Integrated Management of Coffee Berry Borer

INTEGRATED MANAGEMENT OF COFFEE BERRY BORER

FINAL REPORT - INDIA

CFC/ICO/02

Central Coffee Research Institute Coffee Research Station P.O. 577 117 Chikmagalur District Karnataka INDIA

With contributions from:

CABI Commodities United Kingdom

Cenicafé Colombia

and

Jeff Bentley Bolivia CREDITS

EDITORIAL COORDINATION Héctor Fabio Ospina O.

DESIGN AND LAYOUT Carmenza Bacca Ramírez

PRINTED BY FERIVA S.A. Cali, Colombia May 2002

The Commodities Press1

1A joint CABI Commodities-Cenicafé enterprise CONTENTS

1. INTRODUCTION 1

2. COFFEE IN INDIA 6 The Indian economy and coffee Coffee production Coffee zones and growing Coffee farmers Coffee production costs

3. THE CBB IN INDIA 8 Incidence and spread Coffee phenology and CBB attack 8. SOCIOECONOMIC 29 STUDIES ON THE 4. TECHNOLOGY TRANSFER 10 ADOPTION OF IPM AND RESEARCH Methodology Overview of previous work in Extent of adoption of man- India agement interventions Research of the present Projet Constraints in CBB managment technology 5. TESTING, VALIDATION adoption AND INTEGRATION OF 13 Awareness about CBB concepts 29 COMPONENTS 9. CONCLUSIONS 29 6. THE ECONOMICS OF CBB Losses due to CBB 16 REFERENCES AND Cost of CBB management PUBLICATIONS 29 7. IPM INFORMATION APPENDICES DISSEMINATION 29 Appendix 1: Coffee Board Farmer participatory method of India infrastucture Farmer-researcher extensionist meetings Appendix 2: Overseas vis- Impact of the FPM programme its to attend meetings, train- Women empowerment ing and conferences on CBB programme Summary of extension activi- Appendix 3: Visits of con- ties sultants to India under the Group gatherings at IPM plots CFC-ICO project Conclusions Appendix 4: Earlier Indian studies on CBB ecology and cultural control

ince 1990 the Coffee Board of India 1995 for a period of three years. All the ex- (CBI) has initiated various research isting programmes on CBB were integrated Sand extension programmes to combat into this new project and new research and the pest problem of the coffee berry borer extension programmes were also started. (CBB) (Hypothenemus hampei, Coleoptera, This resulted in the emergence of a pack- Scolytidae). As additional financial support age of practices for the management of the was required to carry out the new pest, and the creation of awareness among programmes, the Ministry of Commerce, growers on the importance of taking up Government of India, was approached for timely measures for tackling the pest. The specific projects under the CBI’s Plan. Thus project was continued as an ongoing the first project cleared by the Ministry was programme during the IX Plan Period (1997- a scheme to provide pesticide and plant pro- 98 to 2001-02). tection equipment to the smallholder growers, which was launched during November As the CBB is a serious pest problem in most 1991. Under this project, pesticide and of the leading coffee growing countries in sprayers were supplied to smallholder grow- the world, and the fact that much progress ers at subsidized rates. had been achieved in research in biocontrol of CBB in Colombia and Mexico, the Coffee In order to intensify the studies on devel- Board approached the International Coffee oping an integrated management package Organisation (ICO) to initiate an interna- for coffee berry borer with more focus on tional project on the pest. biological measures, another project, the Biological Control of Berry Borer, was initi- The ICO took an initiative in this direction, ated during May 1993. Studies were con- and during November 1995 a consultant, Dr. ducted on the possibilities of using J. A. Nicholas Wallis was sent to India to biocontrol agents in the management of prepare a draft proposal for an international berry borer and studies were initiated on project. CABI Bioscience was identified as the the indigenous fungal pathogen, Beauveria Project Executing Agency (PEA) for imple- bassiana. Simultaneously, attempts were mentation of the project, and the Common also initiated to import some exotic parasi- Fund for Commodities (CFC) agreed to pro- toids of the borer as there was no parasi- vide funds. The details of the project were toid recorded from India. Two parasitoids, worked out in a meeting of the participat- Cephalonomia stephanoderis and Prorops ing countries (India, Colombia, Mexico, Ec- nasuta were introduced from Mexico dur- uador, Guatemala, Honduras and Jamaica) ing September 1995, and two entomologists with CABI Bioscience and the CFC at the ICO, were trained on the mass multiplication of London during July 1996. The Project was the parasitoids at ECOSUR (El Colegio de la launched in January 1998 for a period of Frontera Sur), Tapachula, Mexico. From three years, which was later extended to De- these activities, a culture of C. cember 2001. This project provided an op- stephanoderis was established in India. portunity for research and extension personnel to have first hand knowledge of the lat- The Scientific Advisory Committee of the est developments in CBB management Ministry of Commerce under the Chairman- worldwide. The technology of mass breed- ship of Dr. M.S. Swaminathan recom- ing CBB parasitoids developed at Cenicafé, mended, during August 1994, to consider Colombia was made available to India the CBB as a national pest and implement a through training to researchers. Addition- project that had a mission mode approach ally, two parasitoids, Phymastichus coffea to intensify the ongoing programmes in a and Prorops nasuta were imported from more effective way. Thus, the Ministry of Colombia under this project. A farmer par- Commerce sanctioned a project, the ‘Na- ticipatory approach was highlighted in the tional Mission on Control and Prevention process of technology development and of Coffee Berry Borer’, during December implementation.

6 7

The detailed information on the various research and extension infrastructure, and of search and extension activities conducted the principal collaborating personnel, are under the Integrated Management of the provided in Appendix 1. A summary of the Coffee Berry Borer Project (CFC/ICO/02) are training and meetings visits made by some the subject of this report. The information of these personnel under the project is given is reported in the context of the coffee in- in Appendix 2. Finally, the visits of consult- dustry in India, and previous work con- ants to the project are summarized in Ap- ducted on the CBB by the Coffee Board. A pendix 3. summary of the structure of the Board’s re-

THE INDIAN ECONOMY sons have been suggested including issues related to the WTO and market AND COFFEE liberalisation, climate problems and government agricultural policies. For example, in ndia’s major agricultural exports are ce- states such as Gujarat, Orissa and Rajasthan reals, spices, cashew, oilcake/meal, to the monsoons have been deficient in the last Ibacco, tea, coffee, and marine products. two years. The main consequences of this The value of agri-exports to total exports have been a fall in the output of essential of the country normally ranges between 15% oilseeds (groundnut, mustard and soybean and 20%. Coffee exports have remained are most affected). On the other hand, food above US$ 400 million during the last 5 grain growth rates have declined from 3.5% years, with the exception of 1999-2000 in the 1980s to 1.8% in the last decade, while when it was US$ 315 million. Table 1 de- in the non-food grain economy the growth scribes the composition of Indian agricul- rates have fallen from 4% in the 1980s to tural exports between 1996-1997 and 1999- 3.1% in the 1990s. Similarly, Menon (2001) 2000. points out that the fall in international prices of robusta coffee as well as pepper From Table 1, it is clear that some products has also affected coffee farmers in have been increasing their share of exports, Karnataka and Kerala states. Pepper is an for instance spices, cashew, sesame, guar important extra-income for many coffee gum meal and seafoods. Other groups have farmers, but its productivity has been low remained more or less constant, including (between 275 and 300 kg/ha), despite its tea, coffee, tobacco, fruit & vegetables, meat potential. Additionally the price has seen & meat preparations. Overall coffee has re- reductions from Rs 22,600/quintal in 1999 mained constant at about 6% of agricultural to Rs 12,000/quintal in 2000. A similar situ- exports, despite low international coffee ation has occurred with areca nuts. Prices prices during recent years. In the states fell from Rs 154/kg in September 1999 to where coffee is grown, coffee contributes Rs 78/kg a year later. This farming crisis between 3% and 4% of GNP (ICO, 1997). deserves a more complete analysis, but the point here is to highlight how the Indian In India there seems to be a consensus that agricultural sector is facing a difficult pe- the agricultural sector is suffering a seri- riod that is affecting many crops. ous crisis (Muralidharan, 2001). Several rea-

10 11

Coffee production are, while in 2000-01 the expected productivity was about 1,084 kg of parchment cof- offee planted areas have increased fee per hectare. The suggestion is that crop markedly in India during the last half technology has been responsible for this century. In 1950/51 there were increase in productivity. Figure 3 shows the C trend in coffee productivity in India (x-scale 92,523 ha, but by 1999/00 the total area had grown to 340,306 ha. Figure 1 shows a not continuous). very significant increase in total area during the 1970s and 1980s. Arabica increased Nevertheless, coffee yield remains close to by about 100,000 ha, whilst robusta rose 1 ton/ha, which is low in comparison to by 146,000 ha. Areas planted to arabica and yields in other coffee countries. So, even robusta are now roughly equal. without expansion in total area, there may still be considerable scope for an increase In 1950/51 production was estimated at in total production. about 18,893 metric tonnes while in 2000/ 01 the forecast is for 295,000 metric tonnes Arabica and robusta are classified accord- (Coffee Board, 2001). Figure 2 shows this ing to the post-harvesting processing clearly. method: “washed” and “non-washed” (naturals), and are further classified into 25 In 1950-51 coffee productivity was approxi- grades based on the size of the bean and mately 255 kg of parchment coffee per hect- on the total number of defects or imperfec-

Figure 1. Area planted with coffee in India, 1950-51 / 2000-01

Figure 2. Coffee production in India, 1950-51/ 2000-01 12

Figure 3. Yield per hectare in India, 1950-51/ 2000-01

tions (ICO, 1997). Separate from this sys- Coffee zones and growing tem, specialty coffee is now the fastest growing segment, and India hopes to export more in the next few years. Organic coffee is also inety two percent of Indian coffee is of increasing interest, especially for coffee grown in the states of Karnataka, grown on tribal land, which represents NKerala and Tamil Nadu, the rest in about 42% of the coffee area in India. Here small sectors of the states of Andhra coffee is managed in a less intensive tradi- Pradesh and Orissa, as well as in some of tional way, which is close to organic pro- the North Eastern states. Table 3 describes duction guidelines (Central Coffee Research production in the main coffee areas in In- Institute, 2000). Some of the grades of spe- dia. Figure 4 shows a map of India, high- cialty coffee and value added coffee are lighting the main coffee states. Mysore nuggets EB, Monsooned Malabar (AA, PB, C), Monsooned Arabica AA, Forty nine percent of coffee in India is Monsooned Robusta AA, Monsooned arabica and fifty one percent robusta. Cof- Arabica Tr., Monsooned Robusta Tr., fee leaf rust (CLR, Hemileia vastatrix) is a Monsooned Basanally, Monsooned Arabica major constraint to arabica production and BBB, Monsooned Robusta C., Monsooned Ro- efforts have been made to obtain resistant busta Bulk, Monsooned Robusta Blacks, varieties. Hence the main goal in arabica Robusta Kapi Royale. Exports for these varietal improvement is CLR resistance, to- coffees have roughly doubled in the last gether with high productivity, wide adapt- decade (Table 2). ability and improved quality. In the case of

