Insecticides disrupt IPM

Some agronomists say broad spectrum insecticides are not needed in Asian rice fields. Here Steffen Johnsen, Le Thi Thu Huong, Kim Thuy Ngoc and Trinh Dieu Thuy examine claims made by manufacturers that two products, fipronil and lambda-cyhalothrin, are compatible with integrated pest management (IPM) in Vietnam.

Asia is home to the most successful integrated pest management programmes in the world. The FAO-led Intercountry Rice IPM Programme and the national programmes collaborating within it, have trained hundreds of thousands of rice farmers in 'farmer field schools'. Here farmers learn through their own experience that insecticides are very rarely needed in pest management in rice. Broad spectrum insecticides do more harm than good, by disrupting the highly effective natural control of pests and by their negative effects on non-target organisms. 
    The International Rice Research Institute (IRRI) recommends that farmers do not use any insecticides for the first 40 days after sowing, and only apply insecticides later after careful assessment of the need.
    Through the concerted efforts of the National IPM Programmes in 14 Asian countries, the Intercountry Programme and IRRI, insecticide use in rice in many Asian countries has been declining. Devastating pest outbreaks, as a result of pesticide overuse that eliminated natural predators, have been avoided leading to more stable yields and this way food security for millions of Asians has improved (see box for example).
    Recently, two relatively new insecticides fipronil and lambda-cyhalothrin  have been introduced and strongly promoted (by Rhône-Poulenc and Zeneca, respectively) for use in rice (and many other crops). Rhône-Poulenc claims that fipronil has "minimal effects on beneficial insects and non-target organisms" and Zeneca says that "lambda-cyhalothrin can be incorporated safely into rice IPM strategies with a good understanding of the crop environment and careful use­-the widely held belief that the use of broad-spectrum insecticides is always incompatible with IPM strategies is untrue."
    A DANIDA development project, sponsored by the Danish Government, about environmental impact of current IPM Programmes in the region, investigated the effects of these insecticides on natural pest control and the rice ecosystem  in farmers' rice fields in Vietnam. All animals inside a hollow sheet-metal cylinder were sampled, identified and sorted into guilds (functional groups), such as 'leaf-eating pests', 'stem damaging pests', 'predators on leaves and stem', 'predators on water surface', 'predators in the water', 'parasitoids', 'detritivores' (free-swimming mosquito larvae), 'neutrals' (for example, some adult flies), zooplankton and annelid worms and mosquito-larvae, living in the mud of the paddy. A total of 20 such samples were taken once per field per week.

Lambda-cyhalothrin - key results
In fields which were untreated with pesticides, natural enemies, such as spiders, always out-number pests by at least two to one and often by 10 to one or even more. This  has been shown by work in Indonesia by Bill Settle, Peter Ooi and co-workers and was confirmed for Vietnam in this study.
    It is obvious that under such conditions pests have no chance of reaching damaging levels. The high numbers of natural enemies are mainly sustained by large populations of midge flies that build up emerging from  larvae living in the water and the mud, and by other insects which are neither pests nor beneficials, but rather 'neutrals'.

Pests out-number predators
In the marketing of lambda-cyhalothrin Zeneca quotes a study performed at the Philippine Rice Research Institute (PhilRice) which documents only transitory and small negative effects of their product on natural pest control. In this study, the ratio between predators and pests was, however about 0.5 that is, the pests outnumbered their predators by two to one. This, of course, is an unstable situation, where pest outbreaks may occur. The Vietnam study obtained results similar to those of Zeneca only when a field had previously received heavy pesticide use. When  lambda-cyhalothrin was used in fields after  there had not been intensive pesticide use, the chemical always had very strong, negative effects on predators which landed on the crop and on the water surface. It could also reduce the ratio between predators and pests to near the 0.5 found in Zeneca's study.
    Figure 1 illustrates these facts.  The vertical axis indicates the ratio between predators and pests. The shaded band shows how this ratio, in fields untreated with pesticides, is always in favour of the predators or 'defenders' of the crop. The line of stars at the bottom of the graph indicate the ratio of c. 0.5, which was the basis for Zeneca's study, and the two lines show our results, from fields undamaged by prior pesticide use. The solid line shows a healthy proportion between predators and pests all through the season, ranging between about two and up to 15 predators per pest, in the untreated fields. The downturned arrows indicate the three applications of lambda-cyhalothrin (Karate 2.5 EC) applied experimentally. The broken line indicates that the very first application dramatically and almost instantaneously reduces the ratio (by killing the predators) and that subsequent applications continues the job. These fields were in Ha Bac province in northern Vietnam.
    In a field in southern Vietnam this chemical apparently induced problems with planthoppers which are the most feared rice pests.
    This study indicates that lambda-cyhalothrin- like any other broad-spectrum insecticide-is damaging to IPM efforts, and possibly to sustainable high yields in rice production. The results directly contradict the claims of the company which are the basis for its marketing of lambda-cyhalothrin as compatible with IPM. It can also adversely affect farmers' incomes because of the high cost of this product.

Fipronil - conflicting results
Under conditions of high pest and predator densities in southern Vietnam, fipronil was also found to be very detrimental to natural pest control, and to reduce the predator/pest ratio in a similar fashion to lambda-cyhalothrin. Here however, the action was much slower and more persistent than that of lambda-cyhalothrin.

Important market
Early in 1998, Rhône-Poulenc announced that it had opened a fipronil formulation plant in Bienhoa, Dong Nai, Vietnam. According to the trade journal Agrow, the company has invested US$7.3 million in developing the site. Fipronil is Rhône-Poulenc's biggest-selling product in Vietnam, and the plant will help to consolidate this position.

