Assembling an ant community: species functional traits reflect environmental filtering.

Species should only persist in local communities if they have functional traits that are compatible with habitat-specific environmental conditions. Consequently, pronounced regional environmental gradients should produce environmental filtering, or a trait-based spatial segregation of species. It is critical to quantify the links between species' functional traits and their environment in order to reveal the relative importance of this process to community assembly and promote understanding of the impacts of ongoing environmental changes. We investigated this relationship using epigaeic ants in an environmentally heterogeneous region of Florida. We found evidence for environmental filtering as environmental conditions such as groundcover, surface temperature, vapor pressure deficit, and plant diversity were strongly correlated with assemblage composition. Certain species traits appeared particularly important to persistence: (1) ants in environments with less groundcover have relatively longer legs but do not differ in size, (2) ants in hotter environments exhibit greater thermal tolerances, and (3) ants in hotter and drier environments do not exhibit greater desiccation resistance. These findings show surface complexity and temperature may interact with morphology and physiology to impact the spatial distribution of ants and underscore the importance of climate change. Climate warming is predicted to alter assemblage composition, competitive dynamics, and consequently impact ecosystem processes. We suggest environmental filters acting at regional scales, as shown here, act in tandem with more frequently studied local-scale competitive interactions to delimit ant community assemblages.

Fast food in ant communities: how competing species find resources.

An understanding of foraging behavior is crucial to understanding higher level community dynamics; in particular, there is a lack of information about how different species discover food resources. We examined the effect of forager number and forager discovery capacity on food discovery in two disparate temperate ant communities, located in Texas and Arizona. We defined forager discovery capacity as the per capita rate of resource discovery, or how quickly individual ants arrived at resources. In general, resources were discovered more quickly when more foragers were present; this was true both within communities, where species identity was ignored, as well as within species. This pattern suggests that resource discovery is a matter of random processes, with ants essentially bumping into resources at a rate mediated by their abundance. In contrast, species that were better discoverers, as defined by the proportion of resources discovered first, did not have higher numbers of mean foragers. Instead, both mean forager number and mean forager discovery capacity determined discovery success. The Texas species used both forager number and capacity, whereas the Arizona species used only forager capacity. There was a negative correlation between a species' prevalence in the environment and the discovery capacity of its foragers, suggesting that a given species cannot exploit both high numbers and high discovery capacity as a strategy. These results highlight that while forager number is crucial to determining time to discovery at the community level and within species, individual forager characteristics influence the outcome of exploitative competition in ant communities.

Roles of human disturbance, precipitation, and a pathogen on the survival and reproductive probabilities of deer mice.

Climate change, human disturbance, and disease can have large impacts on the dynamics of a species by affecting the likelihood of survival and reproduction of individuals. We investigated the roles of precipitation, off-road vehicle (ORV) alteration of habitat, and infection with Sin Nombre virus on the survival and reproductive probabilities of deer mice (Peromyscus maniculatus). We used generalized linear mixed models to estimate the effects of these factors and their interactions by fitting capture-recapture data collected seasonally from 2002 to 2007 at 17 sites in the Great Basin Desert of central Utah, USA. During periods with high precipitation, we found no difference in survival and reproductive probabilities between seasons, but during drier periods, we found a reduction of overwinter survival and fall reproductive activity. Precipitation also interacted with disturbance to affect survival probabilities and female reproduction; in periods with low precipitation, deer mice on highly disturbed sites had extremely low survival probabilities and low reproductive probabilities of females compared to those of individuals from low-disturbance sites. However, high precipitation ameliorated the effect of disturbance on both parameters. Deer mice from sites with high impact of ORV disturbance also had low survival over summer. Additionally, male reproductive probabilities were diminished on highly disturbed sites in both seasons; in contrast, they were reduced only in the fall on low-disturbance sites. Density had an overall negative effect on survival and reproductive probabilities of deer mice. For females, the negative effect on reproductive activity was amplified in highly disturbed sites. We found no effect of hantavirus infection on survival probabilities of deer mice. Overall, this study revealed complexity in the determinants of deer mouse survival and reproduction given by the effects of a number of significant interactions among explanatory variables. Thus, factors that may not appear to have a strong effect when investigated alone can still be influential by modulating the effect of a different factor.

