• Crop establishment increased with on row sowing. Stubble from the previous season helps soil moisture infiltrate into non-wetting sands and into a position closer to the seed.
• Initial germination of brome grass weeds was higher with inter row sowing.
• Late brome grass weed numbers was also greatest with inter row sowing. Sowing on row may also increase crop and weed competition as will a higher seeding rate.

There were no differences in medic regeneration in wheat stubbles either harvested at two heights, left standing or rolled.

Average trial yield was 2.8 t/ha, approximately 1 t/ha below the nitrogen unlimited potential as identified by Yield Prophet® during the season. No significant differences in yield were found between stubble treatments (stubble retained, worked or removed) and nutrient treatments (normal practice, normal practice plus additional nutrients to enhance stubble breakdown). After three years of trial work, no significant differences in soil carbon were found between the stubble and nutrient treatments.

Following two years of grass free medic pasture soil mineral nitrogen levels averaged 158 kg N/ha and all disease inoculum levels were low.

Cutting the medic pasture for hay lowered dry matter production and medic pod set in the following season.

An early light tickle resulted in higher weed germination and lower powdery mildew damage in spring.

Full cut tillage following two years of grass free medic resulted in the highest wheat yields while a discing prior to seeding resulted in the lowest yield.

Working medic residues in the year prior reduced the following wheat yield but grain protein was higher.

Cutting a medic pasture for hay in 2013 reduced medic productivity in 2014 and lowered wheat grain protein in 2015.

• In 2015 the drier start to the season and low soil moisture resulted in lower herbicide efficacy and less chemical damage than expected.
• In different stubble management systems the activity and resulting weed control of specific herbicides will be influenced by the solubility index (movement through the soil profile with rainfall events) of that herbicide. Soil texture and soil chemical properties can affect chemical movement and availability in the soil profile.
• Herbicide performance will vary seasonally due to soil moisture levels, rainfall pattern post application, timing of weed germination, position and number of weed seeds in the profile, etc. Understanding how the various herbicides work can reduce the likelihood of failures.
• Herbicides are only one tool for weed control – always adopt an integrated weed control package that includes non-chemical control, and where possible, consecutive seasons of total weed control.
• Consider the whole farming system when making chemical decisions as the impact may last for several seasons (eg. effects on medic germination and medic seed bank).

• There were no differences in wheat yield at Minnipa in response to stubble architecture, seeding position and nitrogen treatments in 2015.
• In 2015 plant establishment was reduced with cultivation and the addition of nitrogen at seeding compared to standing stubble cut low at harvest. The extra nitrogen applied at seeding also reduced the early dry matter.
• Removing and cutting stubble low decreased the Yellow Leaf Spot disease incidence and snail numbers compared to high cut stubble.
• Stubble management and seeding position had little effect on grass weed numbers.
• Overall at Minnipa, stubble management and seeding position have not impacted highly on crop production, weeds, disease and pests over two years with relatively high stubble loads in low rainfall farming systems.

• An east-west (E-W) sowing direction increased yield over northsouth (N-S) sowing direction in an average season.
• The results showed a decline in yield due to weed competition, but no effect on weed competition due to row direction. So sowing in an E-W direction may give a yield benefit with no difference in weed seed set.
• The wider row spacing of 30 cm resulted in a yield reduction and greater weed biomass at harvest.
• There were no differences in yield with ribbon seeding with either 18 or 30 cm row spacings, but ribbon seeding reduced ‘weed’ biomass.

• Managing herbicide resistance in ryegrass continues to be crucial in maintaining sustainable crop production on Lower Eyre Peninsula.
• Management strategies other than herbicides need to be deployed to ensure sustainable ryegrass control into the future.
• Windrow burning proved to be an effective method in reducing ryegrass seed numbers in 2015.
• Managing ryegrass on differing soil types will prove a challenge into the future.
• Information generated by this project will provide data to simulate how different management strategies can be used to manage ryegrass in a sustainable, cost effective way.

