Comparing resource pulses in aquatic and terrestrial ecosystems

Resource pulses affect productivity and dynamics in a diversity of ecosystems, including islands, forests, streams, and lakes. Terrestrial and aquatic systems differ in food web structure and biogeochemistry; thus they may also differ in their responses to resource pulses. However, there has been a limited attempt to compare responses across ecosystem types. Here, we identify similarities and differences in the causes and consequences of resource pulses in terrestrial and aquatic systems. We propose that different patterns of food web and ecosystem structure in terrestrial and aquatic systems lead to different responses to resource pulses. Two predictions emerge from a comparison of resource pulses in the literature: (1) the bottom-up effects of resource pulses should transmit through aquatic food webs faster because of differences in the growth rates, life history, and stoichiometry of organisms in aquatic vs. terrestrial systems, and (2) the impacts of resource pulses should also persist longer in terrestrial systems because of longer generation times, the long-lived nature of many terrestrial resource pulses, and reduced top-down effects of consumers in terrestrial systems compared to aquatic systems. To examine these predictions, we use a case study of a resource pulse that affects both terrestrial and aquatic systems: the synchronous emergence of periodical cicadas (Magicicada spp.) in eastern North American forests. In general, studies that have examined the effects of periodical cicadas on terrestrial and aquatic systems support the prediction that resource pulses transmit more rapidly in aquatic systems; however, support for the prediction that resource pulse effects persist longer in terrestrial systems is equivocal. We conclude that there is a need to elucidate the indirect effects and long-term implications of resource pulses in both terrestrial and aquatic ecosystems.     

Quantity and quality: unifying food web and ecosystem perspectives on the role of resource subsidies in freshwaters

Although the study of resource subsidies has emerged as a key topic in both ecosystem and food web ecology, the dialogue over their role has been limited by separate approaches that emphasize either subsidy quantity or quality. Considering quantity and quality together may provide a simple, but previously unexplored, framework for identifying the mechanisms that govern the importance of subsidies for recipient food webs and ecosystems. Using a literature review of > 90 studies of open-water metabolism in lakes and streams, we show that high-flux, low-quality subsidies can drive freshwater ecosystem dynamics. Because most of these ecosystems are net heterotrophic, allochthonous inputs must subsidize respiration. Second, using a literature review of subsidy quality and use, we demonstrate that animals select for high-quality food resources in proportions greater than would be predicted based on food quantity, and regardless of allochthonous or autochthonous origin. This finding suggests that low-flux, high-quality subsidies may be selected for by animals, and in turn may disproportionately affect food web and ecosystem processes (e.g., animal production, trophic energy or organic matter flow, trophic cascades). We then synthesize and review approaches that evaluate the role of subsidies and explicitly merge ecosystem and food web perspectives by placing food web measurements in the context of ecosystem budgets, by comparing trophic and ecosystem production and fluxes, and by constructing flow food webs. These tools can and should be used to address future questions about subsidies, such as the relative importance of subsidies to different trophic levels and how subsidies may maintain or disrupt ecosystem stability and food web interactions.     

Linking the brown and green: nutrient transformation and fate in the Sarracenia microecosystem

Linkages between detritus-based ("brown") food webs and producer-based ("green") food webs are critical components of ecosystem functionality, but these linkages are hard to study because it is difficult to measure release of nutrients by brown food webs and their subsequent uptake by plants. In a three-month greenhouse experiment, we examined how the detritus-based food web inhabiting rain-filled leaves of the pitcher plant Sarracenia purpurea affects nitrogen transformation and its subsequent uptake by the plant itself. We used isotopically enriched prey (detritus) and soluble inorganic nitrogen, and manipulated food web structure to determine whether the presence of a complete brown web influences uptake efficiency of nitrogen by the plant. Uptake efficiency of soluble inorganic nitrogen was greater than that of nitrogen derived from mineralized prey. Contrary to expectation, there was no effect of the presence in the food web of macroinvertebrates on uptake efficiency of either form of nitrogen. Further, uptake efficiency of prey-derived nitrogen did not differ significantly among S. purpurea and two congeneric species (S. flava and S. alata) that lack associated food webs. Although upper trophic levels of this brown food web actively process detritus, it is the activity of the microbial component of this web that ultimately determines nitrogen availability for S. purpurea.     