12 13

Figure 4. Main coffee producing states in India

robusta it is known that this species pos- sity can range from 6,900 trees/ha in dwarf sesses high tolerance to leaf rust disease, arabica varieties to 1,000 in robusta variet- white stem borer, nematode attacks and has ies such as SLN.1 R or SLN.3 R. The national good potential to give consistent yields. Re- average is around 1,000 to 1,400 coffee search therefore is concentrated on its trees/ha. shortcomings: lack of drought-resistance, late stabilisation of yields and inferior cup The amount and the frequency of fertiliser quality. use in India varies according to the coffee variety. For instance, there are four key pe- But other factors such as soil characteris- riods to apply fertilisers: pre-blossom, post- tics and climate conditions have an impor- blossom, monsoon and post-monsoon. For tant role in deciding what type of coffee arabica, if a high yield is expected (above should be grown. Table 4 describes the ad- 1,000 kg parchment coffee/ha), coffee farm- equate conditions for each type of coffee. ers should apply fertiliser in each of the above periods. But if the expected yield is The density of coffee trees per hectare var- below 1,000 kg, the monsoon period appli- ies according to several conditions. The den- cation can be avoided. On the other hand, 14

for robusta, if the yield is above 1,000 kg, tries such as Colombia, Honduras or Mexico. farmers should apply fertilisers three times Table 6 below describes the distribution a year (pre-blossom, post-blossom and post- pattern of holdings in different states of monsoon). When the expected yield is be- India (Coffee Board, 2000). low 1,000 kg, it is recommended to apply pre-blossom and post-monsoon only. From Table 6, 98% of farmers have less than 10 ha, and 86% have 2 ha (5 acres) or less. It is generally larger farmers who carry out This characteristic is important since the coffee fertilisation. Large farms apply adoption of technologies is often affected fertilisers near the recommended levels, by farm size. Duque et al. (2000), working while smallholders commonly choose lower in Colombia, found that the adoption of IPM rates or none at all. The use of organic mat- in CBB management was higher with me- ter is also frequent. The expected price for dium and large-scale farmers than with the coffee harvest obviously plays a key role smallholder farmers (with less than 5 ha of in decisions about fertiliser application. coffee).

At the farm level, Indian coffee productiv- If this is true for other coffee countries, the ity is low in comparison with yields reached strategy for transferring IPM should be de- in other countries such as Brazil, Colombia signed to promote good adoption levels or Vietnam. But productivity has been grow- despite socio-economic barriers, but also ing. Table 5 shows coffee productivity ac- the IPM strategy by itself should be made cording to the species grown over the last easy to adopt. Despite most coffee produc- five years. ers being smallholders, from Table 6 it can be concluded that for Indian coffee there is a land concentration and if we take the cu- Coffee farmers mulative percentage of farmers and coffee areas we can quantify this as the Gini coef- n India there are estimated to be 140,293 ficient (Tascon, 1980). This indicator gives coffee farmers while the total area a measure of land concentration, which in Iplanted in this crop is 340,306 ha; so the this case has a value of 0.48 that indicates national average would be 2.42 ha of coffee that there is some degree of land concen- per farmer, approximately 6.05 acres. This tration, e.g. 80% of coffee farmers have just figure is similar to that found in other coun- 40% of the coffee area farmed. The Indian

14 15

value of 0.48 compares to that of Colombia one farm. The coefficient equals the shaded (0.57) and Mexico (0.43), i.e. that land con- area divided by the total area below the 45° centration is lower than in Colombia, but line. This analysis is useful to help under- higher than Mexico. stand the importance of land distribution between owners, in order to design appro- Land concentration can be seen from dif- priate pest management strategies. ferent points of view e.g. land reforms, but it is important for IPM management because As mentioned above, 67% of the Indian the pest management proposal should take workforce works in agriculture. According into account the farmers’ resources (for in- to a study carried out by the Project (data stance, the method might create a high de- in preparation), Table 8 shows typical labour pendency on external inputs). It is logical requirements for 3 Indian states and re- to suppose that labour should be the most gions: Kogadu, Wayanad and Tamil Nadu available resource for these types of farms. (NB this excludes labour required for har- Another key point is that income generated vesting coffee). These are high requirements by small farms is of course not high (Table in comparison to other countries, but are 7), and if the market punishes coffee qual- less than other Indian estimates (e.g. Cof- ity they could be severely affected by CBB. fee Board, 2000). The ICO (1997) calculates that coffee production in India employs The Figure 5 displays the graphical repre- around 367,000 persons equaling 1.07 sentation of the Gini coefficient. The 45° labourers per hectare. Labour costs show sloping line (Gini coefficient = 0) means no differences between producing states. Table land concentration whereas a value of 1 9 indicates the labour cost in four states of would mean all land was concentrated in India. 16

Figure 5. Land concentration in the coffee sector in India

16 17

Hence a contracted temporary worker could ent effects depending on the period of the expect an annual income of around US$ 250 year when income is scarce because some if he works all year. In comparison to many activities have to be delayed to attend to a countries labour is therefore very cheap in more important activity. The farmer has to India (e.g. compared with Colombia, where manage different labour requirements and a labourer could earn the Indian yearly availability to optimise his gross income. In equivalent in about two months), and this the case of CBB management, practices that is no doubt the reason why labour is a can easily be accommodated by the farmer smaller proportion of the total production will be easier to transfer. cost structure. IPM strategies often demand greater amounts of labour than another In order to diversify their income, Indian methods (e.g. chemicals), hence the low cost coffee farmers have developed production of the labour in this country could facili- systems involving more than one crop in tate the adoption of an IPM strategy for man- order to get income from different sources aging CBB. and in different periods of the year. In a survey carried out in the Project on coffee Coffee farmers’ income is seasonal in India, farmers in Karnataka it was found that which is a risky situation because there is 95.6% of them had pepper as an inter-crop. just one peak of income per year. In order In many cases the inter-crop allows genera- to minimise the effect of these seasonal tion of an important part of the total farm variations, farmers can attempt to spread income. For example, during Mr. Duque’s flows of labour and harvest production visit to the Wayanad region, he met Mr. throughout the year. Upton (1996) points Vijayakumar who has a small farm of about out that there are various strategies, such 1 ha. He told him that his income depended as diversification of agricultural production, on three sources - coffee, pepper and areca to establish different on- and off-farm ac- nut. The individual split of each appears in tivities, storing food, seeds and animal fod- Figure 6 below. der, etc. Seasonal income will have differ-

Figure 6. Income composition in a small farm, Wayanad district 18

It is therefore important to stress the con- clean coffee are needed to cover all variable tribution of other crops to the income com- costs. Yield over the break-even point will position of a normal coffee farm. It is quite cover fixed costs and the farm’s profit. The possible that sometimes this alternative in- formula proposed by Kay (1981) can be used come becomes a relevant source to cover for this purpose for robusta coffee, assum- costs related to CBB management. ing a farm gate price of Rs 650/50 kg of cherry coffee (Coffee Board of India, 2001).

Coffee production costs

offee production costs in India for the year 2000-2001 was approxi- Cmately Rs 15,952/acre, around US$ 340/acre. This figure is equivalent to US$ 850/ha/year. Table 10 shows the composi- If the average yield is 468 kg of clean coffee tion of the total production cost in terms per acre (for 2000 – 2001), the average of both fixed and variable costs. break-even point would not be reached by many Indian coffee farmers. This situation From Table 10 it can be inferred that fixed reflects the real world scenario of interna- costs are high, perhaps due to the irriga- tional coffee prices. The Figure 7 elaborates tion equipment that was considered as a on this. fixed cost. When the proportion of fixed costs is high it is more difficult to reach a As can be seen in Figure 7 an ‘average’ cof- break-even point. A break-even analysis can fee farm would not reach the break-even be useful to know how many kilograms of point, due mainly to high fixed costs and

Figure 7. Break-even point for an average farm in the sample

18 19

low coffee prices. Assuming that all other tation levels will lead to higher CBB man- variables analyzed remain constant, there agement costs, which would increase vari- would need to be an increase in coffee price able costs. But on the other hand, depend- of about Rs 35/kg to reach the break-even ing on the infestation level, CBB can reduce point. Thus CBB can play a key role in the yield. If yield reduced so that the total rev- economic performance of Indian coffee enue per acre was also reduced, the break- farms, because on the one hand high infes- even point would be more difficult to reach.

Incidence and spread The Coffee Board, the Directorate of Plant Protection, Quarantine and Storage, Govern- he coffee berry borer was first re- ment of India, the Directorates of Horticul- ture, Govt. of Karnataka and Tamil Nadu, ported from a large plantation in and the United Planters’ Association of Gudalur, Nilgiri District, Tamil Nadu T South India organized the first major sur- during February 1990. Damage to berries vey of coffee berry borer during September was first noticed in the curing works, where 1990. A systematic sampling method was the coffee from this plantation was pro- adopted to assess the infestation level in cessed, and the cause was identified as in- coffee estates as well as at coffee curing sect damage. The matter was reported to centers. In the field survey, 50 plants each the CBI’s Research Department and the in- from five locations in each estate were se- sect was identified as the coffee berry borer, lected. Two branches from infested plants Hypothenemus hampei. Immediate surveys were selected for recording the number of in the surrounding areas revealed the pres- infested berries and the percentage of plant ence of the borer in a few more plantations, and berry infestations were calculated. and interim control measures were suggested based on the information available. At curing centres, composite samples were Simultaneously, surveys and awareness prepared by drawing one kilogram of sample campaign programmes were organized in from each bag, out of 10% of bags selected the infested and surrounding areas. It is from 10% of the lots in storage. A working thought that Sri Lankan refugees who sample of one kilogram was further pre- settled in the area might have brought berry pared out of four composite samples by borer infested coffee from Sri Lanka for dividing them into four equal parts. After domestic use, resulting in an accidental in- screening the samples thoroughly for introduction of this pest. fested beans, the percentage of bean /cherry infestation was calculated.

