1. Large rivers in the central United States experience a variety of ecological interactions. Both short- and long-term temporal changes affect both fish and zooplankton communities in rivers. Variation in diel patterns of predation could affect differences of behavior in prey. Long term temporal changes across seasons can lead to different areas of rivers providing growth or costing energy. Spatial changes also affect habitat quality and behaviors of the aquatic organisms within large rivers. Spatial variation can be dynamic and occur between longitudinally connected habitats or vary through different sections of the reach and can also change seasonally. 2. Invasive species have major economic and ecological impacts. In aquatic ecosystems, plankton are the base of the food web and a planktivore invader can cause major ecological disruptions. The Wabash River is a large free flowing river with a variable hydrology, which causes dramatic changes in habitat availability through time. Therefore, habitat usage, quality, and availability may differ for native and invasive species through time. The objectives of this study were to assess how habitat quality changes in a large, flashy, free-flowing river over time, and to compare the quantity and overlap of quality macrohabitat for invasive and native fish. Growth rate potential (GRP) was used to quantify habitat quality for Silver Carp (Hypophthalmichthys molitrix; invasive) and Gizzard Shad (Dorosoma cepedianum; native). To calculate GRP, bioenergetic models were used with monthly observations of food abundance (zooplankton, phytoplankton, and detritus), water temperature, dissolved oxygen, and water velocity of various habitat types from the Wabash River’s confluence with the Ohio River upstream to Terre Haute, Indiana, approximately 215 river miles. Negative GRP for both Silver Carp and Gizzard Shad occurred during spring and most of summer. However, in the fall, the majority of the river had a positive GRP for both species. Reduced flows and increases in food abundance were the cause for the switch from negative to positive GRP. There was a high degree of spatial overlap in positive GRP areas between Silver Carp and Gizzard Shad, suggesting that there is a high degree of competition. Acoustically tagged Silver Carp selected for areas predicted to have a positive GRP from the model presented in the study. This model showed that regulation of flows in hydrologically altered systems could potentially lower the impact of Silver Carp in those systems. With Silver Carp selecting for predicted habitats, a bioenergetics model can be used to inform commercial fishers where to focus effort to maximize catch.3. Temporal changes in rivers tend to be quantified at broad seasonal scales. However, river conditions including water flow, water temperature, dissolved oxygen, and availability of zooplankton prey may change daily. In large rivers, these conditions may change in a diel fashion between backwater and main channel habitats. To assess diel changes in the Illinois River, zooplankton, paired with mobile hydroacoustic surveys for fish were sampled from three main channel and three off-channel sites in the LaGrange reach every four hours in a twenty-four-hour period at different depths between October 22 and 26, 2018. Water depth, water temperature, water velocity, and fish (planktivore and nonplanktivore) densities were quantified during each sampling time. Water temperature and water velocity differed between the two habitat types. Water temperature was higher and water velocity was lower in the off-channel. Water temperature was colder in the morning in both the main channel and off-channel. Water velocity did not change throughout the day. The majority of the zooplankton community was comprised of rotifers which were less mobile than larger taxa and did not migrate. When evaluating the more mobile, large body zooplankton, time and depth affected copepod density in off-channel environments, where the deeper depth had a higher copepod density than the shallower depth. In contrast, only time of day affected off-channel cladoceran density, where densities increased at night. Fish did not exhibit diel patterns and the off-channel habitat had a higher density. The most likely mechanism behind the diel movements of the large bodied zooplankton was predation by planktivorous fish. The present study suggests that connections to the off-channel habitats in large rivers will promote quality habitat for both zooplankton and fish.

