Leaf breakdown patterns in a NW Italian stream: Effect of leaf type, environmental conditions and patch size

We studied the decomposition process and macroinvertebrate colonisation of leaf packs to determine to what extent leaf consumption and invertebrate abundance depend on the pollution level, season, leaf type and patch size. We exposed 400 leaf packs made of two leaf types, alder and chestnut, at two sites of the Erro River (NW Italy) with different environmental alteration levels. Leaf packs were set out as three patch sizes (alone, or in groups of 6 or 12). A first experiment was carried out in winter and a second in summer. Leaf packs were retrieved after 15, 30, 45 and 60 days of submersion to determine the leaf mass loss and to quantify the associated macroinvertebrates. Natural riverbed invertebrates were collected in the same areas. Patch size, season, leaf type and pollution level significantly affected mass loss. The breakdown process was faster for alder leaves, during summer, at the unpolluted site, and in smaller patches. Leaf type and patch size did not affect macroinvertebrate density and richness, but the highest taxon richness was found in winter and at the unpolluted site. There were more shredders and predators than in the natural riverbed. Our study supports two recent ideas regarding leaf processing in streams: that patch size influences the leaf breakdown rate and that the breakdown rate can be used to evaluate water quality and environmental health.     

Seasonal and interannual dynamics of diatom assemblages in Sacrower See (NE Germany): a sediment trap study

Diatom assemblages from sediment trap samples collected during ten intervals between October 2003 and October 2005 in Sacrower See (NE Germany) were related to limnological and meteorological data. Sacrower See is a dimictic, 38 m deep, hypertrophic lowland lake (29.5 m a.s.l.). We identified distinct seasonal and interannual changes of diatom assemblages for the studied period. Diatoms showed a typical seasonal succession for temperate, dimictic, and eutrophic lakes. Stephanodiscus parvus, S. hantzschii, S. neoastraea, and S. alpinus had high accumulation rates during winter and spring, whereas species of the genera Stephanodiscus, Fragilaria, and Nitzschia were the predominant diatoms during summer and autumn. In a Canonical Correspondence Analysis, precipitation, air and water temperatures, epilimnetic calcium, pH, and total phosphorus concentrations together explained 70% of the variance of the diatom data. Interannual variability in the diatom assemblages during the two sampled years mainly seems to reflect changes in the total phosphorus concentration and temperature and secondarily the onset of the growing season and of stratification. Handling editor: J. Saros     

Ecosystem respiration and organic carbon processing in a large,tidally influenced river: the Hudson River

We estimated whole-ecosystem rates of respiration over a 40-km stretch of the tidally influenced freshwater Hudson River every 2 to 3 weeks from May through November. We measured in situ concentrations of oxygen over depth at dusk and dawn at 10 stations spaced over this interval. The use of multiple stations allowed for the consideration of the influence of tidal advection of water masses. Respiration was estimated from the decrease in oxygen overnight with a correction for diffusive exchange of oxygen with the atmosphere. We estimated this flux of oxygen to or from the atmosphere using the measured oxygen gradient and a transfer velocity model which is a function of wind velocity.Integration of the data for the period of May through November yields an estimate of whole-ecosystem respiration of 591 g C m^–2 (S.E. = 66). That the standard error of this estimate is relatively low (11% of the estimate) indicates that the use of multiple stations adequately deals with error introduced through the advection of water between stations. The logarithm of average daily respiration rate was correlated with average daily temperature (p = 0.007;r² = 0.62). We used this temperature-respiration relationship to derive an estimate of the annual respiration rate of 755 g C m^–2 yr^–1 (S.E. = 72). This estimate is moderately sensitive to the estimated flux of oxygen between the atmosphere and water; using the lower and upper 95% confidence limits of our model relating the transfer velocity of oxygen to wind speed gives a range of annual respiration estimates from 665 g C m^–2 yr^–1 to 984 g C m^–2 yr^–1.The river is strongly heterotrophic, with most respiration driven by allochthonous inputs of organic matter from terrestrial ecosystems. The majority of the allochthonous inputs to the river (over 60%) are apparently metabolized within the river. Any change in allochthonous inputs due to changes in land use or climate patterns would be expected to alter the oxygen dynamics and energy flow within this tidally influenced river.     

