Natural Enemies and Farmer Decision-making

Understanding farmers' decision-making processes is important if outside interventions are to enable farmers to make more informed pest management decisions, and especially if interventions are to result in farmers deciding to adopt IPM practices which are better for farm family welfare and for the environment¹. In order to gain a better understanding of farmers' decision-making in pest management, fieldwork was carried out in 1999-2000 in smallholder vegetable systems in Kenya and cotton systems in India, both characterized by relatively high levels of external inputs, increasing production costs, and pesticide misuse problems. A further objective was to explore the impact of training interventions on farmers' decision-making and the study therefore compared trained and untrained farmers at three sites where IPM training projects were underway or had recently taken place. These were:

• the Insecticide Resistance Management (IRM) project of the Natural Resources Institute and the Central Institute for Cotton Research (CICR) in Maharashtra State, India
• the Farmer Field School (FFS) cotton IPM project run by Voice Trust and Agriculture Man Ecology (AME) in Tamil Nadu State, India
• the Farmer Field School vegetable IPM project managed by CABI Bioscience in collaboration with Kenyan research, extension and NGOs in Central Province, Kenya

Fieldwork was done by a multidisciplinary team in collaboration with training project staff and involved semi-structured individual interviews with men and women farmers, and farmer group analysis and discussion using partial participatory farm management budgets and causal diagrams. Details of methodology are given in the project report². The use of discovery-learning exercises in FFS programmes to help farmers appreciate the role of natural enemies forms a key pillar of the IPM training curriculum³ and the effectiveness of such training in encouraging the conservation of natural enemies has been reviewed⁴. Farmers' awareness of natural enemies was therefore studied in this fieldwork, with attention to how such knowledge might be used in pest management decision-making. Farmers were also asked how they would deal with a hypothetical example of an unfamiliar insect appearing in their crop, in order to explore how training and other information sources influence subsequent problem analysis and decision-making.

Putting Training into Practice

Table 1 summarizes the awareness of natural enemies and recognition capability among the different trained and untrained farmer groups interviewed and their use of natural enemies in decision-making. Trained farmers were consistently more aware of natural control mechanisms than their untrained counterparts and could identify several key groups of important arthropod predators and parasitoids, as a direct result of the training activities. In terms of their use in decision-making, FFS farmers appeared to make much more active use of natural enemy incidence and levels as decision tools than did farmers from the IRM programme.

IRM programme staff explained that previously many cotton farmers would spray indiscriminately against any insects in their fields, including beneficials, and the training had helped them to distinguish between good and bad insects. Decision-making on the need for pesticide applications and timing and product selection by IRM trained farmers was mainly based on scouting for specific pests, observation of spray thresholds (e.g. Helicoverpa armigera eggs present on 10 out of 20 plants sampled) and project recommendations. These recommendations were validated with farmers through demonstrations in their field, evaluating particular products for specific pests, pest stages or crop growth stages. Untrained farmers in the same area relied mainly on pesticide dealers for advice, often spraying insecticides on a calendar basis every 8-15 days. They tended to select products such as monocrotophos which they perceived as 'powerful', i.e. fast-acting, or to use up chemicals left over from earlier applications. Untrained cotton farmers in Tamil Nadu also made very similar decisions.

Table 1. Farmers' knowledge and use of natural enemies (NE) in decision-making

Farmer group	Knowledge of NE	Use of NE in decision-making
IRM cotton farmers, Maharashtra, India	Learnt about biological control as part of their training and aware of insect natural enemies. Able to recognise certain groups (ladybirds, lacewings).	No longer spray against "any insect" but otherwise NE levels did not appear to play a significant role in their decision-making process.
Untrained cotton farmers, Maharashtra	Low awareness of natural enemies, only 2/6 interviewed had heard of "pests that eat other pests."	None.
FFS cotton farmers, Tamil Nadu, India	Learnt about biological control during training and all aware of natural enemies. Could recognise a range (ladybirds, hoverflies, lacewings, spiders, egg and larval parasitoids).	NE population levels, in both main crop and intercrop rows, important in deciding whether or not to spray. Final decision made on basis of pest and natural enemy levels in each field.
Untrained cotton farmers, Tamil Nadu	Concept not totally new but they had either not observed any in their fields, or said they would not be able to recognise them.	None.
FFS vegetable farmers, Kenya	Acquired a detailed understanding and appreciation of natural enemies via training (lacewings, hoverflies and ladybirds (larvae and adults) parasitic wasps, chameleons). Farmers invented their own names for natural enemies that did not have names in the local language.	Pest control decisions on action and methods partly made with respect to NE levels via AESA1. Avoided spraying synthetic insecticides to conserve natural enemies. Pegging rather than ash preferred for cutworm control as it "leaves pests alive for natural enemies to eat."
Untrained vegetable farmers, Kenya	Aware that natural enemy insect species existed, but said they would not be able to identify them.	None. One respondent aware that chameleons were beneficial in coffee but did not want them in her vegetable plot in case they got diseased and contaminated the crop.
1 Agro-ecosystem analysis.