Large scale ecosystem restoration is an important societal issue because significant risks, costs, and benefits can accrue on large landscapes. It is important to understand baseline ecosystem conditions, existing condition, and to the extent possible estimate ecosystem response to alternative management scenarios. Incorporating ecosystem process and function into restoration planning and implementation will make ecosystem restoration projects sustainable. The Upper Mississippi River System is an excellent case study for such issues because it is an important, multiple-use ecosystem with significant ongoing investment in ecosystem, agri-system, and navigation system management. Large-scale geomorphology, hydrology, and land cover information was compared among presettlement, contemporary, and potential future reference conditions to examine ecosystem state and evaluate mechanisms responsible for ecosystem condition. The UMRS was scaled by physiographically similar characteristics into large floodplain reaches several hundred river miles in length, geomorphic reaches 50 to several hundred miles, and a mile-by-mile segmentation of the river floodplain extent. Ecologically relevant geomorphic classes were devised from existing data and evaluated by river reach to characterize presettlement geomorphology, and dams and levees were superimposed to reflect the altered hydrogeomorphology of the contemporary ecosystem. A pre- and post-impact Indicators of Hydrologic Alteration river stage analysis evaluated dam impacts, and pre-dam and post-dam aquatic habitat class distribution was compared. A floodplain inundation simulation analysis provided new information on the potential spatial distribution of frequent floods. Land cover data available for presettlement and modern reference periods were compared at several spatial scales. Multivariate analyses evaluated land cover characteristics among geomorphic reaches, as well as to assess the influence of hydrogeomorphic drivers on land cover for presettlement and contemporary reference periods. The objective of this research was to clearly delineate the divergence of environmental conditions among reference periods to evaluate which drivers need to be, and can be, altered to change ecosystem state. Hydrogeomorphic response to development indicates several restoration objectives that are appropriate system-wide and others that are best suited to specific river reaches. Similar data sources are available for much of the rest of the United States through the Public Land Survey and engineering surveys of any significant civil works projects.

Physical habitat for instream biota derives from a combination of stream system structural and hydraulic phenomena. Consequently, the quantity and quality of physical habitat is dynamic both over time and in space along the river, laterally, longitudinally and vertically. Its characterization through stream assessment and classification leads to a better understanding of factors that determine and limit habitat extent and quality. This thesis investigates the effects of space and time on nearshore aquatic habitat in the gravel reach of Fraser River, British Columbia by employing a large river, stage-adaptive habitat classification system. The distribution and abundance of habitat are spatially quantified at the reach scale (32 km), and temporally quantified through a period of about 60 years at several adjacent gravel bars (7 km), and at approximately 500 m3 s-1 increments in discharge during the declining limb of the flood hydrograph at two well-developed gravel bars. Of the ten habitat types evaluated, the bar edge habitat type is most abundant by length and number of units. However, its relative importance is reduced when weighted by fish-habitat association characteristics. Preferred habitat types (channel nook, eddy pool and open nook) are frequent and available to aquatic organisms, and most common at well-developed bars and in zones of equilibrium long-term sedimentation. Preferred habitat was at a maximum 30 years ago when major new bars developed and the thalweg shifted, effectively increasing the amount of bar shoreline and nearshore habitat. This increase is due to substantial change in river planform morphology following a 30-year period of large annual floods. However, amounts of habitat did not increase exclusively during periods of higher than average flows, or decrease exclusively during periods of lower than average flows. Instead, habitat abundance response to flow may occur with a two- or three-year lag. Short term changes in stage are critical to amount of preferred habitat. Optimal discharge for maximum preferred habitat vailability is in the range of approximately 2500 m3 s-1 to 4000 m3 s-1, which approximates long term mean flow. As flow increases, the proportion of preferred habitat compared with total bar shoreline decreases. Comparison with the 2006 flow duration curve shows that 15 – 30 % of discharges are optimal for maximum fish density and biomass. These discharges occurred during April 27 to May 17 and July 14 to August 7, 2006.

