The Canadian prairies are dominated by millions of depressions, or potholes, that have a significant impact on streamflow generation in the region. It has been difficult to incorporate the dynamic storage, or “fill and spill”, processes of these depressions in hydrologic models of prairie watersheds. Additionally, the region has a cold-climate, where snowmelt and soil thaw processes impact the generation of streamflow, sediment export, and non-point source pollutants. This paper discusses improvements to the hydrological model SWAT (Soil and Water Assessment Tool) for modelling streamflow and sediment export in two pothole-dominated watersheds in Saskatchewan: the Assiniboine and Qu’Appelle watersheds. The Pond module of SWAT was modified to incorporate dynamic storage in the potholes using a probability distribution to assess how many of the depressions would be spilling and therefore contributing to streamflow. Soil erodibility coefficients were adjusted seasonally to improve estimates of sediment export. Improved performance for simulating streamflow was seen over the existing SWAT Pond module for simulating landscape storage. Sediment export estimates improved when using seasonally adjusted erodibility coefficients over annual sediment erodibility coefficients.

This paper presents results from a study to examine the flow and sediment deposition patterns in a new vortex-type stormwater retention pond using physical scale and computational models. The pond, that is circular in plan, has a peripheral inlet and a central outlet, which creates a strong circulation in the pond. Two geometries of the pond were tested. Measurements of the residence time distributions, flow patterns, and sediment deposition patterns were taken at the design flows for the two pond geometries. The effectiveness of an internal berm to improve desirable flow characteristics was also examined. The berm improved flow and sediment deposition in the pond, as the sediment deposition patterns showed that most of the sediments deposited outside the berm. This is considered to be beneficial for pond maintenance. The physical scale model results are also compared to 3D computational fluid dynamics (CFD) modeling results using ANSYS Fluent.

Modelling the hydrology of North American Prairie watersheds is complicated because of the existence of numerous landscape depressions that vary in storage capacity. The Soil and Water Assessment Tool (SWAT) is a widely applied model for long-term hydrological simulations in watersheds dominated by agricultural land uses. However, several studies show that the SWAT model has had limited success in handling prairie watersheds. In past works using SWAT, landscape depression storage heterogeneity has largely been neglected or lumped. In this study, a probability distributed model of depression storage is introduced into the SWAT model to better handle landscape storage heterogeneity. The work utilizes a probability density function to describe the spatial heterogeneity of the landscape depression storages that was developed from topographic characteristics. The integrated SWAT-PDLD model is tested using datasets for two prairie depression dominated watersheds in Canada: the Moose Jaw River watershed, Saskatchewan; and the Assiniboine River watershed, Saskatchewan. Simulation results were compared to observed streamflow using graphical and multiple statistical criterions. Representation of landscape depressions within SWAT using a probability distribution (SWAT-PDLD) provides improved estimations of streamflow for large prairie watersheds in comparison to results using a lumped, single storage approach. Copyright (c) 2016 John Wiley & Sons, Ltd.

Nonpoint source pollution is a critical problem in Canadian prairie watersheds. However, sediment mobilization and export are poorly represented in existing models for these watersheds. The poor representation is partly because the hydrology of the region is highly influenced by the existence of numerous dynamically-connected landscape depressions that vary in storage capacity and because of the complex freeze-thaw cycles in the region. The objective of this research was to improve sediment export simulation modeling in these cold-climate prairie watersheds by incorporating a probability distribution function of depression storage capacity and a seasonally varying soil erodibility factor into the soil and water assessment tool (SWAT) model. The probability distribution function is used to represent the variation in storage capacity of the numerous depressions, whereas the seasonally varied soil erodibility factor is used to account for changes in erodibility as the soil freezes and thaws. Results from two case study watersheds confirm an improvement in sediment export predictions when varying storage capacity is represented and the sediment loss routine includes seasonally varying soil erodibility. (C) 2016 American Society of Civil Engineers.

Non-point source pollution due to agricultural activities is an important problem that has been threatening water resources in Canadian prairie watersheds. The development of strategies to prevent nutrient loss depends on the quantification of nutrient mobilization and transport across a watershed. Integrated eco-hydrological models can play an important role in this regard. However, current model applicability to cold-climate Canadian prairie watersheds is limited due to the complex dynamics of nutrient export under the existence of numerous landscape depressions and freeze-thaw cycles. The aim of this study was to evaluate an eco-hydrological model for nutrient export prediction and assess the impacts of management practices for a cold-climate prairie watershed. To achieve the objectives, a new version of the SWAT model called SWAT-PDLD, which combines SWAT and a Probability Distributed Landscape Depressions (PDLD) model, along with a seasonally varying soil erodibility factor, was applied to a Canadian prairie watershed (the Assiniboine River watershed, Saskatchewan, Canada). The PDLD module is used to simulate the effect of the numerous landscape depressions that exist in these watersheds on streamflow, whereas a seasonally varying soil erodibility factor is used to take into account seasonal variation of sediment and nutrient generation due to the cold climate conditions. Model calibration and uncertainty analysis were performed using the Sequential Uncertainty Fitting (SUFI-2). The study shows that the SWAT-PDLD model with seasonally varying soil erodibility simulates the daily nutrient export in a cold prairie watershed satisfactorily as confirmed by both graphical plots and statistical measures. A sensitivity analysis of sub-watershed discretization revealed that the streamflow is relatively insensitive to sub-watershed discretization but it did affect sediment and nutrient export. Importantly, the model shows that both filter strips and cover crops decreased sediment, phosphorous, and nitrogen export, while conservation tillage increased phosphorous export in the study watershed. (C) 2016 Elsevier B.V. All rights reserved.