• 18 cm (7”) systems showed better plant establishment in a drier seeding than the 30 cm (12”) system.
• Higher seeding rates resulted in higher grain yield but also higher screenings and lower protein.
• Grass weeds were lower in the higher seeding rate and in the 18 cm row spacing indicating crop competition is a non-chemical weed reduction method.
• Single row or spread row seeding boots showed little differences in plant establishment, grain yield and quality or grass weed competition.

There were no differences in grass weed numbers in paddock N7/8 between cereal crop treatments imposed of Sakura, East West and North South sowing, cross sowing and ‘15 cm nudge’.

• Deeper sowing at 3-4 cm achieved better early dry matter and increased grain yield in a later sowing.
• Frost damage may have lowered yield with early sowing, but inter-row placement at TOS1 had the highest yield.
• TOS 3 placement in row achieved better early plant growth and higher grain yield.
• Later sowing reduced brome grass numbers, and the size of the panicles, so it substantially lowered returns to the seed bank.

In the 2014 season there were no significant yield advantages recorded at five sites across upper Eyre Peninsula when using the fungicide products over the nil treatments. This was in a season with an early start and minimal stress during crop establishment, and at sites with high Rhizoctonia inoculum levels. The broad acre farmer demonstrations in the 2014 season showed no visual differences or early plant growth measured at the given sites during the cropping season. There were differences in the level of crown root infection at Cleve, where the nil treatment had a higher number of crown roots infected and a greater % of crown root infection than the fungicide treatments. There were differences between treatments recorded in the mid-May sown barley crop at Piednippie in Rhizoctonia patch score and the seminal root score. Differences in protein, screenings and test weights were recorded between treatments at several sites with a general trend of lower yields having higher protein, higher screenings and lower test weights. Further evaluation of research trials and farmer demonstrations using new fluid products and fungicide placement, will occur over two more seasons to evaluate the economics of using fungicides in low rainfall farming systems.

Average trial yield was 3.8 t/ha, close to the potential as identified by Yield Prophet during the season. No significant differences in yield were found between stubble treatments (stubble retained, worked or removed) and nutrient treatments (normal practice, normal practice plus additional nutrients to enhance stubble breakdown). Changes in Soil Organic Matter fractions over the duration of the trial (2012 to 2014) resulting from the stubble and nutrient treatments will be assessed in March, 2015.

• There was a 0.17 t/ha yield advantage in the 2014 season due to removing or cultivating the previous season’s stubble, possibly due to extra nitrogen being available to the crop during the growing season.
• There was a 0.08 t/ha yield advantage at MAC this season by inter row cropping rather than placing the seed on row.
• There were no differences in either barley or ryegrass numbers at the start of the cropping season or in crop.

The 1.6 t/ha pasture residue had no impact on sowing and plant establishment in drier seeding conditions in 2014. • In 2014 there were no differences in wheat establishment, yield or grain quality due to different pre-sowing treatments with pasture residues. • The wheat stubbles have been harvested at different heights and pasture establishment and will be monitored next season.

At Minnipa in 2013 there were cereal yield responses to fungicide treatments in both wheat (up to 14% better than no fungicide) and barley (up to 12 % better than no fungicide but not all strategies were effective). However there were still visual Rhizoctonia patches present. In-furrow fungicide applications were more effective than seed treatments. Tillage, starter nitrogen and zinc produced similar yields to many of the fungicide treatments. A three week delay in seeding reduced yield by nearly one third. Fungicide treatments did not prevent an increase in Rhizoctonia inoculum levels during a cereal phase.

In 2013 the Rhizoctonia best bets demonstrations using rotations were over sown with cereal and showed break crops which are grass free are the best option to lower Rhizoctonia inoculum levels.

No difference in yield or quality was noted for the stubble or nutrient treatments. It is likely that soil carbon is being increased as the additional nutrients did not increase yield. Increasing soil carbon is a slow process and several years are needed to see whether these treatments do increase soil carbon. A range of sites across southern Australia have been chosen to see if soil carbon can be increased.

• Zn deficiency was evident at mid-tillering at one site only (Streaky Bay) in treatments where zinc had not been applied.
• Manganese deficiency was not identified via tissue testing at Streaky Bay despite treatments with added manganese having better earlier growth.

2012 was a set up year for the Rhizoctonia best bets demonstrations. Fungicides banded with fluid fertilisers at seeding did not reduce Rhizoctonia in the following cereal crop.