Lake to land subsidies: experimental addition of aquatic insects increases terrestrial arthropod densities

Aquatic insects are a common and important subsidy to terrestrial systems, yet little is known about how these inputs affect terrestrial food webs, especially around lakes. Myvatn, a lake in northern Iceland, has extraordinary midge (Chironomidae) emergences that result in large inputs of biomass and nutrients to terrestrial arthropod communities. We simulated this lake-to-land resource pulse by collecting midges from Myvatn and spreading their dried carcasses on 1-m2 plots at a nearby site that receives very little midge deposition. We hypothesized a positive bottom-up response of detritivores that would be transmitted to their predators and would persist into the following year. We sampled the arthropod community once per month for two consecutive summers. Midge addition resulted in significantly different arthropod communities and increased densities of some taxa in both years. Detritivores, specifically Diptera larvae, Collembola, and Acari increased in midge-addition plots, and so did some predators and parasitoids. Arthropod densities were still elevated a year after midge addition, and two years of midge addition further increased the density of higher-order consumers (e.g., Coleoptera and Hymenoptera). Midge addition increased arthropod biomass by 68% after one year and 108% after two years. By manipulating the nutrient pulse delivered by midges we were able to elucidate food web consequences of midge deposition and spatial and temporal dynamics that are difficult to determine based on comparative approaches alone. Resources cross ecosystem boundaries and are assimilated over time because of life-history strategies that connect aquatic and terrestrial food webs and these systems cannot be fully understood in isolation from each other.     

Whole-system nutrient enrichment increases secondary production in a detritus-based ecosystem

Although the effects of nutrient enrichment on consumer-resource dynamics are relatively well studied in ecosystems based on living plants, little is known about the manner in which enrichment influences the dynamics and productivity of consumers and resources in detritus-based ecosystems. Because nutrients can stimulate loss of carbon at the base of detrital food webs, effects on higher consumers may be fundamentally different than what is expected for living-plant-based food webs in which nutrients typically increase basal carbon. We experimentally enriched a detritus-based headwater stream for two years to examine the effects of nutrient-induced changes at the base of the food web on higher metazoan (predominantly invertebrate) consumers. Our paired-catchment design was aimed at quantifying organic matter and invertebrate dynamics in the enriched stream and an adjacent reference stream for two years prior to enrichment and two years during enrichment. Enrichment had a strong negative effect on standing crop of leaf litter, but no apparent effect on that of fine benthic organic matter. Despite large nutrient-induced reductions in the quantity of leaf litter, invertebrate secondary production during the enrichment was the highest ever reported for headwater streams at this Long Term Ecological Research site and was 1.2-3.3 times higher than predicted based on 15 years of data from these streams. Abundance, biomass, and secondary production of invertebrate consumers increased significantly in response to enrichment, and the response was greater among taxa with larval life spans < or = 1 yr than among those with larval life spans >1 yr. Production of invertebrate predators closely tracked the increased production of their prey. The response of invertebrates was largely habitat-specific with little effect of enrichment on food webs inhabiting bedrock outcrops. Our results demonstrate that positive nutrient-induced changes to food quality likely override negative changes to food quantity for consumers during the initial years of enrichment of detritus-based stream ecosystems. Longer-term enrichment may impact consumers through eventual reductions in the quantity of detritus.     

Measuring terrestrial subsidies to aquatic food webs using stable isotopes of hydrogen

Understanding river food webs requires distinguishing energy derived from primary production in the river itself (autochthonous) from that produced externally (allochthonous), yet there are no universally applicable and reliable techniques for doing so. We compared the natural abundance stable isotope ratios of hydrogen (deltaD) of allochthonous and autochthonous energy sources in four different aquatic ecosystems. We found that autochthonous organic matter is uniformly far more depleted in deuterium (lower deltaD values) than allochthonous: an average difference of approximately 100% per hundred. We also found that organisms at higher trophic levels, including both aquatic invertebrates and fish, have deltaD values intermediate between aquatic algae and terrestrial plants. The consistent differences between leaves and algae in deltaD among these four watersheds, along with the intermediate values in higher trophic levels, indicate that natural abundance hydrogen isotope signatures are a powerful tool for partitioning energy flow in aquatic ecosystems.     

