For water flowing in a culvert, it is important to know the critical depth. Though water may not actually be flowing at the critical depth, it is helpful to know if the actual water depth is greater than or less than critical. Critical depth is a quantity of fundamental importance to understanding the flow characteristics. If the actual depth is greater than critical depth, then the flow is considered "subcritical". Subcritical flow is "slow flow" and is impacted by downstream conditions. If the actual depth is less than critical depth, then the flow is "supercritical". Supercritical flow is "fast flow" and is impacted by upstream conditions. It flows faster than the wave speed and is unimpacted by downstream conditions. A detailed discussion of subcritical and supercritical flow can be found on our gradually varied flow calculation page.
As water flows down a culvert, the depth may change and pass through the critical depth if the bottom slope, geometry, or pipe material changes. If the flow is initially subcritical and the channel slope increases, the water may undergo a hydraulic drop if it transitions from subcritical to supercritical. Conversely, if the flow is supercritical and the culvert slope becomes flatter, then there could be a hydraulic jump such that the water passes through the critical depth as it becomes subcritical.
Doubts have been expressed about the validity of the critical depth defined in terms of the minimum specific energy head of the free-surface streamline when dealing with developing open-channel flows. This note examines the two approaches for defining critical flow, that based on the minimum specific energy of the free-surface streamline and that based on the mean energy head of the whole flow section. Large differences for the dimensionless critical depths are obtained with the two methods due to each critical depth proving to be a different control point on the free-surface profile. It is argued that both methods are different alternatives, although the critical depth concept was different in each case. Theoretical support to critical flow computations based on the free streamline is provided. An alternative approach for computing the discharge characteristics of broad-crested weirs based on the energy loss inside the boundary layer is also proposed.
