REGIONALREPORTS Harnessing Butterfly Biodiversity for Improving
Livelihoods and Forest Conservation: TheKipepeo Project IANGORDON
WASHINGTONA YIEMBA The Kipepeo Projectisn
co111m11nity-bnsedbutterfly farmingprojectonthe margins of
Arab11ko-Sokoke Forest onthe north canst of Kenya. This f orest is
n globnllyimportant forest for biodiversityconservation.Inthe
early1990s, 54% to 59% of the local co11111111nitywanted the entire
forest cl enred for settle-ment nnd tire forest wns i11vnded by
farmers onseveral occasions.The Kipepeo Project wns set 11pto
c/rn11ge co1111111111ity nttit11des to tireforest by giving them n
stake initsconservation. Kipepeo trained farmersliving next tothe
forest to renr forest b11tterflies.B11tterflypupae were p11rclrnsed
fro111the farmers for exporttothe live b11tterfly exhibit
ind11stryinE11rope nnd tire United States. C1111111lntivecomm11nity
earnings fro1111994 to2001exceeded $130,000 with
significantpositive effects onbothlivelihoodsnnd attitudes.The
projecthns been
finmrcinllyself-s11stniningsince1999.B11tterflymonitoringindicates
tlrntthere hnve bee11no adverse effectsonwild butterfly
populations. Keywords: Arab11ko-Sokoke Forest; b11tterfly
farming;inco111e generation; 11onti111berforest products;
forest-ndjncent communities; benefit shnri11g Arabuko-Sokoke Forest
(ASF)is an island of unique biodiversity in a sea of human poverty.
Lying on the north coast of Kenya and occupying an area of 42,000
hectares, it is the largest remaining protected fragment of a
forest mosaic that once stretched from southern Somalia to northern
Mozambique. Because of its relictual status (Burgess, Clarke, &
Rodgers, 1998), it is also one of the most important biodiversity
sites in Kenya. It has been ranked as the second most important
forest for threatened bird conservation on mainland Africa (Collar
& Stuart, 1988) and is one of 19 Important Bird Areas (IBAs) in
Kenya that have been prioritized as
criti-calsitesforintensiveandimmediateconservationaction(Bennun&
fo11r11nlof E11 viro11111e11l&Develop111e11 t, Vol.12, No. 1,
March200382-98 DOI: 10.1177 /1070496502250439 2003 Sage Publica
lions 82 Schwartz IIMPACT OF STATECAPACITY81 Skocpol, T.(1990).
Bringing the state back in: Strategies of analysis in current
research. In P.Evans, D. Rueschemeyer, &T. Skocpol (Eds.),
Bringing the state back in (pp. 3-43). New York:Cambridge
University Press. Smil, V.(1998).The e11viro11111entoutlook
forChinn.Unpublished manuscript. State Environmental Protection
Administration. (1998).Selected e11viro11111e11tnlstandards of
thePeople'sRepublic of Chinn(1979-1997) . Beijing:Author. State
Environmental Protection Administration. (n.d.) Report 011the state
of tile e11viro11111e11t
inChinn2000.RetrievedDecember5,2001,fromhttp:/
/www.zhb.gov.cn/english/ SOE/ soechina2000 I english/waterI
water_e.htm State plan for protection of the ecosystem good for
all. (1999, February 15). T/1e Chinn Daily. State Statistical
Bureau. (1987-1997). Zhong Guo To11gji Nin11jin11[State statistical
yearbook of China] . Beijing: Zhong Guo Tongj i Chuban She. Stepan,
A. (1978). The state and society: Peruin co111pnrntive perspective.
Princeton, NJ:Prince-ton University Press. Taihu's water quality
improves. (1999, January 8).T/1eC/1i11nDaily. Toomany nutrients
harm lake. (1998,December 3).The C/1i11nDaily. Turner, J. L.,
&Wu,F.(2001).Development of environmental NGOs inMainland
China, Taiwan and Hong Kong. In Proceedings ofGreen NGOn11d
E11viro11111e11tnl ]oumnlist For11111 (pp. 1-2). Washington, DC:
Woodrow Wilson International Center for Scholars. World Bank.
(1997).Clear water, blue skies: China's enviro11111entinthe new
century. Washing-ton, DC:Author. World Health Organization. (1993).
Guidelines for drinking water quality (2nd ed., Vols. 1 and
2).Geneva: Author. Wu,F.(2002).New partnersor old brothers?GONGOs
intransnationalenvironmental advocacy in China.
ChinnE11viro11111e11tSeries, 5, 45-58. Young, 0.
R.(1979).Co111plinnce and public nut/10rity. Baltimore: Johns
Hopkins University Press. Yu,J.,&Yang, J.
