Les Stanfield¹, Doug Mulholland² and Don Cowan²

¹Ecohealth Solutions, Milford, Canada

²University of Waterloo, Waterloo, Canada

Managing natural channels requires good policy and planning, high quality field data, predictive models, and expertise.  While progress has been made on many facets of this challenge, we have yet to develop integrated models that consider large-scale landscape influences on channel form and function, biota and water quality. In this paper we will show how readily accessible datasets contained within the Flowing Waters Information System (FWIS) and recent GIS approaches offer a potentially powerful base for developing the first integrated landscape models.   It is anticipated that such models that consider landscape and local factors as well as scaling factors, channel morphology and stability, water quality and biologic metrics would support decision making and scenario testing by practitioners.  We will highlight recent biological modeling successes and offer a template for the overall model development in the hopes of encouraging more interest in this initiative.  Besides the challenge of developing these models, the maintenance of the partnerships and databases that facilitate the analysis is also critical.  Moving forward, our community must become even more adept at sharing data to reduce the effort required to compile data sets that will improve model predictions.  Further, the applications that facilitate data sharing must be supported by coordinated funding and stable institutions so that these data are available for future analysis. We cannot rely on good will alone to protect these datasets and we must get better at sharing our data if we are to make more progress in the next 30 years.

John McDonald, M.Sc., P.Geo. (Limited) ¹ and Aaron Farrell, M.Eng., P.Eng., CPM²

¹Matrix Solutions Inc., Mississauga, Canada

²HDR Inc., Righmond Hill, Canada

Characterization and management of surface water features in Southern Ontario has historically classified all of these features as “watercourses”.  Under this approach each feature was assessed by multiple disciplines (i.e. hydrology, geomorphology, hydrogeology, fisheries, and terrestrial ecology), and ultimately assigned a constraint ranking of “high”, “medium” or “low” provided by each.  The final constraint ranking provided direction regarding the management opportunities for each feature.  Under this previous approach, low rated features were typically first order streams or headwater drainage features, whereas medium and high constraint features corresponded to open regulated watercourses. In general, high-constraint watercourses would essentially be protected in situ, medium-constraint watercourses would remain on the surface but may undergo realignment or enhancement, and low-constraint watercourses could effectively be removed from the landscape, provided that local drainage density targets are met.

More recently, the HDF guidelines developed by TRCA and CVC introduced a new category of drainage feature (“HDF”), and established an approach for classifying these types of features.  The protocols focused on the lower order watercourses, hence did not encompass the full suite of drainage features encountered on the landscape.  Consequently, there became a need to harmonize these new protocols with historic practice for characterizing surface water features.

Recently, during the South Milton Urban Expansion Area Subwatershed Study, the approaches for characterizing regulated watercourses and HDFs were harmonized to establish management recommendations for a Conceptual Land Use Plan. per feature. This approach was developed and executed through a collaborative effort between the Town and its consultants, Participating Landowner Consultants, Conservation Halton, Halton Region, and Ministry of Natural Resources and Forestry. The approach developed through this process built upon the HDF classification process per the TRCA/CVC protocols, and refined the approach for establishing constraint rankings for regulated watercourses consistent with the historic practices in the Town of Milton. The ultimate definitions and evaluation system provide a template that may be applied in other jurisdictions, and also allows some flexibility in determining appropriate definitions and management opportunities based on local and regional understanding and policy.

Dr. Corey Dawson¹ and Dr. Peter Ashmore²

¹Dalhousie University (Agricultural Campus), Truro, Canada

²Western University, London, Canada

River design has long been practiced by hydraulic engineers, but natural channel design has more recently become a collaborative effort to meet diverse stakeholder objectives. Designing for natural geomorphic complexity is particularly challenging. Irregularity is generally difficult to consistently measure but has geomorphic and ecological value while too much riverscape complexity may be perceived as messy or unsafe by the public. Advancements in high-resolution 3-D surface data applications are providing more opportunities to quantify morphology aspects for channel design decision-making and post-project assessment, which can identify acceptable trade-offs and increase monitoring investment potential.

