Background

In SA the time at which wheat flowers is very important in determining yield. With farm sizes increasing and sowing opportunities decreasing, getting wheat crops established so that they flower during the optimal period for yield is difficult. Whilst no-till and dry-sowing have been used successfully in SA to get more area of crop flowering on time, an opportunity exists to take advantage of rain in March and April to start sowing crops earlier than currently practiced. This is a tactic which complements dry sowing. Earlier sowing is now possible with modern no-till techniques, summer fallow management and cheaper insecticides and fungicides to protect against diseases associated with early sowing. 

In the last few decades wheat breeding has focused on mid to fast developing cultivars which are only suited to sowing in late April-May. Sowing earlier than is currently practiced requires cultivars which are not widely grown in SA, and which are much slower to mature, either through having a strong vernalisation/cold requirement (winter wheats) or strong photoperiod/day length requirement (slow developing spring wheats).

Research funded by GRDC in New South Wales (NSW) demonstrated that slow developing cultivars sown early yield more than mid-fast varieties sown later when they flower at the same time. This is because early sowing increases rooting depth and water use, reduces evaporation and increases transpiration efficiency. Early sowing slow developing cultivars is a way of increasing yield potential with very little initial investment. A lack of slow developing cultivars adapted to SA is currently limiting grower’s ability to take advantage of the water use efficiency (WUE) benefits of early sowing. However, some advances can be made by using mid (e.g. CutlassA) and mid-fast (e.g. TrojanA) developing cultivars to open up sowing windows.

Methodology

GRDC early sowing trials in SA in 2015 are at four locations (Cummins, Minnipa, Port Germein and Conmurra) and each has three times of sowing (aimed at mid-April, late-April, mid-May) and 10 wheat lines (six commercial and four near-isogenic lines (NILs) in a Sunstate background). There was seed bed moisture at all times of sowing at all sites and seed germinated shortly after sowing. The commercial lines are described in Table 1. At Minnipa, a new winter wheat that is being developed by Australian Grain Technologies (AGT) (RAC2341) was also tested. This line is derived from Mace, and whilst being winter in habit (needs to experience a winter before it will run up to head), it is very fast developing once it has vernalised. This means it is better adapted to the SA environment than existing winter wheats such as Wedgetail which tend to flower too late. The Hart Field Site Group also established a similar early sowing trial funded by members, and there are also trials funded by SA Grain Growers Industry Trust (SAGIT) evaluating different wheat lines for early sowing in the Mid North and upper Yorke Peninsula. 

Table 1: Commercial wheat varieties used in the SA trials at Cummins, Minnipa and Port Germein.

Results

Results from all 2015 experiments are presented in Table 2. At all sites Trojan sown in either mid or late April was the highest or equal highest yielding treatment. Trojan sown in its optimal window (late April) out-yielded Mace sown in its optimal window (mid-May) at three out of four sites where the two were grown (mean yield advantage 0.6 t/ha). Slow developing cultivars bred in other states (e.g. EGA Wedgetail, EGA Eaglehawk and Rosella) showed poor adaptation to all sites. This is in part because even when sown early they flower too late in many SA environments (e.g. see Minnipa flowering dates in Table 3).

The winter wheat RAC2341 showed great promise at Minnipa, where its yield was not significantly different to that of Trojan and Cutlass at all times of sowing, despite flowering much later (Table 2 and Table 3) and therefore taking less frost risk. If released, this cultivar would give SA growers an excellent option for taking advantage of any establishment opportunities arising from March through to late April when it becomes safe to sow currently adapted spring wheats. It would also provide the first ever adapted dual purpose cultivar for SA, which would increase productivity on mixed farms.

At Conmurra, highest yields came from Trojan and Cobra sown 15 May and Manning sown in late April. This is the second year where there has been no demonstrable advantage of sowing earlier than mid-May at this site. In this environment heavy disease pressure from BYDV and Zymoseptoria tritici make sowing earlier than mid-May undesirable, unless crops are to be grazed during their vegetative phase. Growers in this environment wishing to sow early and graze should use Manning for its superior resistance to BYDV, Z. tritici and leaf rust (fungicides will still be required for control of Z. tritici and stripe rust).

Grain quality data were only available for Minnipa at time of writing. At this site protein increased due to yield concentration effects with later sowing, but screenings increased and test weight decreased such that highest paying binned grades (APW and AUH2) were predominantly achieved at the first two times of sowing.

Table 2: Grain yield (t/ha) of five early sowing trial sites in SA in 2015. There was seed bed moisture at all times of sowing and seed germinated shortly after sowing. P-values and LSDs are for time of sowing (TOS) x cultivar interactions.

Table 3: Anthesis date for cultivars at different times of sowing at Minnipa in 2015. Values shaded in grey flowered within 7 days of 1 September which is assumed to be the optimal flowering date for Minnipa

Table 4: Grain quality from the experiment at Minnipa in 2015. Quality data from other sites were not available at time of writing.