(2001).NationaleffortstotacklepollutioninTni/111Lnke.RetrievedMay
12, 2001,fromhttp:/ /www.chinaenvironment.com/ english/
channel/pollution/waste/ taihu.html
YunnanShengTongjiJu.(1986-1996).Y111111n11To11gjiNin11jinn[Statisticalyearbookof
Yunnan].Beijing: Zhong Guo Tongji Chuban She. ]011nthn11Schwartzis
nssistn11t professor of political science nttile State University
of New York nt New Paltz where he teaches courses in
enviro11111e11tpolitics n11di11temntionnl relations.HisP/1.D.
i11political scie11ce is fro111theU11iversity of
Toro11to.Hisresearchinterests include evnl11nti11gthe effect of
various factorsi11J111e11ci11gpolicy e11force111ent,specifically,
the role of civil society groups. Hisregional focusisdeveloping
cou11triesintheAsiaPncific/Chi11n. Gordon, AyiembaITHE KIPEPEO
PROJECT83 Njoroge,1999).ASFis part of the East African Coastal
Forest/Eastern Arc Forest complex that ranks among the top 25
biodiversity hotspots on Earth (Myers, Mittermeier, Mittermeier, da
Fonseca, &Kent, 2000). The forest is home to six globally
threatened bird species and an additional five bird species that
are coastal endemics. Of the six globally threatened
species,two,theSokokeScopsOwlandClarke'sWeaver,areeach known only
from ASF and one other site (East Usambara Mountains in
TanzaniaandtheDakatchaWoodlands,respectively) . Afurthereight bird
species (of a total of 270 species) found in ASF are regionally
threat-ened. ASF also has an exceptionally diverse amphibian fauna,
including
thecoastalendemicBunty'sToad.Therearethreerarenear-endemic
mammals(Ader'sDuiker,Golden-RumpedSengi,andtheSokoke Bushy-Tailed
Mongoose).A small population of around 100 elephants
livesintheforest.Therearesixtaxaofbutterfliesthatarecoastal
endemicsincludingone(Charaxesblanda)thatisvirtuallyaforest endemic.
An unknown number of other invertebrate species must also be forest
endemics. The forestissurrounded on all sides by village
communities. There are 51villages actually bordering on the forest
and having a population of about 110,000 people. These people are
predominantly of the Giriama tribe who settled in the area west of
the forest more than 100 years ago and moved on to areas east of
the forest in the 1950s and 1960s. Average household size ismore
than 13,and 55%of the households consist of multiple families. The
population density ofKilifi District has risen from 47 to 60 people
per km2 between 1989 and 1997 (Government of Kenya, 1997). The
original population of the surrounding area belonged tothe
hunter-gatherer Sanyatribe.Prior tothe protection (gazettement)of
the forest by the colonial government of Kenya in 1932, they used
it freely for their subsistence needs. A small group of Sanya still
lives on the northern side of the forest.Although they grow maize
and other crops, they still
huntintheforestandarefrequentlyinvolvedinillegalcuttingof polewood.
Subsistenceagriculture isthemainoccupationofthesurrounding
population. This includes the production of maize, cassava, and
beans, with income supplemented by cash crops such as cashew,
mango, and coconut. Agricultural land is generally poor, and crop
yields are low. The mean size of farm holdings is 6.9 ha (0.5 ha
per capita), with farms grow-ing an average of 1.6 ha of maize.
Most households own goats (average of fiveper
household),buttsetsefliesandalackof grazingare con-straints
tocattle keeping.Although many uses of the forest for subsis-tence
or income generation are illegal,they still continue. Forest usage
includes collection of water, fuelwood, poles, and herbs; butterfly
farm-ing;andhunting of wildlifeformeat.Participatoryassessmentswith
localcommunities indicatethat building polesareperhapsthe single
most important product used from the forest. Communities on the
west-84JOURNAL OF ENVIRONMENT &DEVELOPMENT em side of the
forest rely on it more heavily than those on the east due to
differences in availability of trees on private land and a greater
degree of comparative affluence on the east. In 1991the average
declared household income in the forest-adjacent community was
estimated at KSh 17,300, giving a per capita income of only
KSh1,470($20at current rates)(Mogaka, 1991). In 1993,an
inde-pendent survey along the wealthier eastern margin of the
forest gave a per capita income of KSh 2,728 ($40)(Maundu, 1994).
By any standards, this is an impoverished community, and it is not
surprising that its mem-bers have had little concern for the
conservation of Arabuko-Sokoke For-est. Instead, the forest has
been seen as the source of many of their prob-lems. In 1991, 96% of
the farmers (N = 32) were unhappy with the forest, and 54% wanted
it completely cleared for settlement (Mogaka, 1991). In a
follow-upsurvey in 1993 (Maundu, 1994), 59% (N = 142)wanted the
whole forest cleared. The reasons for this discontent will be
familiar to any conservationist workingwiththelegacyof
state-controlledforestsinthedeveloping world: poverty,
resourcedenial,wildlifecropraiding, and hunger for land (Maundu,
1994; Mogaka, 1991). Because of poverty, the local com-munities
were forced to rely on the" free resources" of the forest, yet
their legal access to these resources was strictly limited by the
forest depart-ment and the status of the forest as a gazetted
(protected) forest reserve. To gain access, they had to risk arrest
and beatings and/ or pay bribes to corrupt forest guards. Poverty
was worsened by wildlife crop raiding by animals from the forest
(mainly elephants and baboons), causing consid-erable losses of
maize, cassava, mangoes, and coconuts. Elephants also caused
deaths, injuries, and destruction to property and were the single
most important cause of hostility to the forest.There is no
government compensation for loss of crops or injuries caused by
wildlife; only in the event
ofdeathisanycompensationpaid,andthenonlyKSh30,000 ($400), and that
only after persistent claims. Hunger for land resulted in forest
invasions in 1993-1994, led by local councilors, and tosustained
campaigns forthe removal of government protection (degazettement)in
two areas of the forest (Kararacha and Mida). The following
quotations (Maundu, 1994) give a good impression of community
feelings about the forest in the early 1990s: This is Government's
forest, you cannot get inside, if you are caught there with even a
grass twig you are arrested. You receive a thorough beating even
when goats are found browsing at the edge of the forest. It is not
useful to me but tothe Government because it benefits from it.
