Reading slope erosion after a gel installation: what the rills tell you

Four days after a 38 mm storm, a revegetation slope at 1:2.5 showed a pattern the site supervisor described as random. It was not random. The rills started 60 cm below the top of the batter, ran in parallel lines about 1.8 m apart, and stopped dead at a contour of jute matting. Between the rills, the surface was intact. That pattern named the failure in about ninety seconds, and it was not the gel.

Erosion reading is a skill that pays for itself on the first site visit. Rills record where water went, how fast, and what it was carrying. Once you can read them, a slope tells you how the material was applied, whether it was incorporated, and whether the specification was followed.

Where the rills start is the diagnosis

Rills that begin at the top of the batter and run the full length usually mean the slope is receiving water from above, not from the face. Sheet flow from an unshaped crest, a road, or a compacted access track is the usual source. No amount of polymer in the face will fix that, because the water is arriving with energy already in it.

Rills that begin 50 to 80 cm below the crest are more informative. That is often the depth where the topsoil layer ends and the subsoil begins, or where the incorporation pass stopped. If the gel was mixed into the top 100 mm and the batter was then trimmed with a machine, the trimmed surface can be subsoil with almost no polymer in it, and the topsoil plus gel ends up in a windrow at the toe.

Rills that fan out from a single point indicate a concentration, usually a discharged downpipe, a gully, or a low spot in the crest. Again, not a material problem.

Depth and spacing tell you about incorporation

A shallow, wide, braided wash that lifts a few millimetres of surface is what you see on a well-incorporated slope under a storm that exceeded design. The soil stayed together; the water simply ran off faster than it could infiltrate.

A narrow, deep, V-shaped cut with vertical sides means the surface had no structure to resist. That is typical of a slope where dry granules were broadcast on the surface, wetted once, and left. The granules swell, the surface lifts a little, then dries and shrinks, and the crust that forms is brittle. Under the next storm, the crust fails along the lines where the water concentrates.

Spacing is the other clue. Rills at regular intervals every 1.5 to 2 m generally reflect the pass pattern of a machine: a dozer blade, a rotary hoe, or a roller. If the rills align with wheel tracks or blade passes, you are looking at compaction, not at the polymer.

The failure that looked like a gel problem

On the 1:2.5 batter, the specification called for 45 kg/ha mixed through the top 150 mm, followed by hydromulch. The contractor broadcast the granules with a spinner, then ran a rotary hoe over the batter, then trimmed it to grade with an excavator bucket. The trimming pass removed 80 mm of the treated layer and pushed it to the toe, where it formed a band of gel-rich soil that grew weeds all winter. The face itself had roughly a third of the specified dose and no mulch, because the hydromulch was applied after trimming and the crew ran out of tank capacity at the top of the slope.

The result was 38 mm of rain, a set of parallel rills, and a client who was told the polymer was poor quality. Two years later the toe of the slope is green and thick, the face is still patchy, and the pattern matches the application history exactly.

What to do with what you find

If the rills are shallow and the surface is intact, the slope is behaving about as well as an unprotected batter can under that storm. Repair by filling the rills, re-seeding and re-mulching, and leave the gel alone.

If the rills are deep, vertical-sided and concentrated, the treated layer is missing or unincorporated. Repair means re-working the face, not adding more material on top. A second broadcast over a crusted surface will mostly wash off in the next event and will make the next diagnosis harder.

If the rills align with machine passes, you have a compaction problem that no polymer will solve. Deep-ripping the face, where access allows, is the only fix that lasts.

Take photographs before you repair. A slope with a tape measure in frame, a photo of the rill cross-section, and a note of the storm depth is the difference between a warranty argument and a maintenance record. The application guide has a slope checklist that covers incorporation depth and trimming sequence, which is where most of these failures begin.

If you have a slope that failed its first storm, send three photos: the crest, a rill cross-section with a tape, and the toe. We can usually tell you within a day whether the material is still in the face or sitting in the sediment trap.