Sequential decision-making in a variable environment: modeling elk movement in Yellowstone National Park as a dynamic game

We develop a suite of models with varying complexity to predict elk movement behavior during the winter on the Northern Range of Yellowstone National Park (YNP). The models range from a simple representation of optimal patch choice to a dynamic game, and we show how the underlying theory in each is related by the presence or absence of state- and frequency-dependence. We compare predictions from each of the models for three variables that are of basic and applied interest: elk survival, aggregation, and use of habitat outside YNP. Our results suggest that despite low overall forage depletion in the winter, frequency-dependence is crucial to the predictions for elk movement and distribution. Furthermore, frequency-dependence interacts with mass-dependence in the predicted outcome of elk decision-making. We use these results to show how models that treat single movement decisions in isolation from the seasonal sequence of decisions are insufficient to capture landscape scale behavior.     

Balancing predation risk,social interference,and foraging opportunities in backswimmers,<Emphasis Type="Italic"> Notonecta maculata</Emphasis>

Loss of foraging opportunities and intraspecific competition for prey may be important costs of using refuges, because a hiding animal is unable to use or defend its foraging area from conspecific intrusions. Thus, animals should balance antipredator demands with other requirements in deciding when to come out from a refuge after a predators unsuccessful attack. Observations on foraging and social interactions of backswimmers Notonecta maculata suggest that foraging may be costly in terms of intraspecific agonistic interactions. When prey density is low, increasing the probability of finding a prey may require active exploration of a larger area, but this also increases the probability of encountering a competitor. After simulated exposure to predators, unfed bugs resumed feeding positions after a significantly shorter hiding period than recently fed bugs. We hypothesized that hiding time may also be reduced by recent interactions with conspecific competitors, due to an increased perceived need to defend feeding opportunities. Thus, when a predator attack occurred immediately after an agonistic conspecific interaction, backswimmers resumed feeding positions more quickly, and closer to the original position from which they were disturbed, suggesting short-term defense of particular positions. We conclude that when foraging, backswimmers balance the benefits of finding prey with the costs of predation risk and social interference in deciding their foraging strategy.Communicated by P.K. McGregor     

Effects of size on predation risk,behavioural response to fish,and cost of reduced feeding in larval Ischnura verticalis (Coenagrionidae: Odonata)

Summary We used laboratory studies to examine the role of predation risk and cost of anti-predator behaviour in determining the behavioural response of several larval instars of Ischnura verticalis to a fish predator (Lepomis gibbosus). Smaller larvae were less susceptible to fish predation than larger larvae. Smaller larvae depressed movement to a greater degree in the presence of fish than did larger larvae; large larvae were generally less active than small larvae regardless of fish presence. Reduced feeding resulted in smaller larvae suffering more in terms of reduced growth than did large larvae. In general, our results tend to support the hypothesis that individuals that suffer high costs of anti-predator behaviour but little risk of predation may only exhibit anti-predator behaviours in the presence of predators, whereas individuals with a higher risk of predation and a lower cost of anti-predator behaviour may evolve anti-predator mechanisms that are in effect even in the absence of predators.     

Ecosystem dynamics through the past 2000 years as revealed by fossil mammals from Lamar cave in Yellowstone National Park,USA

Lamar Cave, a late Holocene paleontological site in Yellowstone National Park, USA, has yielded 36 fossil mammal species from 10 stratigraphic units representing about 2000 years. The fossil fauna is similar to the mammals in the park today and affords a unique opportunity to investigate paleocommunity and ecosystem dynamics through time.

Remains of an extralimital species in Yellowstone today, Microtus ochrogaster are found only in the oldest cave deposits, dated at 1695 ± 60 yr B.P. Disappearance of this species by about 1550 years ago marks evolution into the ecosystem that has persisted into the present and is recognizable by a fossil mammal assemblage that is virtually identical to the modern one. Relative abundance fluctuations in the mammal populations imply community resilience through the past 2000 years. Trends in relative abundances of small mammals, particularly the inverse relationship between Microtus and Spermophilus are interpreted as evidence of declining grass cover near Lamar Cave between approximately 1500 and 1000 years ago. The initiation of this environmental change may well have stimulated the ecosystem change marked by Microtus ochrogaster extirpation. Support for this hypothesis is found in a pollen record from a nearby lake.     

