Abstract

Canola yield generally declines with a decrease in growing-season rainfall on the eastern margin of the wheatbelt in Western Australia (WA). Canola cultivars vary in their timing of the phases of development and their canopy structure. These differences should permit the development of a canola ideotype with improved water use efficiency and high seed yield for low rainfall areas of the WA wheatbelt. We made field measurements of canopy structure, radiation interception, leaf gas exchange and instantaneous water use efficiency (WUE) in a wide range of current canola cultivars. There was a marked variation in canopy structure and physiological responses among both genotypes and environments. This research will provide clear targets for canola yield improvement programs in this low rainfall environment.

Introduction

Canola is a newly introduced oilseed crop in Western Australia (WA), of which the area under cultivation has grown very rapidly, from approximately 30,000 ha in 1993 to some 500,000 ha in 1998. Farmers in the eastern part of the WA wheat belt are eager to include canola in their farming system as an alternative break crop with the possibility of good economic returns. However, cultivation of canola, besides having advantages, presents a number of difficulties in the drier region of WA wheat belt. These are, (i) an uncertain distribution of rain throughout the cropping season subjects the crops to many periods of unpredictable drought, (ii) a short growing season with a terminal drought affects the development of seeds in growing pods, which leads to relatively low crop yield, and (iii) current canola cultivars with high yield are unsuited, having been developed for high rainfall areas that have a longer growing season. Hence, there is a need to develop high yielding cultivars with high water use efficiency for the short growing season of the drier parts of the WA wheat belt.

Limited water supply is a major problem for crop production in Mediterranean environments (Blum and Pnuel, 1990). The efficiency with which a crop uses water may reflect crop performance and it provides a means for comparing environments (Perry, 1987). Different genotypes perform differently in different environments and their growth habit is one of the principle descriptors of this variation (Kretchmer et al, 1979). Knowledge of how environment affects architectural traits and the strength and stability of their relationship with yield is a necessary pre-requisite of an effective breeding program for increasing yield through improved architecture. Donald (1968) suggested the importance of having a model (ideotype) as a goal for breeding programs. This concept has had an impact on breeding programs and ideal plant types of many crops have been described (Brothers and Kelly, 1993; Denis and Adams, 1978; Mock and Pearce, 1975).

The aim of this study was to find a plant type of canola with suitable phenological and physiological patterns to fit the growth cycle of the plant in a short growing season with limited water supply.

Materials and methods

Experiment 1 was conducted in a low rainfall environment [LRE] at the Dryland Research Institute, Merredin, WA, and a medium rainfall environment [MRE] at the Avondale Research Station, Beverley, WA in 1997. In the Merredin trial 26 Brassica genotypes [2 B. juncea and 24 B. napus] (Table 1) were included while in the Beverley trial a smaller group of 10 Brassica genotypes [1 B. juncea and 9 B. napus] was used (Table 1). The latter group was considered to be representative of the variation in architectural traits present in the former group. Sowing took place on 27 May and 10 June at Merredin and on 26 May and 9 June at Beverley, at a rate of 6 kg/ha. Eight-row plots, 15 m long, were sown with a distance of 20 cm between the rows. Superphosphate (D. super), at the rate of 74 kg/ha was drilled with the seeds and 40 kg/ha urea was topdressed at the same time. Six weeks after each sowing a further 40 kg/ha urea was topdressed. Sprayseed (2L/ha), IBS Sprayseed (2L/ha) and Talstar (100 ml/ha) were applied as pre-emergence herbicides. Diseases and weeds were controlled effectively using post-emergent sprays. Crops were grown to maturity and harvested mechanically.

Table 1: Brassica genotypes used in Experiments 1 and 2.

Experiment 1			Experiment 2
Merredin 1997		Beverley 1997	Beverley 1998
B.juncea	B.napus	B.juncea	B.juncea
82NO22-98	DB 53	82NO22-98	82NO22-98
JK 4	Hyola 42		JK 4
	KEM 169
B.napus	KEM 171	B.napus	B.napus
93-002C5	Monty	93-153C	93-002C5
93-009C5	Narendra	DB 52	93-184C5
93-015C5	Oscar	DB 53	93-53C5
93-107C5	RI27*504	KEM 171	Hyola 42
93-153C	RK 9	Monty	KEM 169
93-184C5	RL 18	Oscar	KEM 171
93-216C5	RL 26	RK 9	Monty
93-383C5	RL 28	RL 18	Narendra
93-53C5	RL 29	TRAP 16	RL 28
DB 52	TRAP 16		TRAP 16

Experiment 2 was conducted in a MRE (Avondale Research Station, Beverley, WA) in 1998. Twelve Brassica genotypes [2 B. juncea and 10 B. napus] (Table 1), selected from experiment 1, were sown on 9 June at 3, 6, 9 and 12 kg/ha. Plant husbandry was the same as in Experiment 1.

