Take home messages

• Stubble reduces yield in canola-wheat rotations, particularly at higher yields

• Crop diversity is one of the main principles of conservation agriculture – inclusion of legumes and barley can improve profitability and sustainability  of stubble retained systems

• Burning wheat stubble in canola-wheat-wheat rotations in C & SNSW increases profit, but can be operationally difficult and comes at a cost to soil health and possibly sowing opportunities

Background

Recent experiments in central and southern NSW using modern zero and no-till methods in farmers’ paddocks (Table  1) and small plot trials (Table 2), as well as longer term experiments have shown that in canola-wheat rotations, cereal stubbles reduce yield. The average yield penalty from retaining stubble across 22 years at the CSIRO Harden field site is 0.3 t/ha for wheat and 0.4 t/ha for canola. The NSW DPI SATWAGL experiment at Wagga Wagga recorded similar penalties, and at both sites yield penalties due to retaining stubble were positively related to growing season rainfall. Experiments investigating these penalties in modern farming systems will be ongoing as part of research by CWFS, FarmLink and CSIRO in the GRDC stubble initiative.

It needs to be remembered that stubble is essential for reducing wind and water erosion, maximising infiltration of summer rainfall events and maintaining good soil structure, and at least 70% stubble cover (2-3t/ha cereal stubble) should be kept on all paddocks at all times during the summer fallow period. However, retaining levels greater than this past sowing is unlikely to provide any yield benefits, and will often lead to yield reductions.

Yield penalties from stubble retention are largely due to N tie up and temperature effects, and identifying and addressing these problems before they occur can improve profitability in stubble retained systems.

Table 1. Yield of canola under different stubble (wheat) treatments in paddock scale trials managed by zero-till farmers conducted by CWFS in 2013 as part of the GRDC Stubble Initiative. Values with a different letter indicate yields are significantly different (P<0.05) within each site.

Table 2.  Mean yield of cereal varieties (wheat, barley, oats) following different stubble management treatments in trials conducted by CWFS in 2013 as part of the GRDC Stubble Initiative.

N tie up

PROBLEM: Soil bacteria and fungi that break down retained stubble compete with crop plants for N, and this has to be taken into account when budgeting N in stubble retained systems. A rough rule of thumb is that every tonne per hectare of cereal or canola stubble will tie-up (immobilise) 5 kg/ha of N. This means that in a paddock with a typical SE NSW cereal stubble load of 5 t/ha, 25 kg/ha of N will be tied up if the stubble is retained, and this must be compensated for with fertiliser N if yield is to be maintained.

SOLUTION: Add more N to paddocks with retained cereal stubbles (5 kg/ha N per 1 t/ha of stubble). Assume that there is no net mineralisation when budgeting N for these paddocks.

Temperature effects

PROBLEM: Retained stubble cools the temperature of the soil and the air immediately above the stubble where plants are growing. This reduces establishment and slows growth rates, particularly of canola, and often results in reduced biomass and yield. This effect is generally worse when stubble is spread evenly across the soil surface rather than retained standing.

During spring, stubble can also insulate the crop canopy from the warmth of the soil during frost events, and in 2013 greater frost damage and lower yields were observed where stubble was retained (Table 3)

Table 3. Grain yield and frost damage for different stubble treatments applied prior to sowing at the FarmLink and CSIRO stubble initiative site at Temora.

SOLUTION: Harvesting high, keeping stubble standing and then inter-row sowing on wider row spacing can reduce this effect somewhat (Table 4), but most experiments suggest that canola establishment, growth and  yield is improved by removing cereal stubble to levels below 2 t/ha.

As well as reducing crop yields, harvesting low and spreading stubble is very costly (Table 5). At a FarmLink and CSIRO stubble initiative site near Wagga, harvesting low and spreading stubble, cost 0.14 t/ha of grain yield even though the header was being driven by an experienced operator according to the yield monitor.  It is estimated that this yield loss plus increased fuel consumption and reduced efficiency, cost the grower approximately $64/ha. This does not take into account the additional risk of weather damage and downgrading due to slower harvest.

Figure 1.  Canola establishment and growth in an experiment conducted by FarmLink and CSIRO in a zero-till system near Wagga Wagga (Bonito sown 19 April with 50 kg/ha MAP). Stubble on the left was harvested high and then burnt whilst the stubble on the right harvested low and spread (John Deere 9770 STS with Power Cast tailboard).

Table 4.  Canola establishment and NDVI on 30 May 2014 following different stubble management treatments applied in a zero-till paddock near Wagga Wagga. Treatments were applied to a 7.6 t/ha stubble, the chopped treatment used a K-Line Trashcutter in December and the burnt treatment was applied in late March.

