Findings from the Middlemount pasture dieback trial site

Location: ‘Gundabah’, Middlemount

Year established: 2020

Key messages

• Annual fertiliser has changed produced the largest yield every year and has changed the species composition from approximately half/half Indian couch and Gayndah buffel grass to nearly 100% buffel.
• Establishing a new pasture with a mix of grass species and legumes has resulted in a more diverse pasture with a large yield.
• Fallowing by chemical application or mechanical cultivation and then rolling seed after planting has been critical to good establishment of seedlings.
• Seed quality had high variability so germination tests were used to calculate seeding rates.

In this article

Site starting point

Trial design and treatments

Results

Information learned

Impact of fertiliser

Starting with a new pasture

Ongoing research

Site starting point

The trial paddock at Gundabah, Middlemount, was dominated by Gayndah buffel after being cropped for a few years in the mid-1970s. Since then, the paddock had slowly become increasingly affected by pasture rundown, which is common in grass-dominated paddocks. Pasture rundown is where soil plant available nitrogen becomes tied up over time in organic matter and as a result, grasses like buffel which have a high nitrogen demand, become less productive. Ripping was conducted in about 2005 to help renovate the pasture. Indian Couch was increasing in the pasture and quickly expanded when pasture dieback occurred in 2017.

Trial design and treatments

Treatments were initially developed to help understand management of pasture dieback, rundown and Indian Couch invasion. At that time (2020), pathogens causing dieback were not confirmed therefore multiple pesticide treatments were included (fungicide and insecticide). A treatment of sowing a break crop of forage oats before going back into pasture was included to see if it would break the disease cycle. A high rate of fertiliser was also added to see if high nutrient supply, especially nitrogen, would encourage pathogens and/or change the grass species composition. The other treatments compared different ways to establish a new mixed legume and grass pasture. The table below outlines the treatments implemented. All treatments are replicated three times and randomly positioned within the trial area to provide a high level of confidence in the results.

* NB Pesticides applied under APVMA research trial permit.

For new grass/legume pasture treatments, the pasture seed mix was primarily chosen based on suitability to the site’s rainfall and soil type. A mixture of grasses with varying susceptibly to pasture dieback were included. Legumes are resistant to dieback and two persistent and productive types suited to this location were also included. All species were tested for germination and the seeding rates applied varied based on these results. As some species had low germination, and some were coated, high seeding rates were needed to ensure sufficient viable seed was planted for a productive pasture.

The diagram below summarizes the activities on site since 2020. Grazing did not commence until the 2024 dry season when a water trough was installed in the paddock.

Results

Pasture assessments are conducted annually in the mid-late wet season (March – June). The dry matter yield is recorded, as well as the species composition. The first graph below (figure 1) shows the cumulative dry matter yield of all treatments since 2020. The second graph (figure 2) shows the dry matter yields and species composition of all treatments in 2024.

The yields recorded in 2020 across all treatments was the starting point for the pasture, before each treatment began. The yields in 2021 were very low across all treatments and there were no differences compared to the control.

In 2021, treatment 5 was regenerating Gayndah buffel and Indian Couch after one pass cultivation and fertiliser. By 2024, this treatment is still approximately half/half buffel/couch.

The yield measured in treatment 6 for 2021 is oats. The first pasture yield was measured in this treatment in 2022 when the treatment was planted back to Gayndah buffel. Despite being planted only to buffel, by 2024 this treatment is starting to have increasing amounts of Indian Couch.

Treatments 7, 8 and 9 which used different methods of fallowing (one cultivation, two cultivations, or herbicide spray) we replanted to the legume/grass seed mix. Despite the lower yields in 2021 and 2022 as the pasture was establishing, these treatments now have higher biomass than treatments 1-6. In 2024, they still have species diversity and it is clear that Indian Couch is being kept to a minimum by the other dominant grasses. Despite being sown as half of the seed mix, the legumes still only make up a small proportion of the biomass. However, this is more legumes than all other treatments.

Treatment 10 which has an annual application of fertiliser, has been the highest yielding treatment every year. It has also gone through a species composition change with nearly all of the yield now Gayndah buffel.

What information has the site provided about pasture dieback management?

While the paddock was affected by dieback prior to the trial being set up, no dieback symptoms have been observed during the trial period. It is now known that the pasture mealybug can cause dieback and mealybug numbers at this site over the trial period have always been very low. The biomass in the insecticide treatment has been the same as the control, indicating no insecticide benefit at this stage, however there has been limited insect pressure at the site.

What has increasing nutrient supply done to the pasture?

In treatment 10, the annual application of fertiliser has made a large difference to pasture yield and composition. This treatment has consistently produced the highest yield over the trial period and resulted in a significant pasture composition shift. At the start of the trial Indian couch made up 67% of the total pasture biomass in this treatment, whereas buffel was 22%. After one application of fertiliser, Indian couch reduced to 22% and buffel increased to 69%. After three years of fertiliser applications, Indian couch was down to 2% and buffel was up to 98%. This demonstrates how responsive buffel is to fertiliser. With the right nutrition buffel can outcompete an invasive grass like Indian couch. While the fertiliser application rates used here may not be commercially viable for everyone, the results demonstrate how rundown (i.e. starved of nutrient) the pasture was and how well buffel grass responds to extra nutrient supply.

Economic studies undertaken by DPI indicate a cost-effective way to improve nitrogen supply long term is via legumes, however these only supply modest amounts of nitrogen (15 – 50kg nitrogen per hectare), and this is based on the amount of legume biomass produced. Therefore, pasture composition benefits will take longer to achieve.

Starting again with a new pasture

All treatments where a new pasture was sown (treatments 7, 8, 9) have grown more biomass than all other plots except the fertiliser treatment. In the new pasture treatments (7, 8, 9) Indian couch not only makes up a small proportion of the pasture, but about 10% of the biomass is legume. As these legumes set seed, this is likely to increase over time.

Excellent new pastures have been established here by fallowing with cultivation or spraying, fertilising, sowing seed at appropriate rates and then rolling the seed in. The soil type at this site is duplex with a loam surface which can get fluffy with multiple cultivations. Using a roller before and after sowing resulted in good seed-soil contact and a high establishment rate.

Plots planted to the new pasture (treatments 7, 8 and 9) were all seeded with a grass and legume mix. Despite planting multiple species, Rhodes and Gayndah buffel are the dominant grasses, and Seca and Caatinga are the dominant legumes.

Research is ongoing

This research site is going into its fifth year, now funded under The Queensland Pasture Resilience Program, a partnership between the Department of Primary Industries, Meat & Livestock Australia, and the Australian Government through the MLA Donor Company.

This site plays a valuable role in understanding pasture dieback, as no dieback has been detected since the trial began. Dieback is cyclic in nature, so can come and go over many seasons. Treatments 6, 7 and 8 that have been resown to a mixture of grasses with more tolerance to dieback and legumes which are resistant, will be an interesting one to watch should dieback come back into the site. Likewise, treatment 4 which is an insecticide treatment will also be watched carefully.

In November 2024 the whole site was burnt in an accidental fire, so it will be interesting to see how each treatment recovers in the 2024/25 wet season.

In 2025, the fungicide treatments (2 and 3) will cease as it is now known that pasture dieback is not caused by a fungus. The plots receiving the annual fertiliser treatment have been soil tested so nutrient budgeting can be done before the next fertiliser application. Grazing will resume at the site after the 2025 measurements have been taken (approx. April-June).