There is great interest in ecology to determine the drivers of species richness. For many taxa, and in natural circumstances, temperature is often found to be a good predictor of richness. The goal of this thesis was to determine, amongst several human-related and natural, environmental and ecological variables, the most important broad-scale predictors of butterfly species richness across Canada. Also, I presented a test of whether spatial relationships are adequate surrogates for the temporal relationship between richness and predictor variables. Using precisely georeferenced and dated butterfly records across Canada, I created butterfly species richness maps for two periods (1900-1930 and 1960-1990), and then related them to candidate predictors. Natural variables such as temperature, precipitation and soil type tended to explain most of the variance in species richness, while human-related variables such as habitat fragmentation, habitat heterogeneity and pesticide density added very little. A comparison of temporal and spatial relationships showed that temperature was a consistent predictor of richness through time and space, but that the impact of human activities on richness differed. My results are consistent with the species-energy hypothesis while showing that human-related variables are not having a large measurable effect on butterfly species richness patterns in Canada at broad scales. Also of critical importance in this thesis is the difference I found between the spatial and temporal relationships of richness vs. human activity level. I show that the assumption commonly made in macro-ecology that spatial variables are appropriate surrogates for temporal ones, is not always correct.

Organisms live in heterogeneous environments but fail to demonstrate optimal adaptation toil the totality of the ecological conditions they experience. This imperfection would result from trade-offs between ecologically relevant traits, which implies that trade-offs are the ultimate cause of major features of life such as biodiversity. Here, I selected bacterial populations in defined laboratory environments to test for the importance of differences in productivity and quality among alternative resources (Chapter 1), as well as the spatial pattern of resources (Chapter 2), in shaping the evolution of ecological niches and genetic diversity. The experiment presented in Chapter 1 provides support for a general model for the evolution of diversity. In Chapter 2 however, I suggest that mutational processes can impinge on the expression of trade-offs, which is not recognized in our current view of the problem of diversity. These findings provide plausible explanations for situations where expected trade-offs are not found.

Understanding the spatial ecology and habitat use of declining species is essential for their management and successful recovery. I examined the movement patterns and habitat selection of map turtles (Graptemys geographica), a species at risk, in St. Lawrence Islands National Park. Adult females moved longer distances and had larger home ranges than juvenile females and males. The longest distances moved by adult females occurred during nesting excursions. Examining movements allowed for critical habitat locations to be found (nesting sites, hibernacula). My data on movements will allow St. Lawrence Islands National Park to mitigate development plans and regulate visitor impacts on these sites. I examined habitat selection at multiple spatial scales. Map turtles generally avoided deep water (>2 m) and selected home ranges in waters <1 m deep with significantly more natural than developed shoreline. Adult females used deep water more often and males preferred areas with surface cover. Management effort should implement regulations concerning further shoreline development.

