Publications

While several studies have shown that invasive plant effects on soil biota influence subsequent plant performance, corresponding studies on how invasive animals affect plants through influencing soil biota are lacking. This is despite the fact that invasive animals often indirectly alter the below-ground subsystem. We studied 18 offshore islands in northern New Zealand, half of which have been invaded by rats that are predators of seabirds and severely reduce their densities, and half of which remain non-invaded; invasion of rats thwarts seabird transfer of resources from ocean to land. We used soil from each island in a glasshouse experiment involving soil sterilization treatments to determine whether rat invasion indirectly influences plant growth through the abiotic pathway (by impairing seabird-driven inputs to soil) or the biotic pathway (by altering the soil community). Rat invasion greatly impaired plant growth but entirely through the abiotic pathway. Plant growth was unaffected by the soil community or its response to invasion, meaning that the responses of plants and soil biota to invasion are decoupled. Our results provide experimental evidence for the powerful indirect effects that predator-instigated cascades can exert on plant and ecosystem productivity, with implications for the restoration of island ecosystems by predator removal.

Feather mosses form a thick ground layer in boreal forests that can intercept incoming litter fall. This interception may influence the decomposition of in

There has been a growing recent interest in how foliar herbivory may indirectly affect the belowground sub‐system, but little is known about the belowground consequences of the identity, species composition or diversity of foliar herbivores. We performed an experiment, utilising model grassland communities containing three plant species, in which treatments consisted of addition of each of eight aphid species in single and in two‐ four‐ and eight‐species combinations, as well as an aphid‐free treatment. While aphid species treatments did not affect total plant biomass or productivity, aphid species identity had important effects on the relative abundance of the three plant species. This in turn affected the abundances of each of three groups of secondary consumers in the soil food web (bacterial‐ and fungal‐feeding nematodes, and enchytraeids) but not primary consumers (microbes, herbivorous nematodes) or tertiary consumers (predatory nematodes). The fact that some trophic levels responded to treatments while others did not is consistent with trophic dynamic theory. Aphid species treatments also affected the community composition within each of the herbivorous, microbe‐feeding and top predatory nematode groups, as well as diversity within the first two of these groups. However, aphid species diversity per se had few effects. There were specific instances in which specific aboveground and belowground response variables in two aphid species combinations differed significantly from those in both of the corresponding single aphid species treatments (apparently as a consequence of resource use complementarity between coexisting aphid species), but no instance in which increasing aphid diversity beyond two species had any effect. Our results provide evidence that the identity of aboveground consumers can have effects that propagate through multiple trophic levels in soil food webs in terms of consumer abundance, and composition and diversity within trophic levels.

1. The effects of the grass species Lolium perenne L. and nine dicotyledonous grassland species, grown in monocultures and two-species mixtures, on (i) the soil microbial biomass, (ii) the respiration: biomass ratio and (iii) plant litter decomposition was investigated in a glasshouse experiment. 2. Microbial biomass was sometimes greater and sometimes less in the two-species mixtures than could be explained in terms of the additive effects of the two component species grown singly; this variation was independent of differences in below-ground plant productivity between monoculture and mixture treatments. 3. The microbial respiration: biomass ratios and plant litter decomposition rates in the two-species mixture treatments were either greater or less than expected based on the monoculture treatments; these differences were dependent on the combinations of species present. Because the respiration: biomass ratio is a measure of ecosystem stability, it is here proposed that stability does not respond predictably to shifts in species diversity. 4. These results provide evidence that increasing plant species richness (from one to two species) has the potential to influence soil processes positively or negatively in a non-additive way. The possible ecological implications of this are discussed.

No abstract is provided for this article.

Aim To evaluate the potential of island systems as tools for investigating how species affect ecosystem properties. This is done by first reviewing the available literature, and then by focusing on a study in an archipelago of lake islands in northern Sweden. Location Islands in lakes Uddjaure and Hornavan in the northern Swedish boreal forest. Results The available literature suggests that islands offer two types of opportunities for investigating how species affect ecosystem properties. The first emerges from the fact that island biotas are often relatively depauperate compared with corresponding continental biotas. This enables the effects of species to be assessed against a less complicated background. The second involves the use of islands as spatially independent replicate ecosystems, which can be used to evaluate the effects of ecosystem drivers over larger spatial scales than is possible through most other approaches that allow proper replication of ecosystems. Examples are presented of the use of each approach. The Swedish lake island system is based on fifty independent islands. A wildfire frequency gradient across these islands favours different plant functional types on different islands; this leads to plant species effects influencing a range of properties both above‐ and below‐ground. Two examples are presented of ecological issues that can be investigated using this island system. First, the difference in fire frequency across islands enables assessment of the effects of fire (and hence fire suppression) on terrestrial carbon storage, and the global change implications of this. Secondly, the system can be used to address theories about the importance of plant community composition and biodiversity in driving ecosystem properties at the across‐island spatial scale. Conclusions Islands provide real opportunities for investigating biological drivers of ecosystem functioning, although they have been vastly underused for such purposes. Further, island studies have significant potential for investigating the effects of global change phenomena on terrestrial ecosystems, ranging from the impacts of biodiversity loss and invasive organisms to issues relating to terrestrial carbon sequestration. Island ecosystems enable questions to be addressed about ecosystems over ecologically meaningful spatial scales that can be addressed in no other way.