Currently our understanding of both system and phytoplankton metabolism in large rivers is somewhat limited. Knowledge of the metabolic balance in such systems is necessary not only for proper management of the river itself, but also for the lakes into which they discharge. The River Continuum Concept proposes that the deep, turbid waters of large rivers have a poor light climate which leads to heterotrophic conditions (respiration > photosynthesis) yet this idea has been challenged. Similarly, it has been predicted that phytoplankton growth in large rivers is limited to areas of unusually favourable light climate and water retention (e.g. margins, backwaters), but the evidence is limited. Through longitudinal and diel measurements of Chl a, nutrient concentrations, dissolved oxygen and stable oxygen isotopes it was shown in this study that the lower Grand River was autotrophic during the two successive summers but either balanced or heterotrophic in other seasons. This implies that large rivers such as the Grand can be a transition zone for nutrients and a phytoplankton source, depending on season. Experimental incubations to measure oxygen production under varying irradiance demonstrated that phytoplankton could indeed grow (i.e., achieve positive net production) in the main river channel. Comparison of system and plankton metabolic rates further indicated that the phytoplankton were responsible for the major portion of the system production, but much less of the respiration. Sediment oxygen demand probably accounted for much of the additional respiration, but interactions with marginal and upstream habitats was probably an additional influence on both consumption and production of oxygen. The results further showed that stable oxygen isotope dynamics did not conform to the steady state model commonly used to infer metabolic patterns from environmental isotope data. A non-steady model was more successful and largely supported independent assessments of metabolism.

Currently our understanding of both system and phytoplankton metabolism in large rivers is somewhat limited. Knowledge of the metabolic balance in such systems is necessary not only for proper management of the river itself, but also for the lakes into which they discharge. The River Continuum Concept proposes that the deep, turbid waters of large rivers have a poor light climate which leads to heterotrophic conditions (respiration > photosynthesis) yet this idea has been challenged. Similarly, it has been predicted that phytoplankton growth in large rivers is limited to areas of unusually favourable light climate and water retention (e.g. margins, backwaters), but the evidence is limited. Through longitudinal and diel measurements of Chl a, nutrient concentrations, dissolved oxygen and stable oxygen isotopes it was shown in this study that the lower Grand River was autotrophic during the two successive summers but either balanced or heterotrophic in other seasons. This implies that large rivers such as the Grand can be a transition zone for nutrients and a phytoplankton source, depending on season. Experimental incubations to measure oxygen production under varying irradiance demonstrated that phytoplankton could indeed grow (i.e., achieve positive net production) in the main river channel. Comparison of system and plankton metabolic rates further indicated that the phytoplankton were responsible for the major portion of the system production, but much less of the respiration. Sediment oxygen demand probably accounted for much of the additional respiration, but interactions with marginal and upstream habitats was probably an additional influence on both consumption and production of oxygen. The results further showed that stable oxygen isotope dynamics did not conform to the steady state model commonly used to infer metabolic patterns from environmental isotope data. A non-steady model was more successful and largely supported independent assessments of metabolism.

Physical habitat for instream biota derives from a combination of stream system structural and hydraulic phenomena. Consequently, the quantity and quality of physical habitat is dynamic both over time and in space along the river, laterally, longitudinally and vertically. Its characterization through stream assessment and classification leads to a better understanding of factors that determine and limit habitat extent and quality. This thesis investigates the effects of space and time on nearshore aquatic habitat in the gravel reach of Fraser River, British Columbia by employing a large river, stage-adaptive habitat classification system. The distribution and abundance of habitat are spatially quantified at the reach scale (32 km), and temporally quantified through a period of about 60 years at several adjacent gravel bars (7 km), and at approximately 500 m3 s-1 increments in discharge during the declining limb of the flood hydrograph at two well-developed gravel bars. Of the ten habitat types evaluated, the bar edge habitat type is most abundant by length and number of units. However, its relative importance is reduced when weighted by fish-habitat association characteristics. Preferred habitat types (channel nook, eddy pool and open nook) are frequent and available to aquatic organisms, and most common at well-developed bars and in zones of equilibrium long-term sedimentation. Preferred habitat was at a maximum 30 years ago when major new bars developed and the thalweg shifted, effectively increasing the amount of bar shoreline and nearshore habitat. This increase is due to substantial change in river planform morphology following a 30-year period of large annual floods. However, amounts of habitat did not increase exclusively during periods of higher than average flows, or decrease exclusively during periods of lower than average flows. Instead, habitat abundance response to flow may occur with a two- or three-year lag. Short term changes in stage are critical to amount of preferred habitat. Optimal discharge for maximum preferred habitat vailability is in the range of approximately 2500 m3 s-1 to 4000 m3 s-1, which approximates long term mean flow. As flow increases, the proportion of preferred habitat compared with total bar shoreline decreases. Comparison with the 2006 flow duration curve shows that 15 – 30 % of discharges are optimal for maximum fish density and biomass. These discharges occurred during April 27 to May 17 and July 14 to August 7, 2006.