The modified niche model: Including detritus in simple structural food web models

Food webs are constructed as structural directed graphs that describe “who eats whom,” but it is common to interpret them as energy flow diagrams where predation represents an energy transfer from the prey to the predator. It is the aim of this work to demonstrate that food webs are incomplete as energy flow diagrams if they ignore passive flows to detritus (dead organic material). While many ecologists do include detritus in conceptual and mathematical models, the detrital omission is still commonly found. Often detritus is either ignored or treated as an unlimited energy source, yet all organisms contribute to the detritus pool, which can be an energy source for other species in the system. This feedback loop is of high importance, since it increases the number of pathways available for energy flows, revealing the significance of indirect effects, and making the functional role of the top predators less clear. In this work we propose the modified niche model by adding a detritus compartment to the niche model. We demonstrate the effect of structural loops that result from feeding on detritus, by comparing empirical data sets to five different assembly models: (1) cascade, (2) constant connectance, (3) niche, (4) modified niche (original in this work), and (5) cyber-ecosystem. Of these models, only the last two explicitly include detritus. We show that when passive flows to detritus are included in the food web structure, the structure becomes more robust to the removal of individual nodes or connections. In addition, we show that food web models that include the detritus feedback loop perform better with respect to several structural network metrics.     

Introduced trout sever trophic connections in watersheds: consequences for a declining amphibian

Trophic linkages between terrestrial and aquatic ecosystems are increasingly recognized as important yet poorly known features of food webs. Here we describe research to understand the dynamics of lake food webs in relation to a native riparian amphibian and its interaction with introduced trout. The mountain yellow-legged frog Rana muscosa is endemic to alpine watersheds of the Sierra Nevada Mountains and the Transverse Ranges of California, but it has declined to a small fraction of its historical distribution and abundance. Although remaining frogs and introduced trout feed in different habitats of alpine lakes, our stable-isotope analyses clearly show that the same resource base of benthic invertebrates sustains their growth. During one period, insect emergence from naturally fishless lakes was nearly 20-fold higher compared to adjacent lakes with trout, showing that fish reduce availability of aquatic prey to amphibious and terrestrial consumers. Although trout cannot prey on adult frogs due to gape limitation, foraging post-metamorphic frogs are 10 times more abundant in the absence of trout, suggesting an important role for competition for prey by trout in highly unproductive alpine watersheds. Most Sierran lakes contain fish, and those that do not are usually small isolated ponds; in our study, these two lake types supported the lowest densities of post-metamorphic frogs, and these frogs were less reliant on local, benthic sources of productivity. Since Rana muscosa was formerly the most abundant vertebrate in the Sierra Nevada, the reduction in energy flow from lake benthos to this consumer due to fish introductions may have had negative consequences for its numerous terrestrial predators, many of which have also declined. We suggest that disruptions of trophic connections between aquatic and terrestrial food webs are an important but poorly understood consequence of fish introduction to many thousands of montane lakes and streams worldwide and may contribute to declines of native consumers in riparian habitats.     