On detection of the pest, the Coffee Board This survey revealed that: immediately initiated a series of steps to combat the new pest problem through its Fourteen estates of 58 estates Research and Extension Departments. All screened in 166 locations had CBB in- the Extension and Research units were festation - one in Sulthan Bathery, alerted and large-scale grower education Wayanad, Kerala and 13 in Gudalur, programmes were launched in all the cof- Tamil Nadu. The range of plant infes- fee zones. Intensive surveys were organised tation varied from 1% to 95% and in the pest-affected areas and infested es- berry infestation from 0.83% to tates identified. The information on the 13.98%. possible techniques for combating the coffee berry borer was collected by literature The presence of CBB infested coffee survey and was passed onto coffee growers was confirmed in two curing works, for immediate field application of suitable namely M/s. Kushalanagar Works, methods. Phytosanitary measures, such as Kudige, and M/s. Karnataka Coffee a clean and timely harvest, proper gleaning Planters Coffee Curing Works (P) Ltd., collection, removal of left-over and off sea- Kushalnagar, on the basis of 1819 son berries and spot application of endosul- samples drawn from 76 lots from 50 fan were immediately advocated for the go-downs in 6 curing centers. The control of the pest. Simultaneously, experi- range of infested beans varied be- ments were initiated by the Research De- tween 0.22% and 1.26% of beans. partment on the biology of the pest and the effectiveness of various methods for its The pest infestation had spread to management (see below for more details). more areas in Gudalur and shown its

22 23

presence in Wayanad, Kerala. How- form a contiguous coffee growing tract. The ever, CBB had not entered into the CBB infestation in this region had spread plantations of Kodagu District, from the Niligiris towards the North cover- Karnataka. ing Wayanad and South Kodagu over a period of four to five years. The pest was first Survey / campaign programmes were orga- noticed outside this region in the nized regularly in all the three districts Perumalmalai coffee zone of Pulney Hills (Nilgiris, Wayanad and Kodagu), and the in- ranges, Tamil Nadu, during September 1995. fested areas identified for follow-up action. This region is quite far away from the As the pest was spreading gradually to other Nilgiris-Wayanad region, and is separated areas, a second joint survey / awareness by a large stretch of land not growing cof- campaign was conducted in Nilgiris, fee. Wayanad and Kodagu in association with the Directorate of Plant Protection, Quar- Further spread of the CBB in Kodagu was antine and Storage during October 1991. noticed during June 1999, when it was recorded from three villages of Somwarpet This survey revealed that as compared to zone in North Kodagu. The pest was also 1990 the pest was widely spread in Gudalur recorded in Hassan District for the first time and had entered into the adjacent Coonoor in October 1999 from one village in taluk of Nilgiris District. Fresh incidence of Sakleshpur zone. the pest was also recorded in 16 estates in Sulthan’s Bathery and Vythiri taluks of Though the pest was first recorded from Wayanad District. The presence of the pest Pulney Hills during 1995, the incidence was was also confirmed in Kutta and Theralu confined to one village of Pethuparai in villages of Virajpet taluk of Kodagu District. Perumalmalai Zone till 1999. During the The berry borer incidence in Kodagu was survey in October 1999, the pest was de- earlier reported from Kutta during Febru- tected in 21 major coffee growing villages ary 1991. of the Pulney Hills. During December 1999, the pest was recorded from Adimali and During 1992 pest build up was confined to Kattappana liaison zones in South Kerala, areas identified in the previous years and which are adjacent to the Pulney Hills. fresh incidences were noticed to a limited extent. In another major survey conducted During August 2000, the pest was recorded during September 1993, covering 1,328 es- in Bodinayakanur, Tamil Nadu. By March tates in Kodagu region, the pest was ob- 2001, the CBB had spread to an area of 118, served in 16 villages of Siddapur and in one 453 ha, accounting for 35% of the total cof- village each of Virajpet, Sunticoppa and fee area of the country. The annual spread Madikeri Liaison Zones. During the March of the pest, and annual cumulative pest in- 1994 survey, screening of 176 estates in fested area, are graphically depicted in Fig- Kodagu region confirmed the presence of ure 8. the pest in two more villages of Siddapur and four more villages of Virajpet Zone. The data show that the CBB spread during Further surveys were conducted during Feb- the first five to six years was rapid with an ruary 1995 in all the three Taluks of Kodagu annual average increase of 35%. However, district and the pest was recorded from four from 1996 onwards, the rate of spread de- more villages of Gonicoppal, six in Virajpet, clined to about 5.1%. The state-wise CBB four in Siddapur, three in Madikeri and two infested area as of March 2001 is presented in Sunticoppa Zones. in Figure 9.

The coffee growing areas of Niligiri District So far the coffee berry borer has not been of Tamil Nadu, Wayanad District of Kerala noticed in Chikmagalur region, which is the and Kodagu District of Karnataka together major coffee growing district of Karnataka 24

State. The other major coffee growing ar- completed by December/January. The grow- eas still free from CBB incidence are Yercaud ers bestow more care on a timely harvest of in Tamil Nadu, Biligiri Hills in Karnataka and arabica as most arabica coffee is prepared Nelliampathy and Attappadi in Kerala. as washed coffee. On the other hand, robusta takes more time for ripening and so An analysis of the nature and spread of the harvesting generally extends up to Febru- coffee berry borer incidence in the affected ary/March. Since robusta coffee is pro- areas points to the fact that the build up cessed by drying the fruits to prepare cherry and spread are faster in robusta areas when coffee, there is a general tendency to pro- compared to arabica areas. The pest prob- long the harvesting due to various reasons, lem is acute in Wayanad and southern parts especially if there is a labour shortage. This of Kodagu where robusta is predominantly offers the berry borer a better chance to cultivated. Inter-cropping dwarf varieties multiply faster as the berries at this stage like Cauvery under robusta could aggravate are best suited for their multiplication. If the problem due to the availability of the berries are left on the plants after ripening, right type of fruits for a longer time as the number of berries falling during harvest Cauvery matures much earlier to robusta. would be more. This leads to increased re- Harvesting of arabica coffee is generally sidual populations of the pest in the fallen

Figure 8. Spread of coffee berry borer in India

Figure 9. State- wise spread of CBB as of March 2001

24 25 berries (gleanings). Furthermore, berries left Coffee phenology over on tall branches of big robusta bushes can also provide a refuge for the CBB. and CBB attack

The incidence level of CBB on Pulney Hills t is important to consider how coffee was found to be higher than in other areas. phenology relates to CBB abundance in This is mainly due to the multiple blossom IIndia. As in all coffee countries, rainfall (i.e. running blossom) experienced in this pattern is a critical factor because of the region, which results in the availability of way in which precipitation triggers flower- fruits almost throughout the year. This fa- ing and hence the berries upon which CBB cilitates a continuous build up of the pest, depends. In Figure 10 we can see that the and adoption of either cultural or phyto- rainfall pattern consists of four main sanitary control measures becomes imprac- phases: winter, pre-monsoon, monsoon and ticable. post-monsoon. Secondly, the main stages of

CBB entering coffee berry

Figure 10. Rainfall pattern, coffee phenology and CBB attack 26

the coffee fruiting phenology is depicted, variable is the variety - for robusta the and finally the horizontal bar represents the flower-to-ripe-bean cycle is about 40 weeks, most critical period for CBB management. whilst for arabica it is less.

Flowering can occur in February or March if Finally in Figure 10 the most critical period a farmer has irrigation facilities; if not, then for CBB attacks is depicted as the period they have to wait until April when natural between August and October, so labour and flowering will follow the start of the rains. other inputs (e.g. spraying of insecticides) Because of this, the coffee harvest occurs are particularly required at this time. in either December or January. A further

26 27

Overview of previous Thus, to address some of the management technology issues, the Coffee Board started work in India a research programme in the early 1990s to evolve components for a management uring monitoring surveys and visits strategy based on integrated pest manage- to CBB affected estates, interactions ment (IPM) principles. The on-farm valida- Dwith growers on management issues tion of these technologies is discussed later highlighted the following key factors: in this report.

Robusta coffee, grown without Various projects were implemented to proper pruning, poses problems in the achieve this objective. In particular, stud- adoption of phytosanitary measures ies were focussed on phytosanitary mea- and effective spraying operations sures, need-based and judicious applications of pesticides, and biological control Sprinkler irrigation, given to robusta utilizing native fungal pathogens: relatively early, leads to faster growth of berries which results in the early Studies on phytosanitary and other cul- establishment of berry borer when tural control methods showed the impor- compared to fruits formed due to tance of these measures and also resulted blossom naturally watered in the development of some simple and effective techniques, e.g. the use of picking Arabica coffee, cv. Cauvery (catimor) mats, - this work is summarized in Appen- interlined in robusta plantations fa- dix 4 cilitates early establishment of the borer in Cauvery, with subsequent A number of pesticides were screened by migration to robusta the Coffee Board but the only effective one proved to be endosulphan; this has been No effective control measures are widely adopted by farmers although it is available once the borer is established now being phased out by the Government within coffee berries of India due to environmental concerns

There are limitations in the use of Fungal pathogen work has largely focussed biopesticides and parasitoids on Beauvaria bassiana. Laboratory and field studies indicated the potential of B. Off-season crop provides a medium bassiana as a biopesticide against CBB when for the continuous breeding of the fresh cultures were applied at the appro- pest priate period. High volume application was found to give better results than low vol- Constraints in timely harvest of crop ume sprays. Low percentage of infectivity due to shortage of labour was observed in some experiments and seemed to be due to unfavourable weather Primitive and inadequate drying yard conditions or due to loss of infectivity of facilities, especially in the small the culture. Whenever a 1 to 1½ month-old grower sector, results in prolonged culture was used during favourable environ- and improper drying of coffee and mental conditions, the results were better. delayed harvest This revealed that purity of material and timing of the application are the two criti- Transporting of infested coffee from cal factors in obtaining better results. Un- place to place for processing and mar- fortunately all the commercial formulations keting, which facilitates spread of the tested were of poor quality, without caus- pest to new areas ing any appreciable infectivity. The major

30 31

bottleneck in the large-scale use of B. signment from Ecosur, Tapachula, Chiapas, bassiana is the constraint of making avail- Mexico was received on 5th September 1995. able large quantities of fresh material at the This consignment consisted of 510 C. appropriate time (Coffee Board of India, stephanoderis and 140 P. nasuta. Twenty- 2001). four (4.7%) of the former and 18 (12.86%) of the latter perished in transit. The live para- Research of the sitoids were quarantined and multiplied at the Project Directorate of Biological Control present Project (Indian Council of Agricultural Research), Bangalore. Two more consignments of para- uring the CFC/ICO/02 Project, re- sitoids were later received from Mexico. The search has focussed on the devel- culture of C. stephanoderis established well Dopment of biological control with in the laboratory, whereas P. nasuta failed insect parasitoids (particularly on the rear- to survive. ing, release and evaluation of species received earlier), and on the mass-trapping of Further, under the CFC/ICO/02 project, CBB adults. cultures of Phymastichus coffea and P. nasuta in CBB infested parchment were air freighted from Cenicafé, Colombia. The first Research on insect parasitoids consignment consisted of P. nasuta and P. coffea arrived Bangalore on 26th October In India, no indigenous parasitoids were 1999. As the first consignment of P.coffea recorded during extensive surveys con- did not establish, three more consignments ducted in all CBB infested areas, indicating were later sent. that the CBB was introduced without any complement of parasitoids. As parasitoids Rearing, field release and evaluation are being used for the biocontrol of CBB in of parasitoids various countries, introduction of promising parasitoids was attempted. The rearing of C. stephanoderis and P. nasuta imported from Mexico was initiated Importation and quarantining at Coffee Research Sub Station, Chettalli, in of parasitoids January 1996.

