Conservation of biodiversity in the Arabuko Sokoke Forest, Kenya


Muriithi, S.; Kenyon, W.

Biodiversity and Conservation 11(8): 1437-1450

2002


Using an economic approach to provide a rationale for rain forest conservation has been a popular exercise in recent years. This paper uses such an approach to assess the net value of the Arabuko Sokoke Forest in Kenya. The economic benefits associated with the forest derived by local and global populations are estimated by combining evidence from existing studies and the results of a contingent valuation study carried out by the authors. These benefits are set against the cost of preserving the forest to the Kenyan Forest Department. Even when the opportunity cost of the forest land is omitted from the costs of forest preservation, and when the revenues generated from the Global Environment Facility funded project are included, the costs of forest conservation outweigh the benefits. It is only when non-use and existence values are included (which are not realized by the Kenyan population) that the forest benefits exceed the costs. The paper concludes by arguing that, although some projects within the Arabuko Sokoke Forest have been successful in capturing some of the economic value associated with the forest, more needs to be done to design additional capture mechanisms so that a greater proportion of the global benefit of the forest can be realized by local populations and local governments.

Biodiversity
and
Conservation
11:
1437-1450,
2002.
'Pf
©
2002
Kluwer
Academic
Publishers.
Printed
in
the
Netherlands.
Conservation
of
biodiversity
in
the
Arabuko
Sokoke
Forest,
Kenya
SAMUEL
MUMITH1
1
and
WENDY
KENYON
2
'
*
'Forest
Department,
P.O.
Box
8020,
Nairobi,
Kenya;
2
Scottish
Agricultural
College,
West
Mains
Road,
Edinburgh
EH9
3JG,
UK;
*
Author
for
correspondence
(e-mail:
fax:
+44-131-
667-2601)
Received
23
October
2000;
accepted
in
revised
form
24
August
2001
Key
words:
Biodiversity,
Contingent
valuation,
Forests,
Global
benefits
Abstract.
Using
an
economic
approach
to
provide
a
rationale
for
rainforest
conservation
has
been
a
popular
exercise
in
recent
years.
This
paper
uses
such
an
approach
to
assess
the
net
value
of
the
Arabuko
Sokoke
Forest
in
Kenya.
The
economic
benefits
associated
with
the
forest
derived
by
local
and
global
populations
are
estimated
by
combining
evidence
from
existing
studies
and
the
results
of
a
contingent
valuation
study
carried
out
by
the
authors.
These
benefits
are
set
against
the
cost
of
preserving
the
forest
to
the
Kenyan
Forest
Department.
Even
when
the
opportunity
cost
of
the
forest
land
is
omitted
from
the
costs
of
forest
preservation,
and
when
the
revenues
generated
from
the
Global
Environment
Facility
(GEF)
funded
project
are
included,
the
costs
of
forest
conservation
outweigh
the
benefits.
It
is
only
when
non-use
and
existence
values
are
included
(which
are
not
realised
by
the
Kenyan
population)
that
the
forest
benefits
exceed
the
costs.
The
paper
concludes
by
arguing
that,
although
some
projects
within
the
Arabuko
Sokoke
Forest
have
been
successful
in
capturing
some
of
the
economic
value
associated
with
the
forest,
more
needs
to
be
done
to
design
additional
capture
mechanisms
so
that
a
greater
proportion
of
the
global
benefit
of
the
forest
can
be
realised
by
local
populations
and
local
governments.
Introduction
In
1991
Pearce
suggested
an
economic
approach
to
saving
the
tropical
rainforest
whereby
monetary
estimates
of
different
'values'
associated
with
forest
preservation
might
be
estimated
and
shown
to
exceed
the
costs
associated
with
rainforest
preservation,
thereby
providing
a
rationale
for
their
conservation.
This
type
of
analysis
has
since
become
popular,
but
one
problem
remains.
Many
of
the
identified
benefits
do
not
accrue
to
the
countries
that
bear
the
cost
of
rainforest
preservation.
