Yersinia enterocolitica are potentially pathogenic bacteria transmitted through the fecal oral route. Typical disease symptoms include those associated with gastrointestinal disease, although infection can also lead to more serious and invasive illnesses, particularly in sensitive populations. Previous surveys have detected Y. enterocolitica in surface water in various parts of the world, and studies have found drinking untreated water to be a possible risk factor for Y. enterocolitica infection. Methods available for the detection of Y. enterocolitica have been developed primarily for food and clinical samples and have not been tested extensively with water. More commonly used methods include culture based isolation of Yersinia spp. and polymerase chain reaction (PCR) based detection of Y. enterocolitica. Reports suggest that culture based methods available for the isolation of Y. enterocolitica may not be effective for environmental samples. Strain isolation using culture based methods is important, so that further subtyping information can be obtained for epidemiological investigations. In contrast, PCR based detection is more rapid, of higher throughput, can be highly specific and can target pathogenic strains within a species. The overall objective of this work was to evaluate culture based and PCR based methods for the detection of Y. enterocolitica in water, and to examine its prevalence in the Grand River watershed in Southwestern Ontario, Canada. Surface water in this watershed is used to provide all or part of the drinking water for approximately 500,000 people, as well as for recreational purposes. It is also one of the most heavily impacted watersheds in Canada by both agricultural and urban activities. Culture based studies compared two selective agars and four enrichment broths. Results showed that Cefsulodin Irgasan Novobiocin (CIN) agar and modified tryptic soy broth (mTSB) had greater potential for recovering Y. enterocolitica from surface water. Consequently, enrichment in mTSB followed by growth on CIN agar was used to isolate Yersinia from the Grand River. Yersinia strains were isolated from 52 out of 200 (26 %) surface water samples collected over a 17 month period. No seasonal trends were observed in isolation rates. Species isolated were typically considered to be non pathogenic species, although recent evidence suggests they may have potential virulence to humans. The majority of these strains have been found by other groups in surveys of aquatic environments. PCR methods developed targeted two Y. enterocolitica virulence genes: the ail gene, located in chromosomal DNA; and the yad A gene, located on a virulence plasmid. In surface water collected from the Grand River, the ail gene target was detected in 121 samples out of 319 (38 %) over a 29 month period and the yadA gene target was detected in 44 samples out of 206 (21 %) over a 20 month period. Both genes were detected more frequently when the water temperatures were colder. PCR-based studies conducted were quantitative, which has not previously been done with water samples. The median and maximum concentrations in samples positive for the ail gene were 40 and 2,000 cells/100 mL, and in samples positive for the yadA gene were 32 and 3,276 gene copies/100 mL, respectively. Overall results demonstrated that culture based methods are less sensitive than PCR based detection methods for specific detection of pathogenic Y. enterocolitica, suggesting that previous culture based surveys may have underestimated their potential prevalence. Furthermore, potentially pathogenic Y. enterocolitica may be present in the Grand River watershed. While Y. enterocolitica is relatively easily inactivated by traditional disinfection methods used in drinking water treatment processes, it is possible their presence poses a concern for recreational users and individuals drinking untreated water. This study suggests that further investigation is necessary to evaluate possible health risks associated with the occurrence of potentially pathogenic Y. enterocolitica in the Grand River. This work assists with the development of methods and information gathering for an emerging waterborne pathogen that has not been surveyed in the Grand River watershed, nor quantitatively surveyed in any water previously. Findings provide important information for drinking water providers and public health investigations.

