Canola yield and heat stress in the northern agricultural region 2017 trial report

Background

Canola is most sensitive to heat stress from a week before flowers open until a week after. The critical temperatures range from 27-30°C (Kirkegaard et al 2017, Morrison and Stewart 2002). Many canola crops in the northern agricultural region flower when temperatures are high enough to limit yield and there is concern that this will become more frequent and severe if spring temperatures increase, as predicted by climate change models.

Two ways to mitigate this risk include: 1) flower early before temperatures are too high by sowing early and using a short season variety; and 2) reduce the heat stress experienced by the crop by managing soil moisture so plants can transpire during flowering and podding to keep cool.

This trial was designed to look at both of these mechanisms by sowing at different dates and implementing +/- irrigation treatments to manipulate temperatures at flowering and plant water stress.

Methods

Canola (cv. Pioneer 43Y23 RR) was sown in 1.0m rows at Geraldton on four dates (18 April, 28 April, 8 May and 18 May). At each TOS there were +/- irrigation treatments, hence there were eight treatments; each replicated five times. 15 seed were sown per row and the plants were selectively thinned to ensure plants in each treatment were at similar plant development stages. Plant growth stage was monitored twice weekly with development stage on the main stem recorded using the APSIM canola phenology rating code. Volumetric soil moisture at 0-10cm was monitored twice weekly using a Campbell scientific moisture probe. From early August to late September, when plants were maturing, moisture was measured to 1.0m using a Sentek 2000 Diviner. Temperature and humidity were recorded hourly.

The entire trial was irrigated several times in April and May to enable seed to germinate and ensure plants were not moisture stressed before flowering. At the commencement of flowering of the earliest sowing time a rainout shelter was placed above unirrigated plots and left in place through to harvest. Irrigated plots had 10mm irrigation treatments applied twice weekly from the end of August through to the third week in September. Stem sap flow was measured on the largest stems of two plants each in the irrigated and unirrigated plots of TOS4 from 12 September to 17 October using SF-3 sap flow sensors (Edaphic Scientific, Australia) logged with a Campbell Scientific CR10X datalogger at 30 minute intervals.

Plants were hand harvested as they reached maturity, from the 22 September to 2 October. Measurements included biomass, number of pods and aborted pods on the main stem and whole plant, seed yield and seed size.

Results

Seasonal conditions

Rainfall was 315mm for the year to November (Table 1). This compares to a long term average for Geraldton of 445mm. Monthly rainfalls for March, April and May were lower than long term average and irrigation was used to sow the trial. May and June were also warmer than average. Recording of temperature from the Bureau of Meteorology Geraldton town site ceased in 1953 and the Geraldton Airport site temperature is only available from 2011-2017. As such Mullewa temperature data is included in Table 1. This indicates 2017 April, May and June maximum temperatures were 2-3°C greater, long term.

Plant development

Flowering on the main stem commenced as early as 6 June and as late as 31 July, depending on sowing date, and finished as early as 26 June and as late as 21 August (Figure 1). 18 April sown plants had the shortest main stem flowering period of 20 days, the rapid plant development corresponding to the higher temperatures. Main stem flowers were recorded on plants sown on 28 April for 31 days, 5 May 32-38 days (-/+ irrigation) and 18 May 21-25 days (-/+ irrigation) (Figure 1). The very early flowering of 18 April sown plants meant they experienced higher temperatures during main stem flowering than other sowing times and in fact the latest sowing time had least exposure to high temperature during main stem flowering (Table 2). However, while temperatures were higher during main stem flowering in TOS1 than later times of sowing, a high proportion of yield was carried on branches which flowered later under cooler conditions. Similarly, branches on TOS4 plants flowered under warmer conditions than the main stem (Table 2).

Thermal time to the opening of first flowers ranged from 1008°Cd for TOS1 to 1280°Cd for TOS4. Irrigation increased the duration of the flowering period; irrigated plants flowered for 535°Cd while unirrigated plants for 434°Cd. Delayed sowing increased the thermal time to maturity; TOS1 2079°Cd and TOS4 2284°Cd.

Figure 1 Main stem flowering period of canola sown at four times of sowing (+/- irrigation) at Geraldton in 2017

Soil moisture

There were clear differences in soil moisture between irrigated and unirrigated treatments during August and September (Figure 2). Irrigation treatments increased water down to a depth of 60-70cm. Between 13-22 September the diurnal pattern of sap flow and sap velocity was almost identical in irrigated and unirrigated plants, suggesting soil water had not been sufficiently depleted to retard transpiration (Figure 3a). Between 7-16 October sap flow was significantly reduced in unirrigated plants (Figure 3b).

