The mechanisms shaping patterns of biodiversity along spatial gradients remain poorly known and controversial. Hypotheses have emphasized the importance of both environmental and spatial factors. Much of the uncertainty about the relative role of these processes can be attributed to the limited number of comparative studies that evaluate multiple potential mechanisms. This study examines the relative importance of six variables: temperature, precipitation, productivity, habitat heterogeneity, area, and the mid‐domain effect on patterns of species richness for non‐volant small mammals along four neighboring mountain ranges in central Utah. Along each of these elevational gradients, a hump‐shaped relationship of richness with elevation is evident. This study evaluates whether the processes shaping this common pattern are also common to all gradients. Model selection indicates that no one factor or set of factors best explains patterns of species richness across all gradients, and drivers of diversity may vary seasonally. These findings suggest that commonality in the pattern of species richness, even among elevational gradients with a similar biogeographic history and fauna, cannot be attributed to a simple universal explanation.

One expected response to observed global warming is an upslope shift of species elevational ranges. Here, we document changes in the elevational distributions of the small mammals within the Ruby Mountains in northeastern Nevada over an 80-year interval. We quantified range shifts by comparing distributional records from recent comprehensive field surveys (2006)(2007)(2008) to earlier surveys (1927)(1928)(1929) conducted at identical and nearby locations. Collector field notes from the historical surveys provided detailed trapping records and locality information, and museum specimens enabled confirmation of species' identifications. To ensure that observed shifts in range did not result from sampling bias, we employed a binomial likelihood model (introduced here) using likelihood ratios to calculate confidence intervals around observed range limits. Climate data indicate increases in both precipitation and summer maximum temperature between sampling periods. Increases in winter minimum temperatures were only evident at mid to high elevations. Consistent with predictions of change associated with climate warming, we document upslope range shifts for only two mesic-adapted species. In contrast, no xeric-adapted species expanded their ranges upslope. Rather, they showed either static distributions over time or downslope contraction or expansion. We attribute these unexpected findings to widespread land-use driven habitat change at lower elevations. Failure to account for land-use induced changes in both baseline assessments and in predicting shifts in species distributions may provide misleading objectives for conservation policies and management practices.

Small piscivores are regarded as important regulators of the composition of coral reef fish communities, but few studies have investigated their predatory ecology or impact on assemblages of juvenile fishes. This study investigated the foraging ecology of a common coral reef predator, the dottyback Pseudochromis fuscus, using underwater focal animal observations. Observations were conducted at two times of year: the summer, when recruit fishes were an available food item and winter, when remaining juveniles had outgrown vulnerability to P. fuscus. During the summer, P. fuscus directed 76% of its strikes at invertebrates and 24% at recruiting juvenile fishes. When striking at fishes, P. fuscus exhibited two distinct feeding modes: an ambush (26% successful) and a pursuit mode (5% successful). Predator activity in the field peaked at midday, averaging 2.5 captures h -1 of juvenile fishes. Monitoring of activity and foraging in the laboratory over 24-h periods found that P. fuscus was a diurnal predator and was active for 13 h d -1 during the summer. The number of hours during which foraging was recorded differed greatly among individuals (n = 10), ranging from 4 to 13 h. The number of predatory strikes did not increase with standard length, but the success rate and consumption rate of juvenile fishes did increase with size. Estimated hourly mortality on juvenile fish ranged from 0.49 fish h -1 in small P. fuscus individuals (30-39 mm standard length, SL; equating to 6.3 per 13 h day) to 2.4 fish h -1 in large P. fuscus individuals (55-65 mm SL; 30.6 per 13 h day). During the winter, P. fuscus struck at invertebrates with a similar rate to the summer period. These observations of the predatory ecology of P. fuscus support the hypothesis that in coral reef systems, small piscivores, because of their high metabolism and activity, are probably important regulators of coral reef fish community composition.

Null model analysis of species co‐occurrence patterns has long been used to gain insight into community assembly but is often limited to identifying non‐random patterns without providing clarity about underlying ecological mechanisms. This challenge is especially apparent when sampling units are spread across a heterogeneous landscape or along an environmental gradient because multiple mechanisms can produce similar co‐occurrence patterns. We developed a trait‐based approach for discriminating between environmental filtering and biotic interactions as the probable driver of co‐occurrence patterns across environmentally heterogeneous sites. We demonstrate our framework by analyzing the co‐occurrence of small mammals over elevation in three independent mountain ranges in the Great Basin of the western United States. Our sampling design accounts for landscape scale environmental variability and within‐site habitat heterogeneity. We identified 52 non‐random species pairs, of which 36 were aggregated and 16 were segregated. For each pair, we determined which mechanism was the likely ecological explanation using a hypothesis‐testing framework based on functional trait similarity. Expectations of biotic interactions were based on similarity of diet and body size whereas habitat affinity and geographic range were used for environmental filtering. Only four pairs were consistent with expectations under biotic interactions, including pairs for which competitive exclusion has previously been documented. In addition to analyzing individual pairs, we used binomial tests of observed versus expected totals of intra‐ and inter‐guild pairs to determine assemblage‐wide deviations from random community structure. Signatures of environmental filtering were consistent across mountain ranges and scales. Despite differences in species composition and significant pairs among data sets, our approach revealed consistent mechanistic conclusions, emphasizing the value of trait‐based methods to co‐occurrence and community assembly.