Elephants have made us poor here. We don' t plant here because of
elephants. Gordon, AyiembaITHE KIPEPEO PROJECT85 The land is
frustrating because whatever you plant is destroyed. Those of us
who live near the forest endure a lot but this is unknown to the
authorities. It [the forest]is a saving for farm land. (p.34) The
situation at Arabuko-Sokoke provided the context and the ratio-nale
for Kipepeo, a project administered by the East Africa Natural
His-tory Society (EANHS, now Nature Kenya [NK]) in partnership with
the National Museums of Kenya (NMK).Kipepeo was started in 1993
with $50,000fromtheGlobalEnvironmentFacility(GEF)SmallGrants
Programme. Its objective was to change local attitudes by enabling
the forest-adjacent community to get cash incomes from rearing
forest but-terfliesforexporttothelive butterfly exhibit industry
inEurope and
America.Thisindustrydependsontheimportationoflivebutterfly
pupaefromtropicalbutterflyfarms.Theadult butterfliesthateclose
fromthese pupae are flown in simulated natural surroundings in
pro-tected enclosures, and members of the public pay gate fees to
view them. From small beginnings in Britain and Japan in the late
1970s, the live but-terfly exhibit industry has grown to become a
multimillion-dollar global enterprise. In the early 1990s, there
were no large-scale suppliers of Afri-can butterflies forthe
exhibit market.
Butterflyfarmingisparticularlywellsuitedasanincome-earning
ventureforforest-adjacentcommunities.Itrequiresrelativelylittle
investment,usessimpleequipmentandmaterialsthatareusually locally
available in one formor another,andthe basic skills are easily
learned.Moreover, because the resource base forbutterfly farming(in
terms of both butterflies and foodplants) depends on the continued
pres-ence of forest habitat, butterfly farmers experience a daily
and compel-ling linkage between their livelihood and forest
conservation. Butterfly farming also has spin-offs for ecotourism
through the publicity and curi-osity that it generates
(particularly when first introduced) and through the opportunity it
provides for living and attractive displays of its own operations
and of forest biodiversity. Project History It was decided to work
with those farmers who lived immediately on the eastern forest edge
because they were easily accessible, nearest the resource, suffered
most fromcrop raiding, and knew most about what wasgoingon
intheforest.PatrickMaundu,an NMKethnobotanist,
organizedmeetingsthroughthelocalchiefswheretheprojectwas explained.
The households immediately next tothe forest were identi-fied.
Maundu then visited each of these households, conducting a
ques-86JOURNAL OF ENVIRONMENT &DEVELOPMENT tionnaireand
recruiting participants.He interviewed 142households, and 133 of
them agreed to participate (Maundu, 1994). Information on
theforestbutterfliescamefroma9-monthinventory andmonitoring
master's degree research project (Ayiemba,1997).Thisshowed which
species were most abundant, where, and when and provided valuable
baseline data on butterfly abundance before the project began.
Between Juneand December1993,thegroundworkforthe project was
completed.Butterfly speciestargeted forexport were selected on
thebasisof likelydemand,seasonalavailability,and easeof rearing.
Local larval food plants were determined and seedlings grown in a
nurs-eryattheprojectheadquartersatGedeRuins.1 Breedingtechniques
were developed and fieldtested using 25volunteers fromthe
commu-nity.A large flight cage and a breeding shed were built. A
contract was
negotiatedwithaU.K.entomologicaldealer,andexportprocedures were
developed in consultation with the Kenya Wildlife Service (KWS)
andacouriercompanyinMombasa(Gordon,1994).Government
involvementinthe projectwasrestrictedtograntingapprovalforits
operations through the District Development Committee and the
issu-ing of export and forest access permits by KWSand FD.