Table 5. Grain yield, harvest efficiency, speed and fuel consumption from the yield monitor of a JD9770 STS header in 7.6 t/ha stubble near Wagga Wagga in 2013. All differences are statistically significant (p<0.05).

Crop diversity – the forgotten main principal of conservation agriculture

In central and southern NSW, there has been strong adoption of two of the three main principals of conservation agriculture as defined by the FAO;  1) direct planting of crop seeds and 2) retention of permanent soil cover, particularly with crop residues. However, the all important third principal 3) crop diversity is neglected, and canola and wheat dominate cropping rotations. Many of the constraints currently experienced in retained stubble systems could be overcome with the inclusion of more crop types, particularly legumes. This includes pulses, pastures and forage/hay crops. Brown manures certainly remove some of the constraints of retained stubble, but can reduce profitability at a paddock scale, although this may be compensated for by whole-farm operational benefits.

Most legumes do not seem to suffer the same yield penalty experienced by canola and wheat in the presence of large cereal stubble loads. Numerous experiments in SA have shown that the yield of pulses is frequently higher with retained stubble. This is due to stubble providing a ‘trellis’ which increases height of lowest pods and improves harvestability, and also reduced splashing during rainfall events carrying soil-borne diseases up onto foliage. In southern Australia, high profile practitioners of zero-till tend to come from regions suited to growing high value pulse crops (e.g. the Wimmera of Victoria and upper Yorke Peninsula of SA) and these play an important role in their crop sequences.

Legume residues are also less antagonistic to canola than cereal residues, primarily because there tends to be less of them and they break down faster. For this reason, canola growing after legumes yields more than canola growing after wheat, even when extra N is added to the canola growing after wheat (Table 6). Due to this yield increase and reduced fertiliser N requirement, a pulse (e.g. lupins) followed by canola can be as profitable as wheat followed by canola, even if the pulse itself is not as profitable as wheat.

Growing canola after a legume also provides a ‘double break’, which is essential for reducing seed-banks of annual ryegrass and black oats, and inoculum of diseases such as crown rot.

Wheat can be grown very cheaply after a double break, has fewer stubble borne disease problems (yellow leaf spot, crown rot) and typically yields well. Recent GRDC research in the Victorian Mallee has found that crop sequences including a double break were more profitable than continuous cereals or sequences with a single break (see Ground Cover 111 July-August 2014 page 42). Growing barley instead of wheat-on-wheat further increases diversity, overcomes more stubble constraints (yellow leaf spot), and is often more profitable than a second wheat. It also helps manage frost risk, contributes to IWM by allowing later sowing with a double knock and improved competition with weeds.

Table 6. Comparisons of grain yields, applied N and gross margins of canola following different crops at Eurongilly..

*site was chosen for high levels of herbicide resistant annual ryegrass

Future research – Sequences for seeders

As outlined above, a crop rotation of legume – canola – wheat – barley has the potential to overcome many of the constraints imposed by retained stubble in canola-wheat rotations.  In theory, it should improve profitability and sustainability of retained stubble systems. However, this has not been experimentally tested. As part of the CSIRO and FarmLink GRDC stubble initiative project, a new farming systems experiment at the Temora Ag Innovation Centre will test this. It will compare the yields, profitability and sustainability of three different sequence systems (Conservative, Aggressive and Sustainable) with zero- and no-till seeding equipment (Excel single-disc with Aricks wheel vs Flexi-Coil tine with Stiletto knife-points and deep-banding/splitting boots). Sequence systems are as follows;

Conservative – TT canola, wheat, wheat (wheat sown at low density with trifluralin/diuron IBS and 20 kg/ha up-front N)

Aggressive – RR canola, wheat, wheat (wheat sown at high density with Sakura®/Boxer Gold® and 40 kg/ha up-front N)

Sustainable – vetch/pea hay, TT canola, wheat, barley (wheat sown at low density with Sakura® and 20 kg/ha up-front N)

As with all systems experiments, the first meaningful results will not be available until the second year of experimentation – 2015.

Conclusion

Retaining cereal stubble past sowing reduces yields in canola-wheat rotations. Yield reductions can be reduced by adding more N to stubble retained paddocks, keeping stubble tall and inter-row sowing, and by growing canola after legumes rather than cereals. Late burning of cereal stubbles in zero- and no-till systems consistently increases subsequent crop yield, particularly canola, but needs to be balanced against the operational and NRM benefits of stubble retention.

Contact details

James Hunt

CSIRO Agriculture GPO Box 1600 Canberra ACT 2601

02 6246 5066

james.hunt@csiro.au