6-8 April 2009 Prague, Czech Republic

Challenges and Opportunities for Further Improvements in Wheat Yield Gustavo A. Slafer ICREA (Catalonian Institution for Research and Advanced Studies), Department of Crop & Forest Sciences, Centre UdL-IRTA,

University of Lleida - Spain

1

0

6

15

4

10

2

5

0

0 0

20

40

60

80

Arable Land (106 ha)

2

World population (billions)

Wheat yield (Mg ha-1)

3

100

Years (from 1900) Adapted from: •Evans (1997) Phil Trans R Soc Lond B 352, 901-906 •Slafer et al. (1996) In: "Increasing Yield Potential in Wheat: Breaking the Barriers" (M.P. Reynolds et al., Eds). CIMMYT: Mexico DF, pp. 101-133 G. A. Slafer (ICREA) Centre UdL-IRTA

Universitat de Lleida

•Yield of cereals has been very strongly increased during the last half century (e.g. Calderini & Slafer, 1998. Field Crops Res., 57:335-347; Slafer & Peltonen-Sainio, 2001. Agric. Food Sci. Finland 10:121-131; Cassman et al., 2003. Annu. Rev. Environ. Resour. 28:315–58)

•This was due to – genetic improvements in both yield potential and in resistance to diseases – as well as to improvements in management (e.g. Slafer & Andrade, 1991. Euphytica, 58, 37-49; Austin, 1999. Crop Science 39:1604-1610)

•Improved yield potential has concomitantly improved yield responsiveness to environment (Calderini & Slafer, 1999. Euphytica 107: 51–59) Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

10 8 6 4 2 0

Yield (Mg ha-1)

Yield (Mg ha-1) Yield (Mg ha-1)

10 Italy 8 Modern Cv 6 4 2 0 2 4 0

Old Cv

6

8

Argentina

3 2 1 0 0

Australia

Modern Cv Old Cv

0.5 1.0 1.5 2.0 2.5 Environmental index (Mg ha-1)

Modern Cv

Old Cv

0

2 4 6 8 Environmental index (Mg ha-1)

Calderini & Slafer (1999) Euphytica, 107:453

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

3

Calderini et al. (1995)

Argentina

8 6

10

Austin et al. (1980; 1989)

8

UK

6 2

4 2

1

0 0 1900 1920 1940 1960 1980 2000

Grain yield (Mg ha-1)

Grain yield (Mg ha-1)

4

4 2 0 1900 1920 1940 1960 1980 2000

Year

7 Abeledo et al. (2003) 6 Barley 5 4 3 2 1 0 1920 1940 1960 Year

1980

2000

Year Similar evidences for maize and soybean in the US can be found in Evans, 1998. Feeding the 10 billion... (Camb. Univ. Press) Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Wheat Adapted from (i) Slafer, Calderini & Miralles (1996)- In: "Increasing Yield Potential in Wheat: Breaking the Barriers" (M.P. Reynolds, S. Rajaram & A. McNab, Eds) CIMMYT: Mexico DF, pp. 101-133.

Grain yield (Mg ha-1)

3

2

(ii) Calderini, D.F. & Slafer, G.A. 1998. Changes in yield and yield stability in wheat during the 20th century. Field Crops Research, 57:335-347

1

(iii) Slafer & Satorre (1999) In: "Wheat: Ecology and Physiology of Yield Determination" (E.H. Satorre & G.A. Slafer, Eds) Food Product Press, New York, pp. 3-12.

0 0

20

40

60

80

100

Years (after 1900) G. A. Slafer (ICREA) Centre UdL-IRTA

Universitat de Lleida

For more than 50 years, the supply of food commodities has grown faster than demand, in spite of increasing population and per capita incomes.