ExportsstartedonscheduleinFebruary1994,8monthsafterthe receipt of
funds to start the project, and by the end of 1994, Kipepeo had
exported more than 10,000 pupae belonging to 14 species, earning a
little under U.S. $16,000 in revenues. The next 4 years saw steady
growth in exportsandrevenues,withmore23,000pupae exportedin1997and
more than $41,000 earned by the project (Figure 1). El Nino rains
starting in 1997 halted growth in 1998, but 1999 was a record year,
with almost 55,000 pupae of more than 30species exported and more
than $81,000 earned. Exports peaked in 2000with gross earnings of
$99,320.Bythe end of 2001, cumulative export earnings had passed
$400,000, more than eight times the initial grant that set upthe
project, and annual income had grown sixfoldsince1994. The project
has been financially self-sustaining since 1999 (see Table 1 for a
breakdown of costs and income in 2000and 2001) . The number of
butterfly farmers increased from 133 in 1994 to 700 in 2002, and
they now completely surround the forest. Limits on the number of
butterfly farmers are set by the market rather than by resource
scarcity, and there are currently no plans to further expand the
number of farmers unless new markets are identified. In 1994, a
third of the pupae were produced by Kipepeo staff members at the
project headquarters at Gede Ruins, whereas by 2000 all of prod
uc-tion was community based. The average price paid per pupahas
also increased so that community earnings have grown fasterthan
those of 1. Gede Ruins is a national monument site containing the
ruins of a 13th-century Swa-hili fortified settlement and a small
coral-rag forest. It belongs to the National Museums of Kenya and
lies 18 kilometres south of the tourist resort of Malindi. Gordon,
AyiembaITHE KIPEPEO PROJECT87 19941995199619971998199920002001 Year
Figure 1:Kipepeo Export Earnings: 1994-2001 Table1 Financial
Breakdown of Butterfly Export Business (2000-2001) Item Number of
shipped pupae Invoice value of shipped pupae (US$) Value of viable
shipped pupae (US$) Loss rate (%$) Exchange rate (KSh/US$) Earnings
from viable shipped pupae (KSh) Production costs for pupae:
purchases (KSh) Production costs for pupae: operational (KSh)
Production costs for pupae: salaries Profit from pupae business
(KSh) 2000 56,023 113,518 99,320 12.51 73.56 7,306,227 2,806,415
2,080,248 1,221,270 1,198,294 2001 50,494 103,471 94,055 9.10 76.30
7,176,432 2,529,666 2,042,078 901,430 1,703,258 Note: The loss rate
is expressed in terms of lost income in dollars and arises because
some of the pupae do not survive the transit process and no
payments are received for these. Pro-duction costs include payments
tothe community forthe butterfly pupae they produce together with
all other costs incurred by the normal operations of the butterfly
export busi-ness.AsacommunityprojectoftheEastAfricaNaturalHistory
SocietyandNational Museums of Kenya, butterfly earnings are tax
exempt. the project (Figure 2).In 1994 they amounted to just over
KSh 0.26 mil-lion (about $5,000 at that time). In 2000, annual
earnings were just over
KSh2.8million($38,000atthattime),atenfoldincreasesince1994.
Cumulative community earnings from1994to2001exceededKSh10 million
($130,000 at today's rates). The growth in community and project
earnings from 1994 to 1999 was
madepossiblebyanexpandingmarket,particularlyintheUnited 88JOURNAL
OF ENVIRONMENT &DEVELOPMENT 2,000,000+-------------------! .c
en 1,500,000+-------------------! c: .E "'
UJ1,000,000+-------------------! 199419951996199719981999 Years
Figure 2:Community Earnings From Butterfly Sales: 1994-2001
20002001 States. In 1994, Kipepeo exported to a single client. By
1999 there were 11 major customers, two based in the United Kingdom
and accounting for morethan 60%of salesandtherest inthe
UnitedStatesand Canada. Since 1999, however, markets have been more
or less static. In 2001they were impacted by the foot and mouth
epidemic in Britain andthe
Sep-tember11thatrocityintheUnitedStates,bothof whichreducedthe
numbers of tourists.Despite the setbacksthat periodically affect
tour-ism, butterfly exhibits have proved to be enduring public
attractions and are now a standard feature of the tourist industry
from Asia to the Amer-icas.The majorityare owned by the private
sector,although some are run by museums and nonprofit
organizations. They have attracted sub-stantial private and
corporate investment: A highly successful exhibit at Niagara Falls
was constructed in 1996 at acost of $11million and was profitable
in its first year of operation. ASelf-Evaluation This brief history
of Kipepeo raises several questions. How much has it added value to
Arabuko-Sokoke Forest? Has it affected local attitudes to the
forest? How much has it succeeded in improving community
live-lihoods? Have there been any unanticipated negative effects?
Have the earnings been evenly spread? Have the socially and
economically
disad-vantagedsectionsofthecommunitybenefitedfromKipepeo?Have there
been any other spin-offs for conservation? Has butterfly
harvest-ing for farming operations affected wild butterfly
populations in the for-Gordon,AyiembaITHE KIPEPEO PROJECT89
est?IsKipepeo sustainable? Can it be replicated? Some of
theseques-tions are more easily answered than others. ADDED VALUE
Kipepeo has certainly added value to the forest. Table 2 shows
official institutional revenues generated from the forest from 1996
to 2001. Over this 6-year period, butterfly sales account for 80%
of all recorded reve-nues from the forest, with 15% coming from
forest department licenses, fines, and royalties for timber
products; 3% from tourism; and 2% from beekeeping. This could be
viewed as further evidence for the high value of non timber forest
products but is best seen as an artifact of unrecorded
andillegaloff-take(particularlyofBrachylaennforcarvingand polewood
