REVIEW ARTICLES

ARTICLES

ARTÍCULOS GENERALES

Integrated pest management in rice

K.D. Gallagher, P.A.C. Ooi, T.W. Mew, E. Borromeo and P.E. Kenmore¹

INTRODUCTION

This paper is a conceptual guide to the recent developments in rice integrated pest management (IPM) within an ecological framework. It does not provide so much a "how to" as a "why to" guide for IPM programmes that are based on ecological processes and work towards environmentally friendly and profitable production.

IPM in rice has been developing in many countries since the early 1960s. However, much of the development was based on older concepts of IPM, including intensive scouting and economic thresholds that are not applicable under all conditions (Morse and Buhler, 1997) or all pests (e.g. diseases and weeds), especially on smallholder farms where the bulk of the world's rice is grown and that often operate under a weak or non-existing market economy. During the 1980s and 1990s, important ecological information on insect populations became available, making possible a stronger ecological approach to pest management and greater integration of management practices that went beyond scouting and economic threshold levels for decision-making (Kenmore et al., 1984; Gallagher, 1988; Ooi, 1988; Graf et al., 1992; Barrion and Litsinger, 1994; Rubia et al., 1996; Settle et al., 1996).

Instead, an ecological and economic analysis approach to management has been adopted that takes into consideration crop development, weather, various pests and their natural enemies. Operationally, this approach has been defined to form the guiding principles for IPM implementation, clearly setting out in simple language the actions to be undertaken. These principles were first articulated in the Indonesian National IPM Programme but have expanded as IPM programmes have evolved and improved. Currently, programmes in Africa and Latin America use the term integrated production and pest management (IPPM), and the IPPM principles are:

• grow a healthy soil and crop;
• conserve natural enemies;
• observe the field regularly (e.g. soil, water, plant, pests and natural enemies);
• that farmers should strive to become experts.

Within these principles, economic decision-making remains at the core of rice IPM but the approach also incorporates good farming practices and active pest control within a production context.

IPM in rice seeks to optimize production and to maximize profits through its various practices. To accomplish this, however, decision-making must always take into consideration both the costs of inputs and the ecological ramifications of these inputs. A particular characteristic of Asian rice ecosystems is the presence of a potentially very damaging secondary pest, the rice brown planthopper (Nilaparvata lugens). In the past, large-scale outbreaks of this small but mighty insect have occurred, resulting in disastrous losses (IRRI, 1979), although these outbreaks were primarily pesticide induced - triggered by pesticide subsidies and policy mismanagement (Kenmore, 1996). In general, however, the brown planthopper remains a localized problem, especially where pesticide overuse and abuse are common, and can therefore be considered as an ecological focal point around which both ecological understanding and management are required to achieve profitable and stable rice cultivation. The brown planthopper has also become the major entry point for all IPM educational programmes because it is always necessary to take precautionary measures against an outbreak during crop management. Other pests that interact strongly with input management are rice stem borers and certain diseases discussed below.

Populations of brown planthopper are usually kept low by the action of a wide range of natural enemies. Outbreaks reported in the tropics during the 1970s were associated with regular use of insecticides. The stronger the insecticide, the faster is the resurgence of brown planthopper populations - which leads to the large-scale dehydration of rice plants, a symptom known as "hopperburn". Insecticides remove both brown planthoppers and their predators. However, eggs laid inside the stem are relatively unharmed and, when these hatch, the nymphs develop in an environment free of predators. In unsprayed fields, the population of brown planthoppers does not increase to any significant level, suggesting the importance of biological control. Farmers learn about predators by carrying out experiments and when they discover the role of these natural enemies, they are less likely to use insecticides. In Indonesia, Presidential Decree 3/86 provided the framework and support necessary for farmers to understand and conserve natural enemies and this, in turn, has helped keep rice fields in Indonesia relatively free from brown planthoppers over the past ten years. Also during this period, an extensive programme to educate farmers based on a farmer field school model has been implemented.

A last major point to keep in mind when considering IPM decision-making is that of paths to rice production intensification. In most cases, intensification means the use of improved high-yielding varieties, irrigation, fertilizers and pesticides - as was common during the "green revolution" period. However, two approaches to intensification should be considered. The first is the input intensification approach: here it is important to balance optimal production levels against maximizing profits - in some combinations higher inputs can destabilize the production ecosystem. The second route to intensification is that of optimizing all outputs from the rice ecosystem for maximizing profits. In many lowland flooded conditions, this may mean systems such as rice-fish or rice-duck that may be more profitable and less risky but require lower inputs (and often resulting in lower rice yields). In areas where inputs are expensive, where the ecosystem is too unstable (e.g. for reasons of drought or flood) to ensure the recovery of input investments, or where rice is not marketed, then such a path to intensification may be more beneficial over time. However, such a system has a different ecology owing to the presence of fish or ducks, and will therefore involve a different level of IPM decision-making.

IPM ON AN ECOLOGICAL BASIS

IPM in rice is now firmly based on an ecological understanding of the crop and its interaction with soil nutrients and varieties. An ecological overview of our current understanding of how the rice ecosystem operates during the development of the crop and consequent ecological considerations for IPM methods is presented below.

The rice ecosystem in Asia is indigenous to the region; rice was first domesticated before recorded history, perhaps more than 6 000 years ago (Ponting, 1991), while reaching cultivation similar to that of modern days in the sixteenth century (Hill, 1977). This lengthy time period means that the rice plant, pests and their natural enemies have existed together and co-evolved for thousands of generations. Rice ecosystems typically include both a terrestrial and an aquatic environment during the season, with regular flooding from irrigation or rainfall. These two dimensions of the rice crop may account for the extremely high biodiversity (Cohen et al., 1994) found in the rice ecosystem and its stability even under intensive continuous cropping - in contrast with the relative instability of rice production under dryland conditions. Irrigated rice systems in Africa, the Americas and Europe also include this aquatic and terrestrial ecosystem with accompanying high levels of biodiversity - and these factors may also provide the relative stability found in these systems.

Insects

Studies by Settle and farmer research groups in Indonesia (Settle et al., 1996) show that flooding of fields triggers a process of decomposition and the development of an aquatic food web, which results in large populations of detritus-feeding insects (especially Chironomid and ephydrid flies). These insects emerge onto the water surface and into the rice canopy in large numbers, very early in the growing season, providing critical resources to generalist predator populations long before "pest" populations have developed (see Figure 1). This model is quite different from the usual predator-prey models taught in most basic IPM courses and provides a mechanism to suggest that natural levels of pest control in tropical irrigated rice ecosystems are far more stable and robust than in purely terrestrial agro-ecosystems. This stability, however, was found to be lower in rice landscapes that are subject to long (more than three months) dry seasons and where rice is planted in large-scale synchronous monocultures, as well as in areas where farmers use pesticides intensively. A corollary to this idea is the result that increasing amounts of organic matter in the soil of irrigated rice fields - by itself a highly valuable practice for sustainable nutrient management in tropical irrigated rice - has boosted the populations of both detritus-feeding insects and insect predators, thereby improving natural levels of pest control (Settle et al., 1996).