Methylmercury (MeHg), the known toxic form of mercury is generally only 1-3% of total mercury (THg) in precipitation. As such it was thought to be an insignificant source to aquatic ecosystems; however, since other forms of mercury can be re-volatilized from the snowpack back to the atmosphere shortly after deposition the proportion and biological significance of MeHg in the snowmelt runoff is indeed an important source to downstream ecosystems. To assess the seasonal contribution of MeHg in runoff to the overall annual load of MeHg, the tributaries of two lakes, Dickie (DE) and Harp (HP) in the Haliburton-Muskoka district of Ontario were sampled on a weekly basis for MeHg, dissolved organic carbon (DOC) and conductivity over two years. Stream discharge was monitored on a continuous basis. In order to assess the amount of MeHg in the snowpack, snow cores from the watersheds were sampled during early spring. Additionally, bulk precipitation samples were collected using a wet-dry automated precipitation collector in the mercury deposition network (MDN) operated station in St. Anicet, Quebec every 4-6 weeks, for 26 months. The samples were analyzed for MeHg, and an annual deposition rate was calculated. THg and MeHg data obtained from the MDN was used to calculate a deposition rate for THg, and the proportion of MeHg. THg in wet deposition was 5.29-6.88 mug m-2 year-1, and MeHg was 0.15-0.19 mug m -2 year-1 and equivalent to 2.7-2.9 % of THg. Although low MeHg concentrations were measured in runoff during spring freshet, the high discharge rates resulted in quantities which were up to 33.9% of the annual amount of MeHg. The seasonal contribution of MeHg in runoff was found to be: spring>summer>fall> winter for the wetland rich DE and fall> spring> winter> summer for the wetland dull HP. Multiple regression analysis illustrated that DOC was the most significant predictor of MeHg (r2=0.32, p<0.01) across all data with a strong seasonal dependence of the DOC-MeHg correlation. The strongest correlation was found during spring freshet (r 2=0.65, p=0.001) followed by summer, winter and fall (r2= 0.37, 0.23 and 0.22 respectively, p<0.05 for all). The constant flow of water during the winter and strong correlation with DOC during snowmelt suggests a terrestrial source of MeHg even during the winter months, however, direct precipitation during spring and snowmelt were also found to contribute to the MeHg load during snowmelt runoff.

Sixty-seven sample sites in 45 streams across the National Capital Region and Gatineau Park were sampled between the summers of 2001 and 2008 in order to describe communities within the Park and assess differences in stream fish and invertebrate assemblages along a gradient of urbanization. Additionally the relationships between nutrient concentration and biomass of algae, invertebrates, and fish were quantified, and changes in the size distribution of benthic invertebrates and fish along a gradient of nutrient and anthropogenic impacts were described. Principal Component Analyses indicated that urban and agricultural sites were easily distinguished from forested sites using water quality parameters. Invertebrate and fish assemblages did not differ as clearly among site categories. There were increased nutrient levels in agricultural and urbanized watersheds, however the biomass of primary producers did not proportionally track the increase in nutrients. Slopes of the relationships between biomass and nutrient concentration declined with increasing trophic level, indicating a reduction of the efficiency of energy transfer in more eutrophic systems. Size spectra from nitrogen enriched streams had elevated intercepts and increased residual variance was noted in perturbed watersheds. The current study has provided baseline data for stream communities within and outside of Gatineau Park and in doing so it is clear that Gatineau Park streams are in good, but variable health. In order to monitor change, protect diversity within the Park, and mitigate further stress to these streams, it may be useful to use aspects of the size spectra such as normalized density and the residual variance.

Periphyton communities are often the dominant primary producers and an important energy source to higher trophic levels in rivers and streams. Empirical models of periphyton biomass (chlorophyll a) that have high predictive power are generally lacking. The goal of this research was to assess and improve the predictability of periphyton in rivers. A historical river monitoring data set from Alberta showed that, in general, land use in the drainage basin was a good surrogate for instream nutrient concentrations in regression models of periphyton biomass. Land use explained up to 34% of the variability in chlorophyll a whereas models based on instream nutrients explained up to 24% of the variability in chlorophyll a . A field study showed that bacterial abundance in periphyton explained an additional 26 to 29% of residual variance of chlorophyll a, after taking nutrients into account. The relationships between algal and bacterial abundance and production estimates were positive, suggesting bacteria and algae coexist in a mutually dependent association. The sampling design for bacteria in the field study was based on the relationships between sample means and variances of published bacterial abundance and production data. The number of replicates needed to sample periphytic bacterial abundance and production was determined from these relationships. A meta-analysis of published periphyton regression models was used to evaluate model predictive power. Once corrected for the number of observations, terms, and sampling replicates in the models, predictive power of periphyton models has not improved over the last 30 years. Geographic extent of the study area and the type of predictor variables used also had almost no effect on predictive power. The theoretical limit of model precision has approximately been reached for models predicting temporally averaged periphyton biomass. In contrast, residual variances of models predicting instantaneous mean chlorophyll a were, on average, 4.5 times higher than theoretical pure error. Precision of temporal mean models will only be improved by obtaining more precise estimates of mean chlorophyll a. Models that predict instantaneous chlorophyll a may be made more precise by including variables that better reflect the recent site history.