In contrast, cotton FFS graduates in Tamil Nadu observed natural enemies and used them in deciding whether the balance between pests and beneficials merited any control action, in addition to making augmentative releases of Trichogramma spp. for bollworm control. They also used pheromone and light traps to monitor adult.bollworm populations and time parasitoid releases, and erected bird perches to encourage predation. Farmers recounted how they used regular field observation and agro-ecosystem analysis (AESA), rather than specific thresholds, to make decisions, building on their group experience gained during and after training. All FFS farmers we met grew cotton as part of an intercropping system, with various combinations of pigeonpea, cowpea, maize, sunflower and castor. The majority viewed intercropping as a valuable pest management technique by maintaining or increasing natural enemy numbers, although one or two farmers expressed concern that pests might 'jump' from cowpea to the cotton plants. Some FFS farmers explained that insecticides exacerbate bollworm infestations by killing off natural enemies, while others were not sure why insecticides failed to control bollworm when they quickly killed other pests.

In Kenya, untrained vegetable farmers relied on extension services and other farmers for their pest management infor-mation. They applied insecticides on a preventative basis for aphids in kale and beans, either by calendar or crop stage. Preventative weekly sprays of contact fungicides were applied for late blight (Phytophthora infestans) in tomato, plus curative products if preventative control broke down. These farmers mainly used synthetic insecticides but one or two had used home-made botanical preparations or wood ash. Inexperienced vegetable farmers admitted that they watched to see what their neighbours were doing and were not confident in making their own decisions. Kenyan FFS graduates growing vegetables also relied on preventative and curative fungicide application for blight, in the absence of effective, proven alternatives for this key production constraint. However, several also used diluted milk solution to delay the onset of blight, a practice which they had learnt via farmer exchange visits during FFS training, with variable success. For other disease control, FFS and untrained farmers alike would rogue out wilt-infected plants when spotted in their fields.

Where Kenyan FFS graduates differed from their untrained counterparts was in the use of a much wider range of pest and disease management options, careful field observation and specific decision-making tools. They used home-made chilli extracts as their first choice against a variety of insect pests, a traditional practice which they had also learnt via farmer exchange visits. Farmers explained that this choice was influenced by their desire to avoid the use of synthetic pesticides where possible and conserve natural enemies but it also had the advantage of being much cheaper and not requiring observance of post-harvest intervals. They demonstrated knowledge of root-knot nematode Meloidogyne spp. and burned plant trash on nursery beds to prevent this and other soil-borne problems. For insect control they also used ash and they pegged thin sticks against seedling stems to prevent cutworms (Agrotis spp.) encircling and chewing through the stalk. This was another farmer practice acquired via exchange visits and tested during training and FFS farmers opted for pegging over ash as it "left pests alive for natural enemies to eat".

Experimenting and Adapting

Farmers' responses to the hypothetical question on what they would do on encountering an unfamiliar insect in their crop were varied, yet indicated that the majority of trained farmers would not spray immediately on sight but seek advice or conduct their own experiments (Table 2). Regular field observation is key to effective decision-making in pest management and all the training programmes studied had succeeded in persuading farmers of the need to check what is happening in the field before deciding what action to take. Most of the trained farmers appeared to be much more confident in their pest management capability than the untrained farmers. The FFS farmers, especially in Kenya, described how they now relied more on their own knowledge or on their group to solve problems than they had before. FFS training certainly appeared to have provided farmers with new decision tools and useful knowledge on pest and natural enemy biology to assist decision-making. The value of these tools was demonstrated in the Kenyan farmers' answers on how they would deal with an unfamiliar insect: they would catch it, place it in a jar and study its feeding behaviour to find out whether it was a pest or beneficial. This experiment is called an 'insect zoo' and is practised as part of the FFS curriculum³. These decision tools and new agro-ecological understanding appeared to have enhanced farmers' analytical capacity, via an ordered framework for observation and decision-making and more criteria for selection of pest control methods. Trained farmers in this study, from both FFS and IRM programmes, also stated how much they had enjoyed learning about pests and natural enemies, a finding common to most evaluations of discovery-learning based IPM training⁵. The discussions also showed that trained farmers were willing to invest considerable time and effort in detailed field observation of arthropods, contrary to the popular perception of many researchers and extensionists.