The first objective of this thesis was to predict the future success of selected tree species under low (B1, 550 CO2 ppm) and moderate (A1B, 720 CO2 ppm) climate change scenarios as defined in the Special Report on Emissions Scenarios (SRES). This was accomplished through the creation of radial-growth forecasts for eastern hemlock (Tsuga Canadensis (L.) Carr.), sugar maple (Acer saccharum L.), white spruce (Picea glauca (Moench.) Voss), and white pine (Pinus strobus L.) in the Grand River Watershed of Ontario, Canada. The forecasts were founded on historic growth-climate relationships between standardized regional dendrochronologies for each species and past climate data from the Guelph OAC weather station. These species-specific growth-climate relationships were then extended to 2100 using modeled climate data from the Third Generation Coupled Global Climate Model (CGCM3) to project radial-growth under both emissions scenarios. Results indicated that eastern hemlock radial-growth will remain stable throughout the 21st-century, sugar maple and white spruce growth will start to decline, and white pine growth will increase. While the radial-growth forecasts were limited by the length of the past climate data, the accuracy of the modeled climate data, and the number and type of variables used in the forecast model, the results were statically significant and strongly supported in the literature. The second thesis objective was to assess the potential impact of the radial-growth forecasts on environmental planning policy and forest management strategy in the Grand River Watershed. Examples of how the forecasts could influence basic management strategies in the watershed were provided to display the conceptual linkages between the results and policy formulation. Next, the radial-growth forecasts were presented to four forest managers working in the watershed to gage the practical implications, perceptions and limitations of the radial-growth forecasting method. While the managers found the radial-growth forecasts interesting, they also noted that the results were of limited use since they could not account for other factors important to the future success of the study species, such as seedling dispersal and establishment rates, as well as the potential effects of pathogens, insects and invasive species. Therefore, it was recommended that future research should work to extrapolate the results of the radial-growth forecasts to other tree species and types in the region, as well as incorporate more variables into the models, so that more accurate and applicable growth projections could be constructed in the watershed.

A numerical modelling analysis of climate change’s precipitation effects on the long-term, averaged surface and subsurface hydrology of the Grand River Watershed (GRW) was undertaken in order to assess possible areas of concern for decision makers in the water management sector. The physically-based, fully-integrated and variably-saturated 3-D surface-subsurface numerical simulator, HydroGeoSphere, was used to drive five mid-21st century climate change scenarios, developed from multiple general circulation models. Calibration involved altering measured and literature-derived hydraulic conductivity and precipitation distribution estimates, resulting in very good matching between observed and simulated long-term average surface flow at all gauge stations. Subsurface head results, too, matched observed heads quite well, though groundwater linkage to neighbouring watersheds was not included. When groundwater linkage to neighbouring watersheds was allowed, via regional Dirichlet boundary conditions used in a parent study, groundwater throughput was deemed to be unrealistic. All but one of the climate change scenarios caused an increase in both river discharge and water table elevation, with the greatest climate perturbations causing the greatest increases. For Scenario 1 (5% less precipitation than the 1960-to-1999 average), percentage discharge changes averaged -15% over all gauge stations. For the other scenarios (more precipitation than average), the inter-scenario discharge response ranged from approximately +12% to +59%. In general the range of inter-subcatchment response was greater than was the range for intra-subcatchment response; the greatest percentage response was consistently in the Speed River subcatchment, while the least was consistently in the Nith and Conestogo subcatchments. The exception was the application of less-than-average precipitation to the Grand River, whose gauge stations reported percentage changes in discharge that varied more substantially from one another. Subsurface hydrology reacted to the climate change scenarios in much the same manner as did the surface hydrology, with all climate change scenarios associated with a precipitation increase unsurprisingly resulting in higher total hydraulic heads throughout the entire domain. Specifically, the minimum and maximum mean head increases among the climate change scenarios were 0.41 m and 1.25 m respectively, while the only decrease was an average of 0.55 m. Similarly, the depth from the ground surface to the water table decreased in most scenarios, the maximum water table rise being 1.08 m and the minimum 0.36 m. When precipitation was allowed to decrease by 5% relative to the long-term average, the average water table elevation decreased by 0.48 m. However the water table’s pattern of high and low points remained very much the same among all climate change scenarios, suggesting that basin-wide groundwater flow patterns may not be among the hydrological measures most sensitive to climate change. Groundwater recharge, like almost all other components of the water budget, changed in linear proportion to the climate forcing and in agreement with GRW recharge estimates developed by others. Evapotranspiration, which met potential evapotranspiration in all scenarios due to the constant application of precipitation, was the only element of the water budget that did not increase, even though the water table was elevated closer to the rooting zone by most of the climate scenarios. On a smaller scale, changes in flow patterns may well be expected, given that zones of infiltration were observed to intensify with most of the climate forcing.