347 Kg pc

2000 - 2005 336 Kg pc

World per-capita production of cereals

Evidence of a slowdown in the long-term path of agricultural productivity growth has been clear in the last c. 15-20 years

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Developed countries share downward trends in agricultural productivity and in public investment in research These are worrying trends when in coming decades agriculture needs to adapt to climate change while world population grows by another three billion Food productivity declining with research Dr John Mullen School of Business, Charles Sturt University, Distinguished Fellow of the Australian Agricultural and Resource Economics

To regain rates of yield gain compatible with rates of growth of food demand investments in Agricultural Gustavo A. Slafer Research has to increase Centre UdL-IRTA Universitat de Lleida

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Past (impressive) production increases came from farmers’ yield gains (cropping land has only marginally increased in the last half century) related to advances in both genetic and management improvements

Genetic gains were mainly due to gains in yield potential, management gains were mainly due to resource use (e.g. fertilizers, water) Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Future gains must come more relevantly from breeding, if the gains are to be compatible with environmental safety and production sustainability Huge challenge for wheat (and other major crops) breeding It has to be more efficient than what the green revolution has been And now with a crop that already has a rather high yield potential

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

As molecular biology has the potential to identify and map particular genes or QTLs related to any trait - even quite complex ones such as WUE, NUE or yield… … and its usefulness in the case of traits controlled by major (or few) genes is beyond any questioning It is tempting to believe that the issue will be resolved with molecular biology tools, by simply identifying Yield-QTLs (or genes) and introgressing these genes into the elite germplasm Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

‘omics • • • •

Genomics Transcriptomics Proteomics Metabolomics

• Agronomics Crop Physiology Crop Ecology Crop Breeding

As discussed already in several talks of this conference, we must substantially improve productivity (yield potential, water use efficiency) These are very complex traits (high GxE, low heritability) and genes for complex traits are not going to be easily found directly We must understand first what we want to improve: a relatively simple trait putatively related to yield under field conditions For that to happen we must improve funding to more traditional agriculture research: this would not go against research in biotechnology; it might actually be essential for making sense of the huge progress done in molecular biology capabilities Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

It seems clear that we can map virtually any trait but when it comes to yield (or any other complex trait, e.g. WUE, NUE, tolerance to complex abiotic stresses) the direct identification of QTLs in a mapping population doesn’t seem quite useful in practice … We must identify “what to map” (phenotyping) And the ‘traditional’ bottom-up approach used by biotechnology that worked perfectly well for qualitative traits has not worked for complex traits so far

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

The figure shows an array of dots. we could study the array by looking for pattern in them. We would note that • most of the dots are circular, • that there is a bimodal distribution of their size, • that dots of a given size tend to be clustered together • and so on… There is plenty of pattern in the figure to make sound conclusions Figures from Structuralism by Jean Piaget Passioura (1979) Search 10:347-350 Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

But is there significant pattern? What does 'significant' mean? Have we been making the right observations? If we look at an extended field of the dots as shown in this figure, the dots “almost disappear”, Gaining significantly in significance, we are no longer looking at dots We are now looking at the face of a man wearing glasses and smoking a pipe

Figures from Structuralism by Jean Piaget Passioura (1979) Search 10:347-350 Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

The visual metaphor I made implies that our understanding a biological phenomenon is incomplete unless we can relate it to (or translate it into) phenomena in the adjoining level of the organizational scale If we cannot relate it to a higher level, the phenomenon may end up being trivial May be this the cause for the scientific ‘curiosity’ that being the literature full of QTLs for yield (or for resource use efficiency or other complex agronomically relevant attributes) there has been not clear examples of using those QTLs for improving yield of major crops

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Crop physiology

Productivity sustainability

Crop ecology

?

Linear reasoning is not directly acceptable. We must be certain of the level of dependence of the trait on the scale Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Scale-independent traits remain relevant when scaling up to the population and community levels, where crop yield is defined, and the ecosystem level, where sustainability is defined (e.g. Bacillus turingensis insecticidal proteins) Scale-dependent traits are buffered in the scaling up process, and the actual relevance at the population and ecosystem levels could be minor, or negative in some cases (e.g. enhancement of Rubisco as a means to increase crop production; Sinclair et al., 2004).

Trait relevance

1.0

0.0

gene

molecule cell

tissue

organ individual ecosystem population

Level of organisation Victor Sadras, SARDI-Australia, pers. commun.

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida

Conclusions Wheat (and other crops) breeding has been impressively successful in the relatively recent past (green revolution); but it has to be even more successful in improving yield potential in the future Genetic gains are imperative, but a strong reliance in biotechnological “magic” solutions may be counterproductive We must increase our understanding of the processes that matter at the crop level of organisation and then try to identify genetic bases that might help rising crop yield No substantial advances in knowledge will be achieved without funding agricultural research, but it may not be only a matter of total funding but also avoiding diverting all funds behind promises Gustavo A. Slafer Centre UdL-IRTA Universitat de Lleida