• Water repellence resulted in low plant numbers on the control plots. Ripping with inclusion plates substantially increased crop establishment which were further increased by spading, with the greatest plant density on spaded plots with lucerne pellets added.
• Physical intervention increased wheat yields on a deep water repellent sand. The controls using best practice agronomy on unmodified sands, yielded 0.5 t/ha. Ripping with inclusion plates to 30 cm doubled yield, ripping deeper to 41 cm delivered an additional 0.4 t/ha and spading to 30 cm delivered a further yield increase of 0.5 t/ha, or 1.4 t/ha better than the control.
• The addition of 5 t/ha of lucerne pellets or high rates of fertiliser increased wheat yields when incorporated by ripping to 41 cm or spading, and delivered the highest yields of 3.1 t/ha.

• Field pea, vetch, faba bean and lentil had 3-125% higher biomass yield and 51-110% higher grain yield than chickpea and canola at Minnipa, 2018.
• Current high demand and grain prices for faba bean meant they were the most profitable break crop species.
• Low grain yield of canola (0.43 t/ha) and chickpea (0.39 t/ha) resulted in these crop species not being profitable as grain crops.
• Field pea and vetch, in particular, have multiple alternative end-use options in dry seasonal conditions that can be utilised to recover crop input costs.

• This trial aims to investigate opportunities for symbiotic improvement.
• Inoculation will be critical to the establishment of new legume species, where they have not previously been grown. The extent to which inoculation can still improve medic nodulation is being investigated.
• There was no response to medic inoculation in the 2018 trial, even when the inoculation rate was doubled.
• Trigonella was better nodulated and produced greater root weights than the medics.
• These results cannot be fully interpreted until the N-fixation data becomes available.
• Medic breeding lines with improved N-fixation capacity are being developed as part of the Dryland Legume Pasture Systems project and will be tested in 2019.

• This is a component of a new five year Rural Research and Development for Profit funded project supported by GRDC, MLA and AWI; and involving Murdoch University, CSIRO, SARDI, Department of Primary Industries and Regional Development; Charles Sturt University and grower groups.
• This trial aims to assess a diverse range of annual pasture legumes in order to determine whether there are more productive and persistent options for the drier areas (<400 mm) of the mixed farming zone of southern Australia.
• Several new legume species established well under very difficult conditions, but in general were not as productive as the medics or vetch. 
• In the 2018 growing season, Caliph barrel medic was the best adapted pasture legume species to the conditions at Minnipa.
• Astragalus was the best adapted alternative legume species, although it was poorly nodulated.
• Zulu arrowleaf clover also performed well, however its peak dry matter production and flowering time may be too late for SA conditions.

• Novel legume species and genotypes have the potential to reduce feed gaps and provide other farming systems and livestock benefits in low to medium rainfall regions of southern Australia.
• The five-year large scale grazing systems trial established at Minnipa in 2018 is the main livestock field site for the national Dryland Legume Pasture Systems project. Five annual legume species are being tested.
• The priority for this trial in 2018 was to optimise seed set given the poor seasonal conditions, and therefore no grazing was undertaken in the establishment year, but instead will commence with sufficient pasture growth in 2019 to determine the best legume option for livestock production. 

• Harvest weed seed collection showed barley grass is harder to capture using harvest weed seed collection techniques due to its habit of shedding seed early.
• Burning windrows decreased the number of weed seeds present in the soil seedbank at Minnipa in 2017.
• Harvest weed seed collection allows for the capture of ryegrass weed seed and cereal screenings, potentially increasing feed utilisation for stock when placed in chaff dumps or rows.

• Nodulation was reduced when pea seeds were coated with P-Pickel T fungicide before inoculation.
• Reductions in rhizobia numbers on fungicide-coated seeds occurred very quickly, within two hours, but the toxic effect of the fungicide continued after sowing.
• Dry soil conditions are likely to have exacerbated fungicide toxicity to the rhizobia.

• The impact of 13 different soil wetter treatments ranged from none to more than a doubling of wheat establishment.
• Grain yield benefits ranged from zero to 21% (or 0.22 t/ha).
• While higher grain yield generally correlated with better crop establishment, some treatments with low crop establishment were able to generate substantial yield responses, suggesting specific chemistries may promote later season effects.
• Combinations of some soil wetters interacted positively to maximize grain yield response.

