The response of phytoplankton to nutrient loading and planktivorous fish additions was investigated in large experimental enclosures installed in lac Croche at the Station de Biologie des Laurentides de l'Universite de Montreal. During the summer period (July and August), phytoplankton biomass significantly increased with fertilization and the presence of fish in the deep, thermally stratified, enclosures. The presence of fish increased algal biomass, particularly under the fertilized conditions. In the shallow, thermally mixed enclosures, total phytoplankton biomass was significantly affected by fertilization but not by the presence of fish. The taxonomic composition at both the species and the division levels differed in all treatments. In the deep stratified enclosures, both the addition of nutrient (p .001) and fish (p =.003) favored a large increase in Chlorophyta biomass which was composed of 1 or 2 species. Pyrrhophytes increased slightly with the presence of fish. In the shallow enclosures, nutrient additions significantly affected the division composition of total algal biomass, whereas presence of fish did not. The presence of planktivorous fish influenced phytoplankton community structure by modulating total algal biomass, the distribution of biomass according to size classes and the overall taxonomic composition. The impacts of fish were more numerous in the deep than in the shallow enclosures. Thermal stratification may thus be an important factor influencing the impact of planktivorous fish on phytoplankton. (Abstract shortened by UMI.)

Twelve streams of the Ottawa-Hull area were sampled between May 9, 1990 and August 1, 1991 to examine how benthic macroinvertebrate biomass and size spectra varied in relation to periphyton biomass. Among sites, non-filter-feeding macroinvertebrate biomass is correlated positively to periphyton biomass. This relation is defined by the regression equation: Log(BNF) = 1.64 + 0.41*Log(peri) (r$\sp2$ = 0.50, F = 9.9, p = 0.01, S$\sp2$ = 0.15, n = 12). The logarithm of the mean annual periphyton biomass explains 50% of the variance encountered in the logarithm of the mean annual non-filter-feeding macroinvertebrate biomass. This figure goes up to 70% if only samples collected between mid-July and mid-August are considered. The normalized size spectra of all benthic macroinvertebrates from city streams is shifted to the right along the abcissa compared to the normalized size spectra of other streams. This shift indicates a greater individual average weight for macroinvertebrates in city streams compared to macroinvertebrates from Gatineau Park. This shift is not solely related to the eutrophic level of the stream. Other studies in controlled environments will be needed to determine the cause of this shift.

The red-throated loon (Gavia stellata) is a circumpolar species. The purpose of this study was to examine adaptations of this species to environmental conditions in the high arctic. I found that the phenology of the loons in my study area was shorter than that of red-throated loons further south, and that this population tended to select nesting ponds much larger than in the south. I also found that pairs nesting near the ocean (within 9.0 km) were able to rear larger broods (2 chicks vs 1 chick) than pairs nesting far from the ocean (9.0-13 km). Finally, I found that pond size did not influence melting rate but that large ponds took longer to freeze. I also observed that 16% of loons moved their chick(s) to larger ponds when natal ponds began to freeze over. This supports the view that the use of large nesting ponds by these loons is an adaptation to the cold climate. (Abstract shortened by UMI.)

No description available.

No description available.

Although predation is the major factor responsible for nest losses in birds, predation patterns in marshes and adjacent uplands are poorly known. This study examined the factors affecting nest predation on marsh-nesting birds found in areas affected by different intensities of human disturbance (urban, agricultural, natural). Artificial nests simulating waterfowl and passerine nests were used for that purpose. The nests, located along transects running from the center of the marshes to the adjacent upland habitats, were set up in the months of May, June and July 1989 and 1990. Eggshell remains were used to identify the predators. Predation was higher on passerine than on waterfowl nests and on nests located in upland habitats. Deep water prevented mammalian predators from foraging deep in the marsh. Predation rates in urban and natural areas were high while nests located in agricultural areas suffered lower predation. Nest camouflage was generally unimportant in reducing chances of predation. Mammals were responsible for most of the predation events. Differences in density and/or diversity of predators in urban, agricultural and natural areas were the main causes of different predation patterns observed in the three areas.