FIGURE 1
Hypothesized flow of energy in tropical rice systems

A second important consideration in rice IPM is the ability of most rice varieties to compensate for damage. The rice plant rapidly develops new leaves and tillers early in the season, replacing damaged leaves quickly. The number of tillers produced is always greater than the number of reproductive tillers, allowing for some damage of vegetative tillers without affecting reproductive tiller numbers. The flag leaf contributes to grain filling but the second leaf also provides photosynthates, while lower leaves are actually a sink that competes with the panicle. Finally, photosynthates appear to move from damaged reproductive tillers to neighbouring tillers so that total hill yield is not as severely affected as expected when a panicle is damaged by stem borers. Thus, early season defoliators (e.g. whorl maggot, case worms and armyworms) cause no yield loss up to approximately 50 percent defoliation during the first weeks after transplanting (Shepard et al., 1990; Way and Heong, 1994), although higher damage can be expected when water control is difficult. Because early tillering is also higher than the plant can ultimately support reproductively, up to 25 percent vegetative tiller damage by stem borers (Scirpophaga spp., Chilo spp. and Sesamia spp.) (dead-hearts) can be tolerated without significant yield loss (Rubia et al., 1996). Significant damage (above 50 percent) to the flag leaf by leaffolders (Cnaphalocrocis medinalis and Marasmia spp.) during panicle development and grain filling can cause significant yield loss - although this level of damage is not common where natural enemies have been conserved (Graf et al., 1992). Late season stem-borer damage (white-heads) also has a less severe impact than previously expected such that up to 5 percent white-heads in most varieties do not cause significant yield loss. (Rubia et al., 1996; Way and Heong, 1994). The conspicuous rice bug (Leptocorisa oratorius) is another major target for insecticide applications. However, in a recent study involving farmers and field trainers at 167 locations, van den Berg and Soehardi (2000) demonstrated that the actual yield loss in the field is much lower than previously assumed. The rice panicle normally leaves part of its grain unfilled as if to anticipate some level of loss (Morrill, 1997). Numerous parasitoids, predators and pathogens present in most rice ecosystems tend to keep these potential pests at low densities (Barrion and Litsinger, 1994; Loevinsohn, 1994; Shepard and Ooi, 1991; Ooi and Shepard, 1994; Matteson, 2000).

Thus, under most situations where natural enemies are conserved, little yield loss is expected from typical levels of insect pests. Until recently, insecticide applications for early defoliators, dead-hearts and white-heads often led to lower natural enemy populations, in turn, leading to massive outbreaks of the secondary pest, brown planthopper (Nilaparvata lugens) (Rombach and Gallagher, 1994). Work by Kenmore et al. (1984) and Ooi (1988) clearly showed the secondary pest status of the brown planthopper. Although resistant varieties continued to be released, the highly migratory sexual populations of this pest were found to have high levels of phenotypic variation and be highly adaptable to new varieties. Although wrongly proposed to be "biotypes", it was found that any given population included significant numbers of individuals with the ability to develop any gene for resistance (Claridge, Den Hollander and Morgan, 1982; Sogawa, Kilin and Bhagiawati, 1984; Gallagher, Kenmore and Sogawa, 1994). Huge outbreaks no longer occur in areas where pesticide use has fallen following either changes in policy on the regulation of pesticides or educational activities.

A few minor pests are predictable problems and therefore preventive action with natural enemies, resistant varieties or specific sampling and control should be considered. Such pests include black bug (Scotinophara spp.), gall midge (Orseolia oryzae), and rice hispa (Dicladispa spp.), which are consistently found in specific regions; thrips (Stenchaetothrips biformis) where drought causes leaf-curling that provides a good thrips habitat; and armyworms (Mythimna spp. and Spodotera spp.) in post-drought areas (generally attracted by high levels of mobilized nitrogen in the rice plant) although panicle-cutting armyworms cause extreme damage. Green leafhoppers (Nephotettix spp.) are important vectors of tungro (see below) but by themselves rarely cause yield loss. White-backed planthoppers (Sogatella spp.) are closely related to brown planthoppers in terms of population dynamics and are not usually a major yield-reducing pest. Rice water weevil (Lissorhoptrus oryzophilus), introduced from the Caribbean area and northeast Asia, is a problematic pest requiring intensive sampling (Way et al., 1991) and merits greater research on its natural enemies (both rice water weevil and rice bug need research on attractants that could be used for traps early and late in the season, respectively). In upland ecosystems, white grub species and population dynamics require further study and are difficult to manage. Way et al. (1991) have provided a good overview of insect pest damage dynamics while Dale (1994) has given an overview of rice insect pest biology.

Diseases

The need to grow more rice under increasingly intensive situations leads to conditions that favour diseases. High planting density, heavy inputs of nitrogen and soil fertility imbalance result in luxuriant crop growth conducive to pathogen invasion and reproduction. This is compounded by genetic uniformity of the crop stand, which allows unrestricted spread of the disease from one plant to another, and by continuous year-round cropping that carries over the pathogen to the succeeding seasons. Reverting to the less-intense, low-yield agriculture of the past may be out of the question, but a thorough understanding of the ecological conditions associated with the outbreak of specific diseases may lead to sustainable forms of intensification.

Brief descriptions are given here of the three major diseases of rice, namely, rice blast, sheath blight and rice tungro disease.

Blast (Pyricularia grisea, Magnaporthe grisea) occurs throughout the rice world but is more prevalent in areas with a cool, wet climate. It is a recognized problem in upland ecosystems with low input use and low yield potential, as well as in irrigated rice ecosystems with high input use and high yield potential (Teng, 1994). Fertilizer application and high planting density are known to exacerbate the severity of infection. Plant resistance is widely used to control the disease, but varieties often need to be replaced after a few seasons because pathogen populations quickly adopt and overcome the varietal resistance. Recent work undertaken by the International Rice Research Institute (IRRI) and Yunnan Agricultural University demonstrated that the disease can be managed effectively through varietal mixtures (see Box 3).