WIEGAND, Mario Cesar. Proposta metodológica para estimativa da produção de sedimentos em grandes bacias hidrográficas: estudo de caso Alto Jaguaribe, CE. 2009. 109 f. Dissertação (Mestrado em engenharia agrícola)- Universidade Federal do Ceará, Fortaleza-CE, 2009. / Submitted by Elineudson Ribeiro (elineudsonr@gmail.com) on 2016-06-24T19:17:44Z No. of bitstreams: 1 2009_dis_mcwiegand.pdf: 10173855 bytes, checksum: 4bf39f0c74e0ccbca5361594889230d0 (MD5) / Approved for entry into archive by José Jairo Viana de Sousa (jairo@ufc.br) on 2016-07-21T20:20:14Z (GMT) No. of bitstreams: 1 2009_dis_mcwiegand.pdf: 10173855 bytes, checksum: 4bf39f0c74e0ccbca5361594889230d0 (MD5) / Made available in DSpace on 2016-07-21T20:20:14Z (GMT). No. of bitstreams: 1 2009_dis_mcwiegand.pdf: 10173855 bytes, checksum: 4bf39f0c74e0ccbca5361594889230d0 (MD5) Previous issue date: 2009 / Estimation of sediment yield in large river basins has great importance for the analysis and monitoring of water availability and, consequently, for water resources management. However, there are several difficulties related to such estimation. Thus, the scope of this study was to propose a methodology to estimate sediment yield in large river basins with a large number of dams. The study basin is located within the Upper Jaguaribe basin and presents an area, obtained by SRTM data, of 20,670 km². The sediment yield estimation was based on a suspended sediment rating-curve, obtained by measurements of liquid and solid discharges at the basin outlet. The bedload was estimated using the model of Einstein and Brown. Also, the estimation of sediment retention in reservoirs located upstream the basin outlet was carried out. They were classified as small dams (< 50 hm³), which were grouped into five classes according to their storage capacity, and strategic dams (> 50 hm³). Note that sediment yield was obtained for a 25-years period (1984-2008). The results obtained for total sediment yield within the basin was about 450 t.km-².year-1. From this total, about 130 t.km-².year-1 reach the basin outlet, of which 79% correspond to suspended sediment and 21% is bedload. On the other hand, approximately 320 t.km-².year-1 are retained in the surface reservoirs: 91% are trapped in small reservoirs and 9% are trapped in the large dams of the basin. It can be concluded that the results obtained in this study for the total sediment yield in a large river basin is in agreement to that obtained by other authors in South America. It should also be pointed out that the small dams play an important role in the region, not only in relation to the accumulation of water, but also to the retention of sediment generated in the basin. Such retention prevents that the sediment reaches the strategic dams (Orós, for instance), reducing their sedimentation rates and thus, minimizing the impact of sedimentation on overall water availability in the basin. / A estimativa da produção de sedimentos em grandes bacias hidrográficas possui vasta importância para a análise e o monitoramento da disponibilidade hídrica e, por consequência, para a gestão dos recursos hídricos. Todavia, vários são os óbices relacionados a tal estimativa. Desta forma, o escopo geral deste estudo foi propor uma metodologia para estimar a produção de sedimentos em grandes bacias hidrográficas com número elevado de açudes. A bacia estudada encontra-se inserida na bacia do Alto Jaguaribe e apresentou uma área, obtida através do SRTM, de 20.670 km². A estimativa da produção de sedimento foi realizada com base na curva-chave de sedimento em suspensão, obtida através de medições de descargas líquidas e sólidas no exutório da bacia. A carga de sedimento de leito foi estimada aplicando-se o modelo Einstein e Brown. Procedeu-se, ainda, a estimativa do total de sedimento retido em