Scour Holes: Fish the Upstream Lip

At low water on a tidal river, one large boulder is easy to read. Coarse sand tails upstream from the rock, the bottom drops away below it, and a low ridge of fresh gravel sits a few metres downstream. One obstruction has rearranged the bed into a hole with a clear upstream lip and a softer downstream rim. The hole is not random.

structure
Cover illustration: Scour Holes: Fish the Upstream Lip

Key facts

QuantityValueWhat it means at the rim
Local scour depth ys2 × K1 × K2 × K3 × K4 × (y1/a)^0.35 × Fr1^0.43A scale estimate, not a survey
Nose shape K10.9 sharp, 1.0 round, 1.1 squareA blunt face digs deeper
Angle K2 at 0 degrees1.0Flow square to the face
Angle K2 at 15 degrees1.5 at L/a = 4, 2.5 at L/a = 12Skew multiplies depth
Angle K2 at 30 degrees3.5 at L/a = 12The worst common case
Bed condition K31.1 plane bed, 1.3 large dunesA dune field deepens the pit
Armoring K4Floor of 0.7 when D50 exceeds 0.06 mA coarse bed resists scour

One rock, three vortices

It is the visible result of a flow pattern that repeats around every piling, boulder and bridge pier in moving water. Learn the pattern and you can predict where the deep water is, which side is firm enough to stand on, and how quickly the pit will fill after the next flood. The engineering description of this flow names three features. A horseshoe vortex wraps the base of the obstruction, its arms trailing downstream on both sides. A wake vortex forms behind the obstacle where the separated flow reattaches. A toe vortex sits at the upstream foot, where the descending flow turns back along the bed. The horseshoe is the feature that usually explains the deepest part of the pit. It scrubs the bed in a ring around the nose, and its strength depends on how much water is turned downward. The downward flow at the face feeds it. The pressure gradient between the slowed water at the nose and the faster water at the sides drives it. A USGS study of streambed stresses notes an important exception: over a bed that is not planar, the horseshoe vortex may be absent altogether. A rock sitting on a slope, a ledge or a rough reef is not the same case as a cylinder on flat sand. The flow can separate in a different pattern, and the scour becomes a patchwork rather than a single bowl. That exception changes what you look for. Where the bed is flat and the obstacle is clean, expect a compact pit at the upstream toe. Where the bed is uneven, expect shallow scours, streaks and deposition instead of one obvious hole. Steep sides and a sorted rim belong to the first case; mixed sizes and irregular depth belong to the second.

The equation as a scale

The local-scour relation used in bridge inspection is ys = 2 K1 K2 K3 K4 (y1/a)^0.35 Fr1^0.43. Here y1 is the depth just upstream, a is the width of the obstruction, and Fr1 is the Froude number of the approach flow. The four K terms correct for nose shape, angle of attack, bed condition and armoring. Read it as a scale, not a promise. The exponents 0.35 and 0.43 make depth grow slowly with flow depth and speed, so doubling either does not double the hole. The multiplier 2 sets the order of magnitude. On a clean, square-nosed obstacle in deep, fast water, a pit twice as deep as the surrounding water is not unusual. A worked example shows how the terms combine. Take a 1-metre-wide round piling in 3 metres of water, with a Froude number of 0.4, a plane sand bed and no armoring. With K1, K2 and K4 at 1.0 and K3 at 1.1, the estimate is about 2.2 metres of scour below the ambient bed. That is a pit deep enough to change where a fish sits and deep enough to swallow a wading angler.

Angle does most of the work

K1 runs from 0.9 for a sharp nose to 1.1 for a square one, with a round nose or a circular cylinder at 1.0. The spread is small. Shape alone will not double a pit. K2 produces the large numbers. At 0 degrees, flow square to the face, the factor is 1.0 for every length-to-width ratio. At 15 degrees it ranges from 1.5 at L/a = 4 to 2.5 at L/a = 12. At 30 degrees and L/a = 12 it reaches 3.5. A long obstruction hit at an angle digs several times the hole of the same obstruction hit square-on. This explains why a boulder that looks harmless at slack water can have a pit on one side only. The ebb or flood approaches at an angle, and the effective shape becomes long and skewed. The hole itself records the angle that dug it. The deeper side faces the direction from which the strongest flow arrived. For a bridge pier, the angle is set by the channel. For a boulder in a tidal river, the angle reverses with the tide. The pit that survives is the one dug by the stronger of the two flows, which is why the deepest part often sits on the ebb side in an estuary and on the flood side in a river mouth.