The effects of seston food quality on planktonic food web patterns

In planktonic food webs, the conversion rate of plant material to herbivore biomass is determined by a variety of factors such as seston biochemical/elemental composition, phytoplankton cell morphology, and colony architecture. Despite the overwhelming heterogeneity characterizing the plant–animal interface, plankton population models usually misrepresent the food quality constraints imposed on zooplankton growth. In this study, we reformulate the zooplankton grazing term to include seston food quality effects on zooplankton assimilation efficiency and examine its ramifications on system stability. Using different phytoplankton parameterizations with regards to growth strategies, light requirements, sinking rates, and food quality, we examined the dynamics induced in planktonic systems under varying zooplankton mortality/fish predation, light conditions, nutrient availability, and detritus food quality levels. In general, our analysis suggests that high food quality tends to stabilize the planktonic systems, whereas unforced oscillations (limit cycles) emerge with lower seston food quality. For a given phytoplankton specification and resource availability, the amplitude of the plankton oscillations is primarily modulated from zooplankton mortality and secondarily from the nutritional quality of the alternative food source (i.e., detritus). When the phytoplankton community is parameterized as a cyanobacterium-like species, conditions of high nutrient availability combined with high zooplankton mortality led to phytoplankton biomass accumulation, whereas a diatom-like parameterization resulted in relatively low phytoplankton to zooplankton biomass ratios highlighting the notion that high phytoplankton food quality allows the zooplankton community to sustain relatively high biomass and to suppress phytoplankton biomass to low levels. During nutrient and light enrichment conditions, both phytoplankton and detritus food quality determine the extent of the limit cycle region, whereas high algal food quality increases system resilience by shifting the oscillatory region towards lower light attenuation levels. Detritus food quality seems to regulate the amplitude of the dynamic oscillations following enrichment, when algal food quality is low. These results highlight the profitability of the alternative food sources for the grazer as an important predictor for the dynamic behavior of primary producer–grazer interactions in nature.     

Trophic levels and trophic tangles: the prevalence of omnivory in real food webs

The concept of trophic levels is one of the oldest in ecology and informs our understanding of energy flow and top-down control within food webs, but it has been criticized for ignoring omnivory. We tested whether trophic levels were apparent in 58 real food webs in four habitat types by examining patterns of trophic position. A large proportion of taxa (64.4%) occupied integer trophic positions, suggesting that discrete trophic levels do exist. Importantly however, the majority of those trophic positions were aggregated around integer values of 0 and 1, representing plants and herbivores. For the majority of the real food webs considered here, secondary consumers were no more likely to occupy an integer trophic position than in randomized food webs. This means that, above the herbivore trophic level, food webs are better characterized as a tangled web of omnivores. Omnivory was most common in marine systems, rarest in streams, and intermediate in lakes and terrestrial food webs. Trophic-level-based concepts such as trophic cascades may apply to systems with short food chains, but they become less valid as food chains lengthen.     

What is hidden behind the concept of ecosystem efficiency in energy transformation?

The number of energy transformation levels in trophic webs is usually below five, but can be extended up to ten when parasites and hyper-parasites are included. Research on the structure and function of food webs is relevant to the complexity–stability–productivity debate. The aim of this theoretical analysis is to link energetic and connectional aspects of ecosystems with information theory. Updating an energetic model reported by Ricklefs [Ecologia, Zanichelli Editore S.p.A., Bologna, Italy, 1993, p. 896], our approach is integrated with a static analysis of food webs. The length of food webs is theoretically associated with the average ecological efficiency which can be empirically correlated with the effective connectance between species. Furthermore, the advantage of greater complexity when applied to a signalling network is qualitatively addressed.The overall efficiency of energy transformation into biomass throughout a trophic web, in an ecosystem with a given number of species, is the resultant of the various ecological efficiencies, η, at the transitions between the trophic levels. However, we propose that an increment in effective connectance and interspecies connectivity based on a superimposed signalling web may increase the η values, despite the fact that signalling per se has an energetic cost. According to this hypothesis, ecosystem stability would not be necessarily reduced by increasing the number of trophic levels, N, whenever stability in terms of persistence is improved by a cost-efficient regulatory network.     