Sheath blight (Rhizoctonia solani) is a problem during warm and humid periods and is also aggravated by dense planting and nitrogen inputs above 100 kg/ha. No plant resistance is known for sheath blight. A number of bacteria (Pseudomonas and Bacillus) isolated from the rice ecosystem are known to be antagonistic to the pathogen. Foliar application of antagonistic bacteria at maximum tillering stage appeared to effect a progressive reduction of the disease in the field over several seasons (Du, 2001; Xie, Yu and Ren, 2001). The incorporation of straw and other organic matter, with its favourable effects on soil fertility, pH level and possibly on beneficial micro-organisms, may reduce sheath blight incidence in the long term.

Rice tungro disease, caused by a complex of two viruses transmitted by the green leafhopper (Nepthettix virescen), is a destructive disease in certain intensively cultivated areas in Asia where planting dates are asynchronous (Chancellor et al., 1999). Overlapping crop seasons provide a continuous availability of the host that enables the year-round survival of the virus and the vector. Controlling the vector population with insecticide does not always result in tungro control. Synchronous planting effectively keeps the disease at manageable levels. When and where planting synchronously is not possible, the use of resistant varieties are recommended. In addition to varieties with a certain degree of resistance to the vector, varieties highly resistant to the virus itself recently became available. Farmers should also employ crop or varietal rotation, and rogue intensively.

Fungicidal control of blast and sheath blight is increasing in many rice intensification areas. It is extremely important that these fungicides be carefully screened not only for efficacy as fungicides but also for impact on natural enemies in the rice ecosystem. One example is the release of iprobenfos (Kitazin), as a fungicide for blast control. Iprobenfos is an organophosphate that was originally developed for brown planthopper control but is highly toxic to natural enemies. Its use in the rice ecosystem will generally cause destabilization and high populations of brown planthopper. Fungicides should also be carefully screened for their impact on fish - both to avoid environmental damage in aquatic systems and to avoid damage to rice-fish production.

In general, clean and high-quality seed with resistance to locally known diseases should be used as a first step in rice IPM of diseases. An appropriate diversification strategy (varietal mixture, varietal rotation, varietal deployment, crop rotation) should be developed to counter the capacity of the pathogen to adapt quickly to the resistance of the host. The management of organic matter must be geared not only towards achieving balanced fertility but also to enhancing the population of beneficial micro-organisms.

Farmers in the Republic of Korea, who face heavy disease pressure in many cases, can learn to predict potential outbreaks using educational activities that combine various weather and agronomic input parameters with disease outcomes. Computer-based models are also being commercially sold to predict disease potential based on meteorological monitoring (e.g. blast and sheath blight models from Pessl Instruments at www.metos.at). With increasing nitrogen levels used in intensification programmes, however, increasing disease incidence can be expected and high levels of management as outlined above should be carefully considered.

Weeds

Weeds have long been a consideration in lowland rice cultivation, beginning with the origin of puddling, which is thought to have been invented and/or developed in order to create an anaerobic environment that kills several weeds including weedy red rice. In most IPM programmes in lowland rice, weed management has therefore been more closely considered as part of the agronomic practices during puddling and, later, during aeration of the soil with cultivators. At least two hand weedings are necessary in most crops, and in many countries are considered as economically viable owing to low labour costs or community obligations to the landless, who are then allowed to participate in the harvest. With rising labour costs, decreasing labour availability and more effective herbicides, this approach is increasingly being replaced by the use of one or two applications of pre- or post-emergence herbicides. As in the case of fungicides, it is important to ensure that herbicide applications do not upset natural enemies, fish or other beneficial/non-target organisms in the aquatic ecosystem, including micro-organisms (see Figure 1). In the case of upland rice, similar changes are occurring, although better dryland cultivators have already been developed for interrow cultivation as an alternative to herbicides.

Non-herbicide but low-labour weed management methods are also emerging from the organic agriculture sector. The International Association of Rice-Duck Farming has been established in Asia to support research and exchanges, mostly among organic farmers. In rice-duck farming, a special breed of duck is allowed to walk through the field looking for food that is either broadcast or naturally occurring: the physical effects of simply walking up and down the rows is sufficient to control most weeds. In Thailand, there is a system for rainfed rice fields in which mung bean and rice are broadcast together with some straw covering. When the rains come, both crops germinate. If there is abundant rain, the mung bean will eventually die and become part of the mulch, but if the rain is insufficient for the rice then the mung bean will be harvested. An approach characterized by no-till, no-herbicide with ground cover from winter barley straw or Chinese milky-vetch is being used in the Republic of Korea in both conventional and organic systems. Organic farmers in California, the United States, use a water management system in which there is a period of deep (30 cm) flooding followed by complete drying: the rice can take the changes but young weeds cannot survive. A widely adopted method in central Thailand involves growing rice from ratoons. After harvest, the stubble is covered with straw and then irrigated, which allows the rice plant to emerge. This method not only controls weeds effectively but also increases organic matter and requires no tilling.

However, in the majority of cases, labour saving in rice cultivation often means moving towards direct-seeded rice and thus more weed problems. Red rice is already the key weed in most Latin American direct-seeded rice production. It is clear that this situation will lead to more herbicide use in rice production. However, herbicide resistance is sure to emerge and there are obvious health and environmental costs associated with herbicides. Thus it is important that IPM in relation to rice weeds be improved and considered in the broadest terms (e.g. promoting consumer acceptance of modern rice varieties that are red in colour may be part of the solution to red rice problems, although early shattering of weedy red rice would need to be overcome). Crop rotations are only feasible in some areas, while simple line sowers or tractor sowing in rows combined with manual or tractor cultivation may provide some solutions for lowland and upland rice. Herbicide-resistant genetically modified rice will eventually reach the market, but consumers do not favour these varieties and the resulting increase in herbicide use will have obvious negative effects on the aquatic systems that are associated with most rice production. A further major risk with herbicide-resistant rice is the possibility of transferring gene resistance to weedy rice, although the specific characteristics of rice means there is no risk of such a transfer to wild grass species. The use of herbicide-resistant rice in monocropping will also create long-term serious problems of glyphosate resistance in weed species previously susceptible to the herbicide. The ecosystem-level interactions of herbicide-resistant rice will need careful assessment prior to its use.

Community pests

Insects, diseases (with the important exception of tungro virus) and weeds in rice ecosystems are largely managed on individual farms or plots. There is also a community aspect in terms of farmers influencing each other with respect to spraying times or runoff of toxic pesticides into community water resources used for bathing, washing, drinking or fishing. However, certain pests - rats, snails and birds - require greater community-level planning and action. Indeed, the management of these pests involves the facilitation of community organizations not generally supported by extension services except possibly some multipurpose cooperatives or water-user associations.

