River rehabilitation has no shortage of design guidance. The challenge is finding the minimum effective dose: how light can a treatment be while still achieving recovery? Existing manuals can specify how to armour a bank but determining the intervention threshold at a given site remains genuinely challenging. Liability pressure, conservative safety factors and institutional risk-aversion create a ratchet effect driving treatments progressively heavier, with nothing pushing back. When the goal is restoring natural river function, overtreating a reach can be as ecologically harmful as undertreating it.
This presentation draws on field observations and practitioner-led trials from restoration projects on the North Coast of New South Wales. Vegetated revetment techniques using topsoil, shot rock and soil-capped configurations were compared across sites. Granular filter layers were trialled as a functional alternative to geotextile. Large wood placement observations explored minimal ballast thresholds for stable placement and optimal habitat. Rock sizing assumptions were tested using witness marks and post-flood observation.
In each case, observed behaviour diverged from conventional predictions, suggesting current practice is often over-engineered. Lighter treatments performed better than expected; standard anchoring and sizing assumptions were more conservative than field conditions warranted.
The discipline needs strong field-based evidence for scaling back treatment. Closer observation, honest documentation of failure and success, and willingness to accept appropriate risk are proposed as foundations for a more experimental, nature-first culture in river rehabilitation.