• Interactions between herbicide and application rate did not have differing responses for grain yield, biomass yield, or gross margin at Minnipa, 2018.
• Combined herbicide and application timing treatments did show differences in grain yield, biomass yield and gross margin.
• Metribuzin and Diuron treatments indicated increased safety levels (58-466% higher biomass and 6-161% higher grain yield) and lower economic risk compared to Terbuthylazine treatments.
• For herbicide applications incorporated by sowing, Metribuzin was of lower economic risk (45% higher gross margin) than Diuron.

• There is the potential to increase plant dry matter production and grazing using mixtures of plant species.
• Kaspa peas produced the greatest dry matter overall with 1.7 t DM/ha, and an oats/medic mix also did well (1.3 t DM/ha). The short season oats variety, Durack, produced the greatest dry matter for oats with 1.1 t DM/ha. Toreador disc medic (bred for sandy soils) produced the best dry matter (0.75 t DM/ha) amongst the pasture species.
• In a poor season like 2018, dry matter production in pastures could be improved compared to the current district practice of volunteer annual medics.
• Further research needs to be done on regeneration and herbicide tolerance of newer pasture species.

• UAV imagery with skilled specialist analysis and programs has the potential to identify and map weed issues in paddocks.
• Grass weed patches are easier to identify in legume and pasture crops than cereal crops.
• Variability in germination and crop growth makes it harder to identify grass weed areas.
• Data capture and analysis for analytical purposes such as grass weed mapping in individual paddocks will be beyond the skills of most farm businesses unless farmers have a special interest in this area.
• Group A resistant barley grass has once again been detected at the Minnipa Agricultural Centre. Barley grass resistance to Group A herbicides has been detected three times in three years, so be aware it is present current farming systems.

• At Cummins, an increase of up to 100% grain yield in wheat to 5 t/ha over the control was reached by applying 120 kg N/ha (two split applications of 60 kg/ha each at emergence and before GS31) and extra irrigation (50 mm) at GS31.
• At Lock, up to a 19% higher grain yield to 5 t/ha was achieved compared to control by adding extra irrigation (50 mm) at GS31.
• At Minnipa, an increase of up to 31% grain yield to 3 t/ha over the control was achieved by adding extra irrigation (50 mm) at GS31 and applying 120 kg N/ha in two split applications of 60 kg/ha at emergence and by GS31.
• Water use efficiency (WUE) improved with N fertilisation by GS31 at Minnipa and Cummins.
• The split application of 120 kg N by GS31 may only be a good investment in paddocks where crops with N requirements (e.g. canola) were grown in the previous season or soil N at sowing was low (less than 44 kg/ha) and with average or above average seasonal soil moisture at GS31.

• Standard inoculation practices are unlikely to deliver satisfactory nodulation where pulses are sown into dry soils.
• Pea and chickpea sown late (26 June) had the higher nodulation and crop production than those sown dry. Where the soil remained dry for 20 days after sowing nodulation was poor, regardless of the inoculation strategy used.
• Nodulation of pea was low regardless of sowing time at Minnipa in 2018. Both peat applied to seed and granule formulations resulted in similar nodule numbers per plant.
• Nodulation can be improved when dry sowing if inoculation rate of peat is increased. Doubling the rate of inoculation significantly improved pea nodulation. 
• The performance the rhizobial granules was inconsistent. A granule produced at SARDI was more effective in one experiment, but not the other. Two commercial granules provided no advantage compared to peat applied to seed.

• Proximal sensing reflectance data predicts soil moisture with reasonable accuracy at depths (0-10, 10-30, 30-90 cm) on upper Eyre Peninsula in 2018.
• Reflectance data may also be useful for predicting the amount of soil nitrogen and crop macronutrients, including but not limited to nitrogen, phosphorus, potassium and sulphur.
• Further experimental data is required to use soil and crop reflectance as a means to predict nutrient content because environmental parameters can confound results.