Two bethylid parasitoids, Cephalonomia As C. stephanoderis and P. nasuta have simi- stephanoderis and Prorops nasuta were im- lar breeding habits, the rearing technique ported from Mexico in 1995. The first con- is common to both. Initially the parasitoids were multiplied by using naturally infested coffee fruits following the Mexican methodology. Infested berries collected from the field were spread on white paper, air dried for a few days and fruits with dark powder coming out from the hole, which is indica- tion of all CBB life stages, were selected for parasitoid rearing. Mexican scientists had found greenish yellow, yellow and yellow- ish red fruits ideal, but red, green, and black unsuitable. However, over ripe, semi-dry robusta fruits also were found most suitable for parasitoid breeding under Indian Cephalonomia stephanoderis eggs laid on CBB prepupa conditions. 32

When naturally infested fruits with imma- The wasps from the collection jars were re- ture stages of CBB were not available, the moved using an aspirator and transferred Colombian method of using artificially in- into small plastic containers. The contain- fested parchment was employed. Fresh ers were covered with pieces of muslin cloth parchment was demucilaged by natural fer- and closed with lids having circular holes mentation and treated with carbendazim at in the centre. A few drops of half diluted 3 g per litre of water. The parchment was honey were provided on the cloth. The con- spread in trays, air-dried in a room for a tainers were then taken to the field and day, and the moisture level brought to 40- wasps released on infested bushes by gen- 45%. Naked, broken, discoloured and dis- tly tapping the containers. When the wasps eased beans were removed and only healthy emerged in large numbers, they were quan- beans selected for CBB rearing. The selected tified by volumetric measuring, each milli- good parchment was spread in metal or liter containing around 2000 individuals. plastic trays of convenient size with ventilation. Beetles collected from the breeding If released on the main crop, berries with stock of naturally infested fruits were then the parasitoids were harvested in a short released in the trays at the rate of 2 per time and processed, before the wasps could bean. establish themselves in the field. Hence, releases were made mostly on robusta plan- The trays were maintained at 25°C and 80% tations having left-over fruits during the relative humidity and examined on the 8th, post-harvest period. 12th and 18th days after beetle release. Beans contaminated with fungi such as Aspergil- For evaluation, infested berries from the lus and Beauveria were discarded, and the release sites were examined at regular in- frass removed by sieving. After 20-25 days, tervals and the percentage parasitism, as the infested beans had developed sufficient indicated by the presence of parasitoid borer stages - eggs, larvae, prepupae and stages, calculated. pupae - needed for parasitoid breeding. Only seeds with 2 or more holes were selected. The moisture level was around 30% at this Early releases and evaluation stage. About 150 or 300 infested beans measured by volume were placed in plastic The first test field releases of C. boxes of convenient size and the parasitoid stephanoderis and P. nasuta were effected released at the rate of 1 per 3 beans. If natu- in January 1996 in Ballykilty estate, rally infested fruits were used, parasitoids Cannoncadoo, North Kodagu. The parasi- were released at the rate of 1 per 3 to 5 toids were confined using cloth sleeve cages fruits, depending on the number of CBB on infested coffee branches. On day 35, the stages. The containers were covered with lids having muslin cloth ventilation and maintained at 25°C and 80% relative humidity.

Parasitoids started emerging 20 to 30 days after release. An emergence cage (a wooden box with blackened sides with transparent plastic jars fixed downwards) was used to collect the wasps. When the parasitoids were ready to emerge, the containers were placed in the emergence box. Tungsten filament lamps (60W) were provided near the jars to Confined release of attract the wasps. Cephalonomia stephanoderis

32 33

coffee fruits were stripped off and main- Mass rearing and field release of tained in the laboratory for adult emer- Cephalonomia stephanoderis gence (Table 11). Under Indian conditions, over-ripe fruits A second test-caged release was done in were found more suitable than artificially February 1996 with C. stephanoderis on infested parchment for rearing the parasi- Achyutha estate, Maldare, South Kodagu toid. Progeny production of the wasp was and P. nasuta on M.K. Kalaiah’s estate, 5.75 (1.60 to 10.50) in naturally infested Byrambada, South Kodagu (Table 12). The fruits and 3.20 (1.30 to 7.50) in artificially results indicated that the recovery of both infested parchment. C. stephanoderis and P. nasuta were poor in the first confined release made on A total of 2,430,536 wasps were produced Ballykilty estate during June. The recovery at CRSS, Chettalli and RCRS, Chundale dur- from second release during Febraury was ing the period from January 1998 to Sep- more promising. tember 2001. A total of 1,157,790 wasps were released into the field, during the post- Furthermore, P. nasuta was released in harvest period, on 14 farms. The details of small quantities in two sites on Pulney the field releases are presented in Table 13. Hills, Tamil Nadu, but was only recovered The establishment of the parasitoid was from the site, with no further recovery studied by examination of left over black made in subsequent observations. berries and in certain cases the ripe/green berries collected from release sites. About Though P. nasuta could be recovered in 100 fruits collected at random from each small numbers from the confined release release site were examined in the laboratory initially, it could not be found later. It did for the presence of various stages of the not survive in the laboratory culture, in parasitoid and the percentage parasitism spite of obtaining one consignment from calculated. A few non-release sites were also Mexico and three from Colombia. Hence, surveyed to study the spread of the parasi- more attention was given to the mass mul- toid. The details of the evaluation tests are tiplication of C. stephanoderis and intro- presented in Table 14. duction of P. coffea from 1998 onwards. 34

34 35

From the observations so far made on the Unfortunately, both the cultures failed to mass-rearing and establishment of the para- survive beyond the 6th generation. A pre- sitoid, the following points emerged: liminary field trial with P. coffea on caged branches revealed parasitism between 4% Naturally infested fruits were ideal and 7.3%, with a mean of 4.7%. for rearing C. stephanoderis under In- dian conditions Three more consignments of the parasitoid were received from Cenicafé during June, There was no consistency in the es- July and September 2001. Progeny emerg- tablishment of the parasitoid on re- ing from all the batches have established in lease sites the laboratory culture, and a total of 6,905 females have been released on four farms Spread of the parasitoid to non-re- in Kodagu. lease sites with better establishment than in release sites was recorded Mass-trapping studies Survival of the parasitoid was observed in 21 out of 52 release sites surveyed. Parasit- These studies were conducted in Wayanad, ism was observed in red and black berries Kerala, and in Pulney Hills, Tamil Nadu. Here while that in green berries was rare. Nine we report the work from Wayanad, as the out of 22 non-release sites also revealed other studies mostly confirmed the results parasitoid activity. In one case, the parasit- of these. ism in leftovers in the release site was just 9%, whereas in two neighbouring farms (non-release sites) 54% and 77% of leftover Evaluation of traps and lures fruits were parasitised. Carry-over of the parasitoid from one season to the other was As mass trapping is reported to be a new observed in 5 cases. technique in the management of CBB, stud-

Phymastichus coffea

The first consignment of P. coffea was received in October 1999, and the second in Febraury 2000. The parasitoid was reared using the method followed in Colombia.

CBB beetle parasitised by P. coffea, Multiple bottle top trap showing characteristic hanging head 36

ies were initiated to evaluate the effective- galvanised iron wire. The multiple fun- ness of various types of traps and lures. In nel structure was provided with a Wayanad, a preliminary study was con- hood made of plastic plate on top for ducted in January and February 2000 in protection against rain and dry leaves which the following four types of traps were falling into the trap. A small plastic tested. Each type of trap was baited with bottle with water (receptor) was fit- both ethanol-methanol and coffee fruit ex- ted to the neck of the bottom funnel tract lures, and each treatment had 2 traps. to collect the beetles. The receptor All the treatments were placed both in non- was provided with a few holes at harvested and harvested fields to find out about 3/4 of its height to allow rain- differences, if any, in the pattern of attrac- water to flow out. The lure dispens- tion. The traps were hung randomly on the ing bottle with a pinhole on its cap coffee bushes at a distance of 80 feet apart, was kept hanging in the second fun- at about 5-6 feet off the ground. The trap nel designs were as follows: Single funnel trap with receptor: A Multiple funnel traps made of plastic single common funnel was fitted with bottle tops: The top portion of min- the receptor and the hood as in the eral water or soft drink bottles was previous case. The bottle containing removed and four such pieces in in- lure was hung from the hood verted position, resembling funnels, were arranged one over the other Single funnel without receptor: In this without touching using thin case, a single funnel was provided

36 37

with a hood. The inside of the funnel The treatments were replicated five times. was coated with “Trappit” insect glue (polybutene), replacing the receptor Traps made of multiple funnel: Four with water of the single funnel trap. common funnels were arranged one The lure-dispenser bottle was hung over the other using thin galvanised from the hood iron wire, provided with a receptor fitting and hood as an alternative to Plastic can trap: A plastic can (liquid container) of 2 litre capacity was provided with windows at the centre of its four sides and the bottom was filled with water, the trapping media. The lure-dispenser bottle was hung inside the container

Data obtained in the preliminary study are presented in Tables 15 and 16.

Data obtained in the preliminary trials showed that a 1:1 mixture of ethanol and ‘Multiple funnel trap’ methanol is effective in luring CBB beetles. The fruit extract in ethanol also lured the beetle, but to a smaller extent compared to the ethanol-methanol mixture. The multiple funnel trap made of plastic bottle top was the most effective trap among those evaluated. Furthermore, more beetles were trapped in harvested fields than in non-harvested fields.

Further evaluation of trap designs

The following models of traps baited with 1:1 mixture of ethanol and methanol were also tested along with the previously tested multiple funnel traps made of plastic bottle tops and single funnel traps with receptor. ‘Bottle trap’ 38

the multiple funnel traps made of designs, but not as effective as the multiple plastic bottle tops. The lure-dispenser funnel traps made of tapering plastic bottle bottle was hung from the hood. tops.

Bottle trap (the “Brocatrap”): Plastic Evaluation of lures mineral water or soft drinks bottles were used in this case. Windows of A study was conducted on the lures using about 1.5 to 2.0 inches width were cut the Brocatrap. The chemical combinations on opposite sides, leaving only two evaluated in the Brocatrap were: small strips to keep the top and bot- Ethanol + Methanol tom together. A receptor with drain Ethanol + Methanol + Methylene Blue holes was fitted to the neck of the Isopropanol + Methanol bottle and a third of it filled with wa- Diethyl Ether + Methanol ter. The lure dispenser was placed in- Ethyl Acetate + Methanol side the bottle. In another modifica- Carbon Tetrachloride + Methanol tion, the bottle cap replaced the re- Coffee Fruit Extract in Ethanol - Methanol ceptor and the trapping media was mixture held in the main body of the trap it- Coffee Fruit Extract in Ethanol self. The data obtained is presented Coffee Fruit Extract in Methanol in Table 17 Coffee Fruit Extract in Diethyl ether Coffee Fruit Extract in Isopropanol As in the preliminary trial, the multiple fun- Coffee Fruit Extract in Ethyl Acetate nel traps made of plastic bottle tops were Coffee Fruit Extract in Carbon Tetrachlo- found superior to all the other trap mod- ride els. The new model Brocatrap was close in effectiveness to the multiple funnel traps made of plastic bottle tops. Being simple and The data is presented in Table 18. This easy to fabricate, the Brocatrap was very study indicated that coffee fruit extract in cheap costing around Rs 5 to 7 compared 1:1 combination of ethanol and methanol, to Rs 14 to 20 for multiple funnel trap made followed by coffee fruit extract in metha- of plastic bottle tops (depending on the nol, were superior to the standard 1:1 com- quality the bottle). bination of ethanol and methanol.

The studies on the trap designs were continued with modifications incorporated in Evaluation of trap heights the Brocatrap to make it comparable in effectiveness with the multiple funnel traps To find out the optimum height for trap made of plastic bottle tops. These studies placement, multiple funnel traps were indicated that a modified Brocatrap with 6 tested at 3, 5 and 7 feet above the ground. vertical windows of 3-inch size was supe- Each treatment was replicated five times rior to the standard Brocatrap and other with a single trap per replication.

38 39

The results indicated no significant differ- were evaluated with five replications per ence in catch at different heights. There was treatment and 5 traps per replication. The a slightly higher catch, though not statisti- results showed that a significantly higher cally significant, at 5 feet height, so further catch was obtained in the transparent placement of traps were made at this height. Brocatrap.