Although
forests
may
maintain
biodiversity,
store
carbon
and
cycle
nutrients
(Pearce
1991),
the
benefits
of
such
activities
are
shared
by
the
world
population
and
are
often
not
realised
in
terms
of
financial
benefits
to
local
populations.
Some
efforts
have
been
made
to
facilitate
the
transfer
of
benefits,
such
as
debt
for
nature
swaps,
and
the
transfer
of
funds
under
global
institutions
such
as
the
Global
Environment
Facility
(GEF).
However,
the
design
and
implementation
of
capture
mechanisms,
alongside
the
measurement
of
economic
benefits,
is
key
if
the
economic
approach
is
to
provide
a
practical
rationale
for
helping
to
preserve
rainforests.
This
paper
assesses
the
economic
case
for
saving
the
Arabuko
Sokoke
Forest
in
1438
Kenya.
Although
it
has
become
unusable
for
commercial
logging
due
to
past
unsustainable
logging
practices,
the
forest
may
still
be
valuable
both
to
local
people
and
global
populations.
Overseas
visitors
to
Kenya
value
the
forest
for
recreational
purposes.
Local
people
value
it
for
its
subsistence
uses
and
the
Forest
Department
gains
revenues
from
licensed
uses
of
the
forest.
In
addition,
global
populations
may
value
the
existence
of
the
forest
and
the
associated
biodiversity
(existence
value)
and
may
value
the
forest
as
an
asset
to
be
passed
on
to
future
generations
(bequest
value).
The
paper
estimates
the
economic
benefits
the
forest
has
for
different
categories
of
people,
and
shows
that
even
in
its
current
degraded state,
the
forest
appears
to
have
considerable
value
both
to
local
and
global
populations.
However,
practical
means
to
capture
these
values
are
urgently
required
if
the
economic
approach
is
to
provide
a
realistic
rationale
for
conserving
the
forest.
The
Arabuko
Sokoke
Forest
Arabuko
Sokoke
Forest
is
located
near
the
North
Coast
of
Kenya
and
occupies
an
area
of
41676
ha
(Wass
1995).
The
forest
is
a
protected
area
and
is
under
the
management
of
the
Kenya
Forest
Department
in
collaboration
with
the
National
Museums
of
Kenya,
the
Kenya
Wildlife
Service,
the
Kenya
Forestry
Research
Institute,
with
technical
assistance
provided
by
Birdlife
International
(an
internation-
al
NGO).
The
forest
is
managed
in
line
with
Kenyan
forest
policy,
which
requires
preservation
of
forest
areas
for
watershed
protection,
timber
and
other
forest
produce,
and
conservation
of
both
flora
and
fauna.
However,
in
practice
this
policy
has
not
been
effectively
implemented
in
the
past.
Arabuko
Sokoke
Forest
is
located
in
a
hot
and
humid
climate
with
an
average
temperature
of
around
29
°
C.
There
are
two
rainfall
seasons
of
over
1000
mm
in
the
wet
part,
declining
to
600
mm
in
the
dry
part
(KIFCON
1995)
*
.
Four
distinct
vegetation
types
are
identified:
mixed
forest;
Brachystegia
woodland;
Cynometra
forest
and
thicket;
and
finally
the
mangrove
forest
around
Mida
Creek
(KIFCON
1992).
This
environment
has
provided
tree
species
of
high
valuable
timber
for
both
the
furniture
and
construction
industry
and
due
to
the
rapid
growth
of
coastal
towns,
the
forest
became
a
major
source
of
fuel-wood,
timber
and
poles
in
the
1970s.
Growth
of
the
international
tourism
industry
along
Kenya's
North
Coast
has
led
to
additional
demand
for
wood
for
use
in
hotel
construction
and
carving
for
tourists.
The
result
of
this
demand
has
been
over-logging
of
the
forest
to
the
extent
that
the
forest
has
latterly
been
reduced
to
a
secondary
forest
with
few
tree
species
of
quality
wood.
