HERBICIDES in Asian rice - IRRI books - International Rice ...

are available in Asian markets, without many restrictions, or at least without enforced
restrictions (Rola and Pingali 1993, Moody 1996b).
Continuous use of pesticides, broadly classified to include the hazardous category
1 and 2 chemicals, tends to decrease biodiversity within rice ecosystems. Continuous
use also leads to the adaptation and possible blooming of individual microorganisms,
especially those residing in energy-poor environments in the soil (Racke
1990). This means long-term herbicide use could change the microbiology of the soil
as well as the ecology of weed populations. Research by Roger and Simpson (1996)
shows that herbicides applied on soil at recommended levels tend to have little impact
on microbial populations in the short term. The limited studies that exist, however,
indicate the relationship between pesticide use and biological variables may not be
consistent over long time periods. Although short-term studies are valuable in elucidating
the effects of herbicides on microflora in the soil, it is clear that additional field
studies are needed to assess more precisely the long-term impact of continuous herbicide
use on microbial systems and soil fertility.
A more serious threat to rice productivity is that continuous herbicide use can
promote shifts in the weed population and the emergence of herbicide-resistant weeds
(Dowler et al 1974, Kim et al 1975, Mahn and Helmecke 1979, Moody 1991, Ho et al
1992). Ecological shifts of weed species from annuals to perennials have occurred in
areas where herbicides have been used continuously in rice for a number of years,
such as in Japan and Korea (Shibayama 1996, Moody 1991, Ueki 1983, Vega 1988).
Examples from the Philippines and Malaysia (Ho et al 1994) indicate that continued
use of postemergence herbicides, such as 2,4-D, to suppress a number of easily controlled
broadleaf weeds and sedges have led to complete dominance by grass weeds,
which are much harder to control. Similarly, use of grass weed killers, such as
pretilachlor, molinate, propanil, and oxadiazon, over a long time has increased the
dominance of broadleaf weeds and sedges. This empirical evidence indicates that,
although herbicides can eliminate dominant weeds in rice at any given time, chemical
weed control that relies over an extended time on the use of a single herbicide or
group of herbicides with the same mode of action will not be effective.
The emergence of new weed types creates a continuing competition for nutrients,
water, and sunlight-the principal resources of agricultural production. Weeds
tend to be more effective competitors than rice in ricefield ecosystems for three reasons:
they have lower dry weight, they require less water and nutrients for growth,
and they produce more seeds. The physiological interactions between weeds and rice
can be modeled to predict the probable extent of competition in any given rice ecosystem.
These models can be useful in estimating potential yield loss and in designing
more effective weed management programs (Weaver 1996, Kropff and van Laar 1993).
A fine line exists between the evolution of weed types in intensive rice systems
and the buildup of resistance to herbicides in weeds. Controlling weeds with a single
herbicide can lead over time to the evolution of herbicide-resistant weeds, through the
selection of naturally occurring herbicide-resistant biotypes or through genetic muta-
Herbicide use in Asian rice production 15