Landscape Patterns of Sapling Density,Leaf Area,and Aboveground Net Primary Production in Postfire Lodgepole Pine Forests,Yellowstone National Park (USA)

Causes and implications of spatial variability in postfire tree density and understory plant cover for patterns of aboveground net primary production (ANPP) and leaf area index (LAI) were examined in ninety 11-year-old lodgepole pine (Pinus contorta var. latifolia Engelm.) stands across the landscape of Yellowstone National Park (YNP), Wyoming, USA. Field studies and aerial photography were used to address three questions: (1) What is the range and spatial pattern of lodgepole pine sapling density across the burned Yellowstone landscape and what factors best explain this variability? (2) How do ANPP and LAI vary across the landscape and is their variation explained by abiotic factors, sapling density, or both? (3) What is the predicted spatial pattern of ANPP and LAI across the burned Yellowstone landscape? Stand density spanned six orders of magnitude, ranging from zero to 535,000 saplings ha^?1, and it decreased with increasing elevation and with increasing distance from unburned forest (r ²?=?0.37). Postfire densities mapped from 1:30,000 aerial photography revealed that 66% of the burned area had densities less than 5000 saplings ha^?1 and approximately 25% had densities greater than 10,000 saplings ha^?1; stand density varied spatially in a fine-grained mosaic. New allometric equations were developed to predict aboveground biomass, ANPP, and LAI of lodgepole pine saplings and the 25 most common herbaceous and shrub species in the burned forests. These allometrics were then used with field data on sapling size, sapling density, and percent cover of graminoid, forb, and shrub species to compute stand-level ANPP and LAI. Total ANPP averaged 2.8 Mg ha^?1y^?1 but ranged from 0.04 to 15.12 Mg ha^?1y^?1. Total LAI averaged 0.80 m² m^?2 and ranged from 0.01 to 6.87 m² m^?2. Variation in ANPP and LAI was explained by both sapling density and abiotic factors (elevation and soil class) (ANOVA, r ²?=?0.80); abiotic variables explained 51%–54% of this variation. The proportion of total ANPP contributed by herbaceous plants and shrubs declined sharply with increasing sapling density (r ²?=?0.72) and increased with elevation (r ²?=?0.36). However, total herbaceous productivity was always less than 2.7 Mg ha^?1 y^?1, and herbaceous productivity did not compensate for tree production when trees were sparse. When extrapolated to the landscape, 68% of the burned landscape was characterized by ANPP values less than 2.0 Mg ha^?1y^?1, 22% by values ranging from 2 to 4 Mg ha^?1y^?1, and the remaining 10% by values greater than 4 Mg ha^?1y^?1. The spatial patterns of ANPP and LAI were less heterogeneous than patterns of sapling density but still showed fine-grained variation in rates. For some ecosystem processes, postfire spatial heterogeneity within a successional stage may be similar in magnitude to the temporal variation observed through succession.     

The Role of Extracellular Polymeric Substances in the Silicification of Calothrix: Evidence from Microbial Mat Communities in Hot Springs at Yellowstone National Park,USA

We provide nanoscale evidence of the role of sheath exopolymers in the silicification of the sheathed cyanobacteria Calothrix. Electron microscope observations of silicified Calothrix cells revealed that silica accretes directly onto EPS sheath fibrils to produce an open web of silica particles that could remain permeable to nutrients and waste products. We also found that silicified Calothrix cells from different microhabitats contained morphologically distinct silica particles. Differences in silicification texture suggest that environmental variables may influence silicification at the nanoscale. We develop a framework based on aggregation kinetics to address silicification processes in Calothrix and other sheathed cyanobacteria.     