Despite
this,
the
forest
has
a
high
level
of
biodiversity.
A
survey
in
1994
identified
nearly
600
plant
species
(KIFCON
1992)
and
230
birds,
of
which
six
are
listed
in
the
Red
Data
Book
for
African
Birds
(KIFCON
1991).
One
of
these
birds,
Clarke's
weaver,
is
endemic,
while
the
Sokoke
scops
owl,
Amani
sun
bird
and
*
Kenyan
Indigenous
Forest
Conservation
Project.
1439
Sokoke
pipit
are
near
endemic
(KIFCON
1995).
The
forest
is
listed
as
the
second
in
importance
for
bird
conservation
in
Africa
(Wass
1995).
In
addition
to
its
impor-
tance
for
birds,
the
forest
also
provides
a
habitat
for
many
species
of
invertebrates
including
250
species
of
butterfly,
as
well
as
large
and
small
mammals
(KIFCON
1993).
Three
globally
threatened
mammals
occur
in
the
forest:
Golden-Rumped
Elephant
Shrew,
Sokoke
Bush
Tailed
Mongoose
and
the
Alder's
Duiker
(Kingdon
1997).
It
is
clear
that
despite
the
lack
of
good
quality
timber,
the
forest's
continued
conservation
is
beneficial
for
biodiversity,
and
for
a
number
of
other
benefits
which
accrue
to
local
and
global
populations.
However,
the
costs
of
conserving
the
forest
to
the
Kenyan
Forest
Department
are
not
inconsiderable,
and
given
that
it
no
longer
provides
revenue
from
timber
production,
a
rationale
for
continued
conservation
of
the
forest
is
required.
The
economic
benefits
which
accrue
to
local
populations might
be
estimated
in
a
variety
of
ways,
using
a
combination
of
data
from
existing
studies
conducted
in
Arabuko
Sokoke
Forest,
data
from
the
Forest
Department
and
from
environmental
projects
which
derive
an
income
from
forest-related
activities.
The
economic
benefits
which
accrue
to
global
populations
may
also
be
estimated
from
a
variety
of
sources,
including
data
from
the
Forest
Department
and
primary
data
collection
on
visitor
values.
Combining
the
various
sources
of
data
allows
a
comprehensive
picture
to
be
drawn
of
the
total
economic
benefit
of
the
forest.
These
benefits
may
come
in
two
forms.
First,
benefits
from
direct
use
of
the
forest,
such
as
revenue
from
the
sale
of
timber
products,
recreational
and
subsistence
use
of
the
forest.
Second,
some
people
may
be
willing
to
pay
for
indirect
use
values
related
to
the
forest
such
as
carbon
storage,
or
for
non-use
values
such
as
existence
value
or
bequest
value.
Table
1
shows
direct
use
value,
indirect
use
value
and
non-use
value,
and
indicates
how
each
might
be
estimated
in
the
Arabuko
Sokoke
Forest
context.
As
the
table
shows,
the
Arabuko
Sokoke
Forest
is
used
for
tourism
and
recreation,
and
although
the
number
of
visitors
to
the
forest
for
recreation
is
currently
small,
there
is
potential
to
derive
economic
benefit
from
this
use
by
charging
an
entry
free.
Currently,
no
entry
fee
is
charged
to
visitors.
In
such
circumstances
contingent
valuation
studies
provide
a
means
by
which
these
values
can
be
estimated
(Kramer
and
Mercer
1997).
One
such
survey
has
been
carried
out
by
the
authors
to
assess
the
maximum
entrance
fee
that
visitors
and
potential
visitors
to
the
park
would
be
willing
to
pay.
A
second
category
of
economic
value
relates
to
a
number
of
legal
Table
1.
The
values
associated
with
the
Arabuko
Sokoke
Forest,
and
methods
of
estimation.