Table 2. Farmers' responses to a hypothetical new insect appearing in their crop

Farmer group	Response on observing a new insect
IRM cotton farmers, Maharashtra, India	4/5 respondents: would consult CICR1 for advice. 1/5 respondents: would spray, but not sure which pesticide to use.
Untrained cotton farmers, Maharashtra	4/6 respondents: would either consult agricultural officer or pesticide dealer. 2/6 respondents: would spray immediately.
FFS cotton farmers, Tamil Nadu, India	5/6 respondents: would ask the Voice Trust for advice, and discuss with neighbours or extensionist. 1/6 respondents: would set up an insect zoo (an observation exercise from FFS) to find out if it caused damage.
Untrained cotton farmers, Tamil Nadu	2/3 respondents: would spray immediately. 1/3 respondents: would see if it is doing damage before spraying.
FFS vegetable farmers, Kenya	Group response: (1) would catch insect, place it in a jar and observe its feeding behaviour to determine whether it was a natural enemy or a pest, (2) discuss outcome with rest of group and extension and research staff and share information with neighbours who had not been trained, and (3) if it proved to be a pest and spraying was required, chilli would be first choice.
Untrained vegetable farmers, Kenya	Experienced respondents: would handpick it if there were only a few, or if there were many, spray and send a sample to extension services. Less experienced respondents: would ask their neighbours for advice.
1 Central Institute for Cotton Research.

While only one of the Indian FFS farmers said they would carry out an insect zoo experiment, case study work from other cotton FFS groups facilitated by AME in Tamil Nadu reveals how trained farmers actively experiment with pest and natural enemy manipulation⁶. Cotton farmers in Tiruchirapalli had learnt about agro-ecological functions of intercropping as part of their FFS training, observing how an intercrop of cowpea served as an alternate host to cotton aphids, Aphis gossypii, which then attracted populations of ladybirds and syrphids. The farmers agreed that cowpea provided a useful build-up of natural enemies to help control aphids in their cotton crop but pointed out that this benefit only lasted for the 60-day cropping period of cowpea, whereas natural enemies are needed throughout cotton's 150 days in the field. They decided to adapt AME's recommended intercrop design by planting cowpea every 10-15 days during the cotton season, to ensure that there would always be food for ladybird predators. This example demonstrates farmers' ability to make in-depth observations regarding a problem and how participatory training and research methods can encourage further experimentation and build individual analysis and decision-making skills. The issue of how to encourage season-long natural enemy populations in cotton has now been taken up by AME and partners for further research and validation.

Perceptions that pesticides are essential to obtain good yields and that fields must be kept insect-free were expressed by all the untrained farmers in this study. For the Indian cotton farmers, their decision making had traditionally been strongly influenced by the advice and opinions of local input supply providers, in terms of choice of product and timing of application. Independent decision making was more difficult under these circumstances, even for trained farmers, especially those who wished to change from dependency on synthetic pesticides. Despite these pervasive influences, all three training projects studied had managed to alter participating farmers' perceptions about pests and their dependency on pesticides, demonstrating the importance of effective training in bringing about attitudinal and behavioural change at farmer-level. Season-long training, where farmers observe field ecology and compare different pest management regimes, as opposed to conventional extension methods, is critical in achieving such change and discovery-learning about biological control should form an essential element in these training programmes.

This publication is an output of a research project (CPP/R7500) funded by the United Kingdom Department for International Development (DfID), for the benefit of developing countries. The views expressed are not necessarily those of DfID.

References

1 Meir, C.; Williamson, S.; Little, T. (2000) Literature review of farmer decision-making in pest management.
Report for DFID CPP project ZA0352 Analysis of farmers' decision making in pest management.
CABI Bioscience, Ascot, UK, 48 pp.

2 Little, T.; Ali, A.; Kimani, M.; Oruko, L.; Williamson, S. (2000) Analysis of farmers' decision making in pest management DFID/CPP Project ZA0352 (R7500).
A report on fieldwork carried out in Kenya and India, December 1999 - February 2000.
CABI Bioscience, Ascot, UK, 54 pp.

3 Vos, J. (1998) Development of decision-making tools for vegetable farmers in South East Asia.
In: Haskell P.; McEwen, P. (eds) Ecotoxicology; pesticides and beneficial organisms.
Dordrecht, Netherlands; Kluwer Academic Publishers, pp. 404-409.

4 Williamson, S.F.J. (1998) Understanding natural enemies: a review of training and information in the practical use of biological control.
BNI 19, 117N-125N.

5 Meir, C. (2000) Learning and changing: helping farmers move to natural pest control.
In: Stoll, G. (ed) Natural crop protection in the tropics. Letting information come to life.
Weikersheim, Germany; Margraf Verlag, pp. 265-279.

6Anand Raj, D.; Suresh, C. (2000) Farmers' research on natural crop protection.
In: Stoll, G. (ed) Natural crop protection in the tropics. Letting information come to life.
Weikersheim, Germany; Margraf Verlag, pp. 291-296.

By: Stephanie Williamson,
International Project Officer,
Pesticide Action Network-UK Eurolink Centre,
49 Effra Road,
London SW2 1BZ, UK
Email: stephaniewilliamson@pan-uk.org

❑