Grazer traits, competition, and carbon sources to a headwater-stream food web

We investigated the effect of grazing by a dominant invertebrate grazer (the caddisfly Glossosoma penitum) on the energy sources used by other consumers in a headwater-stream food web. Stable isotope studies in small, forested streams in northern California have shown that G. penitum larvae derive most of their carbon from algae, despite low algal standing crops. We hypothesized that the caddisfly competes with other primary consumers (including mayflies) for algal food and increases their reliance on terrestrial detritus. Because Glossosoma are abundant and defended from predators by stone cases, their consumption of algal energy may reduce its transfer up the food chain. We removed Glossosoma (natural densities >1000 caddisflies/m2) from five approximately 4 m2) stream sections during the summer of 2000 and measured responses of algae, invertebrate primary consumers, and invertebrate predators. The treatment reduced Glossosoma biomass by 80-90%. We observed a doubling in chlorophyll a per area in sections with reduced Glossosoma abundance and aggregative increases in the biomass of undefended primary consumers. Heptageniid mayfly larvae consumed more algae (as measured by stable carbon isotope ratios and gut content analysis) in caddisfly removal plots at the end of the 60-day experiment, although not after one month. We did not see isotopic evidence of increased algal carbon in invertebrate predators, however. Patterns of caddisfly and mayfly diets in the surrounding watershed suggested that mayfly diets are variable and include algae and detrital carbon in variable proportions, but scraping caddisflies consume primarily algae. Caddisfly and mayfly diets are more similar in larger, more productive streams where the mayflies assimilate more algae. Isotopic analysis, in combination with measurements of macroinvertebrate abundance and biomass in unmanipulated plots, suggested that a substantial portion of the invertebrate community (>50% of biomass) was supported, at least partially, by local algal carbon during midsummer. These data suggest that algae may be more important to community dynamics in headwater streams than their relatively low productivity would suggest. Through their high densities and relative invulnerability to predation, armored grazers may also affect community structure and flow of algal and detrital carbon in headwater streams.     

A simulation analysis of continental shelf food webs

Energy flow through continental shelf food webs was examined using a simulation model. The model structure expands the two traditional marine food chains of phytoplankton-zooplankton-pelagic fish and benthos-demersal fish into a complex web which includes detritus, dissolved organic matter (DOM), bacteria, protozoa, and mucus net feeders. Simulation of energy flux for different shelf systems using the expanded web revealed that heterotrophic microorganisms and their predators account for a significant component of the energy flux in the continental shelf ecosystem. Contrary to previous models, where all phytoplankton were considered to be grazed by zooplankton, our simulation results indicate that only slightly more than 50% of the annual net primary production is grazed. A substantial quantity of the phytoplankton production directly becomes detritus. Bacteria mineralize detritus and DOM produced by phytoplankton and other components of the food web, converting these to biomass with high efficiency. Consequently, the model predicts that planktonic bacterial production is equivalent to zooplankton production. Exclusion of the bacteria requires the assumption that all DOM is either exported from the system or consumed by another component of the food web. Neither of these assumptions can be supported by present knowledge of the dynamics of DOM in the sea. Model simulations were also employed to test the hypothesis that production exceeds consumption on continental shelves, resulting in exports of 50% of the annual primary production. Simulations of shelves with high rates of primary production resulted in a particulate export of 27% and realistic estimates of secondary production. Results of other simulations suggest that shelves with lower primary production cannot export production and still maintain the macrobenthos and their predators. General properties about continental shelves can also be inferred from the model. From simulations of shelves of differing primary production, nanoplankton are predicted to account for a greater proportion of the primary production in nutrient limited systems. Benthic production appears to be related to both the quantity of primary production and the sinking rates of the phytoplankton. The model indicates that zooplankton fecal inputs to the shelf benthos are only a small portion of the total detrital flux, leading to the prediction that fecal pellets are of little significance in determining benthic production. Finally, the model generates production efficiencies that are highly variable depending on the type of system and kind of populations involved. We argue that the assumed ecological efficiency of 10% should be abandoned for continental shelves and other ecosystems.     