• In 2016 only 40% of the barley grass was greater than 17 cm tall and had the potential to be collected by early swathing, or 60% of the barley grass seed had already shed or was below harvest weed seed collection height.
• Sampling of the swathed paddock area showed slightly better barley grass weed seed collection using swathing than from the rest of the paddock which was harvested normally.
• Seed capture at both swathing and harvest is higher with annual ryegrass than with barley grass.

• Over the last four years, trafficking at any level has not decreased crop production on a heavy Minnipa soil so moving to controlled traffic farming on this soil type is not likely to increase crop production due to less trafficking in the paddock.
• Trafficking on a deep sand (trial at Loxton in the northern mallee) has substantially reduced crop production for at least 4 years so CTF may have additional benefits on deep sands, especially if ripping is used to correct existing compaction.
• Similar to many previous trials, ripping has not benefited production on the heavy Minnipa soil.

• All oat varieties tested at Kimba in 2018 produced the same dry matter, hay quality and grain yield.
• Durack was the earliest maturing variety (for potential hay cutting and grain), so a good choice in very dry seasons.
• In a dry season Moby barley performed well for early dry matter production compared to oats, but performed poorly for grain yield.

• Cereal windrow burning achieved temperatures in excess of those required to achieve high levels of weed seed mortality, except in paddocks which had 11mm of rainfall the week before.'
• The open paddock burn with a high stubble load had a quicker, faster burn but still achieved the necessary temperatures of 450 degrees Celsius for longer than 60 seconds.

In 2017, approximately 50% of barley grass weed seed was already on the ground (dropped) before the first swathing opportunity.

• Vetch production in a very tough season was little affected by trafficking imposed in 2015.
• After three years of crop production, low levels of trafficking have had little impact on grain yield on a Minnipa soil, multiple trafficking on wet soil (in 2015) increased grain yields in the second and third crops.
• Ripping conducted in 2015 lifted the soil substantially but has not benefited production in any of the following three crops.

• Intake (on fertiliser) and Jockey (on seed) is current standard practice for lower EP canola and reduced Blackleg stem infection. In 2014 this practice also increased grain yield. This remains the recommended practice.
• The application method for the fungicide, either as seed treatments or in-furrow, did not increase canola yields in 2015 or 2016.
• The source of nutrients and application method, either fertiliser, in-furrow or foliar had no effect on canola yield in 2014, however further research should be undertaken in this area before final conclusions are drawn.

• Phosphoric acid as a liquid fertiliser resulted in 13% and 8% higher yields in 2014 and 2015 respectively at Streaky Bay, and an 8% yield increase at Warramboo in 2016 when compared to the same rate of P delivered in granular fertilisers.
• Despite the yield increases at Streaky Bay and Warramboo, the economics showed granular fertiliser had greater returns in $/ha compared to using phosphoric acid at 9 units of P.
• The guidelines for moving to the adoption of phosphoric acid as a P source exist in the Fluid Fertiliser Manual which adequately covers the principles, economics, recommendations and chemistry of adopting fluid delivery systems. The manual is available at www.fluidfertilsers.com.au.
• Knowing the responsiveness of a soil type to phosphoric acid is an important factor to consider before investing in a fluid delivery system. The responsive soil types are generally higher n calcium carbonate (greater than 25%). P source responses may also be driven by soil moisture conditions because in 2016, which had wetter seasonal conditions, phosphoric acid and granular P performed similarly in the same soil type at Streaky Bay.
• The adoption of fluid fertiliser systems to place fungicides in-furrow for rhizoctonia disease management is not recommended in this environment. The addition of fungicides showed small and variable yield advantages at Warramboo in 2014 and 2016 and Streaky Bay in 2016. Using fungicides in above average seasons the greatest yield benefit was only 0.22 t/ha with an in-furrow application with granular phosphorus and trace elements. This is lower than the benefits achieved in other regions where yield increases of up to 0.49 t/ha in wheat with split applications of fungicides were achieved. Inluding fungicides on wheat up front will increase input costs and risk over a large cropping program. 
• Large scale demonstrations by growers across Eyre Peninsula with in-furrow fungicide applications at seeding did not reduce rhizoctonia or produce yield benefits over three seasons.
• The adoption of fluid fertiliser systems will depend in growers' soil types for phosphorus responses, the existence of trace element deficiences and ease or timing of application (in-furrow at seeding) and machinery ability to deliver fluids.

