At a glance
Australia's 2025–26 winter crop reached 66.3 million tonnes, the second-largest on record, with farm gate value passing $100 billion. The gains came from yield rather than new land — a story written by weed seed control, rust genetics and tree crop intensification programs.
Australian winter crop production is estimated at 66.3 million tonnes for 2025–26, the second-largest harvest on record and 35 per cent above the ten-year average of 49.2 million tonnes. Barley set an all-time high of 16.33 million tonnes, beating the 14.65 million grown in 2020–21. Wheat reached 35.99 million tonnes, the third-biggest crop this country has grown, and canola's 7.68 million was second only to the record set in 2022–23.
Almost none of it came from new paddocks. Area planted to winter crops was a national record at 25.2 million hectares, but only barely above the year before. The tonnage came from yield.
So the number worth sitting with isn't the harvest. It's $101.4 billion, the forecast gross value of agricultural production in 2025–26 — the first time the figure has passed $100 billion, and four years ahead of the National Farmers' Federation's 2030 target. ABARES expects that to ease to around $95 billion in 2026–27 as livestock prices come off and crop volumes normalise.
The base rate is 1.5 per cent a year, and it has been slowing
The cleanest test of whether agricultural research earns its keep is total factor productivity: output divided by every input that went into producing it, from land and labour to diesel and urea. For the Australian cropping industry, ABARES puts long-run growth at 1.5 per cent a year from 1977–78 to 2013–14.
That average hides a hinge. Growth ran near 3.5 per cent a year until the mid-1990s, then about 0.7 per cent a year for the two decades that followed. The step change of the 1980s — conservation tillage, legume break crops, nitrogen rates matched to stored soil water — had been absorbed.
A mature system still grinding out 0.7 per cent a year is not failing. It is working much harder for each increment. Which is why the recent gains trace back to named programs with named adoption problems, rather than to a general uplift in farming skill.
Weeds take $4.3 billion a year; the returns come from stopping seed set
CSIRO research funded by the Grains Research and Development Corporation, reported in 2025, put the annual cost of weeds to Australian grain producers at $4.3 billion, or $203 a hectare — roughly 7 per cent more than the last national estimate a decade earlier. The same work found the yield penalty had fallen, from 2.8 million tonnes in 2016 to 1.2 million tonnes in 2025.
That gap is the argument for harvest weed seed control: catching weed seed at the header before it goes back on the paddock. Across the available methods — narrow windrow burning, chaff carts, bale-direct systems, the Harrington Seed Destructor — annual ryegrass removal averages around 55 per cent, with the best of them above 95 per cent. In Western Australian focus paddocks monitored by the Australian Herbicide Resistance Initiative, ryegrass fell from 183 plants a square metre in 2001 to close to zero by 2008, and stayed there while those growers cropped 88 per cent of the time.
Adoption followed the evidence at its own pace. A survey of 602 growers in the mid-2010s found 30 per cent using narrow windrow burning nationally, with respondents expecting that to reach 46 per cent within five years. CSIRO's Rick Llewellyn, summing up the 2025 cost work, was plain about the mechanism: weed control remains one of the biggest annual costs for growers, and the investment is paying off because growers stay ahead of the seed bank rather than because they found a cheaper chemical.
Photo by no one cares on Unsplash
Rust resistance returns about six dollars for every dollar spent
Genetic resistance to the three wheat rusts — stripe, leaf and stem — is estimated to save the Australian grains industry about $1.09 billion a year. An ACIL Allen analysis of phase four of the Australian Cereal Rust Control Program, covering 2017 to 2023, put the benefit-cost ratio at 6.02 to 1, with a present value of benefits near $264 million. The program's fifth phase, assessed in 2026, returned 5.72 to 1 on costs of $37.94 million.
Those ratios look tidy until you see what the counterfactual resembles. In 2022, two new stripe rust races, designated 198 and 239, emerged into a wet season. Inoculum reached epidemic levels and saturated paddocks kept spray rigs out of the field. Without the resistance ratings in current varieties and the surveillance that flagged the new races early, the analysis concluded wheat crops would have failed.