for house construction), low government royalties for timber
products(especiallyfuelwood),andaforestfromwhichmostofthe
commercially valuable timber has already been removed. One
inciden-talconclusion, however, isunaffected by thiscaveat:
Tourismisa low earner in a forest where the average visitor sees
little but trees. ATTITUDES AND LIVELIHOODS Kipepeo has had an
effect on both attitudes and livelihoods, but it is hard to judge
how great this effect has been.In 1997, Patrick Maundu repeated
hissocioeconomicandattitudinalsurveyof 1993(Maundu, Sojah,
&Kilili, 1997).He foundthat the proportion of Kipepeo farmers
wanting the entire forest cleared and settled had fallenfrom 59% to
16% andthatbutterflyearningscontributedsome73%offarmers'cash
incomes from farm products. In 1999, the average butterfly earnings
of a subset of 217 Kipepeo farmers had risen to KSh 5,854, more
than twice the per capita income recorded in 1993(KSh 2,728). The
problem is of course that Kipepeo farmers constitute a small
frac-tion of the forest-adjacent community(at best, including
family mem-bers, some 8% of the 110,000 people in the 18
sublocations next to the for-est). For this reason, an independent
assessment by the African Wildlife Foundation concluded that
Kipepeo had had little penetration although its impact had been
significant for those who participated in the project. It is clear
that Kipepeo alone cannot solve the problem of local hostility to
the forest, but at another level, its impact has been greater than
is at first apparent. Since Kipepeo started, the local commwtity
has been able, for the first time, to enjoy legal cash revenues
from the forest. The awareness of this, aided by favorable local
press coverage, has spread beyond the target group of Kipepeo
farmers and has influenced public perceptions of the forest, not
least in the consciousness of the district administrations. A more
concrete demonstration of impacts on attitudes has emerged
fromthefarmers'actionsinresponsetothethreatenedexcisionat 90JOURNAL
OF ENVIRONMENT &DEVELOPMENT Table 2 Recorded Revenues From
Arabuko-Sokoke Forest (1996-2001) Item Kenya Wildlife Service
tourism Forest department revenues Beekeeping Butterfly sales Total
KSh 911,653 4,785,753 805,106 25,794,566 32,297,078 % 2.8 14.8 2.5
79.9 Kararachaandthe1994squatterinvasionofthisarea.Someofthe
Kipepeo farmers protested to the local district commissioner
against the
campaignfordegazettement.OneoftheKipepeoSelf-HelpGroups2 (formed
with the assistance of the project) delivered a protest letter to a
presidentialcommissionwhen itvisitedtoinvestigatethematter.In
December 1997, a spokesman for the farmers went to the press,
saying in theDailyNation3 that the forest should be left alone and
that any exci-sions would deprive them of their butterfly farming
income. President Moi has since made it clear that there would be
no excisions at Arabuko-Sokoke. There were several forces behind
this decision, including
vigor-ouslobbyingbyNairobi-basednongovernmentalorganizations,but
thereisawidespread belief onthegroundthat communitydivisions (pro-
and antidegazettement)related to Kipepeo greatly weakened their
claims on the forest and the case for degazettement. This has had
unfortu-nate consequences in the form of death threats to Kipepeo
farmers from community members who supportedthe
degazettementcampaign and who also suspected that the butterfly
farmers were informing the forest management about poaching
activities in the forest. In terms of livelihoods, an important
issue, concealed by average and total earnings is the distribution
of income within the participating indi-viduals and households.
Figure 3 shows this distribution of the subset of 217farmersin
1999.It isregrettably but inevitably J-shaped,broadly reflecting
the amount of effort each individual/household put into but-terfly
farming. Ten percent of the group earned nothing at all,and 53%
earned less than KSh 3,000. Nineteen percent earned between 3,000
and 6,000shillings, whereas 4%earned more than KSh 27,000.The
biggest earners were the representatives of the self-help groups
who are respon-2. Eight Kipepeo self-help groups were formed by the
farmers in 1996. Each group had
itsownconstitution(broadlydedicatedtoconservationanddevelopment)andbank
account and was registered withthe Department of Social Services.
Kipepeoholds back KSh 5 foreach butterfly pupa purchased fromthe
members of each group and pays the accumulated amount into the
self-help group savings account at the end of the year. 3. The
Daily Nation is Kenya's most widely read newspaper with a
circulation of 180,000 and an estimated readership of 3.8 million.
Gordon, AyiembaITHE KIPEPEO PROJECT91 0-30003001-6000600
19CXlO9001-1 200012001 -1500015001-1600016001-21
00021001-2400024001-27000>27000 KSh earned Figure 3:Frequency
Distribution of Farmer Income Levels in 1999 (N = 217Farmers)
sibleforcollectinganddistributingthepayments forthepupae pro-duced
by their groups and who got KSh 5 per pupa for doing so. The top
farmer, who was a representative, earned KSh 78,825 ($1,120). This
sub-set of farmers was not characterized for social status and
wealth ranking, so the data do not address the question of benefits
to the disadvantaged, although it isworth noting that
thetopfarmer,Kidhuku Baya,isdis-abled and has concentrated on
butterfly farming precisely because it is physically undemanding
work. OTHER BENEFITS The major other benefits of Kipepeo have been
in awareness raising
andeducation.Morethan10,000visitorstotheprojecthavelearned about
the forest and its adjacent communities. For most of them this was
their first exposure tothe biological uniqueness of the coastal
forests of Kenya and tothe problems of forest-adjacent communities
(most nota-bly with forest wildlife). More than 50 schools and
polytechnic and uni-versity groups have also visited the project.
