Learning to fight a fly : developing citrus IPM in Bhutan

F. van Schoubroeck

Research output: Thesisinternal PhD, WU


The chinese citrus fly is one of the key pests in Bhutanese mandarin orchards that lays eggs in developing fruit that cause pre-mature fruit drop. In this study it is used as a "model subject" to explore the integration of technical, social and administrative domains of knowledge. The confinement of the study to control of the fly leads to the study addressing a broad set of issues that are linked through their relevance to control the pest. Citrus fly control and its consequent doubling of the mandarin harvest is relatively easily measured and could serve as an indicator for effectiveness of technology development, farmers'organisation, government services and policies.

Part 1 deals with the socio-economic aspects of the study, and deals with the link between technology, learning processes and organisation. Chapter 1 gives the socio-economic and historical background of the area of study. The area falls under Pemagatshel and Samdrup Jongkhar districts in Bhutan, it is mountainous and borders Assam (under India) in the south. The inhabitants are Sharchok-pa or eastern Bhutanese people with an Indo-Burmese origin, they speak a distinct language that is not written. The local religion is dominated by the Nyingma-pa sect of the Mahayana Buddhism that heavily draws upon Tibetan traditions. The "Dzongda" or District Administrator is the representative of the Royal Government of Bhutan in the "Dzong", the District Administrative centre. Forestry, livestock and agriculture sections represent the Ministry of Agriculture and carry out the developmental programmes in the district. Districts are divided in blocks headed by a kind of mayor, blocks consist of five to fifteen villages with village headmen.

The food situation in the area is relatively good, only in a few blocks shortages of food occur in particular seasons. Maize is the main staple crop and farmers turn excessive produce into drinks so that in the present marketing structure, improvement of staple food production is not likely to improve the local economy. Mandarin is the main cash crop that is mostly sold to Assam and Bangladesh with a huge population of potential consumers. The market can easily absorb any increase in Bhutanese mandarin production, although monopolies and trade restrictions often influence prices farmers get negatively. Even then, in terms of money, mandarin is the most voluminous export crop in the country, which still has a vast potential for further expansion in production both in acreage and per unit. The local variety is seed-multiplied, robust and well adapted to the local conditions. So far, there are no major virus problems, although a growing flow of germplasm in the country increases risks. In Nepal for example, the introduction of virus by developmental programmes wiped out citrus cultivation in large areas. Individual trees yield the equivalent of 40 tonnes per hectare, which shows that the average production of 2 to 6 tonnes can be substantially improved. Bhutan can capitalise upon this large potential for production by improving management through local seed selection and better planting practices, minor irrigation, manuring, and pest control. The major pests include mistletoe, trunk borer, the citrus green stink bug and the chinese citrus fly. The stink bug causes premature fruit drop in summer in low-altitude orchards and the citrus fly does the same, in autumn, in middle and high altitude orchards. Both pests result in fruit drop varying from up to 90 % damage by the stink bug, to 35-75 % damage by the chinese citrus fly.

Chapter 2 starts with an exposure on experiential learning theory, in relation to Integrated Pest Management (IPM) as used by the Farmers Field Schools (FFS) in Indonesia. In conventional learning, science is the source of knowledge and the extension system brings knowledge from science to farmers. In experiential learning, each individual goes through a learning cycle in which apprehension and comprehension follow each other. In this study experiential learning is taken as a process of experiencing, formulating the experience in words (that can be communicated) that shape the interpretation of the following experiences. Thus there is a feedback loop between knowledge and experience, rather than a teacher telling about reality as formulated by scientists.

For a citrus IPM programme, the modified training-and-visit extension routine as used in Bhutan was taken as a starting point and gradually experiential learning elements were introduced in the programme. In stead of giving training on a variety of subjects, training was confined to citrus, the training site was swapped from in-door to orchards, training activities were followed by farmers interviews in their orchards so that their actual effectiveness could be established. Then, in the same village training was repeated to make useful additions if farmers did not follow up recommendations.