Rats

Numerous species of rat occur in rice fields and can cause considerable damage throughout a community. Rats often migrate locally from usually permanent habitats to rice areas as the food supply changes throughout a yearly cycle. The rice plant is most preferred after the panicle has emerged. Although natural enemies of rats do exist (especially snakes), pesticides and other measures taken by farmers often suppress their populations and thus make possible the survival of large rat populations. Rat management strategies have comprised determining the main species of rat present in order to ensure that baits are appropriate and then developing community-level mapping methods to plan and carry out continuous trapping along feeding routes, fumigation or digging of rat holes, modification of appropriate habitat and establishing early season bait stations using second-generation anticoagulant baits. (Although the highly toxic zinc phosphide and repackaged and unlabelled aldicarb (Temik), are still commonly seen, they are strongly discouraged in most countries following instances of the deaths of children and small livestock.) Community programmes can include educational activities on rat biology and behaviour to improve strategy development and participation in programmes (FAO, 1988). Year-round community-level management with emphasis on early-season vegetative-stage action (before booting) is considered to be the key to rat management (Buckle and Smith, 1994; Leung, Singleton and Sudarmaji, 1999). An innovative owl habitat programme in Malaysia has been successful in increasing owl levels as a means of reducing rat populations in rice and plantation crops. A trap-and-barrier system with plastic has also achieved good results in rice fields (Murakami, Priyono and Triastiani, 1991).

Golden apple snail

The golden apple snail (Pomacea canaliculata) originated in the Caribbean region and was introduced to rice-growing areas for the production of a caviar look-alike as an income-generating activity. It has since spread widely, from Japan to Indonesia, and is now one of the most damaging of rice pests. It was introduced without appropriate tests in any country, although it was on the quarantine lists of several countries. The snail feeds on various types of vegetation in aquatic environments, including newly transplanted rice seedlings up to about 25 days old, when their stems become too hard. The golden apple snail has no natural enemies and is highly mobile in its early stages, when it is carried by the flow of irrigation water and spreads rapidly throughout communities. Pesticide applications are being used before transplanting or direct-seeding using highly toxic compounds such as endosulfan, organo-tin compounds and metaldehyde. These compounds have serious health implications for humans and also cause the death of potential fish predators and other natural enemies early in the season (Halwart, 1994). The use of bamboo screens as inlets to fields in order to inhibit snail movement is reported as the first line of snail defence. Draining fields that have several shallow ditches where the snails congregate allows for faster collection and facilitates herding ducks in fields to eat the snails. In Viet Nam, snails are reported to be collected, chopped, cooked and used as fish food to the extent that they are declining as a problem. There does not currently appear to be any biological control exploration or programme active for their long-term control.

Birds

Birds can be very damaging to rice, especially when they occur in large flocks. The red-billed quelea (Quelea quelea) in sub-Saharan Africa and various species in Asia are known as persistent problems in rice ecosystems. In most Asian countries and in Chad, netting is used to trap large numbers of birds for sale as food. Mass nest destruction is also possible for some species. In Asia, these methods have effectively reduced pest bird populations to very low numbers. In Africa, the capture method may bring benefits to local people in terms of income or additional dietary protein, but the impact on pest bird populations has been minor. During the ripening period in northeast Asia, some fields are protected by being covered with bird nets, which are widely available. The nets are also often used to protect seed production fields. In both Asia and Africa various forms of bird-scaring are used to try to keep birds out of the fields. Reflective ribbons or used video or cassette tape are widely used to scare birds in Asia. Scaring in the form of people shouting or hurling dried mud at the birds is common in Africa. Sound cannons and owl or hawk look-alikes are also used in many developed countries. Scaring with devices that are not backed up by people is seldom effective for long, as the birds become accustomed to the device. The use of poisoned baits and the destruction of nesting habitat are discouraged because they are seldom effective and because of the potential negative effects on non-target species in adjacent aquatic environments.

DECISION-MAKING

IPM decision-making has traditionally depended heavily on economic threshold levels that are based on three factors (management cost, commodity price and damage coefficients) that are highly variable and thus have not been found useful under most conditions where economic and ecological considerations are of equal importance for stable production. Decision-making is therefore based on weighing potential management costs against potential losses. Costs and losses, however, can include not only economic costs and losses but can also have an impact on natural enemies, health and the environment while taking into consideration the general condition of the crop. Obviously, a crop under drought or flood stress is going to require a different decision from that for one under optimal conditions.

The first level of decision-making therefore begins with the first principle of IPM or IPPM, which is to create a healthy soil and crop through proper soil fertility management, healthy seed and appropriate varieties, strong seedling management, proper soil preparation, correct time of planting, etc. A robust healthy crop has fewer pests in most cases and can recover from pest damage. This principle applies throughout the cropping season and even beyond, when issues such as crop rotation, cover cropping and green manuring are taken into consideration. Latin American rice production, which has few problems relating to arthropods and many problems relating to weeds, depends primarily on many of these first-level methods (Weber and Paruda, 1994) while rice yields and profitability in Africa and Asia benefit greatly by ensuring proper growth conditions.

The second level of decision-making is based not on traditional pest-scouting but on whole-field observation including soil, water, plant, pests, natural enemies and weather patterns. Potential losses are weighed against potential management costs - a traditional aspect of farming that becomes refined with improved observation skills and ecological understanding gained during IPM training courses. The time required for observation and decision-making is typically short because most rice pests do not have clumped distributions and most fields are relatively small (because of water-levelling technical issues) and uniform.

IPM training programmes for extension officers and farmers therefore focus on economically sound decision-making taking into account ecological and toxicological factors. Pesticide application may occasionally be one decision outcome, especially in higher fertilizer input systems, but is by no means the only possibility. Decisions taken may include soil, water and plant management actions such as increasing or reducing fertilizer usage in response to pest or weather damage. Decision processes take into account the impact of activities on natural enemies and plant compensation factors (e.g. will a spray remove natural enemies and actually cause greater pest populations?). Selecting the least toxic solutions in order to prevent health problems is also a factor in decision-making. Of course, because there is a possibility of pesticide application, topics related to reducing exposure during mixing and spraying, proper protective gear, equipment maintenance and calibration, and storage and disposal of pesticides are also included - even in organic agriculture, which also uses organic pest-management methods that include spraying (e.g. Bacillus thuringiensis, viruses and soaps).