Through the efforts of a
Brit-ishvolunteer,TansyBliss,Kipepeoestablished17WildlifeClubsin
forest-adjacent schools, and the project has served as a focus for
some of their activities. The Kipepeo farmers and their families
have also learned about the forest and its global significance as
well as about the butterflies and their foodplants. They now have
an easy familiarity with many of
theLatinnames,andtheirin-depthexperienceinrearinghastaught them
about insect parasites and diseases.Many of thechildren know more
about insect life histories than the average zoology undergraduate
in the country's universities. On a wider front,three short videos
on the project have been broad-cast on international television
(one on more than 80 different TV chan-nelsworldwide),andradio
interviews have been airedontheKenya 92JOURNAL OF ENVIRONMENT
&DEVELOPMENT BroadcastingCorporation,4
BBCWorldService,StudioBrussels,US Public Broadcasting Service, and
the Voice of America African Service. There have been numerous
articles on the project in magazines and the national and
international press. In 1998, Kipepeo was 1of10 winners of the
Dubai International Award forBest Practices in Improving the
Liv-ing Environment,5 and this led to further publicity in the
international print and electronic media. Talks on the project and
the forest have been
giveninKenya,Uganda,Dubai,Britain,CostaRica,andtheUnited
States.Allofthisfreepublicityhashelpedtoraiseawarenessabout
Arabuko-Sokoke Forest. The cost of the airtime alone would far
exceed the money raised forthe project from donors. SUSTAINABILITY
AND REPLICABILITY
ThesustainabilityofKipepeodependson(a)maintainingand
improvingproductionlevels,(b)expandingmarkets,and(c)the resource
base.Current plans aretoincrease production levelsinpace with
expanding markets, but this is an awkward balancing act that leads
to disappointment for the client or the community when it fails.
The big-gest problem is the vulnerability of butterfly operations
to unfavorable
weather.TheeffectofElNinoonproductionwasnotedearlierand resulted
from excessive rains. There was no dry season in 1997-1998, and the
normal boom in production in May and June did not take place in
1998. Figure 4 shows monthly production levels in the 2 years: That
for 1999 is normal, and that for 1998 aberrant. It is likely that
the more or less continual rainsthrough the intervening dry season
allowedpathogen and parasite levels to persist through the dry
season months.Another factor may have been the lack of a nutrient
flush in the forest at the start of the rains in May and June. If
too much rain isa problem, too little is worse. If there were two
or three dry years in succession, Kipepeo would needsupport tostay
afloat.Butterfly farmingisjustasvulnerableto weather as is
conventional agriculture.
Marketspresentafurtherchallenge.Althoughthegeneralmarket trend has
been upward, four clients have been lost: one went bankrupt
(leaving the project with a bad debt of $4,000), two withdrew
because of
anunsuitablespeciesmixfortheirparticularexhibits,andonewas dropped
after pressure from a rival. There is still an overreliance on one
majorclient(StratfordButterflyFarm),andif thisclientwerelost,
Kipepeowould be in trouble(Figure 5). Kipepeoisalsolikely toface
increasing competition as butterfly farming spreads through the
African 4. The KenyaBroadcasting Service has up to 8 million
listeners within Kenya. 5. The DubaiInternationalAwardisjointly
administeredbytheDubai Municipality and the UnitedNations Habitat
programme. It recognizes outstanding achievements in improving the
living environment. In 1998, there were 450 submissions worldwide.
The 10 winning projects each received a cash prize ($30,000) and
atrophy. Gordon, AyiembaITHE KIPEPEO PROJECT93 Month lo 199sa 1999
I Figure 4:Kipepeo Monthly Exports: 1998-1999 $3,192.78 $30,549.00
Figure 5:Kipepeo Earnings From Different Clients in 1999 C
ButterflyPavilion Butterfly Place D Cambridge,Ontario
0Chesterfield,Ohio Day CFlorida London Pupae Supplies 0Michael
Boppre North Carolina D Philadelphia DStratford continent.
Maintaining and improving product quality, continued atten-dance at
the annual International Congress of Butterfly Exhibitors and
Suppliers6,and a Web site (www.kipepeo.org) will all help tokeep
the market edge that Kipepeo currently enjoys, but developing a
local mar-94JOURNAL OF ENVIRONMENT &DEVELOPMENT ket in the form
of a tourist exhibit in Mombasa is essential for longer term market
security. A proposal forsuch an exhibit isbeing developed for
submission to the United States Agency for International
Development (USAID) and the World-Wide Fund for Nature (WWF). The
resource base of Kipepeo is the forest itself, its vegetation, and
its wild populations of butterflies. As long as the forest is
there, the vegeta-tion will not be a problem, but what about the
butterflies? Table 3 shows results fromtwo 9-month periods of
butterfly monitoring, one in 1993 before butterfly harvesting
started and one in 1997 after 4 years of opera-tions.Monitoring
involved walkingtransectsandcounting sightings. This produced data
on 60 species, of which 23 had been harvested and 37 had
not.Asclimaticconditions and overall butterfly abundance were very
different in the two years, comparisons between them of numbers on
an individual species basis would not be helpful. Instead, the
overall ranks are compared for the two periods on the hypothesis
that if harvest-ing had adversely affected wild populations, then
there would be a
sig-nificantdifferenceinrankingandtheharvestedspecieswouldhave
lower ranks in 1997 than in 1993. The results show that although
butter-flies were significantly more abundant in 1997, there was no
significant change in rank abundance forthe set of 60species
andthat harvested speciesactuallyranked nonsignificantlyhigher in
abundancein 1997 than they had in 1993. On the basis of this
evidence, butterfly harvesting seems not to be having any impact on
wild populations, but the project continues to monitor the
situation. The last question is whether Kipepeo is replicable. The
answer is yes, subject to markets, the capacity to get pupae to the
markets quickly (in practice, within 4 to 5 days from pupation),
and rainfall patterns. Costa Rica, where butterfly farming has been
practiced since the early 1980s, currently exports around $1million
worth of live butterflies a year, and the market for East African
butterflies must be at least as large. Even so, the market ceiling
is low. Kipepeo is only an hour's drive away on a tar-mac road from
international couriers, and the paperwork required for export can
now be obtained in a matter of minutes: Similar conditions of rapid
access to couriers and minimal paperwork would be essential for any
replicate operation. The other factor in Kipepeo's favor is the
rainfall pattern:Althoughtherains start toolatetocatchthe
annualsurgein global demand that starts in March and April and ends
in October, peak production (June to August) still coincides with
the main market season. A butterfly farm in an area where the rains
came a little earlier in the year would be well placed to succeed.