Measurements

Plant density per unit area was recorded 45 days after sowing (DAS). Flowering time was defined as the day on which at least 50% of plants in a plot had at least 5 flowers on their mainstem. Photosynthetically active radiation (PAR) intercepted by the canopy before and after anthesis was determined by measuring incident net PAR above and below the canopy with a 1 m line quantum sensor (Li-Cor Inc., Nebraska. USA) near solar noon, before destructive sampling of plants for the determination of leaf area index (L) using a PATON Electronic Planimeter. The extinction coefficient (k) was determined from the fraction of intercepted PAR and L (Monteith, 1965). Measurements of leaf gas exchange were undertaken before and after anthesis using a portable CIRAS photosynthetic system. Three fully expanded, attached, young leaves from the upper strata of the canopy of four middle rows of each plot were used for measurements. Water Use Efficiency (WUE), at leaf level, was calculated as the rate of net photosynthesis per unit of water transpired.

Plant architectural traits (number of leaves on the mainstem (MS leaves) and the number of first order branches (FOB) and plant height) were recorded on 6 plants from the middle row of each plot at maturity. Data from mechanical harvest of individual plots were used to determine the seed yield per unit area. Harvest Index (HI) was calculated as the ratio of seed yield to total dry matter/biological yield (Donald, 1962).

Results

Canopy characteristics

There was significant variation in the number of FOB and MS leaves among the Brassica genotypes tested [P<0.001] (Table 2a). The plants produced more FOB and MS leaves in the LRE than in the MRE, whereas the height of plants was approximately the same in the two environments.

Leaf Area Index (L), extinction coefficient (k) and total dry matter

Variations among genotypes of L measured before and after anthesis were significant [P<0.05 - 0.001 in different experiments] (Table 2a). L was lower in the LRE than in the MRE. The extinction coefficient (k) calculated before and after anthesis was significantly different [P<0.005-0.001] (Table 2b) for different genotypes. Total dry matter production varied significantly among genotypes [P<0.001] (Table 2c).

Flowering time, seed yield and harvest index (HI)

Significant variation in flowering time (P<0.007-0.001) was observed among genotypes in both environments (Table 2c). The time required to achieve 50% flowering was increased about 15 days greater in the LRE. There was a significant variation in seed yield among Brassica genotypes in both environments [P<0.001] (Table 2d). The maximum yield obtained in the LRE was about 50% of the maximum yield obtained in the MRE. Variation among harvest indices of genotypes was significant [P<0.001] (Table 2d).

Leaf gas exchange and water use

In the MRE variables that indicated leaf gas exchange and water use properties showed no significant variation among genotypes (Table 2e) either before and after anthesis. However, in the LRE there was significant variation (P<0.001) among genotypes in transpiration, stomatal conductance, net photosynthesis and water use efficiency, measured before anthesis.

Discussion

These experiments demonstrated wide variation among Brassica cultivars in canopy architecture, radiation interception, leaf area index, dry matter production and the duration of flowering, the key factors that influence the yield and harvest index of crops.

Table 2: Genotypic variation (averaged over sowing time and seeding rate, as appropriate) in some architectural, physiological and agronomic traits of Brassica genotypes in a medium and a low rainfall environment.

Characteristics		Experiment 1			Experiment 2
	Beverley 1997		Merredin 1997		Beverley 1998
	Minimum	Maximum	Minimum	Maximum	Minimum	Maximum
(a) Architectural traits
Mainstem height (cm)	64	118	64	122	109	167
No of leaves on mainstem (MS)	4.06	5.83	6.89	11.94	5.80	10.11
No of branches (FOB)	3.83	5.17	4.78	7.11	3.18	5.33
Leaf area index (L) (cm2) BA*	2.85	5.88	2.12	3.68	2.83	5.26
Leaf area index (L) (cm2) AA*	3.21	6.25	2.05	3.70	3.16	5.54
(b) Radiation interception
Extinction coefficient (k) BA*	0.25	0.59	0.19	0.45	0.68	1.20
Extinction coefficient (k) AA*	0.26	0.68	0.37	0.65	0.74	1.57
(c) Dry matter production and flowering
Total dry weight (gm/plant)	3.3	6.0	2.7	4.5	5.5	14.2
Flowering time (DAS)	59	85	79	96	72	82
(d) Yield and harvest index
Seed yield (t/ha)	1.19	1.64	0.27	0.87	1.27	2.64
Harvest index	0.56	1.06	0.32	0.67	0.35	0.60
(e) Gas exchange and water use
Transpiration (mmol/m2/s) BA*	3.11	4.30	4.71	8.91	3.34	4.18
Transpiration (mmol/m2/s) AA*	2.63	4.32	3.19	4.47	2.75	4.68
Stomatal conductance (mmol/m2/s) BA*	149	273	214	599	160	240
Stomatal conductance (mmol/m2/s) AA*	151	261	151	264	195	284
Net photosynthesis (μmol/m2/s) BA*	11.7	17.0	12.8	23.8	10.7	17.9
Net photosynthesis (μmol/m2/s) AA*	8.2	12.9	8.9	18.8	11.1	14.6
Water use efficiency (mmol/mol) BA*	3.04	4.45	1.86	3.39	3.22	4.35
Water use efficiency (mmol/mol) AA*	2.09	3.51	3.04	4.71	2.90	4.13