It is generally necessary to include topics related to growing a healthy crop alongside traditional pest-management topics because agronomic considerations and practices such as varieties, mulching, water status and plant nutrient status have an impact on pest status. Furthermore, topics such as basic ecological processes - including soil regeneration, predation, parasitism and pollination - are central to the training process using field studies and hands-on learning methods.

DOES IPM WORK FOR RICE FARMERS?

Although there is a great amount of grey literature (for examples, see www.communityipm.org ) related to the impact of rice IPM on farmers, there are few published data. This lack reflects the financial and technical difficulties of carrying out such studies. Longitudinal studies (e.g. data collected over time) in agriculture are notoriously difficult owing to seasonal changes. Horizontal studies (e.g. comparisons across sites) are also difficult because it is impossible to find an identical IPM and non-IPM control in view of the diversity of interactions and social variation. Both types of studies are costly, especially considering the limitations on methodology just noted. Nevertheless, there are strong indicators of IPM benefits for rice farmers.

The first, and perhaps strongest, indicator is the lack, or greatly reduced incidence, of the brown planthopper. Wide area outbreaks accompanied by massive losses have rarely been experienced in the 15 years during which IPM programmes have been implemented in both policy and field training. In most cases, changes in policy involved the removal of pesticide subsidies, restrictions on outbreak-causing pesticides, investment in biological research programmes or educational programmes for decision-makers, extension workers and farmers. These policy changes usually came about as a result of field trials, either by research bodies or within farmer training activities. The FAO Inter-Country Programme for Rice IPM in South and Southeast Asia, headed by Peter Kenmore, followed this method of bringing policy-makers into contact with researchers and farmers who could explain from their own experience the ecological basis of farming under IPM methods. In Indonesia, the banning by former President Suharto in 1987 of 57 pesticides and removal of pesticide subsidies known to cause brown planthopper outbreaks was the result of bringing cabinet officials into a biological dialogue with top Indonesian scientists, IRRI scientists and farmer groups who had demonstrated the outbreak effects of the pesticides and their ability to achieve high rice yields profitably without the use of these pesticides (FAO, in press).

The second indication comes from the grey literature using case studies (FAO, 1998). Table 1 gives a typical result found across hundreds of communities surveyed in rice IPM programmes in Indonesia. The data show the changes in practices, especially the very common result in IPM of investing less in pesticides and more in fertilizers (including phosphorus and potassium) that follow on from the first principle of IPM outlined above (i.e. growing a healthy crop). Other large-scale studies provide similar data, although a recent study in Viet Nam notes a growth in the use of fungicides (FAO, 2000). The authors have noted that with higher levels of fertilizers (as would be found in Viet Nam) this increase in fungicide is predictable but worrisome because of its potential impact on the ecosystem. These data also reveal the multidisciplinary aspect of rice IPM in that farmers are encouraged to look beyond the pest complex into the multiple parameters for achieving a stable and profitable high-yielding crop.

GETTING IPM INTO THE HANDS OF FARMERS

"IPM is not for farmers but is by farmers" is often noted in IPM programmes. Getting IPM into the hands of farmers, however, is not always easy. Several methods have been developed with various levels of information and completeness. Most agricultural extension services now recognize the importance of natural enemies and are quick to point out the need to conserve them - although their promotion of various insecticides, fungicides and herbicides is often at odds with their awareness of natural enemies. Work by Heong and others from the Rice IPM Network (Heong et al., 1998; Heong and Escalada, 1999; Huan et al., 1999) includes the development of a number of interesting radio messages to spread the word on a large scale that early spraying of insecticides during the first 40 days of the crop is not necessary and in fact increases the risk of higher pest populations later in the crop season. In many cases these radio messages are accompanied by field-based plant compensation participatory research groups (Heong and Escalada, 1998). This programme has been very effective in increasing awareness of the negative impact of insecticides on natural enemies and the role of plant compensation in recovering without yield loss from early season pest damage, and has resulted in a reduction of early insecticide spraying where it has been implemented. Furthermore, these mass media programmes are complementary to community study groups necessary for learning ecological principles or community organizing for community-level pests.

Study groups of various types, focusing on the overall production and pest management, are now common in many rice systems, including organic agriculture, rice-duck groups, Australian rice farmer associations and many others. The FAO Community IPM Programme in Asia (Matteson, Gallagher and Kenmore, 1994) has promoted study groups called "farmer field schools" under which structured learning exercises in fields ("schools without walls") are used to study both ecosystem-level dynamics transferable to other crops (e.g. predation, parasitism and plant compensation) and specific rice IPM methods (see Box 4). More than 1.5 million farmers have graduated from field schools of one or more season's duration in Asia over the past ten years. The field schools have proved themselves very cost-effective as an extension methodology, and many continue in some type of participatory research programme (Ooi et al., 2001). In Mali, work under the Global IPM Facility and the Office du Niger with farmer groups using few pesticides has expanded the IPM methods to arrive at "IPPM" (integrated production and pest management) based on experiences from Zimbabwe and focusing on improved production methods balanced with pest management practices².

Community-based study groups, study circles, field schools and other approaches are now being integrated with sustainable funding approaches through greater reliance on community-based organizations such as IPM clubs, water-user groups, women's organizations and local farmers' unions. With the large-scale World Bank-type extension programmes (T&V) generally being phased out in most countries, it will be necessary for local communities to organize themselves in such a way that they can increasingly cover the costs of expert inputs. Primary school programmes on IPM are also emerging in Cambodia, Indonesia, Thailand and other countries as part of an environmental education curriculum related to the Asian rice-culture (Jatiket, personal communication). Programmes such as farmer field schools in many countries or Landcare (www.landcare.gov.au) in Australia and the Philippines are providing innovative models in the direction of community-based and supported study and action.

FUTURE NEEDS

There is still much room for improving the state of IPM in rice. Indeed, the ecological view of rice presented here must achieve a greater consensus among international and national scientists and policy-makers in order to promote more widely the economic and ecosystems benefits already being realized by some - but not all - rice farmers. A new CD-ROM produced by IRRI is indeed beginning to bring together basic information in a widely accessible format.³ In addition to the pests and other problems listed above, post-harvest pests are still a problem and greater research on non-toxic management methods is warranted. Environmentally friendly methods of control for all types of pests, especially weeds and fungal pathogens, are required to reduce their ever-increasing environmental impact as intensification of production continues. Already, some countries are calling for major changes. For example, the Republic of Korea has banned pesticide use in Seoul's watersheds and is promoting massive organic agricultural investments to ensure clean water and high levels of production. Other communities are moving away from grain maximization to diversification such as rice-fish-vegetable culture as a response to food and nutritional security - and this trend is expected to increase as demand for more profitable non-grain products grows and nitrogen levels are necessarily reduced to lower the environmental impact and incidence of expensive-to-control fungal pathogens.