6.The International Congress of Butterfly Exhibitors and Suppliers
is attended by the world's leading butterfly farmers and butterfly
exhibitors and was first held in 1997. It pro-vides a major
marketing opportunity for butterfly farmers to make personal
contacts with their clients. Gordon, Ayiemba ITHEKIPEPEO PROJECT95
Table 3 Counts of Butterflies in Arabuko-Sokoke Nature Reserve in
1993 and 1997 MeanMean CountCount TotalTotal19931997
Cou11tCollnt(189(187RankRank Species19931997Hours)Hours)19931997
Papilio constantinus*5221,2592.766.73108 Papilio
dardm111s*2658321.404.451614 Papilio11ireus*4706392.493.421220
Papilio demodorns*817680.434.113116
Gmplziwnphilonoe*1202550.631.362632
Gmphiumleo11idas*1420.070.014954 Gmphiumkirbyi*70200.370.113446.5
Gmphiwn co/01111a*1355470.712.932424
Gmplzillmantheus*1642700.871.442029
Gmplziwnport/won*642100.341.123534 Dixeaclwrina1586040.843.232122
Catopsilia flore/la7278833.854.72712
Eure111aspecies1,7224,2929.1122.9522 Pinacopteryx
eviphia21280.110.1545.545 Nephroniatlwlassina2344461.242.391926
Eron incleodom1,4062,3557.4412.5936 Colotisregina4957832.624.191115
Colotisione7326293.873.36621 Colotiseuippe9139764.835.2259
Colotiseris2521181.330.631738 Colotis auxo1142650.601.422830
Colotis evagore2400.130.004358 Belenoiscreona3165,1141.6727.35141
Belenois gidica73350.041.7956.527 Belenoisthysa6212,5513.2913.6485
Appias epap!zia489340.254.993610 Leptosia
a/cesta1,8833,3019.9617.6514 Mylothris agathina76340.400.183343.5
Danauschrysippus*1042640.551.412931 Amauris
niavius*750.040.0356.551 A111aurisoch/ea800.040.005458
Melanitisleda*221340.120.724435.5
Bicyc/11ssafitza1422970.751.592228 Ypthimaasterope2800.150.004158
Euryplmm aclilys80360.420.193242
Bebeariacliriemhilda29200.150.114046.5
E11phaedmneophro11*3278561.734.581313
Neptisspecies2351331.240.711837 Byb!iailitlzyia455180.242.773825
E11ryteladryope2540.130.024252.5 (co11ti1111ed) 96JOURNAL OF
ENVIRONMENT &DEVELOPMENT Table 3 (continued) MeanMean
CountCount TotalTotal19931997 CountCount(189(187RankRank
Species19931997Hours)Hours)19931997
Hypolinmasmisipp11s*57414503.047.7597
Hypolimnasdeceptor1029030.544.833011
Hypolimnasanthedon6140.030.0758.548 Salamis
anacardii*121340.060.725035.5 f1111oniaoenone*2797661.484.101517
f1moniahierta21340.110.1845.543.5 f
unonianatalica1415750.753.072323 Junoniaterea640.030.0258.552.5
Phalantaphalantha91942164.8622.5543 Acraeaspecies1257440.663.982518
Pardopsispunctatissima46100.240.053749
Phsycaeneuraleda1196720.633.592719
Charaxesvaranes*432420.231.293933 Charaxescandiope*890.040.055450
Charaxes cithaeron*18950.100.514739
Charaxesprotoclea*17390.090.214841 Euxanthe
wakefieldi*1110.060.005155 Tirwnulapetiverana*4470.020.256040
Pseudacraeaboisduvali1000.050.005258 Hannatheobene800.040.005458
Wilcoxon'sSpeannan's SingedTestCorrelation Unharvestedz = 2.949p =
.003r2 = 0.9772MeanMean rank=rank= 28.6830.18 Harvestedz = 3.832p
< .0011'2= 0.9926MeanMean (marked*)rank=rank= 33.2330.98 All
speciesr2 = 0.9698 Summary and Conclusions
TheKipepeoProjectlinksconservationanddevelopmentat Arabuko-Sokoke
Forest on the north coast of Kenya through
community-basedbutterflyfarming.It wasstartedin1993andhasearnedover
$400,000 since then through the export of live butterfly pupae to
the live Gordon, AyiembaITHE KIPEPEO PROJECT97 butterfly exhibit
industry in Europe and North America. The butterflies have been
reared by the forest-adjacent community using parental but-terfly
stock from the forest and forest food plants for the caterpillars.