Examining the follow-up of training with farmers in orchards revealed that this training programme led to a better adoption of recommendations in comparison with the earlier more general extension programmes. More surprisingly, monitoring farmers' practices revealed that quite a few recommendations given over the last few years had been impracticable or erratic. For example, the recommendation to plug trunk borer holes with kerosene appeared to be ineffective, as many of such holes were larva escape holes rather than entry holes. Following such observations, recommendations were modified according to experiences of successful farmers. Similarly, station-developed research outcomes were monitored for their usefulness. For example, research came up with the concept of natural enemies preying on pests which was taken up as a subject in training. This idea originated from areas in which calendar pesticide applications were the norm, however, Bhutanese farmers hardly used pesticides and such ideas had few, if any, practical consequences.

A shortcoming of the programme set-up was its poor potential to effectuate more fundamental technological innovations. For example, the theoretical notion that the chinese citrus fly could be controlled through food-baiting and hygienic measures needed to be examined in practical conditions, and the on-going research did not yield enough information to effectuate relevant shifts in practice. Another shortcoming originated in the fact that control of pests such as mistletoe and citrus fly needed to be carried out over entire villages. If an innovator farmer experiments with such technology at individual scale, it is bound to fail, and thus it is likely to be rejected even if it holds a great potential to improve yields. Transfer-of-technology extension was not effective in introducing such a village-wide application of control methods. This made clear that the current link between farmers, extension and research had to be reviewed. This led to a major shift in the set-up of IPM research. In 1993 the national plant protection service decided to have research carried out in village research programmes (locally referred to as "pilot village programmes").

In Chapter 3 it is argued that IPM was originally a technically defined methodology but that, for farmers to properly carry out IPM, they must have the learning capability to follow the ecological processes in their crops. In this end, processes must be visualised so that farmers can build intuition on processes leading to damage. Thus, both monitoring and control tools need to be developed to make IPM possible in practice. These tools can be developed through a Participatory Technology Development (PTD) programme, although most literature on PTD shows more concern for "empowerment" of farmers than for technological development as such. The observation that pests are often to be controlled at appropriate scale had substantial theoretical repercussions. Here I mention the social scale at which innovations take place. I take social cognition as a kind of "intuition" at group level. Social cognition theory observes that groups of people create a reality by sharing vocabulary and common interpretation of phenomena, resulting in a self-referential worldview. This social cognition essential for the implementation of activities at communal level, such as citrus fly control.

The citrus IPM programme set out to experiment with citrus fly control in two eastern Bhutanese villages, while on-station research continued under the same programme. Technology developed on-station was utilised in the village programmes, while technological needs felt in the villages, such as traps to monitor the fly population, were developed on-station. The programme success in villages appeared highly dependent on the social organisation of the two villages. In one village, farmers immediately formed a robust organisation that carried out the experiment over the entire village. In the other village various sources of resistance developed, such as the poor will to implement fly baiting in view of the inherent sin in religious terms. After three years, fly control in both villages succeeded and yields hit historical records that greatly enhanced the socio-economic development of the respective communities. In retrospect, the programme had difficulties in maintaining support during the period that the farming community had not yet acknowledged success. In one village the success led to farmers wanting to go further, and build a minor irrigation scheme for mandarin. For citrus cultivators and the extension service, the outcome of the programme was a combination of sociological and technological knowledge useful in a curriculum for Farmers Field School type of extension programmes in which complete villages learn how to implement citrus fly control. For the agricultural research practice, the programme showed how technology and socology can be combined into a practical blend of the two domains of knowledge.

After the village research programmes had been successful, policy makers wanted to capitalise on developed knowledge and scale-up the technology. So far, social cognition and experiential learning theory had been applied to farmers and a few individual staff. For the scaling-up programme, the same theory was applied to the extension service and the group of policy makers. Scaling-up of the socio-technological set-up appeared easy by extension staff who had been involved in the programme: he had four villages adopt the method within two seasons. Other agriculture staff appeared incapable of introducing citrus fly control in villages of their respective blocks. This was caused by particularities in the staff reward system, an inhibited learning process regarding implementing village programmes, and the fact that the input supply system could not supply the (inherent to IPM) small amounts of chemicals. These problems were poorly addressed because mainstream national agricultural and developmental policies were inspired by developments outside Bhutan rather than by successful initiatives inside the country. It is concluded that for a future scaling-up programme of locally successful technology, various actors in the Agriculture Knowledge System should interact to build a common understanding of the problems at stake so that they can co-ordinate their efforts to have large groups of farmers benefit from economically highly viable technology. In village programmes farmers and staff should try to induce small and measurable changes rather than carrying out broad programmes in order to learn from previous experiences rather than carrying out activities for the sake of doing them.