IPM programme development is required in more countries. These programmes should ensure that educational systems (formal and non-formal) respond adequately to the future needs of reducing the impact of agriculture on the environment while improving yields - IPM/IPPM is, of course, a major aspect of such education. Research in connection with field programmes is needed to involve farmers in finding solutions to problems faced in IPM/IPPM systems as economic changes alter the way rice is cultivated.

In the area of IPM policy, significant improvements are needed with regard to removing toxic compounds and subsidies for toxic pesticides. There is a need for subsidizing the commercialization of locally produced products such as pheromones, attractants, natural enemies, pest-exclusion netting (for birds), high-quality seed, improved disease resistance and balanced soil fertility products (see Box 5). High foreign exchange costs for imported pesticides and increasing consumer awareness of the social costs arising from pesticides and chemical fertilizers can be expected to drive rice IPM systems towards lower impact and local production of environmentally friendly pest management, as is already the trend in most Organisation for Economic and Co-operative Development (OECD) countries.

Acknowledgements

The authors would like to thank Dr K.L. Heong, IRRI, for assistance in providing papers and previews of the Rice IPM CD-ROM and to express their appreciation to Ricardo Labrada (FAO), N.H. Huan (Plant Protection Department of Viet Nam), Henk van den Berg (Community IPM/FAO), Clive Elliott (FAO) and Chan Paloeun (Cambodian Agricultural Research and Development Institute) for their valuable comments and inputs to this paper.

REFERENCES

Barrion, A.T. & Litsinger, J.A. 1994. Taxonomy of rice insect pests and their arthropod parasites and predators, In E.A. Heinrichs, ed. Biology and management of rice insects. pp. 13-362. New Delhi, Wiley Eastern. 779 pp.

Buckle, A.P. & Smith, R.H. 1994. Rodent pests and their control. Wallingford, UK, CAB International.

Chancellor, T.C.B., Tiongco, E.R., Holt, J., Villareal, S. & Teng, P.S. 1999. The influence of varietal resistance and synchrony on tungro incidence in irrigated rice ecosystem in the Philippines. In T.C.B. Chancellor, O. Azzam & K.L. Heong, eds. Rice tungro disease management. pp. 121-127. Proceedings of the International Workshop on Tungro Disease Management, 9-11 November 1998, Los Baños, Laguna, the Philippines, International Rice Research Institute. 166 pp.

Claridge, M.F., Den Hollander, J. & Morgan, J.C. 1982. Variation within and between populations of the brown planthopper, Nilaparvata lugens (Stal). In W.J. Knight, N.C. Pant, T.S. Robertson & M.R. Wilson, eds. 1st International Workshop on Leafhoppers and Planthoppers of Economic Importance. pp. 36-318. London, Commonwealth Institute of Entomology.

Cohen, J.E., Schoenly, K., Heong, K.L., Justo, H., Arida, G., Barrion, A.T. & Litsinger, J.A. 1994. A food web approach to evaluation the effect of insecticide spraying on insect pest population dynamics in a Philippine irrigated rice ecosystem. J. App. Ecol., 31: 747-763.

Dale, D. 1994. Insect pests of the rice plant - their biology and ecology. In E.A. Heinrichs, ed. Biology and management of rice insects. pp. 363-486. New Delhi, Wiley Eastern. 779 pp.

Du, P.V., Lan, N.T.P., Kim, P.V., Oanh, P.H., Chau, N.V. & Chien, H.V. 2001. Sheath blight management with antagonistic bacteria in the Mekong Delta. In T.W. Mew, E. Borromeo & B. Hardy, eds. Exploiting biodiversity for sustainable pest management. Proceedings of the Impact Symposium on Exploiting Biodiversity for Sustainable Pest Management, 21-23 August 2000, Kunming, China. Los Baños, Laguna, the Philippines, International Rice Research Institute. 241 pp.

FAO. In press. The history of IPM in Asia, by K. Eveleens. Rome.

FAO. 1988. Integrated management of rice rats in Indonesia, by A.P. Buckle. FAO Plant Protection Bulletin No. 36. pp. 111-118.

FAO. 1998. Community IPM: six cases from Indonesia. FAO Technical Assistance: Indonesian National IPM Programme, Rome. 258 pp.

FAO. 2000. The impact of IPM farmer field schools on farmers' cultivation practices in their own fields. A report submitted to the FAO Intercountry Programme for Community IPM in Asia, by J. Pincus, Economics Department, University of London. 175 pp.

Gallagher, K.D. 1988. Effects of host plant resistance on the microevolution of the rice brown planthopper, Nilaparvata lugens (Stal) (Homoptera: Delphacidae). University of California, Berkeley, California, USA. (Ph.D. dissertation)

Gallagher, K.D., Kenmore, P.E. & Sogawa, K. 1994. Judicious use of insecticides deter planthopper outbreaks and extend the life of resistant varieties in Southeast Asian rice. In R.F. Denno & T.J. Perfect, eds. Planthoppers; their ecology and management. pp. 599-614. New York, USA, Chapman & Hall. 799 pp.

Graf, B., Lamb, R., Heong, K.L. & Fabellar, L. 1992. A simulation model for the populations dynamic of rice leaf folders (Lepidoptera) and their interactions with rice. J. Appl. Ecol., 29: 558-570.

Halwart, M. 1994. The golden apple snail, Pomacea canaliculata, in Asian rice farming systems: present impact and future threat. Int. J. Pest Management, 40(2): 199-206.

Heong, K.L. & Escalada, M.M. 1998. Changing rice farmers' pest management practices through participation in a small-scale experiment. Int. J. Pest Management, 44: 191-197.

Heong, K.L. & Escalada, M.M. 1999. Quantifying rice farmers' pest management decisions: beliefs and subjective norms in stem borer control. Crop Protection, 18: 315-322.

Heong, K.L., Escalada, M.M., Huan, N.H. & Mai, V. 1998. Use of communication media in changing rice farmers' pest management in the Mekong Delta, Vietnam. Crop Protection, 17(5): 413-425.

Hill, R.D. 1977. Rice in Malaya: a study in historical geography. Kuala Lumpur, Oxford University Press. 213 pp.

Huan, N.H., Mai, V., Escalada, M.M. & Heong, K.L. 1999. Changes in rice farmers' pest management in the Mekong Delta, Vietnam. Crop Protection, 18: 557-563.

IRRI. 1979. Brown planthopper: threat to rice production in Asia. Los Baños, Laguna, the Philippines, International Rice Research Institute.