Value
Method
of
estimation
Direct
use
value
Tourism
and
recreation
Legal
uses—royalties
paid
Butterfly
farming
Illegal
subsistence
use
Indirect
use
value
and
non-use
value
WTP
entry
fee
(CV)
Forest
department
annual
accounts
Project
annual
accounts
Participatory
valuation
(Emerton
1994)
WTP
for
conservation
(CV)
1440
uses
of
the
forest
for
which
the
users
pay
royalties
to
the
Forest
Department,
and
for
which
annual
reports
are
available.
A
third
value
derives
from
a
butterfly
farming
project
which
was
set
up
in
1993
with
assistance
from
the
GEF.
This
project
yields
income
in
the
form
of
export
sales
of
butterfly
pupae,
sale
of
guide
books
on
the
butterfly
farm,
visits
to
the
farm,
and
camping
at
the
farm.
Data
on
all
of
these
activities
are
available
through
the
project
annual
reports.
A
fourth
category
of
benefits
relates
to
illegal
uses
of
the
forest,
such
as
hunting
for
bushmeat
and
collecting
poles
by
local
communities.
Emerton
(1994)
made
an
assessment
of
the
value
associated
with
this
activity.
Finally,
evidence
suggests
that
populations
who
may
never
visit
the
forest,
nevertheless
may
be
willing
to
pay
for
its
conservation,
as
they
value
the
existence
of
the
forest
and
the
species
it
provides
a
habitat
for
(Pearce
and
Moran
1994).
The
authors
have
conducted
a
contingent
valuation
survey
to
estimate
these
values.
As
long
as
each
of
these
benefits
can
be
realised
by
local
populations
or
the
forest
managers,
an
economic
rationale
exists
for
the
continued
protection
of
the
forest.
Contingent
valuation
of
Arabuko
Sokoke
Forest
Contingent
valuation
is
a
widely
used
method
to
assess
non-market
values.
It
is
a
survey-based
technique
where
respondents
to
a
survey
are
asked
their
willingness
to
pay
for
environmental
assets
1
in
carefully
described
circumstances
(Mitchell
and
Carson
1989).
In
the
study
of
Arabuko
Sokoke
Forest
a
contingent
valuation
questionnaire
was
conducted
using
mostly
people
who
had
never
visited
the
forest,
to
estimate
two
different
values.
Firstly,
respondents
were
given
information
about
the
forest
and
asked,
if
they
were
to
visit
the
forest,
the
maximum
entrance
fee
they
would
be
willing
to
pay,
allowing
estimation
of
use
value
of
the
forest.
Currently
entrance
to
the
forest
is
free.
Secondly,
respondents
were
asked
to
suppose
that
they
had
no
intention
of
visiting
the
forest
now
or
in
the
future,
and
were
asked
how
much
they
would
be
willing
to
donate
as
a
contribution
towards
conservation
of
the
forest,
allowing
estimation
of
indirect
use
value
and
non-use
value.
The
questionnaire
comprised
four
sections.
The
first
section
requested
general
information
about
whether the
respondent
was
a
visitor
to
Kenya,
their
participation
in
outdoors
and
wildlife
activities,
and
attempted
to
determine
the
general
en-
vironmental
attitude
of
the
respondent.
The
second
section
provided
respondents
with
information
about
the
forest
and
included
maps,
pictures
of
the
flora
and
fauna
in
the
forest,
and
a
description
of
the
problems
facing
the
forest.
The
third
section
comprised
the
valuation
questions.
First,
respondents
were
asked
whether
in
principle
they
would
be
willing
to
pay
to
enter
the
forest.
They
were
then
asked
specifically
'Could
you
inform
me
of
the
most
you
would
be
willing
to
pay
to
use
the
forest
for
recreation
per
day'.
They
were
asked
to
select
their
answer
from
a
payment
card
containing
amounts
from
$1
to
over
$100.
The
second
set
of
valuation
t
Although
these
techniques
are
also
used
to
value
other
non-market
assets
such
as
health
benefits.
