Abstract

Key Words

Introduction

Compared to the medium and high rainfall zones of south eastern Australia, increases in wheat yields and water use efficiency in the low rainfall regions of South Australia have been modest (Hamblin and Kyneur 1993, Black 2004, Blumethal and Walcott 2006). For example, Hamblin and Kyneur’s analysis showed that between 1950 and 1991, average yields in South Australia increased by between 4 and11 kg/ha/yr, with yield improvements significantly less in the low rainfall areas of the State.

The long-term increases in yield have been associated with improvements in agronomy and in the yield potential of new varieties. There is little information on the impact of genetic improvement on the yield of wheat in low rainfall environments. In a study in Western Australia using 28 historic varieties, Perry and D’Antuono (1989) showed an increase in grain yield with the year of introduction of 5.8 kg/ha/year. In comparison in a medium rainfall area in South Australia, Vandeleur and Gill (2004), estimated an average yield increase of 15.5 kg/ha/yr when rust was not a limitation to yield.

The primary aim of the cereal breeder has been and remains greater yield and/or quality (Perry and D’Antuono 1989) Improving the adaptation of wheat to local conditions was recognised early; for example, Prof. Custance of Roseworthy Agricultural College in the 1880’s was ‘…convinced that early maturity and drought tolerance were essential agronomic qualities for the local environment’ (Simmonds 1989). Subsequently, considerable effort and money has been expended in the quest for better adapted varieties for South Australian climatic and edaphic conditions, as well combating serious foliar and root pathogens. The aim of this work was to assess the level of genetic gain for wheat in the mallee environment.

Methods

In 2001, a series of trials was conducted at eight Mallee sites and one Eyre Peninsular site. Twenty eight historically-important varieties were evaluated against two recently released South Australian lines. The lines were selected for inclusion in these trials because of their important historic contribution to the wheat industry in the SA Mallee (Pearson 1957). Trials were planted as unreplicated small plot trials (10 m x 1.6 m plots) with Yitpi as the check plot throughout. Murray Mallee yields were analysed by Colleen Hunt (BiometricsSA) using spatial Multi Environment (MET) analysis. The experiments were sown in mid June at each of the sites using standard seeding and fertiliser rates. Rainfalls at the sites ranged from 248 mm to 416 mm with and average of 309 mm. There was no recorded yield limiting factors such as frosts, foliar or root diseases for most of the sites. Growing season (April-October) rainfall (GSR) was recorded and used to estimate the water use efficiency (WUE) of the crops. It was assumed that soil evaporation was 90 mm and the variety WUE was estimated as: (grain yield)/ (GSR-90).

Results

Average yields ranged from 1.349 t/ha (Steinwedel) to 2.492 t/ha (RAC 892). The modern varieties were consistently higher yielding than the Heritage lines and the oldest varieties generally ranked consistently low across all sites. RAC 892 and Yitpi ranked first or second across all sites. The next best varieties were Halberd (1969) and Insignia (1946). Varieties performing consistently poorly were Steinwedel (1884), Comeback (1901) and Pusa (1916). The variety rankings for Murray Mallee and Minnipa sites were very similar. The average rate of yield increase among these varieties was 7.2 kg/ha/yr, which was the same as that recorded at Minnipa. (Fig. 1, 2). The heritabilities were high, often exceeding 70%. A frost occurred at Geranium in October, which explains why Geranium had the lowest yield and WUE. Yield variability at this site, assessed by the coefficient of variation (cv%), was also high and the heritability was low (Table 2).

The comparative WUE figures for Yitpi and Federation are presented in Table 2. Yitpi consistently had a higher WUE figures than Federation. At no site did Federation achieve a WUE greater than 18 kg/ha/mm. Minnipa’s WUE results were similar to those from the Murray Mallee for both Yitpi and Federation.

Discussion

One of the problems of comparing the yields of an historical set of varieties is that old and new varieties are grown under modern farming methods, whereas management practices have changed significantly in 100 years. For example, in 1907 six wheat varieties were compared at Loxton (Angus 1908). The experiment was sown on June 12, 1907 at a sowing rate of approximately 50 kg/ha with 56 kg/ha of phosphate. The grain yield of Federation was 1.01 t/ha (15 bushels per acre). The annual rainfall in 1908 was 286 mm, of which 219 mm fell between March and October. The estimated WUE was about 8 kg/ha/mm, which is less than that estimated for the current trials. However, heavy frosts during September reduced the yield by about 404 kg/ha which would have provided a WUE of about 11 kg/ha/mm. In comparison, Federation wheat sown at the Mallee sites in early to mid June in 2001 yielded an average of 1709 kg/ha with an average WUE of 12 kg/ha/mm (Table 2), which is the same as that achieved under the farming systems of 1907. This is somewhat surprising as it suggest that changes in farming practices have had little or no effect on WUE.