Kenmore, P.E., Carino, F.O., Perez, C.A., Dyke, V.A. & Gutierrez, A.P. 1984. Population regulation of the rice brown planthopper (Nilaparvata lugens Stal) within rice fields in the Philippines. J. Plant Prot. Tropics, 1: 19-37.

Kenmore, P.E. 1996. Integrated pest management in rice. In G.J. Persley, ed. Biotechnology and integrated pest management. pp. 76-97. Wallingford, UK, CAB International. 475 pp.

Leung, L., Singleton, K.-P. & Sudarmaji, G.R. 1999. Ecologically-based populations management of the rice-field rat in Indonesia. In G.R. Singleton, L. Hinds, L. Herwig & Z. Zhang, eds. Ecologically-based rodent management. pp. 305-38. Canberra, Australian Center for International Agricultural Research.

Loevinsohn, M.E. 1994. Rice pests and agricultural environments. In E.A. Heinrichs, ed. Biology and management of rice insects. pp. 487-515. New Delhi, Wiley Eastern. 779 pp.

Matteson, P.C., Gallagher, K.D. & Kenmore, P.E. 1994. Extension of integrated pest management for planthoppers in Asian irrigated rice: Empowering the user. In R.F. Denno & T.J. Perfect, eds. Planthoppers: their ecology and management. pp. 656-685. New York, USA, Chapman & Hall.

Matteson, P.C. 2000. Insect pest management in tropical Asian irrigated rice. Ann. Rev. Entomol., 45: 549-574.

Morrill, W.L. 1997. Feeding behavior of Leptocorisa oratorius (F.) in rice. Recent Research Developments in Entomology, 1: 11-14.

Morse, S. & Buhler, W. 1997. Integrated pest management: ideals and realities in developing countries. Boulder, Colorado, USA, Lynne Rienner. 171 pp.

Murakami, O., Priyono, J. & Triastiani, H. 1991. Population management of the rice rat in Indonesia. In G.R. Quick, ed. Rodents and rice. pp. 49-60. Los Baños, Laguna, the Philippines, International Rice Research Institute.

Ooi, P.A.C. 1988. Ecology and surveillance of Nilaparvata lugens (Stal) - implications for its management in Malaysia. University of Malaya. 275 pp. (Ph.D. dissertation)

Ooi, P.A.C. & Shepard, B.M. 1994. Predators and parasitoids of rice insect pests. In E.A. Heinrichs, ed. Biology and management of rice insects. pp. 585-612. New Delhi, Wiley Eastern. 779 pp.

Ooi, P.A.C., Warsiyah, Nanang Budiyanto & Nguyen Van Son. 2001. Farmer scientists in IPM: a case of technology diffusion. In T.W. Mew, E. Borromeo & B. Hardy, eds. Exploiting biodiversity for sustainable pest management. pp. 207-215. Proceedings of the Impact Symposium on Exploiting Biodiversity for Sustainable Pest Management, 21-23 August 2000, Kunming, China. Los Baños, Laguna, the Philippines, International Rice Research Institute. 241 pp.

Ponting, C. 1991. A green history of the world: the environment and the collapse of great civilizations. London, Penguin Books. 432 pp.

Rombach, M.C. & Gallagher, K.D. 1994. The brown planthopper: promises, problems and prospects. In E.A. Heinrichs, ed. Biology and management of rice insects. pp. 693-711. New Delhi, Wiley Eastern. 779 pp.

Rubia, E.G., Heong, K.L., Zalucki, M., Gonzales, B. & Norton, G.A. 1996. Mechanisms of compensation of rice plants to yellow stem borer Scirpophaga incertulas (Walker) injury. Crop Protection, 15: 335-340.

Settle, W.H., Ariawan, H., Astuti, E., Cahyana, W., Hakim, A.L., Hindayana, D., Lestari, A.S., Pajarningsih & Sartanto. 1996. Managing tropical rice pests through conservation of generalist natural enemies and alternative prey. Ecology, 77(7): 1975-1988.

Shepard, B.M., Barrion, A.T. & Litsinger, J.A. 1987. Helpful insects, spiders and pathogens. Los Baños, the Philippines, International Rice Research Institute. 136 pp.

Shepard, B.M. & Ooi, P.A.C. 1991. Techniques for evaluating predators and parasitoids in rice. In E.A. Heinrichs & T.A. Miller, eds. Rice insects: management strategies. pp. 197-214. New York, USA, Springer-Verlag. 347 pp.

Shepard, B.M., Justo, H.D., Rubia, E.G. & Estano, D.B. 1990. Response of the rice plant to damage by the rice whorl maggot, Hydriella philippina Ferino (Diptera: Ephydridae). J. Plant Protect. Tropics, 7: 173-177.

Sogawa, K., Kilin, D. & Bhagiawati, A.H. 1984. Characterization of the brown planthopper population on IR42 in North Sumatra, Indonesia. Int. Rice Res. Newsl., 9(1): 25.

Teng, P.S. 1994. The epidemiological basis for blast management. In R.S. Zeigler, S.A. Leong & P.S. Teng, eds. Rice blast disease. pp. 409-433. Wallingford, UK, CAB International.

van den Berg, H. & Soehardi. 2000. The influence of the rice bug Leptocorisa oratorius on rice yield. J. Applied Ecology, 37(6): 959-970.

Way, M.J. & Heong, K.L. 1994. The role of biodiversity in the dynamics and management of insect pests of tropical irrigated rice - a review. Bull. Entomol. Res., 84: 567-587.

Way, M.O., Grigarick, A.A., Litsinger, J.A., Palis, F. & Pingali, P.L. 1991. Economic thresholds and injury levels for insect pests of rice. In E.A. Heinrichs & T.A. Miller, eds. Rice insects: management strategies. pp. 67-106. New York, USA, Springer-Verlag. 347 pp.

Weber, G. & Paruda, O. 1994. Development of an integrated pest management strategy for rice in Latin America. In E.A. Heinrichs, ed. Biology and management of rice insects. pp. 733-747. New Delhi, Wiley Eastern. 779 pp.

Xie, Guanlin, Yu, Xuefang & Ren, Ziaoping. 2001. Biological control agents against major pathogens of rice in the Yangtze Delta area of China. In T.W. Mew, E. Borromeo & B. Hardy, eds. Exploiting biodiversity for sustainable pest management. pp. 91-99. Proceedings of the Impact Symposium on Exploiting Biodiversity for Sustainable Pest Management, 21-23 August 2000, Kunming, China. Los Baños, Laguna, the Philippines, International Rice Research Institute. 241 pp.