1441
questions
asked
whether
respondents
would
in
principle
be
willing
to
make
a
one-off
contribution
towards
forest
conservation,
and
how
much
they
would
contribute
(also
based
on
a
payment
card
with
amounts
from
$0.5
to
$100).
Respondents
were
reminded
of
their
budget
constraint,
and
that
they may
have
already
contributed
to
conservation
in
other
areas.
The
exact
question
was
'I
want
you
to
consider
how
much
you
are
willing
to
give
as
a
contribution
for
conservation
of
the
forest'.
Although
the
dichotomous
choice
(DC)
elicitation
format
was
favoured
by
the
NOAA
panel
(Arrow
et
al.
1993)
for
use
in
CV
studies,
more
recent
evidence
suggests
that
it
may
not
be
the
most
favoured
option,
not
least
because
it
opposes
one
of
the
other
NOAA
recommendations
of
choosing
the
most
conservative
CV
design.
There
is
substantial
evidence
in
the
literature
which
proves
that
the
DC
format
leads
to
higher
values
than
other
CV
elicitation
formats
(Boyle
et
al.
1996;
Ready
et
al.
1996).
It
is
suggested
that
the
payment
card
elicitation
method
facilitates
the
respondents'
valuation
process
(Mitchell
and
Carson
1989;
Ready
et
al.
1996),
and
has
been
found
useful
in
similar
cases
to
this
(Edwards-Jones
and
Kenyon
1998;
Bann
2000).
The
final
section
of
the
questionnaire
included
questions
on
the
socio-economic
characteristics
of
the
respondents,
including
membership
of
conservation
organisa-
tions,
nationality,
age,
gender,
number
of
children
in
the
household,
level
of
education,
employment
and
income.
Since
only
a
small
number
of
tourists
currently
visit
the
forest
for
recreational
purposes,
only
a
small
number
of
questionnaires
were
conducted
within
the
forest
area
itself.
The
majority
of
the
survey
was
conducted
away
from
the
forest
area.
These
off-site
interviews
were
carried
out
at
beach
locations,
approximately
10
km
from
the
forest.
Interviews
were
held
face
to
face
and
were
conducted
in
June
and
July
1999.
A
total
of
318
respondents
were
interviewed.
The
sample
was
composed
of
15%
Kenyans
and
85%
foreigners.
Only
3%
of
the
respondents
knew
about
the
forest.
The
majority
of
the
sample
stated
that
their
main
holiday
interest
was
wildlife
viewing
(89%).
Seventy-seven
percent
were
interested
in
sun
bathing
and
beach
walking,
while
only
38%
were
interested
in
bird-watching.
Of
the
total
respondents,
7%
were
protesters,
indicating
that
they
rejected
the
contingent
scenario.
A
follow-up
to
the
valuation
question
allowed
protesters
to
be
identified,
by
asking
respondents
why
they
had
made
the
response
they
had.
Those
respondents
who
felt
that
funds
for
conservation
of
the
forest
should
come
from
other
sources,
or
that
their
donation
may
not
go
towards
forest
conservation,
were
classed
as
protesters
and
excluded
from
further
analysis
(Hanley
and
Ruffell
1993;
Alfarez-Farizo
et
al.
1999).
Analysis
of
the
willingness
to
pay
(WTP)
questions
was
divided
into
two
categories,
locals
and
foreigners.
Table
2
shows
the
mean,
median
and
range
of
the
bids
as
well
as
the
percentage
of
non-zero
bids
in
each
category.
Interestingly,
the
entire
sub-sample
of
foreigners
was
willing
to
pay
a
fee
to
enter
the
forest,
and
nearly
all
were
willing
to
contribute
to
the
conservation
of
the
forest.
The
sub-sample
of
Kenyans
who
were
willing
to
pay
an
entry
fee
and
to
contribute
to
the
conservation
of
the
forest
was
significantly
smaller,
with
the
mean
amounts
being
bid
also
significantly
smaller.
Responses
to
the
questionnaire
indicate
a
1442
Table
2.