The extra nutrition treatment had no effect on yield or grain quality this season. In the 2012 season the stubble removed treatment had the highest yield.

• Applying N at GS31 increased yield but no response to manganese was measured.

• Canola and medic will reduce Rhizoctonia inoculum but only one year of cereal will result in high levels of Rhizoctonia inoculum again.
• Disease suppression has not developed in this environment despite it happening within the same time frame in other environments.

• No further yield benefit was measured from applying other nutrients (S, K, Mg) and trace elements (Zn, Mn, Cu) over and above that from applying nitrogen and phosphorus alone.

• Compared to 2008 map, rezoning based on 2009 yield monitor map improved variable rate technology (VRT) performance in 2009, 2010 and 2011 seasons.
• The highest gross margin was returned from paddock N1 as a result of zoning whereby no fertiliser was being applied to 89% of the paddock.
• A limited yield response to applied fertiliser has restricted any gross margin benefit from the VRT technology from this paddock with high soil P reserves.

Over 3 years of crop production (2009-2011) applying replacement phosphorus (P) rates have been the most economic.

• The fluid fertiliser system increased yield and dry matter production at both sites.
• The added carbon treatments with increased microbial activity have not increased grain yield in 2011.

• Disease suppression has not developed in this soil type after 8 years.
• Higher nutrition treatments allowed crops to cope better with high disease levels in both 2010 and 2011.
• Canola will reduce Rhizoctonia inoculum but only one year of wheat will result in high levels of Rhizoctonia inoculum again.
• In good seasons the high input treatments have shown that district practice performance is limited by poor nutrition.

• There was a dry matter response measured at anthesis to all treatments with application of manganese but due to dry season this was not translated into yield increases.

• Despite three years of above average rainfall, a red calcareous sandy loam at Minnipa with soil phosphorus reserves of 27 mg/kg of Colwell P has not required any added P fertiliser for maximum cereal grain yields.

Only medic based systems from 2008 and 2009 resulted in higher wheat yields in 2011.

• Wheat yields in the poor, medium and good zones in response to low, medium and high inputs respectively did not differ significantly in 2011.
• Over a 3 year wheat-canola-wheat rotation there has been no measured production benefit from applying 6 kg/ha of P on the good soil type as opposed to nil P on a shallow constrained soil.

• Annual medic plant establishment density limited by 2008 drought, mice and possibly a false break.
• Pasture production restricted by low plant numbers but increased by 2010 P applications.

• Grain yields reached 80% of their potential water use efficiency in a decile 5 growing season.
• Variable rate strategies did not result in improved yields over blanket strategies and the costs of the inputs ensured the blanket nil fertiliser treatment was at least economically comparable to the other options evaluated.

• Axe, Wyalkatchem and Gladius appear to be the wheat varieties that use added P efficiently.
• Scout is able to maintain yield when no P is added.

• A replacement P strategy produced similar yields to 10 and 20 kg/ha P rates in 2010.
• The trial indicated an economic benefit in increasing P rates up to 20 kg/ha on a deep sandy loam in 2010.

• Canola in rotation reduced Rhizoctonia inoculum levels and increased following cereal yields.
• Fluid fertiliser increased plant dry matter and yield on highly calcareous grey soils.
• A barley/vetch brown manure increased soil mineral nitrogen and this appears to have exacerbated Rhizoctonia disease symptoms on a following barley crop.
• Added carbon (10 t/ha/yr) has increased microbial respiration and microbial nitrogen.

• Changing rotation and nutrition have changed the microbial population activity and diversity but not developed disease suppression after seven years.
• One year of wheat in 2009 resulted in high Rhizoctonia inoculum regardless of previous rotation or fertiliser histories.
• Take-all has increased in the continuous cereal district practice treatment to severe levels; higher nutrition treatments give the plants the ability to cope better with these increased disease levels.
• The high input systems have continually yielded higher than the other systems, but inputs of N have not matched production levels completely so soil reserves of mineral N are now lower in these treatments.
• In good seasons the high input treatments have shown that district practice performance is severely limited by inadequate nutrition.