Lotic meiofaunal community dynamics: colonisation, resilience and persistence in a spatially and temporally heterogeneous environment

1.  The spatial and temporal dynamics of lotic meiofaunal communities were examined with a focus on colonisation, maintenance of populations in lotic environments and persistence of meiofaunal communities.
2.  Lotic meiofaunal colonisation of new habitats may take place via a number of mechanisms and is rapid at both the patch scale (within hours–days) and the drainage basin scale (within 20 years). Successional patterns in lotic meiofaunal communities are evident although data are extremely limited.
3.  Lotic meiofaunal communities appear to be resilient to high flow disturbances. Resilience is moderated by the availability of in-stream refugia and habitat hydrology. Lotic meiofauna may also adopt a 'refuge as habitat' approach to such disturbances.
4.  Lotic cyclopoid copepods possess a common suite of life history characteristics that confers resilience to disturbances. Compared to pelagic planktonic species they have short generation times, many descendants per reproductive cycle and long lifespans. Females in 'source' populations are likely to survive disturbances and so could continuously reproduce over a long period of time producing large numbers of offspring which develop rapidly and recolonise 'sink' areas of the stream.
5.  Persistence of lotic meiofaunal communities is highly variable. Meiofaunal persistence does not increase with increasing proportions of in-stream flow refugia.     

Regime shifts in marine communities: a complex systems perspective on food web dynamics

Species composition and habitats are changing at unprecedented rates in the world''s oceans, potentially causing entire food webs to shift to structurally and functionally different regimes. Despite the severity of these regime shifts, elucidating the precise nature of their underlying processes has remained difficult. We address this challenge with a new analytic approach to detect and assess the relative strength of different driving processes in food webs. Our study draws on complexity theory, and integrates the network-centric exponential random graph modelling (ERGM) framework developed within the social sciences with community ecology. In contrast to previous research, this approach makes clear assumptions of direction of causality and accommodates a dynamic perspective on the emergence of food webs. We apply our approach to analysing food webs of the Baltic Sea before and after a previously reported regime shift. Our results show that the dominant food web processes have remained largely the same, although we detect changes in their magnitudes. The results indicate that the reported regime shift may not be a system-wide shift, but instead involve a limited number of species. Our study emphasizes the importance of community-wide analysis on marine regime shifts and introduces a novel approach to examine food webs.     

High assemblage persistence in heterogeneous habitats: an experimental test with stream benthic algae

1. The persistence of biological assemblages is positively affected by spatial heterogeneity. This influence may be indirect, through increased species richness. Another possibility is the increased availability of refuges from disturbances, which would prevent local loss. 2. We conducted a field experiment to test the hypothesis that greater roughness (a form of spatial heterogeneity) on the surface of substrata allows higher persistence of assemblages of stream benthic algae and that this relationship does not depend on species richness. Samples were taken on six occasions from smooth and rough artificial substrata used for algal colonisation. We calculated the persistence of assemblages using two analytical approaches: the mean distance to group centroid and the sum of the Euclidean distances between consecutive sampling occasions, both in a multivariate space. We also subsampled the data to take into account differences in species richness between treatments and thus to evaluate the effect of species richness on persistence. 3. Assemblages on rough substrata were more persistent than assemblages on smooth substrata. The effects detected were not due to the greater species richness on rough substrata, since a higher persistence of the assemblages on rough substrata remained after the subsampling procedures. 4. Our results indicate a strong positive relationship between substratum roughness and the persistence of stream benthic algal assemblages. We suggest that this is due to the presence of physical refuges in heterogeneous habitats.     

Coupled ecological-social dynamics in a forested landscape: spatial interactions and information flow

We develop an agent-based model for forest harvesting to study how interactions between neighboring land parcels and the degree of information flow among landowners influence harvesting patterns. We assume a forest is composed of a number of land parcels that are individually managed. Each parcel is either mature forested, just-harvested, or immature forested. The state transition of each parcel is described by a Markov chain that incorporates the successional dynamics of the forest ecosystem and landowners' decisions about harvesting. Landowners decide to cut trees based on the expected discounted utility of forested vs. harvested land. One landowner's decision to cut trees is assumed to cause the degradation of ecosystem services on the downstream forested parcels. We investigated two different scenarios: in a strongly-connected society, landowners are familiar with each other and have full information regarding the behavior of other landowners. In a weakly-connected society, landowners do not communicate and therefore need to make subjective predictions about the behavior of others without adequate information. Regardless of the type of society, we observed that the spatial interaction between management units caused a chain reaction of tree harvesting in the neighborhood even when healthy forested land provided greater utility than harvested land. The harvest rate was higher in a weakly-connected society than that in a strongly-connected society. If landowners employed a long-term perspective, the harvest rate declined, and a more robust forested landscape emerged. Our results highlight the importance of institutional arrangements that encourage a long-term perspective and increased information flow among landowners in order to achieve successful forest management.     