Much of the prairie region in North America is characterized by relatively flat terrain with many depressions on the landscape. The hydrological response (runoff) is a combination of the conventional runoff from the contributing areas and the occasional overflow from the non-contributing areas (depressions). In this study, we promote the use of a hybrid modelling structure to predict runoff generation from prairie landscapes. More specifically, the Soil and Water Assessment Tool (SWAT) is fused with artificial neural networks (ANNs), so that SWAT and the ANN module deal with the contributing and non-contributing areas, respectively. A detailed experimental study is performed to select the best set of inputs, training algorithms and hidden neurons. The results obtained in this study suggest that the fusion of process-based and data-driven models can provide improved modelling capabilities for representing the highly nonlinear nature of the hydrological processes in prairie landscapes.

Streamflow in small forested watersheds on the Boreal Plain of western Canada can be a challenge to monitor due to high variation in flow, shifting channel morphology, aufeis obstructions, shallow channel depth, and debris in the channel. Intensive monitoring in natural channels can overcome some of these problems, but frequent assessment and recalibration of streamflow-stage relationships are necessary. Experience over 8years indicates that in-channel structures designed to provide a stabilized stream-monitoring section (SSMS) proved beneficial to monitoring efforts during the ice-free season. The SSMS facilitated accurate gauging of the highest and lowest flows encountered during this period, provided a relatively stable foundation against streambank and channel erosion, and allowed passage of fish and the majority of debris and sediments. Functionality of the structures was markedly limited during ice-in conditions, but was improved with the addition of fitted canopies and propane heating systems.

The accurate prediction of snowmelt runoff is a critical component of integrated hydrological and water quality models in regions where snowfall constitutes a significant portion of the annual precipitation. In cold regions, the accumulation of a snowpack and the subsequent spring snowmelt generally constitutes a major proportion of the annual water yield. Furthermore, the snowmelt runoff transports significant quantities of sediment and nutrients to receiving streams and strongly influences downstream water quality. Temperature-index models are commonly used in operational hydrological and water quality models to predict snowmelt runoff. Due to their simplicity, computational efficiency, low data requirements, and ability to consistently achieve good results, numerous temperature-index models of varying complexity have been developed in the past few decades. The objective of this study was to determine how temperature-index models of varying complexity would affect the performance of the water quality model SWAT (a modified version of SWAT that was developed for watersheds dominated by boreal forest) for predicting runoff. Temperature-index models used by several operational hydrological models were incorporated into SWAT. Model performance was tested on five watersheds on the Canadian Boreal Plain whose hydrologic response is dominated by snowmelt runoff. The results of this study indicate that simpler temperature-index models can perform as well as more complex temperature-index models for predicting runoff from the study watersheds. The outcome of this study has important implications because the incorporation of simpler temperature-index snowmelt models into hydrological and water quality models can lead to a reduction in the number of parameters that need to be optimized without sacrificing predictive accuracy.

In this study, a two-dimensional, depth-averaged computational model (River2DMix) was used to predict the flow pattern and residence time distribution for flow through the Calgary Glenmore Water Treatment Plant northeast clearwell. Results are compared to those from flow visualization and tracer studies in a 1:19 scale model of the clearwell, as well as tracer studies conducted at the plant. Tests were carried out for three flow rates that ranged from minimum to maximum operating conditions. A key observation in the physical model was that it was necessary to let the flow fully develop before starting a tracer test to determine the residence time distribution. This flow development time to achieve steady-state results was approximately 10.5 h at the minimum flow rate tested. Results also show that it was unnecessary to model the structural columns either in the simulation or the scale model for developed flow in this clearwell, although for undeveloped or transient flow conditions the columns were important to consider.

Landscape-scale hydrological models can be improved by incorporating realistic, process-oriented plant models for simulating crops, perennial grasses and woody species. The objective of this project was to present some approaches for plant modelling applicable to daily time step hydrological transport models, such as SWAT. Accurate simulation of plant growth can improve the accuracy of simulations of hydrological and biogeochemical cycles. First, we describe some unique aspects of the general plant model ALMANAC. Next, we describe a modification of the original ALMANAC model used to simulate complex successional vegetation changes in the years following disturbance of a variety of different forest ecosystems, such as forest fires, clear cuts and insect infestations. Finally, we discuss alternative physiological and physical process simulation techniques of plant growth that could increase simulation accuracy in landscape-scale hydrological and transport models such as SWAT.