• Applying phosphorus to a soil with low P reserves when establishing a medic pasture boosts shoot and root dry matter, improves root health and improves N fixation.
• The addition of urea at seeding can reduce nodulation in medic pastures.
• Residues of the herbicide Logran can severely stunt medic growth.
• Applying a full label rate of Agritone 750 (late) decreases pasture production and N fixation in actively growing pastures.
• Any management practice that reduces medic biomass will reduce N fixation, such as the late application of certain herbicides, eg Agritone 750.
• In a dry growing season when medic plants are already moisture stressed, herbicides will have little impact on medic production and N fixation.
• Regardless of seasonal conditions, the management of newly established medic pastures with regard to herbicides and nutrition, appears to have no impact on the yield and protein levels of the wheat crop in the following year.

UAV imagery with appropriate analysis has the potential to identify weed issues in paddocks quickly, reliably and cheaply over large areas.

• There was no difference in grain yield (t/ha) between the early season variety (Trojan) and the later varieties (Mace and Scepter) sown 13 days later.

• In 2016 sowing Mace and Scepter earlier on 13 May reduced grain protein by 0.3% compared to sowing it thirteen days later.

• For the later sowing on 26 May, Mace and Scepter had similar grain yield, however, protein was higher for Scepter.
• When sown early, Trojan had higher protein than Mace and Scepter.

• In 2016 there was a small yield response to phosphoric acid over granular P at Warramboo.
• Fungicides and the addition of extra 40 kg/ha urea at seeding separately reduced Rhizoctonia seminal root infection compared to the control at both sites.
• Including fungicides will increase input cost and risk over a cropping program.
• The addition of a trace element mix or manganese did not improve yields at Streaky Bay or Warramboo.

• Fungicides as seed treatments or in-furrow did not increase canola yield in 2015 or 2016.
• In 2016 Intake (on fertiliser) and Jockey (on seed) which is current standard practice, lowered Blackleg stem infection, and in 2014 this treatment increased yield.

• Over nine seasons, incorporating livestock into the rotation improved overall system outcomes in this trial, including; increased nitrogen cycling and water use efficiency, reduction in weed and pest populations and no negative effects on cereal performance or soil health, while value adding to stubble and pastures by grazing.
• The high input grazed farming system had a gross margin of over $100/ha/year more than lower input and ungrazed treatments over the nine-year trial period.

• Trafficking on wet soil in 2015 resulted in substantially less Rhizoctonia in barley in 2016.
• After two years of cereal production, there is little evidence that heavy vehicle trafficking is severely depressing grain yield on a Minnipa soil.

• Growers should be aware that the application of certain broad-leaved herbicides can result in a reduction in shoot dry matter of medic pastures.
• Application of a full label rate of Agritone 750 (late) resulted in the largest reduction of shoot dry matter.
• Applying P to a soil with low P reserves when establishing a medic pasture boosts shoot and root dry matter, improves root health and improves N fixation.
• Growers should also be aware that the use of herbicides can reduce nodulation and N₂- fixation in medic pastures.

• The yield of both pulse crops decreased in general when delaying sowing.
• Modifying sowing time produced changes in the phenology that affected main yield components and grain yield.
• In general, the later sowing dates reduced grain number/m² and produced a lower thousand grain weight.

• The paddock contained a population of effective rhizobia and so the number of nodules on lateral and taproots was not affected by the addition of different rhizobial inoculants.
• Improved soil nutrition did not affect dry matter production and nodulation.
• Grazing (simulated) reduced the total dry matter produced by the medic pasture, however the use of sheep would provide a more realistic assessment of grazing on DM production, nodulation and N₂-fixation.

• At present, group A herbicide resistance in barley grass is relatively low at a district level on the EP. However, growers need to act now to integrate multiple control tactics to prolong the effectiveness of these cheap and effective herbicides.
• In the absence of group A herbicides, it is still possible to achieve large reductions in barley grass seed bank in a legume pasture phase.
• When using moderate efficacy-low cost herbicides (Treflan + Logran) for barley grass control in wheat, effective control of the weed in previous pasture phase was critical.