BA* = Before anthesis
AA* = After anthesis

In the MRE the number of MS leaves and FOB was lower than in the LRE, perhaps unexpectedly, whereas plant height did not differ between the two environments. This is

presumably due to earlier flowering in the MRE environment. Dracup et al (1998) reported a restricted growth habit of Lupinus angustifolius when grown under the long days of spring-summer in northern Europe, and earlier floral initiation, relative to the values observed in Western Australia. This was explained as a possible response to differences in photothermal conditions. The LRE was substantially hotter than the MRE used in this study, but there was almost no difference in photoperiod between the environments. Another possible explanation for restricted growth could be the quick shift from vegetative to reproductive growth in response to earlier flowering. However, the strong relationship found between these variables when the two environments were compared was not present between genotypes within a single environment. A weak negative correlation was observed between flowering (days from sowing) and number of MS leaves (r = -0.24) and number of FOB (r = -0.24). The association between these parameters needs to be explored further.

In the LRE the ranges of L measured before and after anthesis and the values of final dry matter produced by the individual plants were significantly lower than in the MRE. The leaves of canola genotypes started to senesce a few weeks after anthesis and did not contribute much in the final dry matter production values, which mainly represented stem, branches and pods residues. However, both the interception of PAR and dry matter production contributes to yield. The range of k among genotypes before anthesis in experiment 1 was low in the LRE but after anthesis it was similar to the range observed in the MRE. These differences in k are expected to be the result of variation in leaf orientation within crop canopies. The high values of k observed in experiment 2 were due to their compact canopies.

In this study instantaneous measurement of gas exchange did not significantly vary among genotypes or due to husbandry factors in the MRE. However, significant variations observed in the leaf gas exchange and water use of genotypes in the LRE indicate the importance of these features for crop growth in that environment.

The high seed yield and high HI in the MRE relative to the LRE was presumably the result of greater vegetative growth. The higher values of L and k contributed to some extent to the higher final dry matter production and high HI, which appeared to be the most important determinant of seed yield in this environment.

Conclusion

There is wide variation among canola genotypes in many characters related to adaptation to a dry environment, and the genotypes currently grown in the drier parts of the WA wheatbelt have not been specifically selected for adaptation to this region. These two observations indicate that it will be possible to increase substantially the seed yield of canola by identifying the plant type (ideotype) appropriate to this environment, perhaps in combination with some changes in husbandry practices.

Acknowledgements

We are grateful to GRDC for financial support of this research (Project UWA 225 WR) and to Agriculture WA for providing trial sites and management. Thanks are also due to Dr. Phil Salisbury and Dr. Wayne Burton of Agriculture Victoria for provision of some of the seed stocks used.

References

1. Blum, A. and Pnuel, Y., 1990. Physiological attributes associated with drought resistance of wheat cultivars in a Mediterranean environment. Aus. J. Agric. Res. 41: 799-810.

2. Brothers, M.E. and Kelly, J.D., 1993. Interrelationship of plant architecture and yield components in the pinto bean ideotype. Crop Sci. 33: 1234-1238.

3. Denis, J.C. and Adams, M.W., 1978. A factor analysis of plant variables related to yield in dry beans. I. Morphological traits. Crop Sci. 18: 74-78.

4. Donald, C.M., 1962. The design of a wheat ideotype. In: Proceedings of the 3^rd International Wheat Genetics Symposium. pp. 377-387.

5. Donald, C.M., 1968. The breeding of crop ideotypes. Euphytica 17: 385-403.

6. Dracup, M., Thomson, B., Reader, M., Kirby, E.J.M., Shield, I. and Leach, J., 1998. Daylength responses, flowering time, and seed filling in lupins. Aust. J. Agric. Res. 49: 1047-1055.

7. Kretchmer, P.J., Laing, D.R. and Wallace, D.H., 1979. Inheritance and morphological traits of a phytochrome-controlled single gene in bean. Crop Sci. 19: 605-607.

8. Mock, J.J. and Pearce, R.B., 1975. An ideotype of maize. Euphytica 24: 613-623.

9. Montieth, J.L., 1965. Light distribution and photosynthesis in field crops. Ann. Bot., 29: 17-37.

10. Perry, M.W., 1987. Water use efficiency of non-irrigated field crops. Proceedings of the 4^th Australian Agronomy Conference, La Trobe University, Melbourne. pp. 83.