Bioherbicide development for cereals in integrated weed management

• University of Can Tho, Vietnam
• National Institute of Plant Protection, Vietnam

The world's principal food crops are members of the grass family, which also contains many unwanted weeds that grow in the crops. The cost of these weeds is high, both in Australia and Vietnam, because they reduce potential yields considerably. Control using chemical herbicides is difficult and expensive. In many countries, including Vietnam, much weeding is instead done manually - a laborious task usually carried out by women, using up time that could be more usefully spent.
In addition, if synthetic chemical herbicides are used, as in some areas in the south of Vietnam, the herbicides may affect the production of fish in the rice paddies and may also contaminate the water collected for drinking. The development of resistance by the weed - especially if, as with rice, three crops and therefore three sprayings per year are necessary - is another potential problem.
An alternative approach to controlling grassy weeds is the concept of bioherbicides. These are applied like conventional herbicides but the active ingredient is a naturally occurring microbe - usually a fungus. Such mycoherbicides are very specific, so there is less risk of damage to non-target plants, and they also multiply on the target species, so there is less contamination of products and the environment. In addition, the bioherbicide does not persist in the environment beyond the target plant.
There has been little work on bioherbicides in tropical countries, even though the warm, moist conditions suggest that this concept would be more successful in the wet tropics than in dry temperate areas. A previous ACIAR project (CS2/1994/002) identified the major suitable fungal pathogens of the principal rice and wheat weeds. This project built on this work, by developing bioherbicides from the identified fungi.

The main aim of the project was to turn fungal pathogens into commercial herbicides which can be used against major crop weeds in Vietnam and Australia.

The project was divided into four parts. In the first, scientists targeted barnyard grass, Echinochloa crus-galli and red sprangletop, Leptochloa chinensis in rice in Vietnam. The effectiveness of the bioherbicides was evaluated in field trials involving other control methods, such as varying crop sowing time and sowing rate. Toxicology tests on feeding animals were conducted to ensure the safety of the fungi for field use.
In the second part, the team developed a similar bioherbicide for use against wild oat, Avena fatua, the main weed of wheat in Australia. Fungi already identified as pathogenic were used to make a prototype that would also undergo field efficacy tests. The scientists also sought potential bioherbicides for the other major grass weed of wheat, annual ryegrass, Lolium rigidum.
The next phase of the research concentrated on developing mass production methods for the fungi. Low-cost materials, using waste products from agriculture or food processing, were tested for their ability to act as a substrate and a nutrient source on which to grow the fungi.
Once large quantities were available, the scientists conducted tests on the host range of the bioherbicide, using closely related species and other major crops.

A previous ACIAR project (CS2/1994/002) identified specific fungal phathogens for two species of weeds found in Vietnam. These were barnyard grass, Echinochloa crus-galli, and red sprangletop, Leptochloa chinensis. Both are major grassweeds and were targeted for control by bioherbicides based on the identified fungal pathogens.

These fungi, Exserohilum monoceras for barnyard grass and Setosphaeria rostrata for red sprangletop, showed potential for development as bioherbicides for application in the Red River Delta in the north and the Mekong Delta in the south. In the former, barnyard grass (Echinochloa crus-galli) was the target weed and the fungus Exserohilum monoceras was selected as a potential bioherbicide.
Shade-house experiments with these fungi confirmed their suitability and techniques for their mass production were investigated.

The red sprangletop bioherbicide, was successful in large scale repeated experiments at Cuu Long. Mass production methods for this fungus have been achieved using cheap materials, including rice husks. Cuu Long Rice Research Institute intends to register this bioherbicide as a product with the Vietnamese government and develop it into a practical product for rice farmers.

In repeated large scale field experiments at Hanoi and Haiphong, the barnyard grass bioherbicide proved to be insufficiently virulent to warrant further development.

The project's main impact to date has been at institutional level, increasing capacity for biological control work and informing the Ministry of the potential of this approach to weed/pest control.
The adoption of project results in the first instance will be in the Mekong delta with the development of the practical bioherbicide at Cuu Long Rice Research Institute for red sprangletop. However this bioherbicide could be used anywhere in Vietnam where this weed is a problem. The use of the bioherbicide should decrease the need for chemical herbicides with their externality problems.

At both NIPP and Cuu Long capacity for weed science and plant pathology has been enhanced through this project via the provision of capital items, books and training visits of Vietnamese scientists to Australia, IRRI and international conferences as well as regular visits by Australian scientists to Vietnam.

In Australia, research concentrated on wild oats, Avena fatua, the worst weed of wheat throughout the world. The fungus Drechslera avenacea was identified in a previous ACIAR project (CS2/1994/002) as the most suitable pathogen found in Australia to develop as a bioherbicide for this weed. However glasshouse experiments demonstrated that wild oat plants had considerable ability to recover from infection. Moreover, controlled environment experiments indicated that low temperatures in the field at the time of application of the bioherbicide would severely limit its potential to cause significant infection.

Subsequently, research in Australia was directed towards finding potential bioherbicides for the other major grass weed of wheat, annual ryegrass, Lolium rigidum. Extensive surveys and pathogenicity testing failed to discover a potentially useful fungus to develop as a bioherbicide. The work on wild oats has been published and the work on ryegrass is being prepared for publication.