Summary There has been growing recent interest in the relative importance of within‐species trait variation vs. across‐species trait variation in vascular plants in determining total community‐level trait variation across communities and environmental gradients. Recent studies on plant communities have generally found across‐species variation to be more important than within‐species variation, but comparable studies involving other functionally important biota, such as lichens, are largely lacking. We used a fire‐driven chronosequence involving 30 lake islands in northern Sweden to study how declining soil fertility during retrogression affects the functional traits of each of the dominant epiphytic lichen species growing on the trunks of the tree B etula pubescens . We measured several functional traits for the commonest lichen species on each island and used community‐weighted measures to study the community‐level responses of lichens to the gradient. We found that as retrogression proceeds and soil fertility declines, thallus N and P concentrations and specific thallus mass ( STM ) increase, both within species and at the community level. Lichen secondary compounds showed contrasting within‐species responses and were non‐responsive at the whole community level. By decomposing community‐level measures of these traits across the gradient, we showed that for the three most responsive traits (N, P and STM ), within‐species variation was substantially more important than across‐species variation. This emerges in part because lichen species composition was not very responsive to ecosystem retrogression, and because unlike vascular plants, lichens easily absorb elements over their entire surface, meaning that nutrient concentrations within lichen species are likely to more closely reflect nutrient availability. We found that within‐species variability drove the changes in community‐weighted measures of lichen traits across a strong environmental gradient, which contrasts strongly with what we know from studies of vascular plants where across‐species variation and species turnover is much more important. To understand how lichen functional traits at the community level respond to environmental factors, it is therefore essential to consider the responses of individual species, and the application of traits‐based approaches to lichen communities needs to account for their considerable intraspecific variability.

Work on the impacts of herbivores on ecosystems has traditionally focused on aboveground effects, but a growing number of ecologists are beginning to consider how herbivores affect belowground organisms and processes. Human activity has caused considerable changes in densities of mammalian herbivores throughout the world, through the introduction of herbivores to new regions, the creation of conditions that promote high herbivore densities, and the reduction of their population sizes, sometimes to the point of extinction. These human influences on high mammal densities can have major effects on the decomposer subsystem. Whether these effects are positive or negative depends on the mechanisms involved: for example, whether the changes are in the quantity or quality of the decomposers' resources or in the pathway of vegetation succession. In turn, these belowground effects may influence aboveground biota by altering the supply of available nutrients from the soil. Changes in large mammal densities through human activity may represent an important, though frequently underappreciated, element of global change.

Pure swards of each of four pasture species (Lolium perenne L., Trifolium repens L., Senecio jacobaea L., and Carduus nutans L.) were established in glasshouse conditions and subjected to one of three treatments: spraying with 2,4‐D/picloram mix; spraying with glyphosate; or unsprayed. After the sprayed swards died, all above‐ground and below‐ground tissue was harvested, air‐dried, and placed in nylon mesh litter‐bags which were positioned in the field. Decomposition, microbial basal respiration, and substrate‐induced respiration (proportionally related to the glucose‐responsive microbial biomass) of this litter was then monitored over 338 days. Both herbicide treatments inhibited decomposition of T. repens and L. perenne shoot tissue and C. nutans root tissue, but stimulated that of C. nutans shoot tissue, indicating that herbicides may influence decomposition of different species in different ways; the possible reasons for this are discussed. However, the rapid decomposition of most of the tissues considered in this study suggest that herbicides are unlikely to exert substantial long‐term effects on plant litter persistence. Microbial basal respiration and substrate‐induced respiration of most of the litter types considered were initially very strongly enhanced by both herbicide treatments; however, this effect was highly transitory for all tissue types except one, and for some of the tissue types a strong inhibition of these microbial variables in the herbicide treatments followed. It therefore appears that microbial build‐up on litter from herbicide‐killed plants (and the subsequent decline) occurs earlier than that from unsprayed plants, probably because herbicide‐induced plant damage increases the availability of readily utilisable microbial substrates. The retardation of leaf litter decomposition in herbicide treatments was often associated with reduced microbial activity and biomass, indicating strong linkages between soil‐associated microflora and decomposition processes. This study also indicates that newly developed approaches for simultaneously assessing decomposition and the microbial biomass of leaf litter have considerable potential for investigating impacts of ecological factors on plant litter‐microbial interactions. Keywords: basal respiration2,4‐Dplant litterdecompositionglyphosateherbicidemicrobial biomass

Inter-year coefficients of variation for soil microbial variables in relation to island size.

Plant productivity is predicted to increase in boreal forests owing to climate change, but this may depend on whether N inputs from biological N-fixation also increases. We evaluated how alteration of climatic factors affects N input from a widespread boreal N-fixer, i.e. cyanobacteria associated with the feather moss Pleurozium schreberi . In each of 10 forest stands in northern Sweden, we established climate-change plots, including a control (ambient climate) plot and three plots experiencing a +2°C temperature increase, an approximately threefold reduction in precipitation frequency, and either 0.07, 0.29 or 1.16 times normal summer precipitation. We monitored N-fixation in these plots five times between 2007 and 2009, and three times in 2010 after climate treatments ended to assess their recovery. Warmer temperatures combined with less frequent precipitation reduced feather moss moisture content and N-fixation rates regardless of total precipitation. After climate treatments ended, recovery of N-fixation rates occurred on the scale of weeks to months, suggesting resilience of N-fixation to changes in climatic conditions. These results suggest that modelling of biological N-inputs in boreal forests should emphasize precipitation frequency and evaporative water loss in conjunction with elevated temperature rather than absolute changes in mean precipitation.