As urban populations increase globally, the conversion of natural land cover to anthropogenic land use is on the rise. Diffuse perturbations that accompany such land use changes are considered a chief threat to headwater streams, which provide habitat for diverse faunas of macroinvertebrates, fish, and amphibians and are important sources of sediments, nutrients, and organic matter for downstream systems. Although environmental assessments of stream ecosystems generally measure responses to these perturbations as shifts in community structures, functional rather than structural properties may provide a better indicator of land use impacts since measures of ecosystem function can reveal more about the mechanisms that alter running water ecosystems. The primary objective of this thesis was to quantify stream productivity along a gradient of forest cover and assess bottom-up and top-down mechanisms for observed patterns in stream productivity. In the first chapter, I quantified algal, invertebrate, and fish biomass, the single best correlate of productivity, at 38 first-third order streams along a forest cover gradient at 55 spatial scales and determined that subcatchment forest cover was consistently the best predictor of biomass at all trophic levels in terms of the variation explained by forest cover and the unexplained residual variance. When considering the independent effects of forest cover on biomass at three coarse scales, subcatchment forest cover was the only significant predictor of biomass, suggesting that the common practice of maintaining riparian buffers is an ineffective management practice for protecting stream function from catchment-wide impacts. In the second chapter, I developed two time-saving approaches for processing benthic invertebrate samples that reduced laboratory processing time by 37% and 82% and provided production estimates that were strongly related to production estimates obtained from standard processing techniques. These time-saving approaches were used in chapter 3 so that I could measure algal, invertebrate, and fish production at 12 sites along a gradient of subcatchment forest cover, which I determined to be the most influential spatial scale on biomass in chapter 1. Total nitrogen was a significant bottom-up force that explained 34-83% of the variation in stream productivity at all three trophic levels. However, consumer production to consumption ratios indicated that predators and grazers can potentially deplete their food sources, and that top-down forces are therefore also important contributors to observed patterns in productivity and biomass along the forested gradient.

The impacts of waterpower management on select fish in the Ottawa River are examined in this study and are presented in four manuscripts: (1) a systematic review of published literature on water power management; (2) examining potential causes of variation in lake sturgeon abundance among river reaches; (3) community structure and abundance among water management regimes; and (4) evaluating the predictive power of a lake sturgeon habitat suitability model at multiple scales. (1) Three meta-analyses were completed: (i) macroinvertebrate abundance is lower in zones or areas that have been dewatered due to water fluctuations or low flows (overall effect size -1.64 (-2.51/-0.77; 95% CI)); (ii) a hypolimnetic draw reduces the abundance of aquatic communities (overall effect size -0.84 (-1.38/-0.33; 95% CI)) and macroinvertebrates (overall effect size -0.73 (-1.24/-0.22; 95% CI)) downstream of a dam; (iii) altered flows has a negative effect on the abundance fluvial specialists (-0.42 (-0.81/-0.02; 95% CI) but does not affect habitat generalists (overall effect size -0.14 (-0.61/0.32; 95% CI)). (2) Three main stressors were examined to explain variation in lake sturgeon abundance among river reaches: contaminants, commercial harvest and water power management. Results were consistent with water power management impeding lake sturgeon recovery: abundance was significantly greater in unimpounded river reaches; growth rates were significantly greater in impounded reaches suggesting food was not limiting; and size/age structure in managed reaches was skewed to larger, older fish suggesting recruitment was impaired. (3) Variation in community structure and abundance was assessed for 11 fish species among three water management regimes: winter reservoirs, run-of-the-river and unimpounded. Littoral zone benthivores were significantly lower in abundance in winter reservoirs whereas species that are planktivorous for portions of their life were significantly greater. Lake sturgeon was the only fast water spawning species affected in run-of-the-river reaches. (4) The predictive power of a lake sturgeon habitat suitability model was assessed by paired net sets in good (habitat suitability index (HSI)>0.6) and poor (HSI<0.3) habitat. Lake sturgeon catch-per-unit-effort (CUE) was significantly greater for adults and juveniles in good habitat however, predictive power of the model was low (r2 < 0.18). Standard index netting lake sturgeon CUE corresponded with overall foraging component which suggests the model is a good predictor of sturgeon abundance at multiple scales.