Descriptive
statistics
for
willingness
to
pay
(WTP)
a
donation
and
entrance
fee
to
Arabuko
Sokoke
Forest.
%
with
positive
WTP
Mean
US$
Median
US$
Range
US$
Population
Aggregate
US$
Foreigners
(rt
=
251)
Entrance
fee
100
7.77
5
2-40
1200
9324
—Donation
98.84
21.78
12
1-100
800000
17.4
million
Kenyans
(rt
=
44)
—Entrance
fee
76.36
2.23
2
2-3
1200
2676
—Donation
70.91
5.33
2.5
1-20
1
million
5.3
million
number
of
reasons
for
this
difference.
The
economic
circumstances
of
the
sub-
samples
were
different.
Some
Kenyans
felt
they
already
paid
enough
in
taxes
for
forest
conservation,
and
others
were
worried
about
corruption,
fearing
that
an
entrance
fee
or
donation
would
not
be
put
towards
conservation
of
the
forest.
The
Arabuko
Sokoke
Forest
Annual
Report
(1999)
indicates
that
1200
people
visited
the
forest
in
1998.
If
the
mean
tourist
entry
fee
was
aggregated
over
this
population,
potential
income
from
entry
fees
would
be
US$9324.
If
the
mean
Kenyan
entry
fee
was
used
as
an
entry
fee,
the
overall
potential
income
would
be
US$2676,
both
at
current
visitor
levels.
Aggregation
of
the
donations
for
conservation
of
the
forest,
however,
is
more
complicated.
Some
calibration
is
clearly
necessary,
as
it
would
be
extremely
optimistic
to
assume
that
all
tourists
that
visit
Kenya
(approximately
8
000 000
per
year,
giving
an
aggregate
sum
of
approximately
$17
4
million)
would
be
willing
to
donate
to
the
Arabuko
Sokoke
Forest.
Similarly,
it
is
highly
unlikely
that
the
total
population
of
the
coastal
area
of
Kenya
(approximately
1
million,
giving
a
total
of
$22.73
million)
would
be
willing
to
donate
to
the
forest.
There
may
also
be
error
due
to
free
riding.
It
has
therefore
been
suggested
in
the
literature
that
CV
figures
should
be
'calibrated'
to
reflect
the
fact
that
it
is
unlikely
that
hypothetical
donations
would
reach
the
same
level
as
real
donations.
However,
it
is
unclear
exactly
how
values
should
be
calibrated,
since
many
practitioners
are
unwilling
to
accept
the
arbitrary
50%
suggested
by
the
NOAA
panel
(Arrow
et
al.
1993).
Debate
about
the
precise
nature
of
calibration
of
CV
figures
lies
outside
the
boundaries
of
this
paper;
however,
given
the
debate,
it
should
be
noted
that
the
aggregate
figures
donating
to
conserve
the
Arabuko
Sokoke
Forest
should
be
treated
with
caution.
One
means
by
which
validity
of
CV
results
can
be
evaluated
is
through
construct
validity,
which
involves
testing
the
relationship
between
WTP
and
different
vari-
ables
against
a
priori
expectations.
Table
3
shows
the
independent
variables
that
were
used
in
regression
analysis
to
assess
the
validity
of
the
WTP
results
presented
above.
As
might
be
expected,
intuitively
there
is
a
significant
and
positive
relationship
for
both
the
donation
and
the
entrance
fee
for
income
and
educational
level.
Donations
to
conservation
organisations
have
a
statistically
significant
and
negative
relationship
with
WTP,
although
most
CV
studies
(Carson
1994)
show
a
positive
relationship
between
membership
of
conservation
group
and
WTP.
The
negative
1443
Table
3.
Regression
of
log
donation,
and
log
entrance
fee
against
explanatory
variables.
Variable
Explanation
Donation
Entrance
fee
Constant
—1.73
(-0.81)
Wildlife'
Respondents
who
have
participated
in
wildlife
viewing
—0.829
*
(-3.27)
—0.2384
*
(-2.07)
Don.
org
.
