Since land was first cultivated to create a favorable environment for crops, other less desirable plants have exploited the same land more effectively. Control of weeds has been described as humanity’s biggest single occupation, and while herbicides have greatly reduced the effort needed to control weeds in some farming systems, in others, controlling them requires more labor than any other facet of crop production up to harvest. Weed control can also account for a significant portion of crop costs.
Many lowland rice systems—ranging from highly developed irrigated systems to hillside terraces—integrate several “cultural” weed control elements into crop management. Floodwater, on puddled fields, to suppress weed growth is the most widespread. Another, the transplanting of rice seedlings, gives the seedlings a size advantage over any germinating weeds and allows farmers to maintain floodwater and avoid draining the field, which stimulates weed germination. A third approach, thorough land preparation, involves repeated cultivation, which kills existing weeds and depletes the soil’s “seed bank.”
In many areas, however, water and labor shortages are forcing farmers into new approaches. Direct seeding of rice, for example, requires less labor than transplanting and water shortages lead to reduced periods of flooding. Further, reduced tillage systems allow farmers to save energy costs, which have risen steeply in recent years.
Anyone’s field may be home to dozens of plant species other than the crop itself, though only a few of these are likely to be important weeds. Prevailing growing conditions will usually favor a small number of weeds, which may consequently cause serious problems. If a farmer changes the field conditions—by direct seeding or reducing flooding, for example—other species may become dominant. The composition of a field’s weeds and soil seed bank reflect current and past crop management and farmers, particularly in traditional farming systems, often use certain species as indicators of the system’s overall health.
In the modern systems of intensive rice cultivation and repetitive crop management, there may only be a very limited number of species occurring—often annual grasses that have similar growth habits to rice and are able to thrive under existing farming practices. Repeated use of a single set of crop and weed management components will commonly result in the “deflection” of the weed composition to a single weed species. Further, in some rice growing areas weeds have evolved resistance to regularly used herbicides, making them increasingly difficult to manage. To counter such problems, we need long-term weed management strategies that aim to prevent the buildup of problem weeds and to make the most of opportunities for cultural control.
In this light, knowledge of how the weed species react to different management practices over several seasons can be used to predict changes and avoid problems. Any change—such as a move from transplanting to direct seeding or from flooded to dryland conditions—will tend to favor a new set of weeds, while species that thrived in the previous conditions may decline in dominance. By anticipating changes in weed composition, farmers can alter their crop management in response to emerging problems and implement more effective weed management strategies. Further, by rotating through a series of management practices, farmers can limit the chance of species becoming dominant.
Crop management rotations have not yet played a role in many of the modern rice systems, but these may have a greater role in the future. Changing the timing and depth of flooding, altering soil tillage and crop establishment practices, direct seeding rather than transplanting, alternating rice with other crops, and varying herbicides are examples of management practices that can affect the composition of weeds and prevent the build-up of individual species.
Agronomists must better exploit knowledge of how different weeds respond to varying ways of managing a crop. Farmers need knowledge-based crop and water management rotations that can form the cornerstones of future weed control strategies. Weed management is an age-old problem but it remains as relevant and challenging today as ever, and it is one that we ignore at our peril.
_________________________________________
David Johnson is a senior weed scientist in IRRI’s Crop, Soil, and Water Sciences Division.