Contemporary and Phanerozoic Patterns of Global Biodiversity is a global study testing theories explaining variation in taxic richness, and drawing parallels between contemporary- and paleodiversity. The unifying idea of the thesis is that most of the variation in contemporary and past richness depends on analogous factors: (1) sampling effects, (2) richness-energy relationships, and (3) geometric considerations pertaining to isolation and fragmentation. We present a predictive model of global island biogeography, uniting classic ideas of the Equilibrium Theory of Island Biogeography (ETIB) with the Species-Energy Theory (Chapter 1). We quantify insular isolation and propose an equation to measure the isolation reducing effects of neighboring islands. In Chapter 2 we extend our model to continental parcels of land. We show that area-climate interactions are strong both on continents and on islands. Notwithstanding, species-area slopes do not depend either on distance-based or on historical isolation, in apparent contradiction with some predictions of the ETIB. Thus, broad-scale patterns of diversity on islands and continents are commensurable. In Chapter 3 we evaluate the completeness of the global Phanerozoic record of continental (terrestrial and freshwater) metazoans, aiming to use these data in subsequent chapters. Results of this analysis suggest the relative completeness of the continental fossil record is not markedly inferior to that of the marine fossil record at the taxonomic level of the family and at the stratigraphic level of the stage. The exponential diversification curve of suprageneric continental taxa is unlikely to be the result of rock bias. Chapter 4 adopts a multivariate approach to model variation in fossil diversity. Our main finding is that turnovers of metazoan families are strongly correlated with the number of magnetic polarity reversal records per stratigraphic stage over the Phanerozoic. This suggests that either variation in fossil richness and reversal intensity are both strongly influenced by preservation bias, or that extinction and origination rates have a strong and quantifiable perturbation regime. In Chapter 5 we classify stratigraphic stages into alternating extinction and origination cycles to test for supercyclicity. Paleomagnetic and fossil data both support this hypothesis over the Lower Paleozoic, the Upper Paleozoic and the Mesozoic.

The objectives of this study were to investigate the effects of multiple bacterial species and nutrient injection on phenanthrene degradation, cell elution, and cell distribution within a column system. Three indigenous, phenanthrene degrading species, Actinobacter junii, Pseudamonas oleovorans, and Methylbacterium sp., were used. Characterization studies were conducted to determine the relative cell yield and cell hydrophobicity. Transport studies were conducted in a saturated column system. Results show single bacterial species produce relatively stable cell elution and phenanthrene biodegradation curves. In addition, it appears that cell density within the column may be dependent on cell hydrophobicity and possibly column carrying capacity. Conversely, phenanthrene degradation and cell elution was variable in the presence of multiple bacterial species. Results indicate that synergistic and antagonistic interactions occurred among the species. Injection of an alternative carbon source (R2B) enhanced phenanthrene degradation only in the single species column containing Actinobacter junii. It caused increased cell elution in Actinobacter junii and Pseudamonas oleovorans single species columns. No increase in degradation was found in multi-species columns injected with R2B. These studies illustrate that the dynamics of complex microbial communities should be considered when evaluating contaminant biodegradation and transport in subsurface systems. The findings presented here have important implications for representation of biodegradation in computer modeling and in the design of bioremediation strategies.