Here we propose a multi-scaled analytical method for natural channel design and monitoring, defined as the Geomorphic Form Variation (GFV) approach. The GFV applies variety statistics to topographic surface metrics to take advantage of high-resolution data availability to evaluate morphological aspects of 3-D continuous river surfaces. Flume channel experiments were used to test the GFV’s response to morphological feature changes while River Builder, developed by Brown, Pasternack and Wallender (2014), was used to derive synthetic channels to allow systematic planform design scenarios to be comparatively analyzed with variety statistics.

Results showed steeply curved feature margins were major components of topographic variety and the GFV responded to surface roughness at the selected raster pixel size, finding the approach capable of multi-scaled geomorphic form complexity analysis. River Builder channel analysis found width variability, high bend curvature, streambank roughness, and general thalweg irregularity contributed to greater variety, suggesting potential planform design elements of geomorphic and ecological importance.

Because the metric is raster-based, variety values are affected by pixel resolution and limited by the quality of available river surface data. However, the GFV provides a measurable, scalable, and repeatable assessment approach of various morphological attributes and the details of spatial arrangements and extents. The combination of River Builder software and the GFV approach in the preliminary decision-making process may allow community-based planning which can highlight geomorphic complexity aspects that support public aesthetic preferences and balance diverse objectives for natural channel design.

Bryce Molder¹, Jan Franssen², and Paul Villard³

¹Consulting Geomorphologist (GEO Morphix Ltd.), Campbellville, Canada

²Consulting Senior Watershed Scientist (GEO Morphix Ltd.), Campbellville, Canada

³Consulting Geomorphologist (GEO Morphix Ltd.), Campbellville, Canada

Streambed heterogeneity enhances aquatic ecosystem conditions through provision of complex physical habitat structure and diverse flow regimes. Knowledge of naturally occurring bed variability limits is therefore important in the development of natural channel or aquatic enhancement design. The extent of variability is also significant in the planning, design, and/or protection of subsurface or stream valley infrastructure. Infrastructure placement should account for future bed adjustment to mitigate potential impacts associated with infrastructure damage or failure caused by flowing water, ice, or in-channel debris. Over 100 channels of varying channel boundary composition (e.g., alluvial or semi-alluvial type) were assessed to ascertain bed variability and naturally occurring stream depth limits for a range of Ontario streams. Relationships between bed variability indices and potential driving forces were reviewed and discussed. Analyses were distinguished based on the presence or absence of cohesive glacial deposits and/or shale. Positive correlations were exhibited between maximum pool depth (and/or riffle-pool differentials) and bankfull width. The deepest pools were generally found in channels characterized by exposed cohesive glacial deposits and at given channel bend configurations (0.7<Rc/W<4). Ranges in bed variability were significant, when compared to other potential drivers, owing to the inherent complexity of stream hydrodynamics.

Cal Jefferies^1,2, William K. Annable², Ben Plumb¹, and Jeff Hirvonen¹

¹GeoProcess Research Associates, Ontario, Canada

²University of Waterloo, Waterloo, Ontario

Despite research indicating that the resistance of cohesive sediments to fluvial erosion is related to the geologic media, regulators and industry in Southern Ontario do not typically consider the diversity in geologic landscapes when assessing the erosion potential of streams with cohesive boundaries. To assess the potential differences in erosion characteristics between various cohesive geologies in Southwestern Ontario, a field investigation has been undertaken over the past three years employing the in-situ mini-JET methodology across 13 sites and 10 distinct geologic units (245 total tests).

A novel analytic approach is applied to the dataset to account for the preferential weathering of surficial material at test locations. The sensitivity of the mini-JET results to operational and analytic parameters are assessed to ensure the characterization of material properties is representative of site conditions.