A site with high phosphorus (P) reserves needed no applied P fertiliser in 2010 to produce a 2.5 t/ha wheat yield.

• At Minnipa the late sown (resown due to mice damage) wheat gave a yield response to 20 units of up-front N, but not to N applied in-crop in 2010.
• At Mudamuckla there was no response to applied N above base rates on soil with adequate N content in 2010.

• 4 kg/ha of P as a blanket application increased grain yields and gross margins over nil P in all 3 soil zones, however applying a variable rate to the medium soil zone of 8 kg P/ha improved gross margins over the 4 kg/ha rate.

• In an above average season a high input system was the “best bet”.
• Variable rate had reduced production but at a lower risk.

• After two good seasons a variable rate approach is as profitable as a low input blanket approach and more profitable than a standard blanket approach.

• In low rainfall seasons, (2005-08), north-south sowing increased grain yield by an average of 8.4% compared to east-west.
• Narrow row spacing with retained stubble also showed increased grain yield.
• In 2009 (decile 9+ season) there was a 0.24 t/ha yield advantage in wheat with sowing east-west.
• In 2010 (decile 8-9 season) with Kaspa peas, row direction at sowing had no effect on grain yield.
• Ultimately direction of sowing will depend mostly on paddock shape and direction of sand hills.

• It appears that once disease suppression is achieved in low rainfall farming systems of upper EP and inoculum levels are low, it is quite robust and will be maintained despite systems with substantially different amounts of mineral N.
• Fallow, Medic/Fallow and Peas all lowered rhizoctonia inoculum levels in 2010, and Fallow and Peas lowered levels in 2009.
• Continuous cereals produced the highest inoculum levels but all levels were low at this site.
• Even high mineral N (118 kg N/ha) in the top 60 cm (measured in April after spraytopped medic) did not increase Rhizoctonia patches in 2010.
• Trichoderma fungus and the beneficial PEM microbes are present in the MAC N12 system.

• There was no yield response to increased P rates in any variety tested suggesting that adequate available soil P masked any purported difference in P use efficiency between varieties.
• However, there was a trend in that the grain yield of the variety Axe increased in response to higher levels of P in both the replicated and broad acre studies.

• Rhizoctonia solani AG8 inoculum levels were reduced over summer and this may be associated with summer rainfall events.
• Crop rotation does affect Rhizoctonia inoculum levels, i.e. levels were lowest immediately after canola compared to that after wheat.
• The incidence and severity of Rhizoctonia bare patch disease in cereals depend on the amount of pathogen inoculum, soil microbial community activity and crop/root vigour.
• Seed treatments did not reduce rhizoctonia levels in 2009 trials undertaken on EP and the Mallee.
• To reduce risk of yield loss caused by Rhizoctonia:

1. Control summer weeds to stop build-up of inoculum,
2. Encourage early seedling vigour, sow early,
3. Cultivate deep sow shallow (avoid disc seeders).

• Canola can help reduce inoculum for the following wheat crop.
• Barley and wheat are the most intolerant crops.
• Minimise N deficiency at seeding, particularly for cereal on cereal rotations where summer rainfall has been low, by deep banding N and minimising stubble incorporation at seeding.

• Differences in C input (dry matter production plus any additions) between the treatments were 24 t/ha at Minnipa, and 31 t/ha at
• Poochera over the last two seasons.
• Wheat at Minnipa produced between 0.5 t/ha of stubble in a poor season (2008) and 8.8 t/ha in a good season (2009).
• Doubling the seeding rate of wheat at Minnipa produced 1.4 t/ha extra stubble compared to the wheat control in 2009.
• Barley and vetch brown manures produced the lowest dry matter at both sites.

• The higher input brassica break system had higher cereal yields than the other systems in 2009.
• Rhizoctonia inoculum levels were higher under the continuous cropping systems compared to the medic pasture and brassica break crops.
• The different rotations and fertiliser systems have resulted in differences in the microbial community (number) and microbial diversity (their ability to use different sources of carbon).
• Disease suppression is still the same in all rotations.

• In a good growing season, wheat performed equally well in zones identified as good, medium or poor however they achieved only 50% of their yield potential water use efficiency.