Summary of trials 2013-2015 – putting it into practice

Based on three years of trials, growers in SA could improve whole-farm yields by including TrojanA in their cropping program to complement MaceA (Figure 1). TrojanA has an unusual photoperiod sensitivity allele inherited from a European parent which is rare in Australian cultivars. This allele seems to delay flowering from an April sowing relative to MaceA quite successfully (Table 3).

Figure 1: a) Mean yield (% site mean) of Mace and Trojan at nine SA sites where they were both grown in  during 2014 and 2015 (Minnipa 14 & 15, Cummins 14 & 15, Port Germein 14 & 15, Hart 14 & 15, Tarlee 14). Linear regressions for both Mace (- - -, R=0.05) and Trojan (∙∙∙∙, R=0.47) are significant (P<0.001) and are significantly different from each other in gradient (P=0.045) and intercept (P=0.025).

Figure 1:  b) Mean yield (t/ha compared to Mace at its optimal time of sowing at each site) for Trojan and Mace from the above sites averaged across different arbitrary sowing periods. Error bars are standard error of means, and points with overlapping bars are unlikely to be significantly different from each other.

Despite performing strongly from a mid-April sowing in these trials, it is not recommended that Trojan be planted this early in the majority of SA locations as it incurs excessive frost risk. As a rough rule of thumb, it is best suited to being planted approximately 10 days earlier than Mace. As an example of how it may fit in a program, if 10 May is the optimal sowing time for Mace in a given environment, then the optimal sowing time for Trojan is 1 May. If a grower has a 20 day wheat sowing program and wants to grow half Trojan and half Mace, to maximize whole farm yield they should start with Trojan on 25 April, switch to Mace on 5 May and aim to finish on 15 May.

Sowing mid-April in low-frost environments such as Port Germein carries little risk, and as the results from both 2014 and 2015 show significant yield gains can be achieved by sowing in mid-April. At this site in 2014 Trojan had a significant (0.9t/ha) yield advantage over Mace when sown at this time, but in 2015 yields of the two were not significantly different. The difference in 2014 may have been due to either very warm May conditions that would have caused Mace to ‘bolt’ faster than usual, or unusual frosts that the site experienced that year in early August.

For growers in frosty environments who wish to sow earlier than is safe with Trojan or Mace, Cutlass can be sown approximately five days earlier again than Trojan. Sowing earlier than this in frost prone environments requires a winter wheat, and until the release of winter cultivars adapted to SA (e.g. RAC2341) EGA Wedgetail is probably the best option. However, because of its poor adaption to SA even if sown in early-mid April it is unlikely to yield as well as Mace sown in its optimal window. In this set of trials there was an average yield penalty of 0.4t/ha between EGA Wedgetail sown mid-April and Mace sown in mid-May. Grazing early sown EGA Wedgetail would offset some of the reduction in income compared to mid-May sown Mace. 

Three years of trials across multiple environments in SA have shown that yields decline at a rate of 28kg/ha/day once sowing extends past the end of the first week in May. In order to maximize average yields, growers should aim to finish seeding wheat by mid-May. Growers with longer wheat sowing programs will require multiple cultivars of different development types in order to allow them to start early enough. An example of how this might be achieved is presented in Table 5. In years with a late break where seed bed moisture isn’t available to establish slow developing cultivars in their optimal window, yields will be maximized by dry sowing only a fast developing cultivar (e.g. Mace or equivalent) starting from the opening of its optimal window.

Table 5: An example of how slow developing cultivars and early sowing can be used to maximise farm wheat yield depending on the duration of wheat sowing program. In years where there is no seed bed moisture available and sowing starts ‘dry’, yield will be maximised by planting Mace (or equivalent fast developing cultivar) from 1 May onward.

Remember early sown crops require different management in order to get the most out of them:

• Don’t dry-sow slow developing varieties (EGA Wedgetail, Cutlass), they will flower too late if not established early. There needs to be seed-bed moisture and ideally some stored soil water to get them through to winter.
• If growing winter wheat (EGA Wedgetail) and not grazing defer nitrogen (N) inputs until after Z30.
• Pick clean paddocks – winter wheat is not competitive with ryegrass and common root diseases are exacerbated by early sowing.
• Protect against diseases associated with early sowing – barley yellow dwarf virus (imidicloprid on seed backed up with in-crop insecticides at the start of tillering if aphid pressure high), Zymoseptoria tritici in some areas (flutriafol on fertiliser and timely foliar epoxiconazole applications at Z30 and Z39). Many slow developing cultivars also have poor resistance to stripe rust (flutriafol on fertiliser and timely foliar fungicide application at Z39).

Acknowledgement

Funding for this work was provided through the GRDC Projects CSP00178 and CSP00160 and their support gratefully acknowledged.

Contact details 

James Hunt

GPO Box 1600

Canberra ACT 2601

0428 636 391

james.hunt@csiro.au

@agronomeiste