Practical Applications of Bed Scour Calculations: Two Case Studies

Pipelines can be exposed at water crossings where rivers lower the channel bed. Channel bed scour may cause damage to linear infrastructure such as pipelines by exposing the pipe to the flow of water and sediment. Accurate estimation of depth of scour is therefore critical in limiting damage to infrastructure. Channel bed scour has three main components: (1) general scour, (2) bed degradation, and (3) pool depth. General scour is the temporary lowering of the channel bed during a flood event. Channel bed degradation is the systematic lowering of a channel bed over time. Pool depth is depth of pools below the general bed elevation and includes the relocation of pools that result from river dynamics. Channel degradation is assessed in the field using indicators of channel incision such as channel bed armoring and bank characteristics, through the analysis of long profiles and sediment transport modelling. Pool depth is assessed using long profiles and channel movement over time. The catastrophic nature of bed lowering due to general scour requires a different assessment. A design depth of cover is based on analysis of depth of scour for a given return period (eg. 100-years). There are three main steps to predict general scour: (1) regional flood frequency analysis, (2) estimation of hydraulic variables, and (3) scour depth modelling. Typically, four scour models are employed: Lacey (1930), Blench (1969), Neill (1973), and Zeller (1981), with the average or maximum value used for design depth. We provide herein case studies for potential scour for pipeline water crossings at the Little Smoky River and Joachim Creek, AB. Using the four models above, and an analysis of channel degradation and pool depth, the recommended minimum depth of cover of 0.75 m and 0.142 m, respectively, were prescribed. Variability between scour models is large. The general scour model results varied from 0.45 m and 0.75 m for the Little Smoky River and 0.16 m to 0.51 m for Joachim Creek. While these models are more than 30 years old and do not adequately account for factors such as sediment mobility, they nevertheless do provide usable answers and should form part of the usual toolbox in water crossing scour calculations.