Habitat structural complexity mediates food web dynamics in a freshwater macrophyte community

A considerable amount of research has investigated the influence of habitat structure on predator success, yet few studies have explored the implications for community structure and food-web dynamics. The relative importance of macrophyte structure and fish predation on the composition of the macroinvertebrate and periphyton communities in a lowland river was investigated using a multifactorial caging experiment. We hypothesised that: (1) fish predators are less effective in a more structurally complex macrophyte analogue; (2) strong direct and indirect effects of fish predators (e.g. trophic cascades) are less likely to occur in a structurally complex habitat; and (3) the strength of these patterns is influenced by the composition of the prevailing community assemblage. We measured the abundance and composition of the macroinvertebrate and periphyton communities associated with three different-shaped macrophyte analogues, under different fish predator treatments and at different times. Macrophyte analogue architecture had strong, consistent effects on both the macroinvertebrate and periphyton communities; both were most abundant and diverse on the most structurally complex plant analogue. In contrast, the fish predators affected only a subset of the macroinvertebrate community and there was a suggestion of minor indirect effects on periphyton community composition. Contrary to expectations, the fish predators had their strongest effects in the most structurally complex macrophyte analogue. We conclude that in this system, macrophyte shape strongly regulates the associated freshwater assemblage, resulting in a diverse community structure less likely to exhibit strong effects of fish predation.     

Diversification in temporally heterogeneous environments: effect of the grain in experimental bacterial populations

Although theory established the necessary conditions for diversification in temporally heterogeneous environments, empirical evidence remains controversial. One possible explanation is the difficulty of designing experiments including the relevant range of temporal grains and the appropriate environmental trade-offs. Here, we experimentally explore the impact of the grain on the diversification of the bacterium Pseudomonas fluorescens SBW25 in a temporally fluctuating environment by including 20 different pairs of environments and four temporal grains. In general, higher levels of diversity were observed at intermediate temporal grains. This resulted in part from the enhanced capacity of disruptive selection to generate negative genotypic correlations in performance at intermediate grains. However, the evolution of reciprocal specialization was an uncommon outcome. Although the temporal heterogeneity is in theory less powerful than the spatial heterogeneity to generate and maintain the diversity, our results show that diversification under temporal heterogeneity is possible provided appropriate environmental grains.     

The biogeochemistry of pristine,headwater Precambrian shield watersheds: an analysis of material transport within a heterogeneous landscape

The hydrology and elemental transport within five low order Precambrian shield catchments was investigated during 1988–90. Catchments were subdivided and instrumented to examine the vertical and horizontal fluxes of elements within and between two distinct landscape types: open, lichen-covered bedrock outcrops and patches of conifer forest. The dominant hydrologic pathways were Horton overland flow in the lichen-bedrock areas and shallow subsurface flow through organic rich LFH (forest floor) and Ah soil horizons in the forested areas. Annual runoff coefficients ranged from 0.3 to 0.7. Runoff chemistry was acidic (pH 4.01–4.72), with organic anion equivalents (RCOO^-), comprising 60 and 69% of the anion charge total for bedrock and forest runoff, respectively. Forested plots exported more H⁺ (2.6x), DOC (1.4x), Al (1.6x) and Fe (1.8x) and less N (0.40x), P (0.13x), particulate C (0.08x), Ca²⁺ (0.38x), Mg²⁺ (0.83x), Na⁺ (0.85x) and K⁺ (0.32x) per unit area than the bedrock-lichen plots. The catchments exhibited a net export of H⁺ (34), Mg²⁺ (24), Na⁺ (20), K⁺ (4) (units in eq ha^-1 yr^-1) and C (16), Si (5.6), Al (1.6) and Fe (0.47) (units kg ha^-1 yr^-1). The catchments retained N (5.66), P (0.08), Mn (0.03) (units kg ha^-1 yr^-1), and Ca²⁺ (37), and Cl^- (3) (units eq ha^-1 yr^-1). The strong retention of Ca²⁺ within the treed soil islands resulted in extremely low export rates of base cations (-15 to 38 eq ha^-1 yr^-1). The spatial distribution and hydrologic and biogeochemical linkages associated with each landscape unit interact to control element transport within the study catchments.     