Effects of spatial heterogeneity on butterfly species richness in Rocky Mountain National Park, CO, USA

We investigated butterfly responses to plot-level characteristics (plant species richness, vegetation height, and range in NDVI [normalized difference vegetation index]) and spatial heterogeneity in topography and landscape patterns (composition and configuration) at multiple spatial scales. Stratified random sampling was used to collect data on butterfly species richness from seventy-six 20 × 50 m plots. The plant species richness and average vegetation height data were collected from 76 modified-Whittaker plots overlaid on 76 butterfly plots. Spatial heterogeneity around sample plots was quantified by measuring topographic variables and landscape metrics at eight spatial extents (radii of 300, 600 to 2,400 m). The number of butterfly species recorded was strongly positively correlated with plant species richness, proportion of shrubland and mean patch size of shrubland. Patterns in butterfly species richness were negatively correlated with other variables including mean patch size, average vegetation height, elevation, and range in NDVI. The best predictive model selected using Akaike’s Information Criterion corrected for small sample size (AIC_c), explained 62% of the variation in butterfly species richness at the 2,100 m spatial extent. Average vegetation height and mean patch size were among the best predictors of butterfly species richness. The models that included plot-level information and topographic variables explained relatively less variation in butterfly species richness, and were improved significantly after including landscape metrics. Our results suggest that spatial heterogeneity greatly influences patterns in butterfly species richness, and that it should be explicitly considered in conservation and management actions.     

千岛湖栖息地片段化效应对鸟类巢捕食风险的影响

为了解栖息地片段化对鸟类巢捕食风险的影响,我们于2010年4-8月,在千岛湖地区选取16个岛屿,分别在岛屿边缘和内部区域用放置鸡蛋(大型卵)和鹌鹑(Coturnix japonica)蛋(小型卵)的方法进行人工模拟地面巢试验,研究不同体积大小鸟卵的巢捕食率及其差异,分析巢捕食率与岛屿面积、隔离度、形状指数和植物物种丰富...     

The influence of hunger and predation risk on group cohesion in a pelagic fish, walleye pollock Theragra chalcogramma

Variation in the intensity of schooling behavior in fishes suggests that the benefits of aggregation are balanced by certain costs. We examined the proximity of group members to each other in juvenile walleye pollock, Theragra chalcogramma, under different environmental conditions. Food availability, simulated by a gradient of six ration treatments, had a major influence on group cohesion, with increasing dispersion as food level decreased. Group cohesion also decreased at night relative to daytime levels. Small juveniles (x=53 mm TL) maintained on high rations were highly responsive to the potential threat of a predator, with groups becoming more cohesive and remaining so for up to an hour after the initial threat. A chronic threat (continual presence of predators) resulted in tighter group cohesion than an acute threat (single simulated attack). Small juveniles maintained on low rations were less responsive to predation threats and recovered quickly, supporting the hypothesis that hunger induces risk-taking behavior. Large juveniles (x=149 mm TL) did not change their degree of aggregation in response to either type of predation threat. An overall plasticity in the degree of cohesiveness among group members indicates that walleye pollock are capable of gradually modifying their schooling behavior according to the environmental context.     

Effects of tsetse targets on mammals and birds in Kasungu National Park, Malawi

Possible effects on wildlife of targets baited with an attractive odour (acetone), impregnated with deltamethrin and used to control tsetse flies (Glossina morsitans morsitans), were investigated in the Kasungu National Park, Malawi. Mammals and birds were censused simultaneously along transects in 'Miombo' woodland or 'Dambo' grassland/mixed woodland, with and without targets. Mammals were also monitored by surveys of their spoor on experimentally cleared plots (5 m in diameter) and a target relocation experiment was conducted when the targets were removed from the test transects and transferred to the controls. Significantly fewer small antelopes (e.g. common Duiker Sylvicapra grimmia) were detected in plots along transects with targets (tests) than along control transects. The presence of targets affected the frequency of occurrence of antelopes, suids and large herbivores in the experimental plots, but small carnivores, monkeys, rodents and hares were unaffected. A conclusion of the relocation experiment was that the deltamethrin-impregnated cloth was responsible for the observed effects and not the acetone. A total number of 23 species of birds meeting a criterion for their abundance in the areas surveyed were selected for detailed analyses. Lower indices in the test areas than in the controls were recorded for 15 of these 23 species in the Dambos and for 10 of 21 species in the Miombo woodland. Matched paired comparisons revealed significantly lower numbers in the test areas than in the controls in both habitats only for black-headed Oriole Oriolus larvatus, but for Grey Lourie Corythaixoides concolor, little bee-eater Merops pusillus, fork-tailed Drongo Dicrurus adsimilis and combined data on three species of sunbirds in the Dambos. The black-eyed bulbul Pycnonotus barbatus had significantly lower numbers in the tests than in the controls in Miombo woodland. The numbers of three species of dove were higher in the tests than in the controls in both habitats, significantly so for the Cape turtle dove Streptopelia capicola, and numbers of the Flappet Lark Mirafra rufocinnamomea were significantly higher in the tests than in the controls in Dambos. The results are discussed in the light of previous studies on environmental effects of tsetse control, including effects of tsetse targets on pollinators especially non-target horseflies.     