Predator diversity and identity drive interaction strength and trophic cascades in a food web

Declining predator diversity may drastically affect the biomass and productivity of herbivores and plants. Understanding how changes in predator diversity can propagate through food webs to alter ecosystem function is one of the most challenging ecological research topics today. We studied the effects of predator removal in a simple natural food web in the Sierra Nevada mountains of California (USA). By excluding the predators of the third trophic level of a food web in a full-factorial design, we monitored cascading effects of varying predator diversity and composition on the herbivorous beetle Chrysomela aeneicollis and the willow Salix orestera, which compose the first and second trophic levels of the food web. Decreasing predator diversity increased herbivore biomass and survivorship, and consequently increased the amount of plant biomass consumed via a trophic cascade. Despite this simple linear mean effect of diversity on the strength of the trophic cascade, we found additivity, compensation, and interference in the effects of multiple predators on herbivores and plants. Herbivore survivorship and predator-prey interaction strengths varied with predator diversity, predator identity, and the identity of coexisting predators. Additive effects of predators on herbivores and plants may have been driven by temporal niche separation, whereas compensatory effects and interference occurred among predators with a similar phenology. Together, these results suggest that while the general trends of diversity effects may appear linear and additive, other information about species identity was required to predict the effects of removing individual predators. In a community that is not temporally well-mixed, predator traits such as phenology may help predict impacts of species loss on other species. Information about predator natural history and food web structure may help explain variation in predator diversity effects on trophic cascades and ecosystem function.     

Divergent composition but similar function of soil food webs of individual plants: plant species and community effects

Soils are extremely rich in biodiversity, and soil organisms play pivotal roles in supporting terrestrial life, but the role that individual plants and plant communities play in influencing the diversity and functioning of soil food webs remains highly debated. Plants, as primary producers and providers of resources to the soil food web, are of vital importance for the composition, structure, and functioning of soil communities. However, whether natural soil food webs that are completely open to immigration and emigration differ underneath individual plants remains unknown. In a biodiversity restoration experiment we first compared the soil nematode communities of 228 individual plants belonging to eight herbaceous species. We included grass, leguminous, and non-leguminous species. Each individual plant grew intermingled with other species, but all plant species had a different nematode community. Moreover, nematode communities were more similar when plant individuals were growing in the same as compared to different plant communities, and these effects were most apparent for the groups of bacterivorous, carnivorous, and omnivorous nematodes. Subsequently, we analyzed the composition, structure, and functioning of the complete soil food webs of 58 individual plants, belonging to two of the plant species, Lotus corniculatus (Fabaceae) and Plantago lanceolata (Plantaginaceae). We isolated and identified more than 150 taxa/groups of soil organisms. The soil community composition and structure of the entire food webs were influenced both by the species identity of the plant individual and the surrounding plant community. Unexpectedly, plant identity had the strongest effects on decomposing soil organisms, widely believed to be generalist feeders. In contrast, quantitative food web modeling showed that the composition of the plant community influenced nitrogen mineralization under individual plants, but that plant species identity did not affect nitrogen or carbon mineralization or food web stability. Hence, the composition and structure of entire soil food webs vary at the scale of individual plants and are strongly influenced by the species identity of the plant. However, the ecosystem functions these food webs provide are determined by the identity of the entire plant community.     

Avian predators are less abundant during periodical cicada emergences, but why?

Despite a substantial resource pulse, numerous avian insectivores known to depredate periodical cicadas (Magicicada spp.) are detected less commonly during emergence years than in either the previous or following years. We used data on periodical cicada calls collected by volunteers conducting North American Breeding Bird Surveys within the range of cicada Brood X to test three hypotheses for this observation: lower detection rates could be caused by bird calls being obscured by cicada calls ("detectability" hypothesis), by birds avoiding areas with cicadas ("repel" hypothesis), or because bird abundances are generally lower during emergence years for some reason unrelated to the current emergence event ("true decline" hypothesis). We tested these hypotheses by comparing bird detections at stations coincident with calling cicadas vs. those without calling cicadas in the year prior to and during cicada emergences. At four distinct levels (stop, route, range, and season), parallel declines of birds in groups exposed and not exposed to cicada calls supported the true decline hypothesis. We discuss several potential mechanisms for this pattern, including the possibility that it is a consequence of the ecological and evolutionary interactions between predators of this extraordinary group of insects.     