This investigation presents new and aggressive approaches to link the results of scientific endeavor to management of a portion of the Canadian boreal forest, within the framework of the detailed forest management plan (DFMP) process of a forestry company in the province of Alberta. The first component in the DFMP was landscape projection, whereby cumulative impacts of key natural and anthropogenic disturbance agents were modelled under current and altered climate conditions. The second component addressed two types of impact assessment. The Biodiversity Assessment Project (BAP) modelled ecosystem diversity at landscape and habitat levels, as well as developed habitat supply models, relative to changing vegetation composition, management practices, and stand age. Models were used during the development of a preferred forest management strategy to address undesirable ecological predictions. In the Forest Watershed and Riparian Disturbance (FORWARD) project, a variant of the soil and water assessment tool was developed to model the impacts of watershed disturbance on streamflow. In the third component of the DFMP, timber supply scenarios were devised based on maximizing annual allowable harvest in a sustained yield fashion, while incorporating elements of the BAP and FORWARD project as constraints in harvest sequence optimization. This initiative is an example of an industry-led effort to manage forests using a system that is regionally centered, science based, peer reviewed, and considers multiple activities and their cumulative environmental effects.

The Forest Watershed and Riparian Disturbance (FORWARD) project input into the Millar Western Forest Products Ltd. Detailed Forest Management Plan consists of three main components: 1) watershed and stream layer maps and associated datasets; 2) soil and wetland layer maps and associated datasets; and 3) a lookup table that permits planners to determine runoff coefficients (the variable selected for hydrological modelling) for functional first order watersheds, based upon various site factors and time since disturbance. The watershed and stream layer component includes a hydrological network, a Digital Elevation Model, and Strahler classified streams and watersheds for functional first and third order watersheds in the entire Millar Western Forest Management Agreement area. Relatively coarse mineral soils (which drain quickly) and wetlands (which retain water) were the key features that needed to be identified for the FORWARD modelling effort; therefore, the soil and wetland layers represent a combined soil texture and wetland coverage. The runoff coefficient lookup table integrates predictions of hydrologic impacts of harvest into planning. Key words: forest management, watershed, hydrology, stream, soils, wetlands, modelling

Several modifications were made to the Soil and Water Assessment Tool (SWAT) to better represent processes occurring within forested watersheds on the Boreal Plain in Canada. The modified model, called SWATBF, was applied to the Willow Creek watershed (15.1 km2) in north central Alberta. The performance of the model for the calibration period (2001–2003) was good with coefficients of efficiency of 0.89 and 0.81 being achieved for the prediction of monthly and daily runoff, respectively. However, it was found that SWATBF did not perform as well for the validation period (2004–2006) with the monthly and daily coefficients of efficiency being 0.44 and 0.27, respectively. Potential sources of error to explain the decline in model performance for the validation period are discussed. SWATBF has the potential to be used as a tool by forest managers for predicting the effects of land use change on the Boreal Plain provided that it can be satisfactorily validated.

This paper presents results of scale model testing of. the Glenmore Water Treatment Plant NE Clearwell in Calgary, Alberta, Canada. This clearwell has a perforated baffle wall at its inlet, a 4-pass serpentine system, and a weir at the outlet. The model was operated based on Froude similarity and had a scale of 1:19. Tracer studies and flow visualization were carried out for three flow rates. From the residence time distribution developed from the tracer tests, it was seen the baffle factor and Morril Dispersion index were affected by the time between when the flow in the model and tracer test was initiated. It took approximately 10 h of flow development time for the results to reach steady-state values for the minimum flow rate. At higher flows, the baffle factor was not as strongly affected by flow development time, however the Morril Dispersion index better matched prototype values at longer times.

This study examined total dissolved (TDP) and particulate phosphorus (PP) concentration and export in nine headwater streams draining Boreal Plain forests in Alberta, Canada, during the growing season and autumn before and in the first year after winter harvest in four of the watersheds. Mechanical and chemical site preparation was also conducted before and during the post-harvest sampling period. During the growing season, TDP and PP concentrations in streams draining harvested watersheds were higher than in the reference streams at storm peaks (P ≤ 0.08) and along the falling limbs of storm flows (P ≤ 0.03). We postulate that forestry activities enhanced TDP leaching from watershed soils after they became saturated. This promoted the growth of periphyton in streams, which subsequently sloughed off during high flows. In autumn, TDP concentrations were higher in streams draining harvested watersheds along the rising limb, at the storm peaks, and along the falling limbs (P ≤ 0.05). However, PP concentrations in autumn were similar in harvested and reference streams along the rising limb (P = 0.19) and were lower in harvested streams at the storm peaks and along the falling limbs (P ≤ 0.08). No differences in TDP or PP concentration were detected between harvested and reference streams during baseflow-dominated periods. Mean TDP and PP export coefficients tended to be higher in the post-harvest than pre-harvest year in both reference and harvested watersheds. After normalization to precipitation inputs, this change (“impact ratio”) was 153% and 130% higher in harvested than reference watersheds for TDP and PP, respectively (P ≤ 0.08). The TDP impact ratio and percent of the watershed area cut were positively related (r2 = 0.95, P = 0.02). A 30 m riparian buffer strip in two harvested watersheds did not appear to reduce the harvest response.