• There was no increase in wheat (2015) or barley (2016) performance with application of manganese on highly calcareous sands, even though these environments have regularly produced severe Mn deficiency in the past.
• Current industry guidelines still appear to be relevant for diagnosing trace element deficiencies in cereal crops and how they are best managed.

• The recommended field pea industry practice of P-Pickel T^® seed treatment and 2 foliar fungicides of Mancozeb failed to significantly reduce disease infection levels or increase grain yield over untreated control treatments under high blackspot disease pressure in 2016.
• Early disease control applications (four weeks after sowing) were important for reducing initial blackspot infection levels at Minnipa, conversely later spring applications were important at the higher rainfall site of Hart.
• Over two consecutive years, a yield benefit of at least 15% has been obtained from application of new experimental fungicide actives over the current industry practice treatment of two foliar sprays of mancozeb.
• Further research is required to understand the interaction in efficacy between fungicides and timing of disease infection, together with the drivers of Ascochyta blight onset and progression in different field pea growing environments.

• There was no detectable direct effect of sowing direction this season at Minnipa with a mild spring.
• Using a narrow row spacing of 18 cm instead of 30 cm resulted in wheat yield increasing from 3 t/ha to 3.6 t/ha (19% increase).
• Light interception was very sensitive to sowing direction, and not having ‘weeds’ resulted in higher light within the canopy in the north-south direction compared to east-west.
• Knife point and ribbon seeding systems achieved similar crop establishment and crop performance.

• SAGIT funded trials over the last 3 years have shown the potential of vetch on Eyre Peninsula, leading to the selection of an early flowering (90-95 days) line with increased winter vigour (SA 34876) as a new variety for low rainfall mixed farming systems.
• The last three seasons have been particularly good at Minnipa for both grain and hay production of vetch.
• Piednippie trials have not been as successful as Minnipa, but have still shown with the right management vetch could produce good hay, fodder or grazing on the grey calcareous soils.
• On Eyre Peninsula early sowing (mid- April) can achieve good yield and hay production but is heavily reliant on either good subsoil moisture or late follow-up rain.
• Herbicide choices for vetch are very dependent on local conditions so talk to your local agronomist about the best options available.

• Reducing row spacing to 18 cm from 30 cm increased grain yield of wheat by more than 0.5 t/ha in 2015 and 2016.
• Increasing seeding rate also increased grain yield in 2015 and 2016.

• Late grassy weed dry matter was 65% lower, and barley grass weed seed set was 57% lower, with a higher seeding rate.
• 18 cm row spacing had 42% lower grass weed dry matter than 30 cm row spacing.
• Single row or spread row seeding boots showed little differences in plant establishment, grain yield and quality or grass weed competition.

• The optimum agronomic sowing window for field pea coincided with high blackspot risk in many districts of South Australia in 2015.
• Under such high disease risk situations, growers in low rainfall areas may be best suited to choose alternative break crop options to field pea to avoid significant yield losses through delayed sowing or disease infection.
• Experimental fungicide treatments with greater efficacy than Mancozeb showed improved blackspot control and significant yield increases over the Nil and Mancozeb treatments in 2015. Further assessment and application approval is still required.

 

• Field peas have proven to be the most reliable pulse option on the upper Eyre Peninsula with yields remaining the most stable across seasons.
• Under favourable conditions there is potential for alternative pulse crops to be successful.
• Paddock selection, soil type, time of sowing, pulse agronomy, marketing and storage options all need careful consideration when looking at growing alternative pulse crops.
• New varietal options offer earlier maturity as well as improvements in harvestability, disease resistance and herbicide tolerance over older commercial standards, aiding production and profitability.
• Higher value alternatives, such as lentils, may be a high risk option and pulse crops better suited to the region could still prove to be the most profitable option in the long term.

• Phosphoric acid showed a yield response at Streaky Bay in 2014 of 13% and 2015 of 8%.
• Fungicides did not reduce Rhizoctonia infection or significantly increase yield in 2015 at either site despite high levels of inoculum.
• Including fungicides will increase input cost and risk over a cropping program.
• The addition of trace element or manganese treatments did not improve yield at Streaky Bay or Warramboo in 2015.