Stripe rust can take 60 to 80 per cent of yield from a susceptible variety. Leaf rust typically 20 to 30 per cent. Stem rust, currently held in check, has caused total crop failure here before; the 1973–74 epidemic is what produced a national control program in the first place. The 2009 rust year cost the industry an estimated $1.7 billion. Breeding programs have provided markers for more than 25 resistance genes, work that began at the University of Sydney in 1921 and has never really stopped.
In the northern zone, intensity is a bet on stored water
CSIRO's Northern Farming Systems project has run seven long-term experiments across the subtropical grains zone since 2015, accumulating data on 80 site-and-system combinations over eleven years. It tests four management levers against a best-practice baseline:
- crop intensity, meaning the share of time a paddock carries a crop rather than a fallow;
- crop diversity, meaning the range of species available and the break-crop rules that govern them;
- nutrient supply, with annual nitrogen and phosphorus budgets set to a median-yield expectation or a high-decile one;
- soil quality, tracked across the life of the trial rather than season by season.
The results resist a single answer, which is the point. Higher intensity widens the sowing window and lifts production in good years; it also draws down stored water a following crop needed. Earlier work in the South Australian Mallee, by Victor Sadras and David Roget, found a flexible intensive rotation doubled gross margin over four years and halved its year-to-year variability, with neutral or slightly beneficial effects on nitrogen leaching and deep drainage. The same study flagged a trade-off growers know well: water-use efficiency can be pushed up at the expense of nitrogen-use efficiency, because a crop grows best when no single resource is running short.
Orchards show what intensification looks like, and where it doesn't
The National Tree Crop Intensification Program ran from 2020 to 2025 on a $28 million budget channelled through Hort Innovation's Hort Frontiers fund, across almonds, avocados, citrus, macadamias and mangoes. Its most striking result was in mango, where high-density and trellised systems produced up to 310 per cent more yield than traditional orchards. Economic modelling found that despite higher establishment costs, intensive blocks reached positive cash flow quickly. The program closed with a field day at the Queensland Department of Primary Industries research facility at Bundaberg in May 2025.
Avocado told the opposite story. The program found intensification constrained by a shortage of low-vigour cultivars and by the sheer vigour of the trees; canopy management and pruning costs ate the gains, and the promising 'Ashdot 17' rootstock needs more validation before anyone builds a business case on it. That is a useful public result. A research program willing to report where its central premise doesn't hold is worth more than one that reports only the 310 per cent.
Photo by James Baltz on Unsplash
The season still writes the headline
None of this makes production predictable. In 2024–25, New South Wales lifted winter crop output 79 per cent and Queensland more than doubled to its biggest crop ever, while South Australia and Victoria ran well below average after a dry winter and severe September frosts. A year later the pattern flipped: New South Wales down 10 per cent on below-average spring rain in the south, Victoria up 30 per cent to its third-highest crop, South Australia up 68 per cent.
Better practice blunts those swings rather than removing them. ABARES analysts have noted that improved management cushioned some losses in the dry southern seasons. It didn't prevent them. Average farm cash income nevertheless rose in 2025–26, up $67,000 or 42 per cent to about $227,000 a farm in real terms.
The other constraint isn't agronomic. China's January tariff move, at 55 per cent on beef above set quotas, sits behind the forecast slide to $95 billion in 2026–27. Research can lift tonnes per hectare. It cannot lift a trade barrier, and it cannot make a dry spring wet.
What this means for your farm
Productivity gains now arrive in increments: a variety with a better disease rating, one more herbicide mode of action retired from the rotation, a canopy decision made three seasons before the yield shows up. The gains sit in the detail, and the detail is where the money leaks when nobody checks it.
Weed control is the place the arithmetic is most visible, because the cost is annual and the benefit compounds in the seed bank. Disease resistance is the cheapest input on the farm and the least visible, since you never see the epidemic that didn't happen. The intensification case also differs by enterprise — clear for mango, not yet for avocado — and a program result from a Bundaberg trial block is not automatically a result for your paddock.
Here is a next step you can take before the season turns. Pick the paddock with the highest ryegrass count and write down, on one page, the seed-set control it actually received last year: at harvest, through the summer fallow, at crop topping. Not what the program said; what the paddock got. If the honest answer is one tactic, used once, that is the paddock to spend next season's weed budget on.
- crop intensification
- Australian agriculture
- ABARES
- winter crop production
- harvest weed seed control