A surrogate derived from the novel analytic approach applied here is used to investigate differences in the depths of weathered surficial layers at test sites. Subaerial tests are shown to have a higher presence of weathered material compared to submerged tests and tests higher on banks have a greater presence of weathered material compared to tests lower on banks or along the streambed. Halton Till is shown to have a statistically lower mean critical shear stress (τ_c) compared to the grouped results of the other geologic units investigated, however, the difference in mean τ_c of 2.8 Pa has limited implications given the range in τ_c at individual sites (low value Pa ≤ τ_c ≤ high value Pa).

Comparisons between spring and summer conditions indicate that the presence of the surficial weathered layers is higher in the spring, but the critical shear stress of the underlying unweathered material is similar between seasons. The preparation of readily erodible weathered material corresponding to freshet flows suggest that early spring may be responsible for an outsized amount of erosion in streams with cohesive boundaries, particularly in headwater systems with shorter, more frequent hydrographs. This indicates that regulators and industry should place a higher importance on considering how alterations to river systems and various stormwater management techniques change the frequency and magnitude of weathering processes acting on cohesive boundaries, in particular during the early spring.

Chris Muirhead

GHD Limited, Waterloo, Canada

Design and assessment of natural channel systems requires an understanding and approximation of natural conditions in an artificial environment such as through a hydraulic model. As preferred channel designs have transitioned from historically hard-armored channels to more complex, riffle-pool systems over the past decades, additional data and increased hydraulic modelling efforts are required to quantify these complex and dynamic systems. In the absence of data, time or scope for increased efforts, hydraulic modelers have sought to simplify the complicated hydraulics associated with varying bed morphology through approximated assumptions. One such assumption was the development of a roughness value which includes both bed roughness and form roughness to approximate the impact of cascade-pool sequences on the water surface elevation and velocity in the channel in steeper (2-3% gradient) systems.

An assessment has been performed of 3 Southwestern Ontario watercourse designs for cascade-pool sequences to compare the impacts of approximating form roughness, represented by Mannings n, with a model using higher density cross-sections with a roughness value only accounting for bed roughness. Water surface elevation and velocity outputs from the two different modelling approaches were compared and implications for design and flood hazard assessment will be discussed.

Matthew Iannetta^1,2 and Bruce MacVicar²

¹GeoProcess Research Associates, Toronto, Canada

²University of Waterloo, Waterloo, Canada

Many commonly used field-based bedload monitoring methods often yield limited inter-flood or discontinuous data, which restricts our understanding of bedload transport dynamics. Field efforts involved in collecting these data can be difficult, expensive, and dangerous in some circumstances of high flow. In this study, we present a newly developed remote, integrated, automatic, and continuous bedload monitoring station. The station was deployed in a semi-alluvial headwater creek located in a rural watershed in Southern Ontario for the purpose of developing a field-calibrated predictive bedload transport model.

The station design integrates two indirect monitoring devices including an in-situ radio frequency identification (RFID) antenna tracker and “Benson-Type” seismic impact plates. The in-situ antenna automatically tracks RFID pit tag tracer stones that pass over the antenna. The impact plates convert mechanical energy exerted by bedload particles striking the plate into electrical energy recorded as total counts. A Bunte direct sampling device is included in the station design for the purpose of indirect data calibration. The station configuration is unique compared to other existing bedload monitoring stations described in published literature in that it is less costly, easy to deploy in the field, and optimized for remote applications.

400 synthetic RFID tracer stones with a distribution spanning four half-phi size classes were seeded upstream of the station and manually tracked on three occasions over a period of 27 months. Inter-flood tracer movement was automatically tracked on one occasion during a December 2019 flood event. The impact plates yielded variable results over the relatively short study duration. Manual tracer tracking was the most reliable field method and results were used to develop a calibrated predictive model of bedload transport that demonstrated strong alignment with similar studies conducted in a neighbouring watershed. Although limited in this study, indirect sampling results provided novel insight into bedload transport dynamics. Lessons learned during field implementation were equally as valuable as study results, and can be used to make improvements for future monitoring station deployment.