Evaluation of trap colour Trap spacing

To study the preference of colour in combi- To study the spacing or trap density per unit nation with the lure, Brocatraps were area, multiple funnel traps at distances of painted orange, red, green, yellow, white, as 20 feet (109 traps/acre), 40 feet (27 traps/ well as a transparent version. All versions acre), 80 feet (7 traps/acre) and 120 feet (4 40

trap/acre) were evaluated in a randomised Evaluation of trap density block design. Each treatment was replicated and effect on CBB incidence five times. The study commenced in June, the time when the residual population of The trial was conducted by installing the CBB in the gleaning / leftover fruits migrate traps in coffee field after harvest at 10, 20, to the new crop. 40, 80 and 120 feet spacing corresponding to trap densities of 435, 109, 27, 7 & 4 re- Before imposing the treatments, the total spectively per acre. The post-harvest catch residual fruits and total population of the of CBB per acre and the fresh incidence of borer were estimated by sampling the fruits CBB in the new crop were recorded, as left on the plant and on the ground. Num- shown in Table 20. ber of infested fruits left on the plant was estimated by sampling all the fruits in 10 The data revealed that catch per acre was randomly selected plants and the number significantly higher with close spacing or of fruits on the ground was estimated by higher trap density, which is in conformity sampling 10 sites at random in an acre us- with the result of the previous experiment. ing quadrats of 1 square foot size. The num- The higher trap density reduced the infes- ber of CBB beetles trapped was recorded tation considerably by around 67% to 73% after a month, in July, and then again in in the new crop as compared to the control. September. The weather parameters dur- But, considering economic factors, a trap ing the study were recorded. The data is density of 27 per acre seems to be optimal. presented in Table 19. From the studies on mass trapping conducted so far the following inferences could The data indicated that more beetles are be drawn for further investigation: trapped when the traps were set up at closer spacing. However, the percentage of beetles The coffee fruit extract in 1:1 com- trapped was very low compared to the esti- bination of ethanol and methanol, fol- mated CBB population in the field and hence lowed by coffee fruit extract in metha- the real impact of trapping on beetle popu- nol, were superior to the standard 1:1 lation needs further investigation. Some combination of ethanol and methanol progress was made in the following study.

40 41

The multiple funnel trap made of Higher trap density reduced the plastic bottle tops is the most effec- infestation considerably by around tive trap, followed by the modified 67% to 73% in the developing new crop Brocatrap with 6 vertical windows compared to the control. However, a each of 3-inch size. Considering its trap density of 27 per acre seems to simplicity, economy and its effective- be optimal ness, the modified Brocatrap with 6 vertical windows is preferable for field Additional studies indicated that commonly applications available alcoholic beverages could be used in place of ethyl alcohol in the lure.

espite some differences in CBB man Collection of gleaning agement between arabica and ro Removal of infested berries Dbusta the following calendar of op- Removal of off-season and left over erations represents, ideally, the most impor- berries tant practices in CBB management in ro- Use of picking mats busta coffee. Table 21 describes these Drying coffee to the standard test monthly operations. As may be concluded weight from Table 21, there are some key months Early disposal of the coffee pro- in CBB management, such as those from the duced beginning of the year, and those from Au- Maintenance of trap plants around gust to December. drying yard Spot spraying of endosulfan wher- By the start of the CFC/ICO/02 CBB project ever absolutely necessary, as a last a number of cultural control and other man- resort agement interventions were being promoted by the CBI in line with the above calendar; The data on the borer incidence was col- these were based on the earlier research lected at regular intervals and necessary conducted on the CBB (e.g. see Appendix measures were suggested accordingly. Un- 4). In order to test more fully the impact of til March 2000 the data collection was done the suggested interventions, particularly bi-monthly and since April 2000 on monthly their integration, on-farm studies of IPM basis. The monthly percentage incidence were set up during the Project across the levels for each plot was worked out and whole coffee-growing region affected by the average levels of incidence of all IPM plots CBB. for three different regions are presented in Table 22. A total of 52 such plots of approximately one hectare each were established in differ- The average incidence-level from August to ent agro-climatic zones. Small, medium and December 2000 varied from 0.6% to 1.1% in large-scale growers were all included in the IPM plots of Kodagu region, 1.0% to 1.8% in study. The farmers or farm managers were Wayanad region and 5.9% to 7.4% in Pulneys fully engaged in the study in an attempt to region. The level of incidence came down make the study as participatory as possible. to less than 0.5% from December onwards in both Kodagu and Wayanad regions, but A systematic sampling technique was adopted to assess the incidence levels in the on-farm plots by taking periodical counts and recording the data regularly. A sample size of 10 plants in the case of robusta and 20 plants for arabica plants was fixed for assessing the incidence level. The total number of berries and the number of infested berries on five randomly selected branches on the selected plants were recorded before suggesting the measures or interventions required to keep the pest under control. The following were the major activities suggested for the IPM plots depending on the conditions prevailing in each plot:

Timely harvest Picking mat used whilst harvesting Thorough and clean harvest

44 45

it continued to be as high as 5% in Pulneys. The data revealed that the CBB incidence The level of inoculum in leftover berries was could be kept at low levels in Wayanad and significantly low and stood at 0.1% to 0.35% Kodagu by adopting IPM measures, particu- during June and July 2000 in Kodagu re- larly cultural and phyto-sanitary measures. gion, at around 0.5% in Wayanad and at 5.5% in Pulneys. The weighted average of the in- The incidence level on Pulney Hills was com- cidence level (worked out from June to Feb- paratively higher, mainly due to the preva- ruary), indicated the monthly rate of inci- lence of more favourable conditions, espe- dence at 0.6% in Kodagu, 1.0% in Wayanad cially multiple blossoms which allows the and 7.7% (excluding June/July incidence) in borer to have uninterrupted generations. Pulneys.

Losses due to CBB 0.9916). So the model can forecast up to 99% of the variations in percentage of weight n a study conducted by the Central lost due to different infestation levels of Coffee Research Station (2001), an CBB. The main features of the model appear I estimation of crop loss due to CBB was in Table 24. carried out. Crop loss was estimated at different levels of infestation from 0% to 100%. The model can be expressed mathematically The study included loss in weight, but not as follows: in quality, although it is known that CBB affects quality as well. Unfortunately, up to now, the internal coffee market in India does not have a clear standard of rules for both so as to punish or to reward coffee depend- Where: ing on quality. However, the study is required from the point of view of yield loss. INF = Percentage of infestation The data collected are: the loss in forlits and in percentage of clean coffee is shown in The following graph (Figure 11) shows the Table 23. model. This function represents an increasing return function, which means in this From this information a loss function has case that every successive increase in the been estimated, where the independent vari- infestation level results in higher additional able is the percentage of infestation caused loss. by CBB, and the dependent variable the loss in percentage of clean coffee. From the farmers point of view this kind of function is quite dangerous because the The model of loss obtained was of a qua- losses will increase more than proportion- dratic type, (Probability > F = 0.0001) and ally for a certain level of pest attack. So they has a very high coefficient of determination should adopt the techniques and compo- (R-square = 0.9944; Adjusted R-square = nents needed to ensure infestation levels

48 49

that can be managed easily and to avoid sud- invest in pest control is more related to the den high pest levels. pest level.

With reference to the model, a simulation In Figure 12 are shown the potential losses of crop loss was carried out assuming an due to different infestation levels, expressed average yield of 468 kg of clean coffee per in Rupees/acre. acre (see the section on coffee production above) and a coffee price of Rs 28.5 per kg Figure 12 shows that potential losses can of clean coffee. The simulation involved in- be severe and thus even a moderate infes- festation levels between 0% and 100%, and tation of CBB can make it more difficult for the percentage of loss was calculated (Table farmers to reach the break-even point, let 25). alone generate a profit.

From Table 25 it is clear that when infestation levels are low, expected losses are also Cost of CBB management low. So farmers should carry out those control measures needed to keep the pest at n the last year of the Project, an assess this level or below. However, at this point, ment was made of 97 farms/estates it is quite possible that the willingness to Iinvolving both IPM plots (described ear-

Figure 11. Loss function due to CBB attacks in India 50

lier) and non-IPM plots. From a general point imply that these small farmers are invest- of view the average cost for CBB manage- ing the equivalent of 5.3 man-days per acre ment was 426 Rs/acre/year (approximately per year in pest control. Farmers belonging US$9.06), which equates to around to category 3 are spending 10.3 man-days - US$22.60 per hectare/year. The maximum almost twice the smaller farm sizes. Figure cost observed was about 2,344 Rs/acre, 13 describes the frequency distribution of which equals US$124.00 a hectare. On av- CBB cost per acre/year. erage the share of CBB management over the total production cost was 3.8% - this is According to Figure 13, the distribution is lower than observed in other countries, like biased to the right. It is evident that to find Colombia, where the share is about 7%. coffee growers investing more than 1500 However the CBB cost is quite variable ac- Rs/acre when controlling CBB is rare. The cording to variations in farm size. Table 26 majority of farmers tend to spend up to 600 shows the CBB cost estimated for four farm Rs/acre, since 78.3% of the sample is in that sizes. As can be seen, the observed tendency range. Taking into account that the survey is for CBB management to be more costly was carried out in 2000-2001, when inter- with increasing coffee farm size with the national coffee prices were better than 2001- exception of the largest ones where the cost 2002, it is likely that this investment capac- was lower than in categories 2 and 3. This ity has been reduced. If this hypothesis is might indicate that some economies of scale true then the willingness to spend large are present. What is very clear is that small- amounts of money when managing CBB will holder farmers tend to invest less in CBB be lower. Farmers will thus choose the most management. If we assume that cost of a adequate practices that they believe will re- man-day is about 60 Rs/day, this would duce production costs.

Figure 12. Simulation of crop losses due to CBB infestation levels (Rupees/ acre)

50 51

Another element of CBB management cost In this way the investment required to con- is that when productivity is higher, the pest trol CBB should be higher. Figure 14 depicts problem tends to be more difficult to deal this tendency, despite the large scatter of with. Thus the investment required might points. also be higher. Table 27 describes coffee productivity according to holding size, in- From the data above it can be observed that dicating that when a holding is larger, pro- when productivity is above 700 kg of cof- ductivity is higher. fee/ha CBB cost tends to be higher, above

Figure 13. Cost of CBB management - histogram of frequencies

Figure 14. Relationship between cost of CBB management and holding size 52

Rs 600/acre. However, for the sample farms In summary, in general terms, the cost of in the study, there were both IPM and non- CBB management/acre is relatively low (Rs IPM plots. Table 28 shows the CBB cost 426/acre). Given the steep increase in loss breakdown for both types. in revenue with increasing CBB infestations, farmers should consider investing in man- Although the costs are apparently different, agement strategies at low CBB infestation a statistical analysis for mean differences rates. the results show that the CBB cost is equal (P value = 0.3062), whether or not the plot was in the Project.

