Some of the challenges of our landscape regeneration efforts have been both working to restore eroded badlands in some parts of the property, as well as to arrest continuing erosion. To make matters more complex, some sections of the property have “dispersive” subsoil. While the topsoil is relatively stable, when the subsoil becomes saturated it dissolves and begins to move, hollowing out tunnels beneath the surface that ultimately collapse into sinkholes, and eventually gullies. Of course, because it’s all happening underground, tracking or predicting which areas are vulnerable, or already cavernous is tricky.
Over the last seven years, we’ve been gradually working our way around the property installing erosion control structures, based on the work of permaculture-based landscape restoration folk like Craig Sponholtz and Bill Zeedyk. Hailing from the dry southwest of North America, Sponholtz and Zeedyk advocate applying small and slow interventions that use local materials to slow water flows and gradually heal erosion. In the permaculture spirit of viewing problems as solutions, our junk-filled erosion gullies have provided a seemingly inexhaustible supply of demolition rubble, concrete chunks and segments of brick walls that are ideal material for these structures.
One of the sections that we’ve avoided is a notorious headcut that our DIY soil tests showed us boasted slaking soil that disintegrates when moist and subsoil that was also more saline than other locations around the farm. With saline subsoil, most plants wanted to avoid getting their stabilising roots down. Tree-planting was also fraught, survival rates in the saline subsoil and cracking grey topsoil were low and the very process of digging a small hole for a seedling can be enough to expose the subsoil so that next time there was a downpour, a sinkhole would open and swallow the seedling, tree guard and all. With the headcut moving about a metre a year and also widening, we thought it was time to get stuck in. At the very least, we’ve discovered that doing something is better than doing nothing and the act of getting in there and digging around allows for more accurate monitoring and effective intervention.
For this spot we thought we’d try a Craig Sponholtz strategy called the “Rock Mulch Rundown”, a way of sculpting and armouring eroding headcuts with a low amount of surface water flows. We started by chipping the headcut back from it’s unstable, vertical form to a more stable angle. This process is also useful for spreading stable topsoil and using this to cover and cap the exposed and unstable subsoil. We stomped it down, and spread gypsum. As noted in this Western Australian information on managing dispersive soils:
Structurally unstable soils are often sodic, because a large proportion of the cations attached to the clay particles are sodium. This reduces the stability of the soil aggregates (soil crumbs or clods) and makes the soil prone to dispersion… Apply lime or gypsum. Calcium ions displace some of the sodium ions on the surface of soil particles, allowing it to leach out, thus improving the structural stability of the soil. Lime should only be applied to acid soils (pHCa < 4.8) as it will not dissolve in alkaline soils and will have no benefit. Apply gypsum on the surface of alkaline soils. Gypsum improves the structural stability of dispersive top soils quite quickly; however it may take several years to reach the subsoil.
We raked and spread the gypsum, then set to work armouring the surface with chunks of demolition rubble. Once completed, we planted native perennial grasses into the cracks, and even tossed a few of our friend Steven Hoepfner‘s local provenance seed bombs into sheltered spots to germinate. We’ve found that the layer of rubble acts as a mulch, protecting the soil and holding moisture after surrounding exposed areas have dried out.
In the week following completion, southeastern Australia was hit with a cold snap that involved torrential downpours, Antarctic winds and snow in the highlands. We checked our intervention and, while the rain had clearly washed around the rocks, carrying some of the gypsum and gravel, the headcut still appeared stable. Meanwhile, as another strategy, we’re trying to intercept and harvest water higher up in the catchment. In the case of this gully complex, we’ve started planting silvopasture uphill which we hope will further stabilise the soil and intercept water before it hits the gullies. As is key with this kind of lo-fi intervention, we’ll keep watching, learning and tweaking where we need to!