Yersinia enterocolitica are potentially pathogenic bacteria transmitted through the fecal oral route. Typical disease symptoms include those associated with gastrointestinal disease, although infection can also lead to more serious and invasive illnesses, particularly in sensitive populations. Previous surveys have detected Y. enterocolitica in surface water in various parts of the world, and studies have found drinking untreated water to be a possible risk factor for Y. enterocolitica infection. Methods available for the detection of Y. enterocolitica have been developed primarily for food and clinical samples and have not been tested extensively with water. More commonly used methods include culture based isolation of Yersinia spp. and polymerase chain reaction (PCR) based detection of Y. enterocolitica. Reports suggest that culture based methods available for the isolation of Y. enterocolitica may not be effective for environmental samples. Strain isolation using culture based methods is important, so that further subtyping information can be obtained for epidemiological investigations. In contrast, PCR based detection is more rapid, of higher throughput, can be highly specific and can target pathogenic strains within a species. The overall objective of this work was to evaluate culture based and PCR based methods for the detection of Y. enterocolitica in water, and to examine its prevalence in the Grand River watershed in Southwestern Ontario, Canada. Surface water in this watershed is used to provide all or part of the drinking water for approximately 500,000 people, as well as for recreational purposes. It is also one of the most heavily impacted watersheds in Canada by both agricultural and urban activities. Culture based studies compared two selective agars and four enrichment broths. Results showed that Cefsulodin Irgasan Novobiocin (CIN) agar and modified tryptic soy broth (mTSB) had greater potential for recovering Y. enterocolitica from surface water. Consequently, enrichment in mTSB followed by growth on CIN agar was used to isolate Yersinia from the Grand River. Yersinia strains were isolated from 52 out of 200 (26 %) surface water samples collected over a 17 month period. No seasonal trends were observed in isolation rates. Species isolated were typically considered to be non pathogenic species, although recent evidence suggests they may have potential virulence to humans. The majority of these strains have been found by other groups in surveys of aquatic environments. PCR methods developed targeted two Y. enterocolitica virulence genes: the ail gene, located in chromosomal DNA; and the yad A gene, located on a virulence plasmid. In surface water collected from the Grand River, the ail gene target was detected in 121 samples out of 319 (38 %) over a 29 month period and the yadA gene target was detected in 44 samples out of 206 (21 %) over a 20 month period. Both genes were detected more frequently when the water temperatures were colder. PCR-based studies conducted were quantitative, which has not previously been done with water samples. The median and maximum concentrations in samples positive for the ail gene were 40 and 2,000 cells/100 mL, and in samples positive for the yadA gene were 32 and 3,276 gene copies/100 mL, respectively. Overall results demonstrated that culture based methods are less sensitive than PCR based detection methods for specific detection of pathogenic Y. enterocolitica, suggesting that previous culture based surveys may have underestimated their potential prevalence. Furthermore, potentially pathogenic Y. enterocolitica may be present in the Grand River watershed. While Y. enterocolitica is relatively easily inactivated by traditional disinfection methods used in drinking water treatment processes, it is possible their presence poses a concern for recreational users and individuals drinking untreated water. This study suggests that further investigation is necessary to evaluate possible health risks associated with the occurrence of potentially pathogenic Y. enterocolitica in the Grand River. This work assists with the development of methods and information gathering for an emerging waterborne pathogen that has not been surveyed in the Grand River watershed, nor quantitatively surveyed in any water previously. Findings provide important information for drinking water providers and public health investigations.