Part two of the thesis deals with the biology and control of the chinese citrus fly (Bactrocera minax). The data collected in village research programmes were often not complete, and the technical articles are mostly based on station-led research activities. Although the chinese citrus fly is a member of the family of fruit flies, its biology is quite peculiar in a few aspects. The fly has one generation per year and employs two rest periods (rather than only one in most other tephritid species): both as pupa in soil and as egg in fruit. The female's long ovipositor enables it to lay eggs in the pulp rather than in the peel of citrus fruits (where most other tephritid flies oviposit), thus avoiding contact of eggs with toxins in the skin. The fly oviposits early in the fruit development season (rather than during fruit ripening). After May, female flies puncture fruits of over 11-mm diameter. Fruits that develop earlier into critical sizes attract more ovipuncturing than later developing fruits. This mechanism explained uneven ovipuncture distribution over fruit populations within trees, between trees, as well as between different citrus species. Moreover, it allows for trap cropping, as in the early season an orchard's fly population flocks on a local lime species that develops its fruits weeks earlier than mandarin. The oviposition season lasted until September. The development of eggs into active maggots took one to four months, in October and November maggots started feeding and induced early ripening and premature fruit drop. In all four years of observation the median of fruit drop was in the third week of October, and was slightly earlier in a high-altitude than in a mid-altitude orchard. Possibly, the period of hatching related to shortening light periods, which may explain the uniform timing of fruit drop over the years.

Fruit populations could be checked for ovipunctures, which allowed for pre-drop quantification of oviposition activity. Through such checking the effectiveness of food baiting activities in villages was evaluated. Not all punctured fruits dropped; comparison of in-tree pre-drop and post-drop puncture density suggested that 76 % of ovipunctures contained viable clutches. In high-altitude orchards puncture density of fruit just before fruit drop appeared a reliable measure for damage, more so if post-drop density was established as well. In lower-altitude orchards, non-tephritid fruit puncturing agents blurred the evidence of oviposition activity.

In theory, the disposal of infested fruits from orchards could be part of an IPM strategy to suppress the fly population. However, monitoring puparia densities in soil revealed that disposing infested fruits hardly adds to the natural control system in soils. If the soil is not disturbed by tilling, natural enemies, mostly ants, prey upon puparia. Various traps were developed to establish critical periods in the fly's life cycle such as the fly's emergence and oviposition period and the best period for proteinaceous food-bait attraction. The chinese citrus fly is best lured during the emergence period in April when flies of both sexes are attracted, unlike in most other tephritid flies, in which baiting is usually carried out just before the oviposition period when mainly female flies are attracted. In one experiment in one isolated orchard, proteinaceous food-baiting during the emergence and oviposition period controlled the fly; after two years of bait application the fly was near extinct. Influx of flies from orchards at about 1,200-m distance made that the population recovered in two years. This results in a recommendation to combine proteinaceous baiting during fly emergence with the use of catch crops in the early oviposition season. Part 1 of the book then showed how fly monitoring and tools can be used to enable farmers to identify critical periods of the fly life cycle and build a social cognition for a joint effort to implement fly control activities.

Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Röling, N.G., Promotor
  • van Lenteren, Joop, Promotor
Award date15 Dec 1999
Place of PublicationS.l.
Print ISBNs9789058081131
Publication statusPublished - 1999


  • citrus
  • citrus fruits
  • mandarins
  • insect pests
  • bactrocera minax
  • integrated pest management
  • bhutan


Dive into the research topics of 'Learning to fight a fly : developing citrus IPM in Bhutan'. Together they form a unique fingerprint.

Cite this