As expected, modern Australian wheat varieties yielded more than a range of Heritage Australian wheats and there appears to have been a gradual improvement in yield from wheats released in the 1800s to the latest line RAC 892. The rates of yield improvement in the Mallee and the other low rainfall site on the Upper Eyre Peninsula were the same, but were greater than the estimated yield gains from Western Australia, but both were much less than genetic gains estimated in a higher rainfall region (Vandeleur and Gill 2004). Roseworthy is a high rainfall site (329 mm GSR per annum and averaging 331 mm GSR over the three years of the trials). The Mallee is a low rainfall zone where average annual and GSR rainfall for Loxton SA is 273 and 190 mm respectively. Using Vandeleur and Gill (2004) data it is possible to compare yields from the two zones. Calculating a per cent yield gain per year using the rate of increase (kg/ha/yr) from 1880 in the case of Mallee and Minnipa, and 1860 from Roseworthy, the Mallee and Minnipa trials show gains of 0.50 % and 0.43 % per year respectively, whereas the Roseworthy trials indicate an average gain (years 1999 and 2001) of 1.21 % per year. The comparison is stark: not only are absolute yield gains lower in the low rainfall Mallee environments but relative yield gains are too. Whilst there has been significant progress in wheat breeding over time, the low rainfall zones of South Australia are lagging behind higher rainfall areas in both real and relative yield gains.

The conclusion from these trials, based on yield alone, shows there has been significant progress in wheat breeding over the last 100 years. South Australian growers have benefited from wheats bred for better adaptation to South Australian soils, with improved disease and pest resistance, tolerance to soil toxicities and deficiencies and finally the unchanging seasonal factors such as rainfall and poor seasons. The increase in wheat yields with time is further demonstrated by the increased water use efficiency of modern varieties over the Heritage lines. In the Murray Mallee Yitpi was 42% more efficient in its use of water for grain

Table 1 Yield, rank, and % Yitpi yield of varieties across eight South Australian Mallee sites and a site at Minnipa, South Australia.

Variety	Year released	Origin	Grain yield
			SA Mallee sites			Minnipa
			t/ha	rank	% Yitpi	t/ha	rank	% Yitpi
Bencubbin	1929	WA	1.961	6	82	2.192	5	85
Caliph	1912	RAC	1.747	17	73	1.952	15	76
Claymore	1956	RAC	1.998	5	84	2.143	6	83
Comeback	1901?	Farrer	1.349	30	57	1.630	27	63
Dirk_48	1949	SA	1.884	10	79	1.906	22	74
Early_Gluyas	1894	SA	1.618	23	68	1.884	24	73
Federation	1901	Farrer	1.709	19	72	2.039	10	79
Ford	1916	RAC	1.730	18	73	1.900	23	74
Gabo	1945	NSW	1.939	8	81	2.003	13	78
Glaive	1966	RAC	1.948	7	82	2.013	11	78
Halberd	1969	RAC	2.168	3	91	2.429	3	94
Hard_Federation	1914	NSW	1.564	27	66	1.963	14	76
Insignia	1946	Vic	2.098	4	88	2.366	4	92
Javelin_48	1949	WARI	1.786	15	75	1.924	18	75
Kite	1973	NSW	1.868	11	78	2.109	8	82
Majestic	1900?	SA	1.649	22	69	1.922	19	75
Marathon	1939	SA	1.618	24	68	1.920	21	75
Pusa	1916	India	1.461	28	61	1.601	28	62
RAC_892	2002	RAC	2.492	1	105	2.847	1	110
Ranee	1924	Vic	1.707	20	72	2.069	9	80
Ridley_48	1949	WARI	1.835	13	77	1.922	19	75
Sabre	1952	RAC	1.845	12	77	1.882	25	73
Scimitar_48	1949	WARI	1.825	14	77	2.009	12	78
Steinwedel	1884	SA	1.375	29	58	1.468	30	57
Sword	1923	RAC	1.771	16	74	1.930	17	75
Ward's_Prolific	1882	SA	1.610	25	68	1.838	26	71
Warigo	1941	WARI	1.666	21	70	1.941	16	75
Wongoondy	1948	WA	1.896	9	80	2.135	7	83
Yandilla_King	1907	SA	1.591	26	67	1.488	29	58
Yitpi	1999	WARI	2.383	2	100	2.577	2	100
site mean			1.803			2.000
No. of sites			8			1