Grain size and refill

K3 raises the estimate from 1.1 over a plane bed to 1.3 where large dunes are present, because a dune field already stores coarse material that the vortex can move. K4 is the armoring correction, with a floor of 0.7 when the bed material is coarse, measured as a D50 above 0.06 metres. A cobble or gravel bed resists scour; a sand bed does not. Grain size also decides what happens after the flood. A coarse pit fills slowly, and the refill leaves a hard rim of mixed gravel and sand. A fine-sand pit can fill within a single tide, and the new surface is smooth and soft. The same hole can be permanent in one river and seasonal in another. Flood flows do two things at once: they enlarge the pit and deliver the material that later fills it. As the flood spreads and slows, sand settles in the hole. By the next low water the pit may hold a new layer of clean sand over old gravel, which is why a depth reading from last season cannot be trusted. The refill pattern is readable. Fresh sand on the downstream rim means the hole is active and recently filled. A rim of mixed gravel and shell means the hole is older and the fine material has been winnowed away. A flat, smooth floor with no rim means the pit has filled completely and the obstruction may have moved.

Reading the rim at low water

The rim is the reliable part of the mark. The upstream side is usually firmer, because flow sweeps past rather than dropping its load. The downstream side can be soft, especially where the wake has just laid down fresh sand. Test with a rod before you put weight on it. Look for the lip. The upstream lip is the edge where the bed breaks downward, and it usually survives the next tide because it is made of the coarser material the flow could not move. The downstream rim is lower and softer, and it moves with every flood. Check the approach angle. A symmetrical pit with even rims was dug by flow square to the face. An asymmetric pit with one steep side and one long tail was dug at an angle. The steep side is the side to fish, because it is the side the vortex scrubbed hardest. Check the grain. A rim of uniform fine sand means a mobile bed and a pit that changes shape quickly. A rim of mixed gravel, shell and sand means a stable bed and a pit that persists.

Fishing the pit

The fishing geometry follows the shape. Bait on the upstream lip drifts over the deep water before the current lifts it. Bait cast past the pit and retrieved slowly drops down the downstream slope. Fish that use the hole tend to sit on the rim, facing upstream, rather than in the deepest point. The rim gives them a view of the flow and a short dash into the pit. The boat version is the same shape from a different angle. Hold outside the pit and let the bait fall down the slope. Anchoring over the hole puts the line in the roughest, most mobile water and swings the boat over the fish. The upstream lip is the casting position. It is firm, it is shallow enough to stand on, and it gives a clean line to both the deep water and the downstream rim. The downstream rim is the place to avoid, not because it is deep but because it is soft and the wake is still moving it.

Where the equation stops

The HEC-18 relation was developed for bridge piers in rivers, with a steady approach flow and a bed of uniform sand. A tidal river reverses, a rocky bed armors unevenly, and a boulder is not a cylinder. The equation gives the order of magnitude, not the depth under your feet. The USGS exception is the important one for anglers. On a sloping or rocky bed, the horseshoe vortex may not form at all. The equation assumes it does. Where the bed is not planar, use the visible rim and the grain size instead of the number. The equation also assumes a single obstacle. A cluster of boulders, a piling row or a reef with several peaks produces interacting vortices, and the scour can be deeper between obstacles than around any one of them. Fish the gaps, not just the faces.

A check at the rim

Find the upstream lip at low water. It is the firm edge where the bed breaks downward, and it usually survives the next tide. Probe the rim with a rod. Soft refill and firm original bed feel different through the tip, and the difference tells you where you can stand. Cast across the pit rather than into it. Let the current carry the bait over the deep water. Check the approach angle. A skewed flow makes an asymmetric pit, and the deeper side is the side to fish. Re-check the depth after every flood. One event can move the lip or fill the hole. Keep your feet off the downstream rim. Fresh sand there is the least stable ground around the obstruction.

Sources

  1. Bridge Inspection: Scour Physics and Changes to HEC-18 — Michigan Technological University · T2
  2. Streambed stresses and flow around bridge piers — U.S. Geological Survey · T1
  3. Evaluating Scour at Bridges, Hydraulic Engineering Circular No. 18 — Federal Highway Administration · T1