The influence of logging and topography on the distribution of spruce-fir forests near their Southern limits in Great Smoky Mountains National Park,USA

We studied the effects of logging history, topography, and potential insolation on the lower-elevation limit of existing spruce-fir forest in Great Smoky Mountains National Park (GRSM). Dummy-variable regression, analysis of variance, and classification trees were applied to environmental data within a geographic information system framework. The effect of logging history on the lower limit of spruce-fir depended on aspect. On north-facing slopes (270°– 90°), the presence of spruce-fir was independent of both logging history and potential insolation. On south-facing sites (90°–270°), the elevation of spruce-fir was significantly higher (by 122 m) in areas that had been logged historically. Classification-tree models suggested an even greater logging effect, indicating that both the lower limit and the upper dominance zone of spruce-fir forest are, on average, nearly 200 m higher in historically logged landscapes. Presence of spruce-fir on south aspects was also significantly related to potential insolation, but the strength of this effect was not dependent on logging history. Classification-tree models, developed separately using data from logged sites versus unlogged sites, were used to estimate the current area of spruce-fir forest in the park expected under the hypothetical scenario that no spruce-fir had been logged (38,675 ha) versus the alternate scenario that it had hall been logged (11,727 ha). At present the area of spruce-fir forest in the park is 21,242 ha. We found greater prevalence of spruce-fir on the Tennessee side of the divide on south aspects and historically logged sites, possibly due to greater occurrence of westerly winds and associated cloud cover.     

Ecosystem recovery in terra firme forests after cutting and burning: a comparison on species richness, floristic composition and forest structure in the Jaú National Park, Amazonia

Six hectares, three in a primary forest and three in a 40 year old secondary forest were inventoried for all trees with Diameter at Breast Height (DNH) of 10 cm or greater in a terra firme forest 200 km north-east of Manaus, central Amazonia in order to compare the difference between structure, species richness and floristic composition. Both species richness and tree density were significantly higher in the upland forest than in the secondary forest. The forest structure pattern analysed (DBH, basal area and estimated dry biomass) did not differ significantly between the two forest types. Similarity indices at species level were only 14%. In the 3 ha of primary forest the number of species varied from 137 to 159, the number of individuals from 639 to 713, total basal area from 32.8 to 40.2 m² and estimate total of above-ground dry biomass (AGBM) from 405 to 560 tons per ha. In the 3 ha of secondary forest, the number of species varied from 86 to 90, the number of individuals from 611 to 653, total basal area from 28.8 to 39.9 m² and estimated total AGBM from 340 to 586 tons per ha. Family Importance Value (FIV) is the sum of relative density, dominance and richness of a family. The most important families in relation to FIV were Burseraceae, Chrysobalanaceae, Lecythidaceae, Myristicaceae, Bombacaceae, Fabaceae and Mimosaceae in the 3 ha of primary forest, while Burseraceae, Lecythidaceae, Sapotaceae, Arecaceae and Cecropiaceae were the most important families in the 3 ha of secondary forest. Importance Value Index (IVI) is the sum of relative density, dominance and frequency of a species. Alexa grandiflora (Caesalpiniaceae), Sckronema micranthum (Bombacaceae) and Pourouma guianensis (Cecropiaceae) were the most important species in relation IVI, in the primary forest, while Eschweilera grandiflora (Lecythidaceae), Protium apiculatum (Burseraceae) and Bertholletia excelsa (Lecythidaceae) were the most important species in the secondary forest. We conclude that species richness was significandy different between the two forests, but that forest structure patterns analysed in this study (DBH, basal area and dry biomass) were similar. This demonstrates that 40 years was sufficient time for the secondary forest to recover the original structure of the primary forest, but not the original species richness. The low species similarity between the two forests indicates that the floristic composition was quite distinct and that the mixture of primary forest and disturbed forest has led to an increase in total species diversity.