52 53

The farmer participatory To find out the adoption levels of recommended packages of prac- method (FPM) tices To improve the growers’ decision- he Coffee Board’s extension network making abilities has been unique in many ways from To identify constraints in adopting Tthe viewpoint of working in close technologies proximity with the growers as field units are To identify innovative, locally located in plantation areas. This enables a adopted, technologies friendly rapport to be maintained with all To improve the productivity and growers in a Liaison Zone through personal quality of coffee grown contact with extension staff. Conventional To improve the overall socio- extension methods have been successfully economic status of the farmer followed for the past half a century, transferring coffee cultivation technologies and FPM is a three way interactive process dealing with the welfare of coffee growing wherein researchers, extensionists and communities. farmers meet on a bi-monthly basis to dis- cuss not only CBB issues, but also all other With a view to strengthening the extension aspects of coffee cultivation. It involves a delivery system, Farmer Participatory Meth- combination of conventional extension ods (FPM) were adopted under the CFC/ICO/ methods such as group gatherings and 02 CBB Project in order to widen the scope meetings, as well as field visits to conduct of dealing with CBB in relation to all aspects demonstrations, in addition to FPM tech- of coffee husbandry. Under the auspices of niques. In the FPM mode, the active partici- the Project, Dr. Falguni Guharay of CATIE, pation of the farmer needs to be ensured Nicaragua, an expert in FPM techniques, vis- for the successful adoption of technologies, ited India during January 2000 and con- unlike other conventional extension meth- ducted master training sessions with 126 ods. research and extension personnel of the Coffee Board. These master trainees later on conducted FPM programmes in their re- Farmer-researcher- spective zones. extensionist meetings The FPM technique was adopted with the following objectives: armer-researcher-extensionist meetings at periodic intervals, at pre-se To bring the farmer, researcher Flected estates, were conducted. Partici- and extensionist together on ‘one pating farmers were divided into groups to platform’ visit the estate and to identify problems. These groups collected specimens and made observations from the field. The leaders of each sub-group presented the group’s views on the standard of maintenance by using flip charts. Deliberations on the observations and various technologies adopted were held to analyse the gap between locally adopted and recommended practices. The facilitators and researchers then helped the groups todiscuss the problems identified, and to find solutions using the basket Dr Falguni Guharay conducting of technologies available. FPM workshop

56 57

The problems identified were ranked, based Opportunity for verification of field va- on merit. The most urgent and need-based lidity of different technologies/practices problems were then selected for discussion. Continuous feedback from the field for Finally, feedback from all growers of a group researchers/extensionists were collected and analyzed for further FPM as a new technique - a refreshing improvement. process

In addition, bi-monthly workshops involv- 2. On extentionists: ing researchers and extensionist were con- Providing an opportunity to exhibit ducted at research centres to deliberate on their talent and abilities the feedback of FPM meetings and to em- Improving upon their communication power extensionists. Under FPM activities, and other extension skills 24 regional technical workshops were held Sharpening their technical knowledge at different research stations during this and skills Project. Interacting at regional research workshops and updating knowledge fre- In summary, FPM groups, consisting of 20- quently 30 growers each, were formed in all the liai- Feeling better, to have first hand son zones in the three states of Karnataka, feedback from many farmers Kerala and Tamil Nadu during the years Developing more of a team-work 2000 and 2001. Details are presented in attitude Table 29. Using a wider variety of tools (like flip charts) to put across their ideas Opportunity to develop new ideas to convince the farmers Impact of the FPM programme 3. On growers: Opportunity to express their views freely he various activities conducted under Chance to learn what other growers this new extension method made the are doing following impressions on the exten- T Verifying field validity of different sion programme, extensionists and coffee technologies/practices growers: Learning latest techniques Focusing more sharply on current 1. On the extension programme in gen- problems eral: Participating and learning continu- Location specific technologies/practices ously coming to the fore Knowing about cost reduction oppor- tunities 58

Thus, the FPM programme launched as a part of the CBB management activity under this Project resulted in extension activities being made more visible, in motivating growers to actively participate in such activities and it inspired extensionists to adopt new extension tools made available to them.

Women empowerment programme (WEP) WEP participants he Indian women empowerment programme (WEP) came about as a The WEP meetings gave women an oppor- result of a visit by Dr. ST Murphy of T tunity to get motivated and understand that CABI Bioscience in April 2001, when it was they have a greater role to play in the man- decided to initiate special activities on the agement of estates. They were also inter- empowerment of women in CBB manage- ested to train on coffee cultivation, prefer- ment and other coffee cultivation aspects. ably at local level and in the local language. This was as a direct result of the fact that most women could not attend FPM activities due to other chores. Two workshops were held at Kalpetta and Somawarpet to Summary of extension gauge the response of the women and the activities carried out during degree of their involvement in managing coffee plots. Encouraged by the positive 1999 to 2001 response from these meetings, similar programmes were organized in all CBI liai- he details of various extension activi son zones. A total of 26 meetings were held ties conducted by the Coffee Board in different areas between April and Octo- Tof India to combat Coffee Berry Borer ber 2001 (Table 30). between 1998 and 2001 are summarized in Table 31. The assessment reports received from field units revealed that the knowledge of women/spouses of estate owners, on cof- FPM-style group fee cultivation varied from 60% to 80%, but gatherings at IPM plots with regard to technical aspects it varied between 20% and 30%. Their involvement he FPM style was adopted to dissemi- in decision-making on the management of nate the inferences of IPM observa estates varied from 10% to 20%, and they tions in the IPM plots. FPM-style had expressed that men generally take de- T group meetings at IPM plots were held in- cisions.

58 59

volving neighbouring farmers to dissemi- nate the knowledge generated at on-farm IPM plots. A total of 17 such meetings were held in different regions, eight in Kerala, five in Karnataka and four in Tamil Nadu.

Conclusions he various extension programmes carried out over the last twelve years Thave been successful in creating an awareness about the coffee berry borer cide use was higher in the initial years, its problem and motivating the growers to take use has now reduced considerably as grow- up control measures, which is evident from ers realize that strict adoption of phyto-sani- the low levels of CBB incidence reported tary measures can bring down CBB pest in- from many of the estates and the reduced cidence to very low levels. The establish- rate of spread of the pest in the recent years. ment of IPM plots in growers’ fields under The incentive provided to the growers, es- the CFC/ICO/02 CBB Project has played a pecially pesticides at subsidised rates dur- significant role in bringing about this ing the early years, played an important role change. The extent of adoption, and the cur- in convincing the growers about the possi- rent constraints to this, is considered in the bility of controlling the pest if control mea- next chapter. sures are adopted on-time. Though pesti-

Methodology Biological control methods Use of plans traps surrounding the study was made of socioeconomic drying yard Use of CBB traps constraints to the adoption of CBB management interventions in 97 es- A All 97 estates were located in three states tates including both IPM plots and non-IPM involved in the Project, and distributed as plots during the 2000 and 2001. This study shown in Table 32. was conducted by the Coffee Board and a consultant to the Project, Dr Hernando The following analysis is aimed at highlight- Duque of Cenicafé. ing some of the results obtained during this study. The ratio of IPM to non-IPM plots was 1:2, so as to include more plots that were not directly linked to project activities. In the selected plots several interventions had Extent of adoption been made under the guidance of research- of management interventions ers and extensionists working on the Project. In non-IPM plots the extent of any he adoption of the interventions pro- adoption of interventions was purely the posed for CBB management showed result of farmer decision-making. The main important differences. For instance, interventions used in the IPM plots and be- T the components such as gleaning collections ing recommended in general have already and the use of picking mats were the most been referred to, and were listed above in adopted. However the fungus B. bassiana the chapter on ‘Testing, Validation and In- was the least adopted intervention by farm- tegration of Components’. They are listed ers, as was the use of sticky traps. Table 33 again here for convenience: summarizes the adoption of all the components comprising the IPM strategy for con- Timely harvesting trolling CBB in India. Thorough and clean harvest Gleaning collections From Table 33, it is clear that the most Removal of infested berries adopted components are those related to Removal of off-season and leftover cultural control, such as the use of picking berries mats and gleaning collections. Use of picking mats The adoption of these components showed Drying coffee up to the accepted different tendencies according to the state standards analyzed. Early disposal of the coffee produced (early selling) A statistical analysis was done in order to Spot spraying of endosulfan wher- test for differences between proportions ever necessary as last option

62 63

according to the procedure suggested by The adoption of B. bassiana was very Walpole et al. (1998) and Levin et al. (1994). low across the board, as shown in Table 33. Firstly, for spot spraying, the observed adoption in the states of Kerala Finally, the component with a high and Tamil Nadu was statistically rate of adoption was gleaning col- higher than in Karnataka, but none- lection, which was statistically simi- theless appeared equal between them. lar in all the states. Secondly, the adoption of blanket spraying was highest in Tamil Nadu, indicating that this region is charac- terized by a way of using of insecti- Constraints in CBB cides that is some distance from the IPM concept. Then there were the management technology farmers from Karnataka State and fi- adoption nally the lowest adoption of blanket spraying was in Kerala. It is interest- he coffee farmers involved in the sur- ing to note that farmers from Kerala vey comprised 30% IPM plots, and the have adopted well the idea of spot Tother 70% were farmers close to these spraying. plots but not taking an active part in the Project. Additionally 83% were owners while The adoption of hot water treatment 17% were paid managers. was statistically equal between Karnataka and Kerala, and between Some constraints to the adoption of inter- Karnataka and Tamil Nadu. However ventions were identified during the study. Tamil Nadu showed a higher adoption For instance, in the case of spot spraying in this component than coffee grow- the coffee growers mentioned a number of ers in Kerala State. constraints, which are shown graphically in the next figure. Sticky traps had a very low rate of adoption, but this rate was similar From Figure 15, it is clear that problems among all three states. related to labour is a major restriction. Fur- thermore, spot spraying operations are seen In general, the use of picking mats as costly. A low percentage of growers did has been well adopted in all states. not know about the use of this practice. The highest adoption was in Karnataka. followed by Kerala with The main constraints in the adoption of the lowest rate in Tamil Nadu. blanket spraying are depicted in Figure 16. 64

In this case the main difficulty is the high edge to do it, as shown in Figure 17. As ex- cost. Of course blanket spraying requires plained above, gleanings collection is the the whole coffee plot to be covered and the most adopted component of the IPM strat- amount of labour needed and the input cost egy against CBB in India. The main con- (insecticide) is expensive to carry out this straints to its adoption appear in Figure 18, operation. Additionally, the lack of knowl- where it can be concluded that the majority edge is another important factor acting as of the coffee growers see this component a barrier. as a beneficial and effective intervention. However, gleaning collection is costly for For hot water treatment the main problems more than 20% of coffee growers. But un- when adopting this component were the doubtedly this component is playing a key cost of the activity and the lack of knowl- role in the control of CBB.

Figure 15. Adoption of CBB components by State

Figure 16. Main constraints in adoption of blanket spraying

Figure 17. Main constraints in hot water treatment adoption

64 65

In the same way the adoption of picking the supply of the fungus itself, but also mats, which are spread under the coffee maybe another inputs such as mineral oil trees to reduce gleanings, was also very and spray equipment. As in the other com- good. Currently many coffee growers think ponents there is a lack of knowledge about that this is very useful intervention because the technology, and in some cases it is seen in farms where it is used, new crops seem as difficult to use and not practical at all as to become less infested. Figure 19 describes a control measure. the main barriers in the adoption of this component.