Variation in active and passive resource inputs to experimental pools: mechanisms and possible consequences for food webs

1. Cross‐ecosystem movements of resources, including detritus, nutrients and living prey, can strongly influence food web dynamics in recipient habitats. Variation in resource inputs is thought to be driven by factors external to the recipient habitat (e.g. donor habitat productivity and boundary conditions). However, inputs of or by ‘active’ living resources may be strongly influenced by recipient habitat quality when organisms exhibit behavioural habitat selection when crossing ecosystem boundaries. 2. To examine whether behavioural responses to recipient habitat quality alter the relative inputs of ‘active’ living and ‘passive’ detrital resources to recipient food webs, we manipulated the presence of caged predatory fish and measured biomass, energy and organic content of inputs to outdoor experimental pools of adult aquatic insects, frog eggs, terrestrial plant matter and terrestrial arthropods. 3. Caged fish reduced the biomass, energy and organic matter donated to pools by tree frog eggs by ～70%, but did not alter insect colonisation or passive allochthonous inputs of terrestrial arthropods and plant material. Terrestrial plant matter and adult aquatic insects provided the most energy and organic matter inputs to the pools (40–50%), while terrestrial arthropods provided the least (7%). Inputs of frog egg were relatively small but varied considerably among pools and over time (3%, range = 0–20%). Absolute and proportional amounts varied by input type. 4. Aquatic predators can strongly affect the magnitude of active, but not passive, inputs and that the effect of recipient habitat quality on active inputs is variable. Furthermore, some active inputs (i.e. aquatic insect colonists) can provide similar amounts of energy and organic matter as passive inputs of terrestrial plant matter, which are well known to be important. Because inputs differ in quality and the trophic level they subsidise, proportional changes in input type could have strong effects on recipient food webs. 5. Cross‐ecosystem resource inputs have previously been characterised as donor‐controlled. However, control by the recipient food web could lead to greater feedback between resource flow and consumer dynamics than has been appreciated so far.     

Seasonal variation in the trophic structure of a spatial prey subsidy linking aquatic and terrestrial food webs: adult aquatic insects

Research over the past decade has established spatial resource subsidies as important determinants of food web dynamics. However, most empirical studies have considered the role of subsidies only in terms of magnitude, ignoring an important property of subsidies that may affect their impact in recipient food webs: the trophic structure of the subsidy relative to in situ resources. This may be especially important when subsidies are composed of organisms, as opposed to nutrient subsidies, because the trophic position of subsidy organisms may differ from in situ prey. I explored the relative magnitude and trophic structure of a cross-habitat prey subsidy, adult aquatic insects, in terrestrial habitats along three streams in the south–central United States. Overall, adult aquatic insects contributed more than one-third of potential insect prey abundance and biomass to the terrestrial habitat. This contribution peaked along a permanent spring stream, reaching as high as 94% of abundance and 86% of biomass in winter. Trophic structure of adult aquatic and terrestrial insects differed. Nearly all adult aquatic insects were non-consumers as adults, whereas all but one taxon of terrestrial insects were consumers. Such a difference created a strong relationship between the relative contribution of the prey subsidy and the trophic structure of the prey assemblage: as the proportion of adult aquatic insects increased, the proportion of consumers in the prey assemblage declined. Specific effects varied seasonally and with distance from the stream as the taxonomic composition of the subsidy changed, but general patterns were consistent. These findings show that adult aquatic insect subsidies to riparian food webs not only elevate prey availability, but also alter the trophic structure of the entire winged insect prey assemblage.     