My research focused on (1) describing the vegetation of Canadian alvars, (2) examining ecological patterns in this system and (3) measuring the intensity of plant competition which occurs there. In chapter 1, ordination and classification techniques were used to describe alvar vegetation at two scales: local (within one alvar) and regional (among 4 alvar sites). At the local scale, changes in species composition corresponded to changes in soil depth and biomass. At the regional scale species composition was related primarily to geographic location--the southern site was distinct from the eastern and northern sites. Relationships between species distributions and soil depth and biomass could also be detected at this scale. In chapter 2, I examined the empirical relationships among important alvar state variables (soil depth, biomass and species richness) in order to test the predictions of the species richness model (forwarded by Grime and others) in a low biomass terrestrial system, and compare the pattern among 4 alvar sites. At all sites biomass was positively correlated with soil depth, suggesting that the soil depth gradient represents a gradient of increasing below-ground resources and decreasing above-ground resources. At each alvar, species richness varied curvilinearly with both biomass and soil depth; maximum species richness occurred at intermediate levels of these factors, as predicted by species richness models. However, the predicted decline in species richness at higher levels of biomass or soil depth was evident at only one site. In chapter 3, I present the results of a field experiment which measured the intensity of total, root and shoot competition along the gradient of soil depth at one alvar. Over all sites, the intensities of total and root competition were greater than zero; that of shoot competition was not. I suggest a model which relates the various effects of competition and of mutualism to the biomass range of interest. (Abstract shortened by UMI.)

The general objective of this thesis was to study the interactions between AM fungi and turfgrass species in the presence of natural or chemical fertilizers. We had hypothesized that this symbiosis is possible, functional and compatible in the presence of fertilizers. To test this hypothesis, we have set three objectives: (i) estimate colonization rate for three AM fungi: Glomus intraradices, G. mosseae and G. aggregatum on two turfgrass species Poa pratensis L. cv Park and Agrostis stolonifera L. cv Penncross; (ii) evaluate the compatibility between natural and chemical fertilizers and AM fungi, and (iii) compare the effects of AM colonization on different physiological parameters in turfgrass. This study was performed over a period of 17 months between June 1996 and October 1997 on two different sites, Clarence Creek and the Central Experimental Farm (Ottawa), respectively low (46ppm) and high (270ppm) in soil phosphorus status. In total, 96 plots were randomly distributed on each sites. To investigate the dynamic of turfgrass community, we performed a greenhouse experiment following the same experimental design as in the field. (Abstract shortened by UMI.)

The nitrogen cycle is the key process in marine mesocosms. This cycle involves mineralizing and nitrifying bacteria which are preyed upon by meiofauna. Large populations of meiofauna (20--40 g·m--2 in the top 2.5 cm of sand) are found in the sand filters of the St. Lawrence marine mesocosm at the Montreal Biodome. A nitrogen budget was prepared for 1995 to quantify the importance of mineralization and nitrification in the mesocosm. Of the input nitrogen, 15% to 31% were transformed into ammonia through bacterial mineralization, and 71% passed through the nitrification process. Filtration and cleaning removed 20% of the input nitrogen, and only 5% were transformed into animal biomass. Experiments were then conducted in microcosms to see how filter meiofauna affect the apparent mineralization rate (AMR) and apparent nitrification rate (ANR). Microfauna and meiofauna abundances were manipulated to quantify their effect on AMR and ANR by linear regression analysis. Harpacticoid copepods dominated in number (87%) and biomass (98%). Neither AMR nor ANR were related to microfauna densities. AMR was inversely related to meiofaunal mass. Particle size affected the effect of meiofauna whereas the C:N ratio affected the AMR0, i.e. the AMR in absence of meiofauna. Meiofauna biomass had no effect on nitrition but had a variable effect on nitration. This effect varied from --20% to +571%. The slope of the ANR vs Meiofauna regression was related to the particulate organic nitrogen content of the sediment. Meiofauna seems to exert a direct action by preying on nitrifiers and an indirect action by reducing the mineralizing bacteria through grazing. Results suggest that meiofauna plays an important role in water treatment by reducing the production of nitrate and more rapidly eliminating nitrite which are toxic to fish.

Rivers have long been considered physically driven ecosystems, and the physical environment varies between streams and large rivers. The objective of this thesis was to investigate the importance of river size for suspended algal development. In the first chapter, I examine the seasonal variability of discharge, water chemistry, and suspended algal abundance to evaluate the accepted (but rarely tested) assumption that streams are more temporally variable than large rivers. Temporal variance of discharge and water quality parameters was determined for 114 stations on 104 rivers in eastern United States with data measured between May to October of 1978 by the U.S. National Stream Water Quality Accounting Network (NASQAN). In the second chapter, I examine suspended algal communities in 46 rivers in Ontario and Quebec to determine the influence of river size and nutrient status on algal biomass and size structure. Chlorophyll a estimates of size structure were compared with greatest axial linear dimension estimates (GALD) determined by microscope enumeration of suspended algal samples collected in 31 Ontario and Quebec rivers. (Abstract shortened by UMI.)