Hydrology and grazing jointly control a large-river food web

Inputs of fresh water and grazing both can control aquatic food webs, but little is known about the relative strengths of and interactions between these controls. We use long-term data on the food web of the freshwater Hudson River estuary to investigate the importance of, and interactions between, inputs of fresh water and grazing by the invasive zebra mussel (Dreissena polymorpha). Both freshwater inputs and zebra mussel grazing have strong, pervasive effects on the Hudson River food web. High flow tended to reduce population size in most parts of the food web. High grazing also reduced populations in the planktonic food web, but increased populations in the littoral food web, probably as a result of increases in water clarity. The influences of flow and zebra mussel grazing were roughly equal (i.e., within a factor of 2) for many variables over the period of our study. Zebra mussel grazing made phytoplankton less sensitive to freshwater inputs, but water clarity and the littoral food web more sensitive to freshwater inputs, showing that interactions between these two controlling factors can be strong and varied.     

Nutrient enrichment reduces constraints on material flows in a detritus-based food web

Most aquatic and terrestrial ecosystems are experiencing increased nutrient availability, which is affecting their structure and function. By altering community composition and productivity of consumers, enrichment can indirectly cause changes in the pathways and magnitude of material flows in food webs. These changes, in turn, have major consequences for material storage and cycling in the ecosystem. Understanding mechanisms and predicting consequences of nutrient-induced changes in material flows requires a quantitative food web approach that combines information on consumer energetics and consumer-resource stoichiometry. We examined effects of a whole-system experimental nutrient enrichment on the trophic basis of production and the magnitude and pathways of carbon (C), nitrogen (N), and phosphorus (P) flows in a detritus-based stream food web. We compared the response of the treated stream to an adjacent reference stream throughout the study. Dietary composition and elemental flows varied considerably among invertebrate functional feeding groups. During nutrient enrichment, increased flows of leaf litter and amorphous detritus to shredders and gatherers accounted for most of the altered flows of C from basal resources to consumers. Nutrient enrichment had little effect on patterns of material flows but had large positive effects on the magnitude of C, N, and P flows to consumers (mean increase of 97% for all elements). Nutrient-specific food webs revealed similar flows of N and P to multiple functional groups despite an order of magnitude difference among groups in consumption of C. Secondary production was more strongly related to consumption of nutrients than C, and increased material flows were positively related to the degree of consumer-resource C:P and C:N imbalances. Nutrient enrichment resulted in an increased proportion of detrital C inputs consumed by primary consumers (from -15% to 35%) and a decreased proportion of invertebrate prey consumed by predators (from -80% to 55%). Our results demonstrate that nutrient enrichment of detritus-based systems may reduce stoichiometric constraints on material flows, increase the contribution of consumers to C, N, and P cycling, alter the proportion of C inputs metabolized by consumers, and potentially lead to reduced ecosystem-level storage of C.     

The probabilistic niche model reveals substantial variation in the niche structure of empirical food webs

The structure of food webs, complex networks of interspecies feeding interactions, plays a crucial role in ecosystem resilience and function, and understanding food web structure remains a central problem in ecology. Previous studies have shown that key features of empirical food webs can be reproduced by low-dimensional "niche" models. Here we examine the form and variability of food web niche structure by fitting a probabilistic niche model to 37 empirical food webs, a much larger number of food webs than used in previous studies. The model relaxes previous assumptions about parameter distributions and hierarchy and returns parameter estimates for each species in each web. The model significantly outperforms previous niche model variants and also performs well for several webs where a body-size-based niche model performs poorly, implying that traits other than body size are important in structuring these webs' niche space. Parameter estimates frequently violate previous models' assumptions: in 19 of 37 webs, parameter values are not significantly hierarchical, 32 of 37 webs have nonuniform niche value distributions, and 15 of 37 webs lack a correlation between niche width and niche position. Extending the model to a two-dimensional niche space yields networks with a mixture of one- and two-dimensional niches and provides a significantly better fit for webs with a large number of species and links. These results confirm that food webs are strongly niche-structured but reveal substantial variation in the form of the niche structuring, a result with fundamental implications for ecosystem resilience and function.