Breeding dispersal in a heterogeneous landscape: the influence of habitat and nesting success in greater snow geese

Despite numerous studies on breeding dispersal, it is still unclear how habitat heterogeneity and previous nesting success interact to determine nest-site fidelity at various spatial scales. In this context, we investigated factors affecting breeding dispersal in greater snow geese (Anser caerulescens atlanticus), an Arctic breeding species nesting in two contrasting habitats (wetlands and mesic tundra) with variable pattern of snowmelt at the time of settlement in spring. From 1994 to 2005, we monitored the nesting success and breeding dispersal of individually marked females. We found that snow geese showed a moderate amount of nest-site fidelity and considerable individual variability in dispersal distance over consecutive nesting attempts. This variability can be partly accounted for by the annual timing of snowmelt. Despite this environmental constraint, habitat differences at the colony level consistently affected nesting success and settlement patterns. Females nesting in wetlands had higher nesting success than those nesting in mesic tundra. Moreover, geese responded adaptively to spatial heterogeneity by showing fidelity to their nesting habitat, independently of snowmelt pattern. From year to year, geese were more likely to move from mesic to high-quality wetland habitat, regardless of previous nesting success and without cost on their subsequent nesting performance. The unpredictability of snowmelt and the low cost of changing site apparently favour breeding-site dispersal although habitat quality promotes fidelity at the scale of habitat patches.     

Exploitation ecosystems in heterogeneous habitat complexes II: Impact of small-scale heterogeneity on predator-prey dynamics

Summary The model of exploitation ecosystems was re-analysed, assuming that habitat patches are so small that they form only parts of the home range of an individual predator. For habitat complexes where productive patches abound, the results suggested that predation will strongly spill over from productive patches, which set the tune for population dynamics within the whole landscape, to barren ones. This result conforms to the one obtained by T. Oksanen by assuming despotic habitat choice and essentially larger patch sizes. For habitat complexes heavily dominated by the barren habitat, spillover predation was predicted to be weak, as was the case in her large patch model. Unlike in her analysis, however, predation pressure was substantially reduced also within the productive habitat. In habitat complexes where patches are so small that they are exploited in a fine-grained manner, predation pressure was always found to be more intense in the barren habitat, contrary to the predictions of the original model of exploitation ecosystems. This analysis thus suggests that their model is applicable mainly on the landscape level. On the level of individual habitats, the applicability of their results depends on the habitat configuration (at its best for the prevailing habitat of the landscape and for moderate-sized patches of an essentially more productive habitat) and generally decreases with decreasing patch sizes.     

A mathematical model for the dynamics of malaria in mosquitoes feeding on a heterogeneous host population

We describe and develop a difference equation model for the dynamics of malaria in a mosquito population feeding on, infecting and getting infected from a heterogeneous population of hosts. Using the force of infection from different classes of humans to mosquitoes as parameters, we evaluate a number of entomological parameters, indicating malaria transmission levels, which can be compared to field data. By assigning different types of vector control interventions to different classes of humans and by evaluating the corresponding levels of malaria transmission, we can compare the effectiveness of these interventions. We show a numerical example of the effects of increasing coverage of insecticide-treated bed nets in a human population where the predominant malaria vector is Anopheles gambiae.     

Food web structure and climate effects on the dynamics of small mammals and owls in semi-arid Chile

Population dynamics of small mammals and predators in semi-arid Chile is positively correlated with rainfall associated with incursions of El Niño (El Niño Southern Oscillation: ENSO). However, the causal relationships between small mammal fluctuations, predator oscillations, and climatic disturbances are poorly understood. Here, we report time series models for three species of small mammal prey and two species of owl predators. The large differences in population fluctuations between the three small mammal species are related to differences in their respective feedback structures. The analyses reveal that per capita growth rate of the leaf-eared mouse is a decreasing function of log density and of log barn owl abundance together with a positive rainfall effect. In turn, per capita population growth rate ( R -function) of the barn owl is a negative function of log barn owl abundance and a positive function of leaf-eared mouse abundance, suggesting a predator–prey interaction. The dramatic population fluctuations exhibited by leaf-eared mouse ( Phyllotis darwini ) are caused by climate effects coupled with a complex food web architecture.     