reservatórios a montante do exutório da bacia monitorada. Estes foram classificados em pequenos açudes (<50 hm³), os quais foram agrupados em cinco classes de acordo com sua capacidade total de acumulação, e açudes estratégicos (>50 hm³). Salienta-se que a produção de sedimento foi obtida para uma série histórica de 25 anos (1984-2008). O resultado encontrado para a produção de sedimento total dentro da bacia foi da ordem de 450 t.km2.ano1. Desse total, cerca de 130 t.km-2.ano-1 passam pelo exutório da bacia, dos quais 79% correspondem a sedimento em suspensão e 21% a sedimento de leito. Por sua vez, 320 t.km-2.ano-1 ficam retidos nos reservatórios, sendo que 91% ficam presos nos pequenos reservatórios e 9% ficam presos nos grandes açudes da bacia. Conclui-se, desta forma, que o resultado encontrado neste estudo para a produção total de sedimento em grandes bacias aproxima-se do obtido por outros autores para a América do Sul. Ressalta-se, ainda, que a pequena açudagem exerce um papel de grande relevância na região, não somente em relação ao acúmulo de água para dessedentação e irrigação, mas também, quanto à retenção de sedimento erodido na bacia. Tal retenção evita que o sedimento chegue aos açudes estratégicos (Orós, por exemplo) diminuindo sua taxa de assoreamento e, portanto, minimizando a redução da disponibilidade hídrica na bacia.

This paper presents an adaptation of the classical model of Streeter-Phelps modeling of Dissolved Oxygen (DO) and Biochemical Oxygen Demand (BOD) in the basin of the Upper Jaguaribe (25,000 km2), State of Ceara, Brazil. The adaptation of the model consisted of the numerical solution of differential equations Streeter-Phelps, considering the effect of incremental flows and sewage releases over the sections, as well as the variability of the sections of rivers and tributaries. For model calibration, including the adjustment of reaeration coefficients (K2) and removal of BOD (Kd), we used the data of the Plan of Management of Waters of the Rio Jaguaribe Basin. Calibration results showed that this simplified model represented well the balance between DO and BOD in a large semi-arid basin, with a good fit for both parameters. For OD, the average deviation was 8.44% and 6.04% by the end and beginning of the rainy season, respectively. As for BOD, the deviations were 18.51% and 30.43% for the two seasons, respectively. In both periods, the OD remained within the standards for Class 2 of Resolution CONAMA 357/2005 throughout the stretch and the BOD breached this limit on a short stretch near the city of Taua. With the already calibrated model were simulated three scenarios: a large full, using as reference flow Q10 of a historical series of Jaguaribe; drought, using the Q50 of the series; and implementation of a WWTP with 80% removal of BOD in all seats. The simulations showed consistent results and that serve as a basis for management of water resources of the study area. / Este trabalho apresenta uma adaptaÃÃo do modelo clÃssico de Streeter-Phelps para modelagem de OxigÃnio Dissolvido (OD) e Demanda BioquÃmica de OxigÃnio (DBO) na bacia do Alto Jaguaribe (Ãrea de 25.000 km2), Estado do CearÃ, Brasil. A adaptaÃÃo do modelo consistiu na resoluÃÃo numÃrica das equaÃÃes diferenciais de Streeter-Phelps, considerando o efeito de vazÃes incrementais e lanÃamentos de esgoto ao longo dos trechos, assim como a variabilidade das seÃÃes dos rios e tributÃrios. Para calibraÃÃo do modelo, incluindo o ajuste dos coeficientes de reaeraÃÃo (K2) e remoÃÃo de DBO (Kd), foram utilizados os dados do Plano de Gerenciamento das Ãguas da Bacia do Rio Jaguaribe. Os resultados da calibraÃÃo mostraram que esse modelo simplificado representou bem o balanÃo entre OD e DBO em uma grande bacia semiÃrida, apresentando um bom ajuste para os dois parÃmetros. Para