But-terfly monitoring has shown that there have been no negative
effects on the wild populationsof butterflies in the forest.
Community earnings have accumulated to more than $130,000since1993
and have had signif-icant positive effects both on livelihoods and
attitudes toward the con-servation of the forest.Kipepeo has shown,
albeit on a small scale, that integrating conservation and
development can be a practical and effec-tive strategy for natural
resource management. Acknowledgements An earlier versionof
thisarticlewas presented atthe International
UnionforConservationofNatureandNaturalResources(IUCN) Regional
Biodiversity Forum in Mombasa in 2000. The authors
acknowl-edgefinancialsupportfromtheGlobalEnvironmentFacilitySmall
GrantsProgramme,theIUCN-NetherlandsTropicalRainForest Programme,
the Biodiversity Support Program, a consortium of World Wildlife
Fund, the Nature Conservancy, and the World Resources Insti-tute,
with funding by the United States Agency for International
Devel-opment,theChicagoZoologicalSocietyandBrookfieldZoo,the
National Museums of Kenya, and the EU-funded BirldLife
International Arabuko-Sokoke Forest Management and Conservation
Project. Above all, we thank the farmers of Arabuko-Sokoke who made
Kipepeo a real-ity.The opinions expressed herein are those of the
authors and do not necessarily reflect the views of the U.S.Agency
for International Devel-opment or any of the organizations or
funding bodies. References Ayiemba,W.0. (1997).AstI1dyof
tlzebutterflydiversityof Arnbllko-SokokeForest,Ke11yn. Unpublished
master's thesis, University of Nairobi. Bennun, L.A., &Njoroge,
P.(1999).lnzportnntBird ArensinKenya.Nairobi, Africa: Nature Kenya.
Burgess, N. D., Clarke, G. P., &Rodgers, W.A.(1998). Coastal
forests of eastern Africa: Sta-tus, endemism patterns and their
potential causes. Biological foumnl of tlze Linnnenn Soci-ety,
64,337-367. Collar, N. J., & Stuart, S. N. (1988). Key forests
for tlzrentened birds in Africa (Monograph Num-ber 3). Cambridge,
UK:International Council forBird Preservation. Gordon, I.
J.(1994).Kipepeoproject:Consultant'sreport.Unpublished manuscript.
Government of Kenya.(1997). KilifiDistrictdevelop111entplan,
1997-2001. Nairobi, Africa: Government Printers. 98JOURNAL OF
ENVIRONMENT &DEVELOPMENT Maundu, P.(1994). Socio-eco110111ic
survey nnd forest attitude report of tlze co111111unityborderi11g
Arnbuko-SokokeForestn11dGn111eReserve. Unpublishedreport
oftheKipepeoProject, Nairobi, Africa. Maundu, P., Sojah, L.,
&Kilili, G. (1997). Report 011tlze i111pncl of Kipepeo project
011tlzeeco110111ic status nnd forest attitude of tlze co111111unity
bordering Arnbuko-Sokoke ForestReserve. Unpub-lished manuscript.
Mogaka,H.(1991).Localutilisntionof
tlzeArnbuko-SokokeForestResen;e.Unpublished manuscript.
Myers,N.,Mittermeier, R.A.,Mittermeier, C.G.,daFonseca,
G.A.B.,&Kent, J.(2000). Biodiversity hotspots for conservation
priorities. Nature,403,853-858. ln11Gordo11cn111eto Africa
fro111Britain to work ns n u11iversity lecturer
i111971.Si11cetlze11,lze hns tnug/11 nl five
AfricanU11iversitiesiiiGhnnn, Zi111bnbwe,n11dKe11yn.lll1993 he left
tlze university tenchi11gprofessio11n11d111ovedtoKilifi onthe
Ke11yn11canst to sin rttheKipepeobutterfly fnn11i11g project nt
Arnbuko-Sokoke Forest. lll1998, Dr. Gordon left Kipepeo to beco111e
coordinator of the
EU-fu11dedBirdLifel11ter11ntionnlArnbuko-SokokeForestCo11servntio11n11dMn11nge111e11tProject,
which ended in March 2002. He currently works ns the liend of
theE11viro111nentnlHen/thDivision ntthe l11temntio11nlCentre of
lllsectPhysiology nndEcology. Wnshi11gto11 Ayie111bn hns been n
research scientist with the Nntio11nl Museums of Kenya since 1994.
Currently lieisthe 111a11ngerof the KipepeoProject, n position held
since 111id-1998.Hetrni11ed as n co11servntio11biologist
nltheUniversity of Nairobi n11d/msbee11actively involved
inbiodiversity co11servntio11 n11d sustni11able use activities in
the Arnbuko-Sokoke Forest area.He isn 111e111ber of the
Arnbuko-Sokoke ForestMn11nge111e11tTen111thnl comprises four
gover11111e11tdepart111entstlmt col-lnbornte closely witlz a
ForestAdjacent Dwellers Associntio11, n11d he ischair tothe
RuralDevelop-111e11tWorki11gGroup of thetea111.