WARI = Waite Agricultural Research Institute, SA; WA = Western Australia; Farrer = Farrer Institute; SA = South; Australia; NSW = New South Wales; RAC = Roseworthy Agricultural College; Vic = Victoria

Table 2 Yield, coefficient of variation (CV %), heritability (%), annual and growing season rainfalls (GSR) and water use efficiencies (WUE) of Yitpi and Federation wheats across the eight sites.

	Ave Yield	CV	Heritabiility	Rainfall (mm)		WUE (kg/ha/mm)
	(kg/ha)	(%)	(%)	Annual	GSR	Yitpi	Federation	Difference
Geranium	1.392	29.7	2.9	416	296	9	6	2
Lameroo	2.020	2.7	93.5	366	255	17	12	5
Nangari	1.919	8.1	69.5	264	199	23	17	6
Pinnaroo	2.616	3.0	91.8	327	254	21	15	5
Tepko	1.363	7.0	87.3	308	228	16	9	7
Waikerie	1.763	12.2	18.7	248	191	21	17	4
Wanbi	1.424	6.6	82.8	293	219	15	10	4
Wunkar	1.928	3.2	90.1	250	195	24	17	7
Mean	1.803			309	230	17	12	5
Minnipa	2.000	2.5		354	267	15	12	3

Figure 1 Mean grain yield (t/ha) for 30 wheat cultivars at eight sites in the Murray Mallee, plotted versus year of introduction. Y axis is grain yield in t/ha and X the years since 1884.	Figure 2 Mean grain yield (t/ha) for 30 wheat cultivars at one site at Minnipa, plotted versus year of introduction. Y axis is grain yield in t/ha and X the years since 1884.

production than Federation, producing on average an extra 5 kg/ha of grain per millimetre of effective rainfall. At Minnipa comparing the same varieties there was 25% increase in WUE producing an extra 3 kg/ha of grain per millimetre of effective rainfall.

Australian wheat breeders are currently using conventional breeding and/or marker assisted selection techniques targeting single genes. Kuchel (2008) states genomic analysis of wheat should give breeders better understand of the genetic basis for drought stress adaptation. Combined with an effective selection system genomic analysis will hopefully lead to greater rates of genetic gain for drought tolerance and therefore provide wheats better suited to the South Australia Mallee cropping regions.

References

Angus W (1908) Agricultural Experiments (1907). J. Agric, South Australia, Number 39, February 1908.

Black I (2004) The regional impact of research and extension in increasing the productivity of South Australian broadacre agriculture: results and recommendations from econometric analyses. South Australian Research and Development Institute: Adelaide.

Blumethal M and Walcott, J. (2006) Strategic Approaches to Growing More Grain. Science for Decision Makers Seminar Series, Bureau of Rural Science. http://www.daff.gov.au/brs

Ferns GK, Fitzsimmons RW, Martin RH, Simmonds DH and Wrigley CW (1975) Australian Wheat Varieties. Identification according to Growth, Head and Grain characteristics, CSIRO.

Hamblin A and Kyneur G (1993) Trends in wheat yields and soil fertility in Australia, Bur. Res. Sciences.

Kuchel H (2008) Breeding for drought tolerance, GRDC Update, Adelaide, February 2008.

Pearson FB (1957) Choosing wheat varieties: Department of Agriculture, South Australia. Leaflet No. 9/57

Perry MW and D'Antuono MF (1989) Yield improvement and associated characteristics of some Australian spring wheat cultivars introduced between 1860 and 1982. Australian Journal of Agricultural Research 40(3) 457 – 472

Simmonds DH (1989) Wheat and Wheat Quality in Australia, CSIRO.

Vandeleur RK and Gill GS (2004) The Impact of plant breeding on the grain yield and competitive ability of wheat in Australia. Australian Journal of Agricultural Research 55: 855-86

• "Global Issues. Paddock Action." Edited by M. Unkovich. Proceedings of 14th Agronomy Conference 2008, 21-25 September 2008, Adelaide, South Australia.