Getting more bang from our crops

William James Farrer, best known for developing the "Federation" breed of wheat which started to impact Australian agriculture back in 1903, had a career in wheat breeding that catalysed a century of progress in the Australian wheat industry, with wheat yields rising from 0.5 to 1.8 tonnes per hectare between 1900 and 2000, accompanied by a huge increase in production, from about 1 to 20 million tonnes and, of course, farmers’ incomes.

Memorialised by statues in Queanbeyan and on his Canberra property “Lambrigg”, and with a Canberra suburb and an Australian electoral division named in his honour, he still gets some attention with the annual Farrer medal that I am proud to have won this year. Perhaps it is a good opportunity to consider the range of influences on wheat yield in Australia, in order to make better decisions on where research investment should be in the future.

I believe we can attribute about half of the Australian wheat yield improvement over the last century to breeding and half to better crop management (agronomy), bearing in mind that part of the increase arises from the positive interaction between these two routes of innovation.

Modern varieties are earlier, shorter, with higher grain to straw ratios, and having more grains per unit area of land, smaller more erect leaves and possibly greater photosynthetic activity. Agronomic improvements over the last century comprise better crop nutrition (through soil improvement and use of chemical fertilisers), improved timeliness of cropping operations (especially through mechanisation and tillage advances), almost fully effective weed control based on herbicides, and reduced soil disease and better soil structure with improved crop rotations and sequences. Crop physiology played a modest role in underpinning about one half of the breeding and agronomic innovations.

It is imperative that yield progress be maintained, both in Australia and especially in developing nations. This will be achieved through closing the gap between actual yields and best-practice ones, and through raising best-practice yields: prospects for the latter are examined in some detail later.

With future agronomic possibilities, better sustainability and cost savings are evident, but yield enhancements appear fairly limited for three main areas of current agronomic research, namely more soil moisture available for the crop, biological soil additives, and seasonal climate forecasts.

The rate of wheat yield progress with conventional breeding is probably between 0.5 and 1.0  per cent pa (with the lower figure for dryland wheat): this rate of progress may be slowing (but certainly it has not ceased) and it may be becoming more difficult. Modern tools such as biometrics, robotics, computers and more recently molecular marker aided selection (MAS) and physiological selection criteria will help, but future progress will be more expensive to achieve.

Given the imperative, and the difficulties of conventional breeding for yield, the new field of functional genomics has been promoted as a way forward. It refers to enhancing plant and crop function through better understanding, and then engineering, at the gene and nucleic acid level, delivering more yield through transgenic varieties.

While transgenic crops have been remarkably successful where simple traits such as herbicide, insect and virus resistance are involved, my assessment is that progress with quantitative traits like yield, or yield under stress, has been disappointing and well short of early claims. Complexity, some unanticipated, at the molecular level is a major reason for this.

At the same time as glamorous functional genomics continues to attract substantial research investment, especially in the private sector, globally there is a struggling interest in approaches which pay more attention to the physiological level of organisation in the crop, seeking the key processes for yield determination, and where appropriate linking them to molecular studies at the gene level.

This approach needs more attention, especially as it is likely to help deliver the benefits of functional genomics. At the same time, we must not overlook investment in conventional breeding and agronomic research which continue to deliver progress.