers (chapter VI). The model takes the various stages of above ground litter decomposition into account. The heterogeneity of the decomposing litter was described by a number of successive quality classes. Decomposition was considered to be primarily a microbial process. The microbial population was assumed to consist of a number of successional species each possessing a unique preference for the different quality classes. Grazers were all considered as a single species grazing upon all microbial species. Three mechanisms by which grazers may stimulate decomposition were evaluated using the data from the laboratory study presented in chapter IV. In the first place: if the microbial population grows to a certain maximal density than removing microbial biomass by grazers may stimulate decomposition since space is created for growth of new microbes at the expense of organic substrate. In the second place: the excretion of highly nutritive mucus by grazers may stimulate bacterial growth. In the third place: reworking of the sediment-detritus-microbial mixture in the grooves of the leaves (see also chapter II), or in the upper layer of the sediment may increase the oxygen availability and may, by mechanical force, enlarge the surface of the substrate on which the microbes attack. The model calculations suggested that removing of microbial biomass by grazers has some stimulatory effect on the decomposition rate of detritus, but not enough to account for the total effect. Recycling of organic matter by excretion of mucus seemed to have no effect at all.
According to the model, bioturbation or reworking contributed most to the stimulation of the decomposition rate.
The model was validated with field data. The model could describe field data obtained from a variety of locations. The biomass of bacteria and grazers estimated by the model were in the same order of magnitude as those found in the field. The model is useful to evaluate decomposition data from different studies and calculate an approximate amount of microbes and primary grazers available for higher trophic levels.
When the model calculations were performed over a period of about a year the stimulating effect of grazers gradually seemed to vanish. This is in agreement with the experiments described in chapter IV, which show that the effect of nematodes on decomposing yellow leaves were less pronounced than on green leaves. Thus, any stimulatory effect of nematodes on decomposition of Spartina anglica in the salt marsh may be restricted to the first stages of the decomposition process.