OD, o desvio mÃdio foi de 8,44% e 6,04% para o fim e inÃcio da estaÃÃo chuvosa, respectivamente. Jà para DBO, os desvios foram de 18,51% e 30,43% para as duas estaÃÃes, respectivamente. Nos dois perÃodos, o OD manteve-se dentro dos padrÃes para Classe 2 da resoluÃÃo CONAMA 357/2005 em todo o trecho e a DBO infringiu este limite em um pequeno trecho prÃximo à cidade de TauÃ. Com o modelo jà calibrado, foram simulados trÃs cenÃrios: uma grande cheia, utilizando como vazÃo de referÃncia o Q10 de uma sÃrie histÃrica do Jaguaribe; estiagem, utilizando o Q50 da sÃrie histÃrica; e, implantaÃÃo de uma ETE com remoÃÃo de 80% de DBO em todas as sedes. As simulaÃÃes apresentaram resultados coerentes e que servem como base para o gerenciamento dos recursos hÃdricos da bacia estudada.

Estimation of sediment yield in large river basins has great importance for the analysis and monitoring of water availability and, consequently, for water resources management. However, there are several difficulties related to such estimation. Thus, the scope of this study was to propose a methodology to estimate sediment yield in large river basins with a large number of dams. The study basin is located within the Upper Jaguaribe basin and presents an area, obtained by SRTM data, of 20,670 kmÂ. The sediment yield estimation was based on a suspended sediment rating-curve, obtained by measurements of liquid and solid discharges at the basin outlet. The bedload was estimated using the model of Einstein and Brown. Also, the estimation of sediment retention in reservoirs located upstream the basin outlet was carried out. They were classified as small dams (< 50 hmÂ), which were grouped into five classes according to their storage capacity, and strategic dams (> 50 hmÂ). Note that sediment yield was obtained for a 25-years period (1984-2008). The results obtained for total sediment yield within the basin was about 450 t.km-Â.year-1. From this total, about 130 t.km-Â.year-1 reach the basin outlet, of which 79% correspond to suspended sediment and 21% is bedload. On the other hand, approximately 320 t.km-Â.year-1 are retained in the surface reservoirs: 91% are trapped in small reservoirs and 9% are trapped in the large dams of the basin. It can be concluded that the results obtained in this study for the total sediment yield in a large river basin is in agreement to that obtained by other authors in South America. It should also be pointed out that the small dams play an important role in the region, not only in relation to the accumulation of water, but also to the retention of sediment generated in the basin. Such retention prevents that the sediment reaches the strategic dams (OrÃs, for instance), reducing their sedimentation rates and thus, minimizing the impact of sedimentation on overall water availability in the basin. / A estimativa da produÃÃo de sedimentos em grandes bacias hidrogrÃficas possui vasta importÃncia para a anÃlise e o monitoramento da disponibilidade hÃdrica e, por consequÃncia, para a gestÃo dos recursos hÃdricos. Todavia, vÃrios sÃo os Ãbices relacionados a tal estimativa. Desta forma, o escopo geral deste estudo foi propor uma metodologia para estimar a produÃÃo de sedimentos em grandes bacias hidrogrÃficas com nÃmero elevado de aÃudes. A bacia estudada encontra-se inserida na bacia do Alto Jaguaribe e apresentou uma Ãrea, obtida atravÃs do SRTM, de 20.670 kmÂ. A estimativa da produÃÃo de sedimento foi realizada com base na curva-chave de sedimento em suspensÃo, obtida atravÃs de mediÃÃes de descargas lÃquidas e sÃlidas no exutÃrio da bacia. A carga de sedimento de leito foi estimada aplicando-se o modelo Einstein e Brown. Procedeu-se, ainda, a estimativa do total de sedimento retido em reservatÃrios a montante do exutÃrio da bacia monitorada. Estes foram classificados em