Is fipronil a fertiliser?
In the marketing of fipronil the company stresses that this pesticide often improves yield in paddy rice, especially on acidic soils. We made the same observation in one location, while in another there was no yield increase. In the case where the yield was increased and fipronil had no effect on pests the gain was so dramatic-almost 50%- that it called for further investigations. How could this be?
    The soil in the treated fields was analysed and compared to that of the control plots, and  slightly elevated levels of several micro-nutrients were found. The pesticide in the granular formulation (Regent 0.3G) was also analysed and  contents of the same micro-nutrients were found. The amounts were large enough to compare with those applied when using a commercial foliar fertiliser, as this formulation of fipronil is applied at the rate of 10kg per hectare. When presented with draft reports of these findings the company vigorously denied adding micronutrients to their formulations. They suggested that the nutrients may enter the formulation through 'inerts' (in this case the volcanic sand used in the granular formulation), and promised to investigate the matter. 
    It would be of benefit if the issue is also addressed by neutral scientists with access to proper equipment and procedures for analysis. In Vietnam this is difficult, and the results of this study are therefore only suggestive. The study compared applications of Regent 0.3G with a foliar fertiliser in a well fertilized and a poorly fertilised farmer's field, immediately next to the field where, the season before, the strong positive effect of Regent 0.3G on yield  had been seen. In the well fertilised field neither Regent 0.3G nor the foliar fertilizer had any effect on yield, while in the poorly fertilised one the foliar fertiliser had the same effect as Regent 0.3G: both increased yield by about 15% (see Figure 2). This finding suggests that micronutrient-contents in Regent 0.3G does play a role in the yield-increasing effect. It also indicates that this effect can be substituted by using a foliar fertiliser, which is harmless to the environment.

Aquatic effects
Fipronil is not harmless, but has some disturbing and long-lasting effects on several groups of aquatic organisms, mainly crustaceans. There is evidence that farmers often use it for killing crabs, which destroy the rice field bunds (walls between fields) by digging into them. This study found the chemical to be detrimental to populations of zoo-plankton and aquatic chironomid flies, which can lead to a decrease in predator numbers. Thus important food for predatory fish in the ecosystem disappear. The effect further calls for extreme caution towards use of fipronil in paddies, because their floodwater is more or less continuous with water used in shrimp farming-in many areas it can flow into natural mangroves which harbour breeding populations of commercial shrimps or crabs.

Conclusion
The authors hope this presentation may have provided useful ideas for other work on ecological effects of insecticides.
    Existing reports, which claim no or negligible effects should be carefully reviewed with respect to the pesticide-use history of the fields which were used.
    It is very commonly seen that in yield-comparisons between pesticide-treated and untreated plots, the former yield more, even if they often have higher pest densities. The authors would like to suggest that such trials also include a foliar-fertiliser-treatment, to mimic possible contents of nutrients in the pesticide-formulations.

Full reports of the project and further information may be obtained from Steffen Johnsen the advisor who headed it: stefulla@post8.tele.dk.

Rice problems - a historical lesson
National rice production programmes in many developing countries have suffered in the past from the almost complete reliance on pesticides as a control tactic. Over-use to control rice insects has often resulted in ecological imbalances with consequent severe effects on rice production. High levels of insecticide use have resulted in the build-up of insect-resistant strains and insect resurgence.
    Insecticides are often more toxic to the natural enemies of planthoppers such as the spider Lycosa pseudoannulata, the mirid bug Cyrtorbinus lividipennis and the veliid bug Microvelia airolineata.
    Resurgence of rice planthoppers has had a significant effect on rice production in Asia. Resurgence-a significant increase in planthopper populations after insecticide application-was especially common in the 1970s and 1980s, occurring in every rice producing country in tropical Asia.

Denno. R.F., and Perfect, T.J, Planthoppers their ecology and management, Chapman & Hall, London 1997. 

Corporate response
In 1994, Zeneca conducted a large scale field trial with PhilRice to investigate effects of lambda-cyhalothrin on natural enemy populations in rice. Applications resulted in a limited reduction in the total number of natural enemies immediately after treatment, but numbers rapidly recovered within 15 days post-treatment. In all lambda-cyhalothrin treated plots the predator to pest ratio's remained similar to the control.

Based on findings from this and other studies. Zeneca promotes lambda-cyhalothrin for control of rice pests on a needs-based use, spraying only when pests exceed economic threshold levels. Following this use pattern the grower will achieve good economic pest control, populations of natural enemies will not be disrupted for long periods of time and no effect on fish production will occur in rice-fish-culture.

These are clearly different conclusions to those made by research scientists in Vietnam, particularly in the differences between predator/pest ratios. There may be a number of reasons why results differ, in particular, a lower pest pressure will result in higher predator to pest ratios and lower economic benefits from applications. More detailed population dynamics is being conducted by Zeneca to assess any differences in community structure between treatments.

Rhône Poulenc disagrees with the conclusion in the DANIDA study that micronutrients, which may naturally occur in the volcanic sand carrier, are responsible for the yield increases. Other fipronil formulations which do not contain any volcanic sand give similar yields. This suggests that the increased yields are due solely to the unique properties of the insecticide, and not the micronutrients.

Rhône Poulenc concludes, after seeing the DANIDA study, that Regent has no adverse effects on zooplankton or aquatic insects, except mosquito larvae. Crustaceans are generally more sensitive than fish, and it is stated on the label in Vietnam that Regent is harmful to crabs.

A position paper is available from Rhône Poulenc.

[This article first appeared in Pesticides News No. 39, March 1998, pages 12-13]