Number
of
donations
made
to
other
conservation
organisations
—0.0185
(1.63)
—0.158
**
(-2.62)
Age Age
of
respondent
0.0198
*
(2.58)
0.003
(0.45)
Edu.
level
Level
of
education
of
respondent
0.206
*
(2.79)
0.160
***
(3.37)
Child
Number
of
children
respondent
has
0.252
*
(2.96)
0.015
(0.36)
Income
Log
of
gross
annual
household
income
0.380
*
(3.04)
0.342
***
(5.44)
R
2
0.293
0.399
'Coded
yes
=
1,
no
=
2.
T-statistic
in
parentheses.
*
Significant
at
the
90%
level.
Significant
at
the
95%
level.
***
Significant
at
the
99%
level.
relationship
may
be
because
respondents
are
already
contributing
to
other
organisa-
tions
and
are
not
prepared
to
add
to
their
charitable
budget.
One
of
the
more
interesting
points
to
note
is
the
fact
that
the
number
of
children
of
the
respondent
has
a
positive
relationship
with
the
donation
and
the
entrance
fee,
although
this
relationship
is
only
statistically
significant
in
the
case
of
the
donation.
An
explanation
for
this
may
lie
in
the
fact
that
those
who
are
willing
to
donate
for
future
conservation
may
be
keen
to
ensure
that
such
habitats
are
available
to
their
children
in
the
future
(bequest
value).
Those
respondents
with
children
would
therefore
be
more
likely
to
donate
more
to
ensure
the
future
preservation
of
the
forest.
Respondents'
participation
in
wildlife
viewing
has
a
statistically
significant
and
negative
relationship
with
WTP
for
an
entry
fee.
This
implies
that
availability
of
alternative
recreation,
such
as
wildlife
viewing
in
other National
Parks,
reduces
the
motivation
of
the
respondent
to
visit
and
pay
to
enter
Arabuko
Sokoke
Forest.
This
might
be
expected,
since
most
of
the
foreign
visitors
to
Kenya
are
interested
in
wildlife safaris
in
National
Parks
where
they
can
watch
big
mammals,
which
are
not
available
in
the
Arabuko
Sokoke
Forest.
Overall
the
results
of
the
regression
confirm
a
priori
expectations.
The
overall
fit
of
the
models
is
good,
with
a
R
2
of
29%
for
the
donation
and
40%
for
the
entry
fee.
In
both
cases
this
is
above
the
15%
recommended
by
Mitchell
and
Carson
(1989),
showing
that
the
results
can
be
considered
reasonably
sound.
Forest
department
royalties
and
revenue
The
Arabuko
Sokoke
Forest
not
only
provides
benefits
to
Kenyans
and
the
rest
of
the
world
through
use
and
non-use
values,
as
the
evidence
above
suggests,
but
generates
revenues
for
its
managing
institution,
the
Forest
Department.
Revenue
is
currently
generated
from
direct
use
of
the
forest
and
is
realised
in
the
form
of
timber
royalties,
fuel-wood
licences,
grazing
fees
and
compensation
for
offences
commit-
ted
against
the
forest
laws.
Such
revenues
are
reported
each
year
by
the
Forest
1444
Table
4.
Forest
Department
royalties
and
revenues
(US$).
1994
1995
1996
1997
1998
Timber
royalties
-Mangrove
5578.62
1792.00
3914.46
2925.61
1765.56
-
Natural
forest
5698.76
3053.19
2058.14
1413.05
284.16
Poles
34.28
496.05
509.4
1914.17
1316.90
Fuel
wood
721.03
399.80
1164
1858.5
2989
Compensation
102.71
145.3
1513.91
366.67
255.70
Grazing
73.82
1.71
53.57
Seedlings
1739.28
1934.90
170.57
165.5
205.94
Rents
and
fees
70
154.36
119.29
Christmas
trees
502.11
694.29
265.03
195.50
545.93
Total
14
446.79
8545.53
9669.33
8995.07
7536.05
Source:
Forest
Department
Annual
Reports
1994-1998.