• There were no differences in canola Blackleg infection or yield using fungicides as seed treatments or in-furrow in 2015.
• In 2014 combined protection of a fungicide on seed and in the furrow as a banded fluid reduced Blackleg infection and increased yield.
• The selection of Blackleg resistant varieties in the rotation is important.

• Grazing sheep have not damaged soil health over eight years of several crop/pasture rotations.
• In 2015 total annual biomass was greater in higher input and grazed rotations. High input grazed systems carried twice the stocking rate of a low input system.

• Green cumin (Cuminum cyminum) is a tap-rooted herb grown extensively in India and the Middle East for its grain.
• It survived trials on Eyre Peninsula in 2015 sufficiently well to encourage another look.
• Alternaria fungal disease is a major threat.
• Its low and slow growth will complicate harvesting and weed control.

• Nodulation and shoot dry matter in medics is depressed by poor phosphorus (P) nutrition and in-crop application of certain herbicides.
• Applying P when establishing a medic pasture boosts shoot and root dry matter and improves root health.
• The timing of application of certain herbicides can have an important effect on maximum shoot dry matter.
• High soil nitrogen levels can reduce nodulation in medic pastures.

• The paddock utilised for the trial had a significant number of effective naturalised medic rhizobia. There was no rhizobial inoculation response.
• Total annual biomass was higher in the regenerating medic plots versus sown treatments due to earlier germination and growth of naturalised medic.
• Grazing increased overall medic production.

• Early sowing (22 April) had the largest positive impact on canola yield when comparing a range of treatments trialled in 2015, similar to results observed in 2013 and 2014. Sowing on 22 April improved yields up to 12%, depending on variety, compared to 29 April sowing date.
• Different management practices are needed to establish canola on different soil types under marginal moisture conditions. On a loamy soil at Minnipa deeper sowing (to 4 cm) produced the highest yields, whereas shallower sowing (1 cm) produced the highest yields at Piednippie, particularly with the smaller seeded variety Stingray.
• Trials at Piednippie in 2015 showed that yields up to 0.8 t/ha are achievable in a below average rainfall season, provided the crop is established early and achieves around 50 plants/m² in open-pollinated varieties.
• Results from similar trials in 2013, 2014 and 2015 have all shown that the highest yielding canola has consistently come from plots with plant establishments of over 50 plants/m² for open pollinated varieties and 40 plants/m² for hybrid varieties.

• Early sowing (22 April) had the largest positive impact on canola yield when comparing a range of treatments trialled in 2015, similar to results observed in 2013 and 2014. Sowing on 22 April improved yields up to 12%, depending on variety, compared to 29 April sowing date.
• Different management practices are needed to establish canola on different soil types under marginal moisture conditions. On a loamy soil at Minnipa deeper sowing (to 4 cm) produced the highest yields, whereas shallower sowing (1 cm) produced the highest yields at Piednippie, particularly with the smaller seeded variety Stingray.
• Trials at Piednippie in 2015 showed that yields up to 0.8 t/ha are achievable in a below average rainfall season, provided the crop is established early and achieves around 50 plants/m² in open-pollinated varieties.
• Results from similar trials in 2013, 2014 and 2015 have all shown that the highest yielding canola has consistently come from plots with plant establishments of over 50 plants/m² for open pollinated varieties and 40 plants/m² for hybrid varieties.

• The trials across agro-ecological zones have highlighted differences between varieties in their ability to respond to a range of production drivers over varied environments and seasons.
• There was no varietal difference between Mace and Trojan at Minnipa with both yielding an average of 1.7 t/ha, and the ungrazed treatment yielded 0.2 t/ha higher than the grazed. There was no difference in grazing response between cultivars.
• Nitrogen (N) can provide assistance in plant recovery through higher yields and grain quality with in-season application of N after grazing.

• Canola yields of 2 t/ha are possible in an average growing season (decile 5) at Minnipa Agricultural Centre.
• The total application of 85 kg/ha N and 108 kg/ha N exceeded the 2 t/ha target, producing 2.13 t/ha and 2.22 t/ha respectively, and 62 kg/ha N very nearly achieved 2 t/ha (1.96 t/ha).
• Soil tests determined that 70-80 kg/ha N was available in the soil, including mineralisation, which predicted a potential yield of 1 t/ha.