Lindsay Davis¹, Jaclyn Cockburn² and Paul Villard¹

¹GEO Morphix Ltd., Campbellville, Canada

²University of Guelph, Guelph, Canada

Winter is a stressful season for freshwater, stream-dwelling fish due to ice decreasing habitat area and creating habitat fragmentation. In addition to the impacts from ice, cooler water temperatures lower fish metabolism causing slow swim speeds and decreased swimming duration. Small streams regularly become completely ice-covered, and common techniques used to study fish habitat developed for open-water conditions are not easily modified. Winter conditions make it difficult to accurately observe fish behaviour in their natural habitat. This study evaluated remote underwater video cameras used to observe minnow communities, habitat, and behaviour in overwinter conditions. Waterproof action cameras (e.g., GoPros^TM and Sony^TM action cameras) were lowered into the water column from the channel ice and set to record for 30-minute intervals, and with a modified antenna, Bluetooth^TM connected cameras facilitated real-time observations. Advances in video camera technology have allowed high quality video to be captured with inexpensive equipment (~$500 CDN for camera, case, memory card and back-up batteries), such as small, portable action cameras that are now readily available. This technique was effective at observing fish behaviour, communities, and habitat preference during the winter in small, ice-covered streams, which is important for water resource and fisheries management, conservation biology, and stream restoration.

Our current understanding of overwinter habitat mainly comes from studies focused on salmonid species. Minnows exhibit different behaviours and typically have a more sensitive flight response compared to larger species. This methodology was used in several sites within the Greater Toronto Area (GTA) to determine the effectiveness of observing minnow communities and behaviour within small ice-covered streams in relation to physical habitat characteristics. Many stream restoration design projects aim to provide habitat for endangered minnow species within the GTA and a better understanding of overwinter habitat in small streams allows designers to incorporate preferred habitat within the designs.

Ian Smith and Ahmed Siddiqui

GEI Consultants Ltd., Kitchener, Canada

Topographic surveys are a useful tool in fluvial geomorphology workflows, which have been used to measure variation in bed elevation and riparian corridors across space and time.  These include surveys of the longitudinal profile of the bed, cross-sections, and topographic elevations in the floodplain and valley complex.  One typical application of a channel survey is to collect detailed geomorphic information to determine erosion thresholds, supporting the development of a stormwater management plan through the identification of target release rates for stormwater management facilities.  Additionally, surveys of a “reference reach” are used to inform channel design projects, through identification of bankfull discharge, as well as using topographic information to identify tie-in points, existing ground elevations, etc.  Topographic surveys of watercourses can also be used to measure rates of adjustment (both laterally and vertically) over time.  Traditionally, survey methods have included the use of low-tech equipment such as dumpy levels or theodolites, and recently more sophisticated tools such as total stations and real-time-kinematic (RTK) positioning.  These tools are typically time-consuming to use, and some may require additional staff resources to support the collection of data.  Until recently, aerial surveys have been expensive.  But with the use of remotely piloted RTK-equipped drones, aerial surveys have become cheaper.  The coupling of drone technologies with high spatial resolution differential Global Positioning Systems (dGPS) has allowed for more accurate and repeatable spatial data collection at rates that far exceed that of more traditional approaches.  Often, field surveys that took days in the past can be completed in mere hours. Similarly, limitations of drones in the past have included their inability to survey underwater topography or shallow bathymetry.  However, recent advancements in drone and camera technology as well as post processing image analysis software has allowed one to include surveys of the topography within watercourses.  We explore the use of Remotely Piloted Aerial Systems (RPAS, aka drones) to complete topographic surveys in and around watercourses and showcase some recent projects where RPAS have been used to support the collection of topographic information to support stream analyses and fluvial design.