The first objective of this thesis was to predict the future success of selected tree species under low (B1, 550 CO2 ppm) and moderate (A1B, 720 CO2 ppm) climate change scenarios as defined in the Special Report on Emissions Scenarios (SRES). This was accomplished through the creation of radial-growth forecasts for eastern hemlock (Tsuga Canadensis (L.) Carr.), sugar maple (Acer saccharum L.), white spruce (Picea glauca (Moench.) Voss), and white pine (Pinus strobus L.) in the Grand River Watershed of Ontario, Canada. The forecasts were founded on historic growth-climate relationships between standardized regional dendrochronologies for each species and past climate data from the Guelph OAC weather station. These species-specific growth-climate relationships were then extended to 2100 using modeled climate data from the Third Generation Coupled Global Climate Model (CGCM3) to project radial-growth under both emissions scenarios. Results indicated that eastern hemlock radial-growth will remain stable throughout the 21st-century, sugar maple and white spruce growth will start to decline, and white pine growth will increase. While the radial-growth forecasts were limited by the length of the past climate data, the accuracy of the modeled climate data, and the number and type of variables used in the forecast model, the results were statically significant and strongly supported in the literature. The second thesis objective was to assess the potential impact of the radial-growth forecasts on environmental planning policy and forest management strategy in the Grand River Watershed. Examples of how the forecasts could influence basic management strategies in the watershed were provided to display the conceptual linkages between the results and policy formulation. Next, the radial-growth forecasts were presented to four forest managers working in the watershed to gage the practical implications, perceptions and limitations of the radial-growth forecasting method. While the managers found the radial-growth forecasts interesting, they also noted that the results were of limited use since they could not account for other factors important to the future success of the study species, such as seedling dispersal and establishment rates, as well as the potential effects of pathogens, insects and invasive species. Therefore, it was recommended that future research should work to extrapolate the results of the radial-growth forecasts to other tree species and types in the region, as well as incorporate more variables into the models, so that more accurate and applicable growth projections could be constructed in the watershed.

A numerical modelling analysis of climate change’s precipitation effects on the long-term, averaged surface and subsurface hydrology of the Grand River Watershed (GRW) was undertaken in order to assess possible areas of concern for decision makers in the water management sector. The physically-based, fully-integrated and variably-saturated 3-D surface-subsurface numerical simulator, HydroGeoSphere, was used to drive five mid-21st century climate change scenarios, developed from multiple general circulation models. Calibration involved altering measured and literature-derived hydraulic conductivity and precipitation distribution estimates, resulting in very good matching between observed and simulated long-term average surface flow at all gauge stations. Subsurface head results, too, matched observed heads quite well, though groundwater linkage to neighbouring watersheds was not included. When groundwater linkage to neighbouring watersheds was allowed, via regional Dirichlet boundary conditions used in a parent study, groundwater throughput was deemed to be unrealistic. All but one of the climate change scenarios caused an increase in both river discharge and water table elevation, with the greatest climate perturbations causing the greatest increases. For Scenario 1 (5% less precipitation than the 1960-to-1999 average), percentage discharge changes averaged -15% over all gauge stations. For the other scenarios (more precipitation than average), the inter-scenario discharge response ranged from approximately +12% to +59%. In general the range of inter-subcatchment response was greater than was the range for intra-subcatchment response; the greatest percentage response was consistently in the Speed River subcatchment, while the least was consistently in the Nith and Conestogo subcatchments. The exception was the application of less-than-average precipitation to the Grand River, whose gauge stations reported percentage changes in discharge that varied more substantially from one another. Subsurface hydrology reacted to the climate change scenarios in much the same manner as did the surface hydrology, with all climate change scenarios associated with a precipitation increase unsurprisingly resulting in higher total hydraulic heads throughout the entire domain. Specifically, the minimum and maximum mean head increases among the climate change scenarios were 0.41 m and 1.25 m respectively, while the only decrease was an average of 0.55 m. Similarly, the depth from the ground surface to the water table decreased in most scenarios, the maximum water table rise being 1.08 m and the minimum 0.36 m. When precipitation was allowed to decrease by 5% relative to the long-term average, the average water table elevation decreased by 0.48 m. However the water table’s pattern of high and low points remained very much the same among all climate change scenarios, suggesting that basin-wide groundwater flow patterns may not be among the hydrological measures most sensitive to climate change. Groundwater recharge, like almost all other components of the water budget, changed in linear proportion to the climate forcing and in agreement with GRW recharge estimates developed by others. Evapotranspiration, which met potential evapotranspiration in all scenarios due to the constant application of precipitation, was the only element of the water budget that did not increase, even though the water table was elevated closer to the rooting zone by most of the climate scenarios. On a smaller scale, changes in flow patterns may well be expected, given that zones of infiltration were observed to intensify with most of the climate forcing.