The answers were similar to the gleanings Awareness about case since 58% of the farmers did not see any restriction in adopting this component. CBB concepts Nonetheless, some of them have the opin- ion that picking mats are costly. he socioeconomic study aimed to establish the level of awareness of some In the case of B. bassiana, the main obstacles Tbasic concepts related to IPM. Table (Figure 20) are the lack of inputs, perhaps 34 describes the results obtained. Despite

Figure 18. Main constraints in gleaning collection adoption

Figure 20. Main constraints in Beauveria bassiana adoption

Figure 19. Main constraints in picking mat adoption 66

the effects of irrigation on the homogene- than 96% of the farmers are aware of the ity of the blossom, most of the farmers do effects that CBB can have on coffee quality, not believe (73%) this operation has any re- which is very important since they should sult in CBB management. In this instance be interested in producing good quality cof- steps could be taken to teach farmers how fee, as prices are normally better. Addition- important it is to have a single blossom in- ally, the majority sees that this pest can stead of several flowerings when managing cause damage on young berries. However, CBB. despite the fact that the majority think that CBB can damage young coffee berries, there A key factor that motivates the adoption of is scope to emphasize this kind of loss in any technology is the importance of the those that are not aware of it. problem it helps to solve. In this way more

66 67

t present, coffee berry borer is For the sample of farms analyzed during present in approximately 35% of the Project, the average total production A the total coffee growing area in In- cost per acre/year was about Rs 15,952 (ap- dia. Taking into account the experience of proximately US$339.40). Fixed costs make- CBB in other countries, further spread of up 34.7% of total production costs while this pest to other coffee growing areas is variable costs contribute 65.3%. From a gen- likely. Hence, there is every need to continue eral point of view, the fixed cost percent- with the programme of CBB control in India age is high. While the average yield was 468 and to use the experiences gained so far to kg of clean coffee per acre, the break-even improve strategies in the future. Some key point has been estimated at 559 kg at cur- points arising out of the CFC/ICO/02 Project rent (2002) prices. are outlined below. Many Indian coffee growers are simply not In general there is a good level of aware- reaching this point, and are thus facing a ness among farmers about CBB and its im- very difficult economic situation. From the plications. Farmer participatory methods sample analyzed, some 60% of farmers were adopted during the Project have helped with not reaching this break-even point. Two the dissemination of information to farm- possible ways for the farmers to reach the ers, and have also been educational for ex- break-even point have been identified. The tension workers and researchers as well. For first is to get a price of Rs 35 per kilogramme instance, coffee farmers are clear about the of clean coffee. The second is to improve type of damage CBB can cause and some of coffee productivity. However, both of these the methods that exist to manage the pest. are hard to attain. A key aspect here is to However, they are not very clear about how avoid any yield reduction as a result of CBB. factors such as irrigation can affect CBB incidence. The average cost for CBB management was about Rs 426 per acre, which is a relatively With respect to the adoption of the avail- low cost. However some farmers are invest- able IPM components outlined in the chap- ing more than Rs 1,000 per acre, and in ter above, it is clear that those related to some cases CBB management costs reached cultural control (such as gleaning collection US$ 124 per hectare. CBB management costs and the use of picking mats) have been suc- tend to increase as coffee productivity in- cessfully adopted in many areas. Spot spray- creases – a tendency that can be seen in ing and blanket spraying have been less other countries, e.g. Colombia. adopted largely because of high costs; labour costs are a particularly important As high costs are the main barrier for tech- constraint. However, in the IPM context, spot nology adoption, and most of the compo- spraying is clearly preferable, and should nents are necessarily linked to use of labour, be promoted as such. the future strategy for CBB management should be based on management compo- The most common constraint identified by nents that are less labour dependent. In this the farmers was the high cost when imple- context, more emphasis should be placed menting the IPM components for CBB man- on biological control. agement. This problem is clearly linked to current low coffee prices - and these low Work on the development of an artificial diet prices are the principal problem faced by for rearing CBB and its parasitoids is un- Indian coffee farmers at present. Nonethe- derway in different laboratories. less, it should be highlighted that in the case Cephalonomia stephanoderis has shown of gleaning collection and use of picking some promise in the evaluation tests and mats, the majority of coffee growers did not this parasitoid is now being distributed in identify any real constraints in adopting many CBB affected areas. Also a culture of them. Phymastichus coffea has now been estab-

70 71

lished in India and release and assessment More generally, farmers have a marked dif- studies are underway. Any further develop- ficulty in assessing crop losses due to CBB ment of Beauvaria bassiana as a biocontrol attacks. A participatory approach is needed agent will have to address critical issues to devise farmer-friendly ways of sampling such as mass production techniques and and calculating losses as this will help en- quality control management. The timing of sure the long-term stable adoption of an IPM applications and the right local climate con- strategy ditions are important factors for success and thus these factors combined could limit the usefulness of this agent. 72

REFERENCES Abdul Rehiman, P., and Vijayalakshmi, C.K., 1999, “Influence of rain on the mortality of coffee berry borer”, Insect Environment, Vol. 4, Jan –March 1999. Central Coffee Research Institute, “Coffee berry borer in India”, Central Coffee Research Abdul Rehiman, P., and Vijayalakshmi, C.K., Institute, Coffee Research Station, 1999, “Kerosene as a synergist of endosul- Karnataka, India, 2001, 112 p. fan”, Insect Environment, Vol. 4, Jan-Mar 1999. International Coffee Organisation, “Monograph on Indian Coffee”, London, International Abdul Rehiman, P., and Vijayalakshmi, C.K., Coffee Organisation, 1997, 28 p. 1999, “Evaluation of low volume and high volume application of endosulfan for the Kay, R., “Farm Management: Planning, Control & control of coffee berry borer, Implementation”, Texas A & M University, Hypothenemus hampei (Ferrari), in India”, McGraw Hill Company, United States of Insect Enviroment,Vol. 5, Oct-Dec 1999. America, 1981, 370 p. Abdul Rehiman, P., and Vijayalakshmi, C.K., Leach, A., “Technical Report: Collection of Eco- 2000, “Rearing technique of the berry nomic and Management Data for Model- borer parasitoids imported from Colom- ling the Coffee Berry Borer in India”, Un- bia”, Indian Coffee, March 2000. published, Imperial College, Ascot, UK, 1998, 38 p. Abdul Rehiman, P., and Vijayalakshmi, C.K., 1998, “Efficacy of varying dosages of in- Levin, R. I., Rubin, D. S., “Statistics for manage- secticides against coffee berry borer”, In- ment (6th edition)”, Prentice Hall, New Jer- sect Environment, Vol. 49 (3), 1999. sey, 1994, 918 p. Balakrishnan, M.M., Sreedharan, K., and Bhat, Menon, P., Krishnakumar, R., Subramanian, T. P.K., 1994, “Occurrence of S., “A round-up from States”, Frontline, Vol. entomopathogenic fungus Beauveria 18, No.2, 2001, pp. 4 –6. bassiana on certain coffee pests in India”, Journal of Coffee Research, 24 (1), pp. 33- Muralidharan, S., “The Farming Crisis”, Frontline, 35. Vol. 18, No.2, 2001, pp. 4 –6. Balakrishnan, M.M., Sreedharan, K., and Bhat, Pant, G. B., Rupa Kumar, K., “Climates of South P.K., and Naidu, R., 1994, “Possible use of Asia”, John Wiley & Sons, England, 1997, Beauveria bassiana as a biopesticide 320 p. against the coffee berry borer”, Indian Coffee 58 (8), pp. 43-44. Walpole, R., Myers, R., Myers, S., “Probabilidad y estadistica para ingenieros”, Pearson Balakrishnan, M.M., Sreedharan, K., and Bhat, Educación, México, 1998, 739 p. P.K., 1995, “New fungal associates of coffee berry borer”, Journal of Coffee Re- search, 25 (1), pp. 52-54.

PUBLICATIONS Balakrishnan, M.M., Prakasan, C.B., and Sreedharan, K., 2000, “Biocontrol of coffee berry borer in India”. Paper presented The following is a list of Indian publications on at the International Scientific Symposium the ecology and management of the CBB that on Coffee held at Bangalore, 4th Decem- have arisen out of the Project, and earlier work ber 2000. on the CBB by the Coffee Board of India. Balakrishnan, M.M., and Sreedharan, K., 2001, Abdul Rehiman, P., and Vijayalakshmi, C.K., “Cof- “Evaluation of Metarhizium anisopliae fee berry borer in fruits of tea”, Kisan against coffee berry borer and shot hole World, July 1996, p. 15. borer”, Proceedings of the Second National

72 73

Symposium on Integrated Pest Manage- 19th October 2001, Bangalore, India, 117- ment in Horticultural Crops: New Mol- 118. ecules, Biopesticides. 17th-19th October 2001, Bangalore, India, p. 202. Ramaaiah, P.K., and Krishanmoorthy Bhat, P., 1992, “Coffee berry borer in India”, Indian Balakrishnan, M.M., Prakasan, C.B., and Coffee, Vol. LVI (1), pp. 33-35. Sreedharan, K., 2000, “Introduction and establishment of Cephalonomia Sreedharan, K., Balakrishnan, M.M., Prakasan, stephanoderis (Hymenoptera: Bethylidae) C.B., Bhat, P.K., and Naidu, R., 1994, “Bio- for the biocontrol of coffee berry borer, ecology and management of coffee berry Hypothenemus hampei (Coleoptera: borer”, Indian Coffee, 58 (8), pp. 5-13. Scolytidae)”, Paper presented at the PLACROSYM XIV International Conference Sreedharan, K., Balakrishnan, M.M., 1999, “Cof- on Plantation Crops, 12th-15th December, fee berry borer, Hypothenemus hampei, in 2000, Hyderabad, India. India - Current status and integrated management strategies”, Paper presented at Bheemaiah, M. M., and Dhruva Kumar, H. K., the Colloquium on Post-harvest Technol- 1997, “Spread of coffee berry borer and ogy of Coffee, held at Mangalore Univer- consequential extension programmes”, sity, Mangalore, India. Indian Coffee, LXI (7), pp. 20-93. Sreedharan, K., Prakasan, C.B., Balakrishnan, Bheemaiah, M.M., Sreedharan, K., Ananda Rao, M.M., Samuel, Stephen D., Vinod Kumar, L.V., and Dhruva Kumar, H.K., 1997, “Poly P.K., Rehiman, P.Abdul., and mat as one of the tools in the integrated Vijayalakshmi, C.K., 2000, “Development management of coffee berry borer”, Indian of an Integrated Management Programme Coffee, LXI (11), pp. 19-20. against coffee berry borer in India”, Paper presented at the PLACROSYM XIV Inter- Kumar, P.K.V., Prakasan, C.B., and Vijayalakshmi, national Symposium on Plantation Crops, C.K., 1990, “Coffee berry borer, 12th-15th December, 2000, Hyderabad, Hypothenemus hampei (Coleoptera: India. Scolytidae): first record from India”, Jour- nal of. Coffee Research, 20 (2), pp. 161- Vijayalakshmi, C.K., Abdul Rehiman, P. and 164. Reddy, A.G.S., 1994, “A note on the alternative shelters of coffee berry borer Kumar, P.K.V., Prakasan, C.B., and Vijayalakshmi, beetles”, Journal of Coffee Research, 24 (1), C.K., 1994, “Record of entomopathogenic pp. 47-48. fungi on Hypothenemus hampei (Ferrari) from South India”, Journal of Coffee Re- Vijayalakshmi, C.K., Abdul Rehiman, P., and search, 24 (2), pp. 119-120. Vijayan, V.A., 1997, “Cultural Methods - an effective component in the management Prakasan.C.B., Sreedharan, K., and Sambamurthy of coffee berry borer”, Insect Environment, Reddy, A.G., 2000, “Trap - a promising tool Vol. 5 (3), pp. 70. in the integrated management of coffee berry borer, Hypothenemus hampei (Co- Vijayalakshmi, C.K., Abdul Rehiman, P., Suresh leoptera: Scolytidae)”, Paper presented at Kumar, V.B., and Reddy, A.G.S., 1999, “In- the PLACXROSYM XIV, International Sym- cidence of coffee berry borer and control posium on Plantation Crops, 12th-15th measures in India”, 9th Swadeshi Science December, 2000, Hyderabad, India. Congress, Kerala, 5th – 7th November, 1999. Prakasan, C.B., Sreedharan, K., Sambamurthy Reddy, A.G., and Gokuldas, M., 2001, “Mass Vijayalakshmi, C.K., Abdul Rehiman, P., 1999, trapping- a new component in the IPM of “Trapping of Hypothenemus hampei coffee berry borer”, Presented at the Sec- (Ferrari) by retaining coffee crop near dry- ond National Symposium on Integrated ing yard”, Insect Environment, Vol. 5, Oct- Pest Management in Horticultural Crops: Dec. 1999. biopesticides and new molecules. 17th- 74

