PhD thesis. Larsen, T. 2007. Plant Nutrition and Soil Fertility Laboratory, University of Copenhagen and Dept. of Terrestrial Ecology, National Environmental Research Institute, Aarhus University. 34 pp.
Soil Collembola are important for decomposition and cycling of carbon and nitrogen but only few studies have quantified collembolan release of these elements. Quantifying the release rates and understanding the mechanisms mediating them might be a key for modelling how collembolans alter the biochemical cycling of nutrients in soil.
The objectives of the thesis were to: i) investigate turnover of carbon and nitrogen by collembolans in relation to their physiology and dietary requirements; and ii) use collembolans as a tool to assess biological mediated fluxes of carbon and nitrogen in soil.
The soil food web is a nitrogen limited environment which also limits the availability of the nitrogen rich amino acids. No studies have yet addressed how amino acid imbalances in the diet affect collembolan productivity and physiology. An insight into physiological responses can be obtained by measuring tissue composition of carbon, nitrogen and phosphorus that are key elements in biochemical compounds. Information on internal physiological processes can also be obtained by analyzing 15N because nitrogen stress is likely to lead to an increased recycling of nitrogen and consequently retention of 15N in the tissue.
This thesis is composed of four separate studies that progressively address specific aspects of collembolan physiology and uptake/ release of carbon and nitrogen. To link the studies with each other, two species of Collembola, the small hemiedaphic Proisotoma minuta and the large euedaphic Protaphorura armata, were included in all the studies allowing us to compare different aspects of collembolan physiology.
Release of carbon and nitrogen by Collembola has been estimated by mass balance models and by direct measurements but these methods have in their current forms some disadvantages. Instead we implemented an isotope massbalance approach that would allow us to follow how two collembolans with contrasting life history strategies, P. minuta and P. armata, allocated carbon and nitrogen to growth, reproduction and metabolism. We also examined how two diets with different amino acid imbalances affected elemental turnover and isotope discrimination. Finally, two plant microcosm studies were included to i) investigate how collembolans (and soil mesofaun) affected plant growth and nutrient release, and to ii) determine the sources of carbon and nitrogen in collembolans using a linear mixing model.
The results showed that life history strategies of collembolans were key factors in explaining carbon and nitrogen allocations. We estimated that metabolism contributed to replacement rates of tissue of ~10 % day-1 for P. minuta, and ~6 % day-1 for P. armata (for carbon and nitrogen). In comparison, reproductive tissue replacement rates were ~2 % day-1 for P. minuta and ~0.5 % day-1 for P. armata. The relative expenditures to metabolism relative to reproduction were thus lower for P. minuta than P. armata.
The study on amino acids imbalances revealed that low protein quality increased C:N and C:P ratios significantly (p<0.05) indicating that carbon rich storage compound (lipids and fatty acids) increased relative to nitrogen-phosphorus rich compounds (RNA and proteins). However, P. armata had a much weaker response compared to the two other species. We also measured isotope compositions and found significantly lower 15N values (p<0.05) in the tissue of collembolans raised on the low than high protein quality diet. This contrasts the hypothesis proposing that amino acid deficiency increases retention of 15N in tissue. A possible explanation for this result might be that collembolans preferentially digested proteins with amino acids depleted in 15N. In the microcosm studies, we did not find a significant increase in plant growth in treatments with Collembola. Analysis of source contributions indicated that photosynthate (root derived C) was the main carbon source for collembolans (54–79 % of total C) supporting recent results that the rhizosphere channel is very important for collembolans in addition to the detritus based channel.
Full report in pdf-format (658 kB).
