Rivers are inherently dynamic environments with fluctuations in water quality, hydrology, connectivity and geomorphology. Though geomorphology has long been recognized as an important driver defining biological, ecological, and physical habitat characteristics of rivers, a readily applied classification tool that links such characteristics has been lacking. The Geomorphic Response Unit (GRU) method provides a novel approach to identifying large scale patterns in geomorphic character that provide a link between the hydrological regime and different habitat types to which species respond. Specifically, I investigated whether Geomorphic Types and GRUs are related to the distribution and abundance of different fish species, reflecting unique physical habitat characteristics of individual GRUs. The thesis chapters are manuscript based. The second chapter identifies relationships between specific Geomorphic Types, identified using the Geomorphic Response Unit (GRU) methodology, and Lake Sturgeon overwintering locations in the South Saskatchewan and Saskatchewan Rivers. Habitat selection ratios suggest that Lake Sturgeon in the Upper South Saskatchewan River significantly selected for one of seven possible Types for overwintering. Logistic regression results found both Type 0 and Type 4 predicted significantly higher Sturgeon presence than all other Types (P = < 2e-16 for both). The third chapter examines relationships between GRUs and abundance of both mature and immature Carmine Shiner in the Birch River, Manitoba. Differences in the median mature Carmine Shiner CPUEs among the GRUs are not statistically different (Kruskal-Wallis test H =1.723; df = 3, p value = 0.632), though interesting qualitative relationships were identified which may inform further studies. The fourth chapter investigates whether GRUs derived using a large scale network approach are linked to the abundance of specific fish species in the Assiniboine River, Manitoba. A Kruskal-Wallis test identified significant differences in CPUE among GRUs for 10 of 14 tested species. Post-hoc pairwise multiple comparisons using Dunn’s Method with Bonferroni p-value correction for multiple paired tests isolated the GRUs that were different from one another. Overall, my findings suggest that Geomorphic Response Units (GRU) are an effective means of identifying patterns in geomorphic structure within Prairie Rivers at both reach and segment scales. Further, I identified links between both Geomorphic Types and GRUs and patterns in abundance of various fish species covering a wide range of life history traits. These findings suggest that GRUs have potential as a valuable fisheries habitat management tool, increasing efficiency of monitoring efforts through quantification of habitat availability, connectivity, and complexity in Prairie River systems.

>Magister Scientiae - MSc / Several attempts have been made to assess the impact of post-fire soil erosion; however, erosion occurs as a result of the complex interplay between many factors, such as climate, land cover, soil and topography, making precise estimation difficult. Additionally, these factors are far from constant in space and time, and often interact with one another. To assess the impact of wildfire on soil erosion and factors influencing its variability, the post-fire soil erosion response of two mountainous headwater sub-catchments namely Langrivier and Tierkloof, with different vegetation cover in the Jonkershoek Valley was examined using a systematic approach that combines efforts in field and laboratory work, spatial analysis and process-based numerical modelling. Geospatial modelling shows high spatial variability in erosion risk, with 56 % to 67 % of surfaces being highly susceptible excluding rock cover. The model highlights the importance of terrain and vegetation indices, with predicted erosion being more severe on steep slopes with lower vegetation cover. / 2021-08-30