pequenos aÃudes (<50 hmÂ), os quais foram agrupados em cinco classes de acordo com sua capacidade total de acumulaÃÃo, e aÃudes estratÃgicos (>50 hmÂ). Salienta-se que a produÃÃo de sedimento foi obtida para uma sÃrie histÃrica de 25 anos (1984-2008). O resultado encontrado para a produÃÃo de sedimento total dentro da bacia foi da ordem de 450 t.km2.ano1. Desse total, cerca de 130 t.km-2.ano-1 passam pelo exutÃrio da bacia, dos quais 79% correspondem a sedimento em suspensÃo e 21% a sedimento de leito. Por sua vez, 320 t.km-2.ano-1 ficam retidos nos reservatÃrios, sendo que 91% ficam presos nos pequenos reservatÃrios e 9% ficam presos nos grandes aÃudes da bacia. Conclui-se, desta forma, que o resultado encontrado neste estudo para a produÃÃo total de sedimento em grandes bacias aproxima-se do obtido por outros autores para a AmÃrica do Sul. Ressalta-se, ainda, que a pequena aÃudagem exerce um papel de grande relevÃncia na regiÃo, nÃo somente em relaÃÃo ao acÃmulo de Ãgua para dessedentaÃÃo e irrigaÃÃo, mas tambÃm, quanto à retenÃÃo de sedimento erodido na bacia. Tal retenÃÃo evita que o sedimento chegue aos aÃudes estratÃgicos (OrÃs, por exemplo) diminuindo sua taxa de assoreamento e, portanto, minimizando a reduÃÃo da disponibilidade hÃdrica na bacia.

Physical habitat for instream biota derives from a combination of stream system structural and hydraulic phenomena. Consequently, the quantity and quality of physical habitat is dynamic both over time and in space along the river, laterally, longitudinally and vertically. Its characterization through stream assessment and classification leads to a better understanding of factors that determine and limit habitat extent and quality. This thesis investigates the effects of space and time on nearshore aquatic habitat in the gravel reach of Fraser River, British Columbia by employing a large river, stage-adaptive habitat classification system. The distribution and abundance of habitat are spatially quantified at the reach scale (32 km), and temporally quantified through a period of about 60 years at several adjacent gravel bars (7 km), and at approximately 500 m3 s-1 increments in discharge during the declining limb of the flood hydrograph at two well-developed gravel bars. Of the ten habitat types evaluated, the bar edge habitat type is most abundant by length and number of units. However, its relative importance is reduced when weighted by fish-habitat association characteristics. Preferred habitat types (channel nook, eddy pool and open nook) are frequent and available to aquatic organisms, and most common at well-developed bars and in zones of equilibrium long-term sedimentation. Preferred habitat was at a maximum 30 years ago when major new bars developed and the thalweg shifted, effectively increasing the amount of bar shoreline and nearshore habitat. This increase is due to substantial change in river planform morphology following a 30-year period of large annual floods. However, amounts of habitat did not increase exclusively during periods of higher than average flows, or decrease exclusively during periods of lower than average flows. Instead, habitat abundance response to flow may occur with a two- or three-year lag. Short term changes in stage are critical to amount of preferred habitat. Optimal discharge for maximum preferred habitat vailability is in the range of approximately 2500 m3 s-1 to 4000 m3 s-1, which approximates long term mean flow. As flow increases, the proportion of preferred habitat compared with total bar shoreline decreases. Comparison with the 2006 flow duration curve shows that 15 – 30 % of discharges are optimal for maximum fish density and biomass. These discharges occurred during April 27 to May 17 and July 14 to August 7, 2006. / Arts, Faculty of / Geography, Department of / Graduate