Protection intégrée contre les ravageurs du riz

La protection intégrée (PI) contre les ravageurs du riz s'est développée dans de nombreux pays depuis le début des années 60. Pendant les années 80 et 90, il a été possible d'avoir accès à des renseignements importants, du point de vue écologique, sur les populations d'insectes, ce qui a permis de faire une plus large place à des méthodes écologiques de lutte contre les ravageurs. Pour ce qui est de la lutte contre les insectes, on reconnaît aujourd'hui que l'inondation des rizières provoque un processus de décomposition et la création d'une chaîne trophique aquatique favorisant l'apparition de grandes populations qui se nourrissent de détritus. On tient également compte de la capacité de la plupart des variétés de riz de compenser les détériorations subies. L'intensification de la riziculture favorise les maladies, toutefois il n'est pas possible d'envisager un retour à une agriculture moins intensive, à bas rendement, comme par le passé. Une compréhension des conditions écologiques qui favorisent l'apparition de maladies peut conduire à une intensification durable. Dans la culture du riz, les économies de main-d'oeuvre se traduisent souvent par des semis directs en place, et le riz rouge est la principale plante adventice. Il est donc important d'améliorer la lutte intégrée contre les adventices du riz et de l'envisager dans la perspective la plus large possible.

Dans le domaine de la PI contre les ravageurs, les décisions sont prises traditionnellement en fonction d'un seuil économique. Aujourd'hui, le premier stade décisionnel de la protection intégrée concerne l'obtention de cultures saines. Le deuxième stade repose sur l'ensemble des observations faites sur le terrain concernant notamment le sol, l'eau, les plantes, les ravageurs, les ennemis naturels et le climat. Les documents relatifs aux répercussions de la PI contre les ravageurs du riz sur les agriculteurs sont rares. Le premier indice, et peut-être le plus important, est l'absence ou la forte réduction de la présence des delphacides brunes du riz. Certaines études de cas ont montré que les agriculteurs qui pratiquent la protection intégrée contre les ravageurs investissent moins en pesticides mais davantage en engrais. Il est maintenant fréquent que des groupes d'étude soient organisés dans le cadre des programmes de protection intégrée contre les ravageurs. Le Programme FAO de protection intégrée contre les ravageurs en Asie, a encouragé la formation de groupes d'étude appelés «Ecoles pratiques d'agriculture» dans lesquels des exercices pédagogiques effectués sur le terrain sont utilisés pour étudier la dynamique des écosystèmes et les méthodes de protection intégrée contre les ravageurs pour le riz. Mais il reste encore beaucoup à faire dans ce domaine. On a encore besoin pour tous les types de ravageurs de méthodes de lutte, qui respectent l'environnement afin de diminuer l'impact sur le milieu. Des améliorations sont encore nécessaires pour supprimer les composantes toxiques et éliminer les subventions en faveur des pesticides toxiques. Il convient aussi de subventionner la commercialisation de produits locaux comme les phéromones, les pièges attractifs, les ennemis naturels, les filets de protection, les semences de qualité supérieure, ainsi que des produits dont la résistance aux maladies est améliorée et qui favorisent de manière équilibrée la fertilité des sols.

Manejo integrado de plagas en el arroz

El Manejo integrado de plagas en el arroz (MIP) se ha desarrollado en muchos países desde comienzos de los años sesenta. Durante los años ochenta y noventa, se empezó a disponer de importante información ecológica sobre las poblaciones de insectos que permitía aplicar un enfoque ecológico más sólido en el manejo de plagas. Para el control de los insectos, el MIP reconoce hoy que la inundación de los campos pone en marcha un proceso de descomposición y desarrollo de una red alimentaria acuática que da lugar a grandes poblaciones de insectos que se alimentan de detritos. Asimismo, considera la capacidad de la mayoría de las variedades de arroz para compensar los daños. La intensificación del arroz conduce a condiciones que favorecen las enfermedades. Sin embargo, no es cuestión de volver a la agricultura menos intensiva y de bajos rendimientos del pasado. Se puede lograr una intensificación sostenible gracias al conocimiento de las condiciones ecológicas asociadas con el brote de las enfermedades. En el cultivo del arroz donde el ahorro de mano de obra significa en muchos casos orientarse hacia la siembra directa, el pulgón de la raíz es la plaga principal. Por ello, es importante mejorar el MIP para las malas hierbas del arroz y considerarlo en los términos más amplios.

La adopción de decisiones con respecto al MIP ha solido depender decisivamente de los niveles del umbral económico. Hoy en día, el primer nivel de decisión en relación con el MIP es el de obtener un cultivo sano. El segundo nivel de decisión se basa en la observación completa del terreno, incluyendo el suelo, el agua, la planta, las plagas, los enemigos naturales y las condiciones meteorológicas. Existen pocos datos publicados sobre el impacto del MIP del arroz entre los agricultores. El primero, y quizás el principal indicador, es la falta o la gran disminución de la incidencia de la cicadela parda. Estudios de casos seleccionados mostraron que los agricultores que aplican el MIP invierten menos en plaguicidas, pero más en fertilizantes. Existen ahora grupos de estudio de varios tipos en los programas de MIP. El programa comunitario de MIP de la FAO en Asia ha promovido grupos de estudio llamados «Escuelas de campo de agricultores» en el ámbito de los cuales se utilizan ejercicios estructurados de aprendizaje sobre el terreno para estudiar la dinámica a nivel de ecosistema y métodos de aplicación del MIP en el arroz. Hay todavía mucho margen para mejorar la aplicación del MIP en el arroz. Se necesitan aún métodos ecológicos para combatir todos los tipos de plagas a fin de reducir el impacto ambiental. Hacen falta aún mejoras para eliminar compuestos tóxicos y las subvenciones a los plaguicidas tóxicos. Es necesario subvencionar la comercialización de productos locales como feromonas, atrayentes, enemigos naturales, redes protectoras contra las plagas, semillas de alta calidad, mayor resistencia a las enfermedades y productos equilibrados para fertilizar los suelos.

¹ Respectively, IPM Specialist, Global IPM Facility, FAO Plant Protection Service, Rome; Coordinator of Regional Cotton IPM Programme, FAO Regional Office, Bangkok; Head of Plant Protection Department, IRRI, Los Baños; Plant Protection Department, IRRI, Los Baños; and Coordinator, Global IPM Facility, FAO Plant Protection Service, Rome.

² See www.fao.org/globalipmfacility for country information.

³ For information, see www.irri.org/pubcat2000/newtitles.htm.