Vijayalakshmi, C.K., Abdul Rehiman, P., borer”, registered under the University of Prakasan, C.B., and Sreedharan, K., 1999, Calicut, Kerala. “Record of a new species of Hypothenemus on Gliricidia maculata”, Insect Environ- Vijayalakshmi, C.K., “Bio-ecology and management, Vol. 5, Oct-Dec.1999. ment of coffee berry borer in India”, degree awarded by the University of Mysore, Vijayalakshmi.C.K., Abdul Rehiman, P., and Mysore, Karnataka. Vijayan, V.A., 2000, “Observation on the position of attack of coffee berry borer, Hypothenemus hampei (Ferrari)”, Insect Video documentaries Environment, Vol. 6 (3). Short video film on CBB in English and vernacu- Ph.D. Theses lar languages (Kannada, Malayalam and Tamil). Prakasan, C.B., “Economic threshold and eco- friendly management of coffee berry

74 75

APPENDIX 1:

Coffee Board of India infrastructure and main personnel engaged in the Project

n the early 1940s the Indian Board for coffee market. This liberalisation requires the Expansion of the Coffee Market was the Board to focus on research and devel- Icreated and later renamed the Coffee opment, quality control, market informa- Board of India, renamed again in 1955 as tion, promotion, extension and development the Coffee Board (ICO, 1997). The Coffee of a free market. Board has several functions: Extension offices are dispersed throughout To promote sales and coffee con- all coffee regions giving technical advice sumption in India and around the according to the farmers’ requirements. In world total there are 45 extension units covering To commercialise coffee through a an average of 5,800 ha each. Most exten- common fund (this function has re- sion activities are related to technical sup- cently changed) port, farm visits, transfer of technical knowl- To research coffee with the aim of edge through rural meetings and other me- developing new production tech- dia. The extension service has a set of model nologies coffee plots to carry out demonstrations To offer technical advice in order about correct practices and new technolo- to improve coffee farming with em- gies. Other services provide seeds of new phasis on smallholders varieties, link farmers with research, fore- To improve coffee quality cast harvests, as well as estimating the inci- To manage and lead the coffee sec- dence and importance of pests and diseases. tor at both national and international levels The research studies on the CBB were initi- After the market liberalisation the Coffee ated during 1990 at the Regional Coffee Board reduced its role in coffee Research Station, Chundale, which is situ- commercialisation. Thus coffee farmers are ated in the berry borer affected region in now free to sell their coffee in whatever way Kerala State. Later on, studies were con- they want. However, the Coffee Board has a ducted at the Coffee Research Sub-Station, set of responsibilities that it carries out to Chettalli, Kodagu District, Karnataka and benefit the coffee sector. the Regional Coffee Research Station, Thandigudi, Tamil Nadu as the CBB spread The president is the executive director of to these regions. The personnel involved in the board and he reviews all structural various studies are as follows: changes resulting from the opening of the

78 79

APPENDIX 2:

Overseas visits to attend meetings, training and conferences on CBB 80

APPENDIX 3: Visits of consultants to India under the CFC-ICO project

80 81

APPENDIX 4: Earlier Indian studies on CBB ecology and cultural control

ultural and phytosanitary measures The data showed that a 26-day delay in har- are known to be effective in the man- vesting (beyond the period when 60% of the Cagement of CBB in various countries. fruits were ripe) caused about a 35% in- Hence, experiments were conducted to crease in CBB incidence. evaluate the effect of various components of cultural and phytosanitary measures In another study in Kodagu, the effect of such as timely harvest, gleaning collection, delayed harvest on the coffee berry borer etc., on CBB ecology. Some of this work is build up (‘flare-up’) on the standing crop was reported here. studied by comparing the infestation in crops harvested on time, and in standing Effect of delayed harvest on the popu- crops allowed to over ripen and dry on the lation build up of CBB plant (in the same estate). A 17-day delay in harvesting resulted in about 6-fold in- CBB infestation in the field was enumerated crease of fruit infestation. In a study in at fortnightly interval from December to Wayanad, the level of infestation on new February (i.e. until harvest) on an estate in crop, and berry drop due to CBB, were re- Kodagu and the pest build up recorded until corded from timely and delayed harvested the coffee was harvested. The data re- estates. The data are presented in Table corded are presented in Table A4-1. A4-2. 82

The data indicated that the percentage berry were evident in fallen berries, followed by infestation and fruit drop due to coffee berries on the plant. Further, the off-sea- berry borer was higher in late harvested son berries showed more number of eggs estates than in timely harvested estates. compared to other types of berries. These observations revealed that CBB infestation could build up to very high levels if harvest- Study on CBB build-up on a non-har- ing is not completed on time. vested estate

A study was carried out on Rippon estate, Field study on the impact of gleaning Arappatta, where harvesting was not done on the build up of CBB until May 1996, due to unavoidable reasons, to assess the build up of incidence. Differ- A replicated field trial was conducted to ent types of berries, i.e. berries from the study the impact of infested fallen berries plant, fallen berries and off-season berries, on the build up of CBB in the new crop, by were collected from the infested field at dif- placing different numbers of infested ber- ferent locations and examined in the labo- ries under caged plants. The resultant in- ratory. festation in the new crop was recorded after one month. The experiment was con- The counts of the infested and non-infested ducted for three seasons and the results are berries were recorded. Among the three presented in Table A4-3. types of berries examined, the highest percentage of coffee berry borer infestation was It is evident from the data that the incidence recorded on fallen berries (97.60%) followed of CBB in the new crop significantly in- by the berries on the plant (70%) and off- creases with increase in the number of in- season berries (13%). Furthermore, the fested fallen berries present on the ground counts of different stages of coffee berry (gleaning). borer were also recorded by cutting open 25 berries of each type. The observations In another study, data on percentage fruit showed that the maximum number of adults infestation by CBB in the new crop was re-

82 83

corded from ungleaned and gleaned estates. on an experimental basis, and it resulted in In this study, three ungleaned plots were reduction of pest incidence in the new crop compared with a gleaned plot. In each test in the village. As a result of this, use of pick- plot 10 plants and four branches per plant ing mats gained wider acceptance among were taken to arrive at the infestation level. the growers and the Coffee Board contin- There was a marked difference in infesta- ued to supply the picking mats at subsidised tion level in gleaned and ungleaned estates. rate. The average infestation on the three ungleaned estates was 13.56% compared to In order to quantify the effect of using pick- that of 1.49% on gleaned estates. ing mats on CBB incidence, a replicated trial was carried out at four locations in Use of picking/harvesting mats Wayanad. At each location, two sets of treatments, i.e. harvesting with picking mats Removal of fruits that fall on the ground spread under the plant and without pick- (gleaning) and those remaining on the plant ing mats, were imposed. The number of (left over) after harvest is the most impor- fallen berries on the ground after the har- tant practice for the management of CBB. vest in the treatment plots was assessed by taking counts of fallen berries in a quad- Manual gleaning is often laborious and ex- rate of one-foot size. Counts of 5 quadrates pensive. In this context, harvesting coffee each was taken from 20 sites in each loca- with mats spread under the plants was tion and the mean number of fallen fruits thought to be an effective method to reduce in a square foot area was calculated. The gleaning in coffee fields. This method was results are presented in Table A4-4. Further, demonstrated in Kodagu in 1993. The Cof- the infestation of CBB in the new crop in fee Board supplied picking mats to all the the experimental plots was assessed, and small growers in Byrambada village, Kodagu, the data is given in Table A4-5. 84

The data showed that the use of picking In order to quantify the beetle population mats helped in collection of fruits falling in the left over berries, a study was con- on the ground thereby reducing the glean- ducted on three estates on Pulney Hills, ing by 72.13%. This has resulted in the re- Tamil Nadu for two years. From each plot, duction of CBB infestation in the new crop 100 left over berries each were collected by 71.47% compared to that in the plot har- from five sites in May and the berries clas- vested with out picking mats. sified, based on the number of beetles present in each berry. The percentage of Finally, the data on the picking efficiency, berries containing different ranges of beetle quantity of gleaning collected and the labour populations (e.g. zero, 1-5, 6-10) was worked requirement for gleaning with and without out. Results are presented in Table A4-7. the use of picking mats were collected from the above mentioned trial locations and the The data show that beetles were present in same are given in Table A4-6. all the berries except in one case in a plot during the year 2000 in which beetles were The use of picking mats increased the pick- absent in 1% of the berries examined. The ing efficiency of workers by 48.56%. The population level of more than 10 beetles per amount of gleaning was reduced by about berry was recorded from 72.00% and 89.67% 58% when the harvest was done using the of berries examined during 2000 and 2001 picking mat. As a result the manpower re- respectively. The percentage of berries con- quired for gleaning was also reduced by taining more than 50 beetles per berry was 43.33%. 13% and 28.33% during the years 2000 and 2001 respectively. It is evident from this Beetle population in left over berries study that the left over berries harbour large numbers of beetles, which could form the Berries left on coffee plants after the har- main source for carrying over the inoculum vest are found to harbour large numbers of to the new crop. Hence, removal of these beetles, which can then infest the new crop. berries is essential to reduce pest build up.

84 85

Effect of hot and cold water treatment minutes was sufficient to kill all the stages of infested berries on CBB inside the berry.

Hot water treatment Cold water treatment

The studies were conducted at RCRS, An experiment was conducted to study the Chundale during 1994. Infested fruits were effect of immersing infested berries in wa- dipped in boiling water for 0.5, 1.0, 1.5, 2.0, ter for 12, 24, 36 and 48 hours. Each treat- 2.5 and 3.0 minutes. There were six treatment was replicated four times with 5 kg of ments and each treatment was replicated fruit per replication. The percentage mor- three times with 25 kg of fruit per replica- tality was assessed after cutting open 250 tion. Mortality was calculated after cutting treated fruits per replication. The data are open 500 treated fruits from each replica- presented in Table A4-9. The results indi- tion. The data are presented in Table A4-8. cated that 100% of adult and larvae were The data revealed that dipping infested killed in 48 hours of immersion in cold wa- fruits in boiling water for a period of 2.0 ter.