Figure 2 Soil moisture (% v/v) to 1.0 metre depth as measured in August and September 

Figure 3a Stem sap flow (cm/hr) over two 10 days periods in September and October of irrigated and unirrigated canola plants at Geraldton in 2017

Figure 3b Stem sap flow (cm/hr) over two 10 days periods in September and October of irrigated and unirrigated canola plants at Geraldton in 2017

Plant growth

Earlier sown plants had greater stem diameter (P <0.001) and total plant weight (P <0.001, Table 3). Irrigation did not affect stem diameter but appeared to increase total biomass, especially in early sowing times, even though this was not statistically significant.

Yield and yield components

Irrigated early sown plants produced highest yield. Yield declined with each delay in sowing date and TOS4 produced only 18% as much yield as TOS1 (mean of irrigation treatments), consistent with previous sowing time trials in the region. The proportion of total yield carried on the main stem was very low, especially in the earliest sowing times where it was 1.5% in irrigated and 2.3% in unirrigated plants (Table 4). Hence whilst temperatures during main stem flowering of mid-April sown plants were higher and pod abortion rates were higher (Table 4) compensating seed production from later flowering branches resulted in higher yield potential and harvest index (P <0.001) (Table 3).

There was a clear trend for irrigation to increase yield, particularly on earlier sown plants, but the variability in single plant yield meant this was not statistically significant (Figure 4, Table 3). Because irrigation main effect was not significant the TOS irrigation interaction was also considered not significant despite a low P value.

Figure 4 Plant yield (g/plant) as affected by sowing date and irrigation at Geraldton in 2017

Plant podding

Averaged across irrigation treatments TOS1 plants initiated almost twice as many pods (fertile plus aborted) as TOS4 plants (Table 3). Irrigation increased pod initiation by 6% compared to unirrigated but this was not significant. The proportion of total initiated pods that aborted (aborted %) was not affected by sowing date but was reduced by irrigation (P <0.001). Averaged across all sowing times irrigation reduced pod abortion by around 15%.

There were more fertile pods per plant at earlier sowing dates (P <0.001) and when irrigated (P <0.05). Averaged across irrigation treatments TOS1 plants produced twice as many fertile pods as TOS4 plants. Irrigation increased fertile pods by 40% compared to control plants (Table 3).

There was less main stem pod abortion in later sown plants (Table 4). Although we expected early sown plants to have less main stem pod abortion this observation is consistent with higher temperatures during the main stem flowering period of early sown plants. It is interesting that there was little difference between sowing times in total initiated main stem pods (90-103 averaged across irrigation treatments) (Table 4).

Seed size

Earlier sowing times produced larger seed (P <0.001). Irrigation had no effect on seed size, perhaps due to the increased yield of irrigated plots from earlier sowing times.

Seeds per pod

Earlier sowing times produced more seeds per pod (P <0.001) (Table 3). Irrigation had no effect on the number of seeds per pod.

Conclusions

Yield was highest from early sowing despite heat stress reducing main stem pod set of mid-April sown plants due to pod set occurring on branches at a later date. Hence, it was best to sow early even though autumn conditions were unusually warm.

It is challenging to separate the effects of heat stress (the effect of short term temperature spikes) and terminal drought on yield as in practice the two usually occur together. However, we showed that maintaining high soil moisture levels with irrigation reduced pod abortion by an average of 15%. This suggests that any practice that can conserve soil water until the reproductive stage, such as conservation tillage, stubble retention and altering row spacing and plant density combinations may reduce pod abortion.

The results from this trial reinforce the idea that in the northern agricultural region heat stress across the total flowering period is most likely to be avoided by sowing in April and by using short season varieties. Conserving soil moisture using agronomy to manage the crop canopy such as wide row spacing and low plant density is also likely to improve the plants ability to tolerate high temperatures.

Acknowledgments

Thanks to the Geraldton DPIRD Research Facility for the trial site, Belinda White for technical assistance and GRDC and DPIRD for project funding. Dr. Michael Forster of Edaphic Scientific for supplying sap flow sensors and help with calculations.

References

Kirkegaard, J. A. Lilley J. M. Bril R.D. Sprague S.J. Fettell N.A, and Pengilley G.C. (2017). Re-evaluating sowing time of spring canola (Brassica napus L.) in south-eastern Australia—how early is too early? Crop & Pasture Science, 2016, 67, 381–396 http://dx.doi.org/10.1071/CP15282

Morrison, M. J. and Stewart, W. (2002). Heat Stress during Flowering in Summer Brassica. Crop Sci. 42:797–803 (2002).

Guo, Y.M., Samans, B., Chen, S. et al. (2017). Drought-Tolerant Brassica rapa Shows Rapid Expression of Gene Networks for General Stress Responses and Programmed Cell Death Under Simulated Drought Stress Plant Mol Biol Rep 35: 416. https://doi.org/10.1007/s11105-017-1032-4

GRDC project number: DAW00227