Clay, compost and spading at Walyoo Farm: WMG’s new soil amendments trial

By Simon Kruger and Chanel Duggan, WMG

WMG’s long-term stacked soil amendments trial at Wathingarra, Badgingarra, concluded at the end of 2025 after five seasons. The site tested the combined effects of mechanical amelioration and organic and mineral amendments on deep brown sand with severe soil water repellence. Results from each season are available on WMG’s website project pages.

Two findings from Wathingarra stood out consistently. The first was the durability of the amelioration effect, with mouldboard ploughing and rotary spading applied in 2021 continuing to produce meaningfully higher yields four years later without further mechanical intervention. The second was the performance of organic carbon amendments, particularly compost, and of local clay in the amendment strips. Both lifted yield above the site mean in 2025 and showed a pattern of improved nutrient supply and soil function across the trial period.

Those two amendments are what the new trial at Regans Ford carries forward. The site is on deep sand at Walyoo Farm near Dandaragan, managed by Mark Drake of Lawson Grains, and takes compost and clay from the Wathingarra treatment list into a replicated trial on a different deep sand under commercial grower conditions. Five seasons of results at one site establishes that the amendments worked there. Whether they perform the same way on a different soil in a different part of the district is the question the new trial exists to answer.

Trial design

The Regans Ford trial is structured as a randomised complete block design with five treatments and four replicated blocks, giving 40 plots in total. Each plot is 12m in seeded length with a 10m cutback, and 4 m wide in 2026. In 2027 the plots will be split to 2m width to allow fertiliser treatment variations to be overlaid, adding a second dimension to the experiment in the second season. The total trial footprint runs approximately 201m along the seeding direction and 60m wide, with spaded buffer zones between treatment blocks and additional buffers to account for clay overthrow during spreading.

The crop is Jurien lupin, sown on 7 May 2026 at 100kg/ha to a depth of 4cm on 22cm row spacing. The previous rotation was wheat in 2025 and canola in 2024. Fertiliser across the full site in 2026 is MacroProPlus and Agflow Mn at 100kg/ha, with fertiliser treatment splits deferred to 2027 once the baseline amendment effects are established.

Treatments

The five treatments test spading alone, the two most promising individual amendments from Wathingarra applied with spading, and their combination, against an untreated control.

All spaded treatments were incorporated using an Imants Plot Spader to approximately 30cm depth prior to seeding. Compost was weighed and spread by hand before incorporation. Clay was applied using Mark’s grader spreader before incorporation by spading, with the spread pattern confirming an average application rate close to the 300t/ha target. Commercial clay spreaders tend to deliver a peaked distribution across the spread width, and sampling across the plots confirmed that was the case here. Measurements through the season and at harvest will account for the within-plot distribution.

The untreated control provides the baseline against which amendment and amelioration effects will be assessed. The spading only treatment isolates the effect of mechanical soil disturbance from any nutrient or physical contribution of the amendments. Treatments 3 and 4 test compost and clay individually in combination with spading, while Treatment 5 asks whether applying both amendments together with spading delivers an additional benefit beyond either alone.

Soil baseline

Pre-treatment soil sampling was carried out across 20 plots in May 2026 by Ethan Sirr of Moora Soil and Plant Contracting, with samples taken to 50cm depth in three increments and analysed by CSBP. The results confirm the site is typical deep sand country for the Dandaragan district. Texture across the profile is uniformly sandy to the full sampling depth, transitioning from a brown-grey surface layer to an orange subsoil below 30cm. Organic carbon in the top 10cm averaged around 0.65% across sampled plots, declining to around 0.45% in the 10 to 30cm layer and below 0.30% in the 30 to 50cm layer, consistent with the low organic matter retention characteristic of this soil type.

Soil pH at the surface is moderately acidic to neutral, ranging from around pH 5.6 to 6.9 (CaCl?) across sampled plots in the top 10cm, becoming more acidic with depth and dropping to around pH 4.6 to 6.2 in the 30 to 50cm layer. Colwell phosphorus at the surface was modest across all plots, generally in the range of 12 to 23mg/kg, and Colwell potassium was variable but broadly adequate at the surface. The baseline soil data will serve as the reference point against which any amendment-driven changes in soil chemistry are assessed in subsequent seasons.

Soil biological sampling was also conducted prior to seeding using the SARDI Predicta B platform, providing a pre-treatment picture of the nematode and soil pathogen community across the site. Results showed no detectable cereal cyst nematode or root lesion nematodes (Pratylenchus thornei and P. quasitereoides) across any of the sampled plots. Low levels of a third root lesion nematode species, P. neglectus, were detected in some control and compost treatment plots, though counts were very low and unlikely to be of agronomic significance at this stage. The Predicta B baseline will allow any treatment effects on the soil biological community to be tracked as the trial develops.

Crop establishment

Plant establishment counts were conducted across all 40 plots on 8 June 2026, approximately four weeks after seeding. Mean plant density by treatment is summarised below.

All treatments recorded mean establishment well above the DPIRD-recommended optimum range of 40 to 45 plants/m² for narrow-leafed lupin in Western Australia, above which there is generally no yield penalty and which provides good competitive suppression of weeds. The high establishment across all treatments reflects good seedbed conditions at the time of sowing and provides a clean agronomic starting point for assessing treatment effects through the season. Some variability in establishment was noted within the clay treatment plots, consistent with the peaked spread distribution described above.

At the time of writing, early biomass sampling and plant tissue testing were yet to be completed.

What the trial is testing

The question at Regans Ford is whether the amendment combinations that performed well at Wathingarra translate to yield and soil improvements under a different set of conditions: a new deep sand soil type, a different location and in 2026 a lupin crop rather than wheat.

The spading only treatment gives the clearest early read on what mechanical amelioration alone contributes on this soil, establishing the baseline improvement from soil disturbance before any amendment contribution is considered. The compost and clay treatments then sit on top of that, with the stacked treatment asking whether combining both amendments with spading produces an outcome that neither delivers on its own.

Lupins bring a further dimension. As a legume, the crop contributes nitrogen fixation back to the system. The 2027 season, when fertiliser treatment splits are overlaid on the existing amendment plots, will allow the trial to examine whether amended soils respond differently to fertiliser inputs, which is relevant to how growers would manage nutrition in crops following amelioration investment.

As with the Wathingarra site, the economics will be worked through at the end of the season using a partial budget that sets the cost of each amendment treatment against any measured yield benefit, giving growers a basis for assessing the investment case on comparable sandy country in the district.

What comes next

First flower biomass and tissue sampling will be completed in July and August, with results reported through WMG’s enewsletter and website shortly after. A crop walk at the site is planned as part of the CSBP & WMG Agronomy Event on the 4th of August. Peak biomass sampling is planned for September to October, with harvest yield data collected at the end of the season. Post-harvest soil sampling and the partial budget analysis will follow in early 2027.

Growers with questions about the trial or about soil amendment options on sandy soils in the Dandaragan district are welcome to contact Chanel Duggan at projects@wmgroup.org.au.

This research is supported by the Soil CRC through funding from the Australian Government’s Future Drought Fund. The project is led by Murdoch University in partnership with WMG. WMG thanks Mark Drake and Lawson Grains for hosting the trial and for their support of the project.