How host population dynamics translate into time-lagged prevalence: an investigation of Sin Nombre virus in deer mice.

Human cases of hantavirus pulmonary syndrome caused by Sin Nombre virus are the endpoint of complex ecological cascade from weather conditions, population dynamics of deer mice, to prevalence of SNV in deer mice. Using population trajectories from the literature and mathematical modeling, we analyze the time lag between deer mouse population peaks and peaks in SNV antibody prevalence in deer mice. Because the virus is not transmitted vertically, rapid population growth can lead initially to reduced prevalence, but the resulting higher population size may later increase contact rates and generate increased prevalence. Incorporating these factors, the predicted time lag ranges from 0 to 18 months, and takes on larger values when host population size varies with a longer period or higher amplitude, when mean prevalence is low and when transmission is frequency-dependent. Population size variation due to variation in birth rates rather than death rates also increases the lag. Predicting future human outbreaks of hantavirus pulmonary syndrome may require taking these effects into account.

The origin of human pathogens: evaluating the role of agriculture and domestic animals in the evolution of human disease.

Many significant diseases of human civilization are thought to have arisen concurrently with the advent of agriculture in human society. It has been hypothesised that the food produced by farming increased population sizes to allow the maintenance of virulent pathogens, i.e. civilization pathogens, while domestic animals provided sources of disease to humans. To determine the relationship between pathogens in humans and domestic animals, I examined phylogenetic data for several human pathogens that are commonly evolutionarily linked to domestic animals: measles, pertussis, smallpox, tuberculosis, taenid worms, and falciparal malaria. The majority are civilization pathogens, although I have included others whose evolutionary origins have traditionally been ascribed to domestic animals. The strongest evidence for a domestic-animal origin exists for measles and pertussis, although the data do not exclude a non-domestic origin. As for the other pathogens, the evidence currently available makes it difficult to determine if the domestic-origin hypothesis is supported or refuted; in fact, intriguing data for tuberculosis and taenid worms suggests that transmission may occur as easily from humans to domestic animals. These findings do not abrogate the importance of agriculture in disease transmission; rather, if anything, they suggest an alternative, more complex series of effects than previously elucidated. Rather than domestication, the broader force for human pathogen evolution could be ecological change, namely anthropogenic modification of the environment. This is supported by evidence that many current emerging infectious diseases are associated with human modification of the environment. Agriculture may have changed the transmission ecology of pre-existing human pathogens, increased the success of pre-existing pathogen vectors, resulted in novel interactions between humans and wildlife, and, through the domestication of animals, provided a stable conduit for human infection by wildlife diseases.

Changes in sin nombre virus antibody prevalence in deer mice across seasons: the interaction between habitat, sex, and infection in deer mice.

We examined the impact of season and habitat on Sin Nombre virus (SNV) seroprevalence in deer mice (Peromyscus maniculatus) in Utah's Great Basin Desert from May 2002 through summer 2003. Low mouse captures in 2002 limited analysis for that year. In two seasons during 2003, mouse density and sagebrush cover were positively linked (spring: r = 0.8, P = 0.01; summer: r = 0.8, P = 0.04). In the spring, seroprevalence was negatively correlated with density (r = -0.9, P< 0.01); male and female antibody prevalence did not differ; and scarring was unrelated to antibody status. In the summer, density and antibody prevalence were unrelated; male seroprevalence was higher (chi(2) = 3.6, P = 0.05); and seropositive mice had more scars (t = 2.5, P = 0.02). We speculate nesting behavior could maintain SNV over the winter, whereas summer territoriality could be responsible for transmission.