Terrestrial detritus supports the food webs in lowland intermittent streams of south‐eastern Australia: a stable isotope study

• 1 Large amounts of terrestrial detritus enter many low‐order forested streams, and this organic material is often the major basal resource in the metazoan food webs of such systems. However, despite their apparently low biomass, algae are the dominant food of organisms in a number of aquatic communities which conventionally would have been presumed to be dependent on allochthonous detritus, particularly those in the tropics and also in lowland intermittent streams in arid Australia.
• 2 The dual stable isotope signatures (δ¹³C and δ¹⁵N) of potential primary food sources were compared with the isotopic signatures of common aquatic animals in lowland intermittent streams in south‐eastern Australia, in both spring and summer, to determine whether allochthonous detritus was an important nutritional resource in these systems. The isotopic signatures of the major potential allochthonous plant food sources (Eucalyptus, Phalaris and Juncus) overlapped, but were distinct from algae and the dominant macrophytes growing in the study reaches. The isotopic signatures of biofilm were more spatially and temporally variable than those of the other basal resources.
• 3 Despite allochthonous detritus having relatively high C : N ratios compared to other potential basal resources, results from isosource mixing model calculations demonstrated that this detritus, and the associated biofilm, were the major energy sources assimilated by macroinvertebrate primary consumers in both spring and summer. The importance of these energy sources was also reflected in animals higher in the food web, including predatory macroinvertebrates and fish. These resources were supplemented by autochthonous sources of higher nutritional value (i.e. filamentous algae and macrophytes, which had relatively low C : N ratios) when they became more prolific as the streams dried to disconnected pools in summer.
• 4 The results highlight the importance of allochthonous detritus (particularly from Eucalyptus) as a dependable energy source for benthic macroinvertebrates and fish in lowland intermittent streams of south‐eastern Australia. This contrasts with previous stable isotope studies conducted in lowland intermittent streams in arid Australia, which have reported that the fauna are primarily dependent on autochthonous algae.

     

Nutrient dynamics and pelagic food web interactions in oligotrophic and eutrophic environments: an overview

The concept of limiting nutrients is a cornerstone of theories concerning the control of production, structure and dynamics of freshwater and marine plankton. The current dogma is that nitrogen is limiting in most marine environments while freshwater ecosystems are mostly phosphorus-limited, although evidence of phytoplankton limitation by either N or P has been found in both environments.However, the same considerations apply to the availability of phosphorus in freshwater as to nitrogen in oceans. In resource-limited environments the plankton dynamics depend mostly on the internal mechanisms which act to recycle the limiting nutrient many times over within the surface waters. As the overall productivity increases, this dependence on nutrient regeneration decreases.The relationship between the stock of limiting nutrient, rates of supply and plankton dynamics must therefore be seen in the light of the processes operating within the entire food chain over quite different time scales. There is strong evidence that process-rates are mostly size-dependent and that food web interactions at the microbial level (picophytoplankton, bacteria, microheterotrophs) strongly effect the production of carbon and the regeneration of nutrients in the pelagic zone.     

Population dynamics of a generalist rodent in relation to variability in pulsed food resources in a fragmented landscape

1. Pulsed food resources are often considered equivalent in their potential impact on the reproduction and population dynamics of consumers, but differences in the attributes of food pulses and their distribution in the landscape may cause differences in their effects. 2. We tested whether a perishable pulsed resource (periodical cicadas, Magicicada spp.) had similar effects on the population dynamics of a generalist forest rodent, Peromyscus leucopus, as have been reported for a cacheable pulsed resource (acorn mast). 3. Because the availability of periodical cicadas may vary between edge and interior habitat, we also tested whether habitat type (edge vs. interior) and fragment size affected the abundance of cicadas and P. leucopus. 4. Nearly 90% of the variation in the relative population densities of P. leucopus was explained by the variation in the relative densities of periodical cicadas, and fragments with more cicadas tended to have more reproductive female mice and litters. 5. We found more cicadas and more P. leucopus in edge than interior habitat, but no differences in the relative densities of either in relation to fragment size. 6. Data from a non-emergence year revealed no differences other than the presence of periodical cicadas that could explain the 50% higher relative densities of P. leucopus in the emergence year. 7. At the beginning of the emergence of periodical cicadas, the three fragments with the highest numbers of emergence holes had three times more mice than the fragments with the lowest numbers of emergence holes, suggesting that P. leucopus is able to anticipate the emergence of periodical cicadas and increase reproduction prior to the pulse. 8. Hence, despite differences in perishability, seasonal timing and nutritional quality of pulsed food resources in a fragmented landscape, they appear to have similar positive effects on the population dynamics of the generalist rodent, P. leucopus and, in fact, P. leucopus may be able to anticipate resource pulses.