Strains of insect-pathogenic fungi with high virulence toward certain pest insects have great potential for commercial biological control applications. Identifying such strains has been a central theme in using fungi for biological control. This theme is supported by a persistent paradigm in insect pathology which suggests that the host insect is the predominant influence on the population genetics of insect-pathogenic fungi. In this study, a population genetics analysis of the insect-pathogenic fungusMetarhizium anisopliaefrom forested and agricultural habitats in Ontario, Canada, showed a nonrandom association of alleles between two distinct, reproductively isolated groups (index of multilocus association = 1.2). Analyses of the mitochondrial DNA showed no differences between the groups. The two groups were associated with different habitat types, and associations with insect hosts were not found. The group from forested areas showed an ability for cold-active growth (i.e., 8°C), while the group from the agricultural area showed an ability for growth at high temperatures (i.e., 37°C) and resilience to UV exposure. These results represent a significant paradigm shift; habitat selection, not host insect selection, drives the population structure of these insect-pathogenic deuteromycetous fungi. With each group we observed recombining population structures as well as clonally reproducing lineages. We discuss whether these groups may represent cryptic species. Worldwide,M. anisopliaemay be an assembly of cryptic species, each adapted to certain environmental conditions. The association of fungal genotypes with habitat but not with host insects has implications on the criteria for utility of this, and perhaps other, fungal biocontrol agents.

Simple hydrological models, such as the Seasonal Water Yield Model developed by the Natural Capital Project (InVEST SWYM), are attractive as data requirements are relatively easy to satisfy. However, simple models may produce unrealistic results when the underlying hydrological processes are inadequately described. We used the variation in performance of the InVEST SWYM across watersheds to identify correlates of poorly modeled outcomes of InVEST SWYM. We grouped 749 watersheds from across North America into five bioclimatic regions using nine environmental variables. For each region, we compared the predicted flow patterns to actual flow conditions over a 15-year period. The correlation between the modeled and actual flows was highly dispersed and relatively poor, with 92% of r² values less than 0.5 and 42% less than 0.1. We linked cryospheric variables to model performance in the bioclimatic region with the poorest model performance (the Low elevation Boreal Sub-humid region—LeBSh). After incorporating cryospheric conditions into the InVEST SWYM, predictions improved significantly in 30% of the LeBSh watersheds. We provide a relatively straightforward approach for identifying processes that simple hydrological models may not consider or which need further attention or refinement.

Inherent differences between naturally-formed lakes and human-made reservoirs may play an important role in shaping zooplankton community structure. For example, because many reservoirs are created by impounding and managing lotic systems for specific human purposes, zooplankton communities may be affected by factors that are unique to reservoirs, such as shorter water residence times and a reservoir’s management regime, compared to natural lakes. However, the environmental factors that structure zooplankton communities in natural lakes vs. reservoirs may vary at the continental scale and remain largely unknown. We analyzed data from the 2007 U.S. Environmental Protection Agency’s National Lakes Assessment and the U.S. Army Corps of Engineers’ National Inventory of Dams to compare large-bodied crustacean zooplankton communities (defined here as individuals retained by 0.243 mm mesh size) in natural lakes and reservoirs across the continental U.S. using multiple linear regressions and regression tree analyses. We found that large-bodied crustacean zooplankton density was overall higher in natural lakes compared to reservoirs when the effect of latitude was controlled. The difference between waterbody types was driven by calanoid copepods, which were also more likely to be dominant in the >0.243 mm zooplankton community in natural lakes than in reservoirs. Regression tree analyses revealed that water residence time was not a major driver of calanoid copepod density in natural lakes but was one of the most important drivers of calanoid copepod density in reservoirs, which had on average 0.5-year shorter water residence times than natural lakes. Reservoirs managed for purposes that resulted in shorter residence times (e.g., hydroelectric power) had lower zooplankton densities than reservoirs managed for purposes that resulted in longer residence times (e.g., irrigation). Consequently, our results indicate that water residence time may be an important characteristic driving differing large-bodied zooplankton dynamics between reservoirs and natural lakes.

Null models exploring species co-occurrence and trait-based limiting similarity are increasingly used to explore the influence of competition on community assembly; however, assessments of common models have not thoroughly explored the influence of variation in matrix size on error rates, in spite of the fact that studies have explored community matrices that vary considerably in size. To determine how smaller matrices, which are of greatest concern, perform statistically, we generated biologically realistic presence-absence matrices ranging in size from 3–50 species and sites, as well as associated trait matrices. We examined co-occurrence tests using the C-Score statistic and independent swap algorithm. For trait-based limiting similarity null models, we used the mean nearest neighbour trait distance (NN) and the standard deviation of nearest neighbour distances (SDNN) as test statistics, and considered two common randomization algorithms: abundance independent trait shuffling (AITS), and abundance weighted trait shuffling (AWTS). Matrices as small as three × three resulted in acceptable type I error rates (p < 0.05) for both the co-occurrence and trait-based limiting similarity null models when exclusive p-values were used. The commonly used inclusive p-value (≤ or ≥, as opposed to exclusive p-values; < or >) was associated with increased type I error rates, particularly for matrices with fewer than eight species. Type I error rates increased for limiting similarity tests using the AWTS randomization scheme when community matrices contained more than 35 sites; a similar randomization used in null models of phylogenetic dispersion has previously been viewed as robust. Notwithstanding other potential deficiencies related to the use of small matrices to represent communities, the application of both classes of null model should be restricted to matrices with 10 or more species to avoid the possibility of type II errors. Additionally, researchers should restrict the use of the AWTS randomization to matrices with fewer than 35 sites to avoid type I errors when testing for trait-based limiting similarity. The AITS randomization scheme performed better in terms of type I error rates, and therefore may be more appropriate when considering systems for which traits are not clustered by abundance.