Department
in
its
annual
report.
Table
4
shows
the
revenues
and
royalties
paid
to
the
Forest
Department
from
1994
to
1998
for
a
range
of
items
such
as
timber,
poles
and
fuel-wood.
The
figures
show
a
decline
in
revenues
over
the
reported
period,
which
may
be
attributed
to
the
decreased
availability
of
wood
over
the
years, due
to
over-logging
and
the
subsequent
poor
regeneration
of
the
forest.
However,
the
total
amount
remains
a
significant
source
of
income
to
the
Forest
Department.
The
figures
in
Table
4
show
the
benefit
relating
to
certain
licensed
activities,
however,
most
forest
exploitation
takes
place
outside
the
licensing
system.
Such
activity
has
been
historically
difficult
to
measure,
however,
work
by
Emerton
(1994)
has
attempted
to
place
values
on
the
unlicensed,
subsistence
uses
of
the
Arabuko
Sokoke
Forest.
Subsistence
use
of
the
forest
by
the
local
community
Much
of
the
subsistence
use
of
Arabuko
Sokoke
Forest
by
local
people
is
not
reflected
in
the
current
forest
licensing
system
and
is
considered
illegal.
This
section
reports
results
from
a
study,
commissioned
by
the
Forest
Department,
carried
out
around
Arabuko
Sokoke
Forest
in
1994
with
the
objective
of
measuring
the
value
of
subsistence
use
of
the
forest
by
local
communities
(Emerton
1994).
The
forest
uses
were
identified
using
a
social
economic
survey
of
households
residing
within
a
5
km
radius
of
the
forest.
After
ascertaining
the
quantity
of
forest
goods
used
by
each
household,
economic
benefit
was
calculated
using
the
forest
edge
values
(Table
5).
The
forest
edge
values
were
derived
from
market
prices
of
close
substitutes
less
transport,
labour
and
the
price
of
other
inputs.
For
example,
the
price
of
water
was
inferred
from
the
charges
levied
on
piped
water,
while
the
price
of
bush
meat
was
inferred
from
the
price
of
meat
of
domestic
animals.
To
obtain
the
aggregate
values,
population
density
was
multiplied
with
the
area
and
then
multiplied
by
the
average
consumption
of
products
per
household.
As
Table
5
shows,
the
value
of
subsistence
use
was
estimated
at
US$21183,
with
the
largest
proportion
of
value
coming
from
1445
Table
5.
Subsistence
use
of
forest
by
the
local
community.
Source
Illegal
value
(US$)
Percentage
of
total
illegal
value
Water
493.07
2.33
Land
poles
533.3
2.5
Wild
meat
11
725
55.35
Fuel-wood
8062
38.06
Honey
102.65
0.48
Utility
items
43.57
0.21
Medicine
223.92
1.06
Total
21
183.51
100
Source:
Emerton
(1994).
wild
meat
and
fuel-wood
(Emerton
1994).
These
results
should
be
seen
as
estimates
only,
due
to
difficulties
in
calculating
shadow
prices
of
the
subsistence
goods
used
in
attributing
values
to
the
illegal
forest
products.
However,
evidence
from
the
literature
endorses
these
estimates.
Pearce
et
al.
(1994)
suggested
that
there
would
be
high
dependence
on
forests
for
fuel-wood
by
the
local
communities
in
Kenya,
since
71%
of
the
population
derive
their
domestic
energy
requirement
from
fuel-
wood.
The
Kenya
Development
Plan
estimated
this
dependence
to
be
73%,
which
implies
significant
use
of
local
forests
for
fuel-wood.
The
importance
of
forests
in
Kenya
as
a
source
of
meat
has
been
reported
by
Fitz-Gibbon
et
al.
(1995)
who,
like
Abelson
(1996),
suggested
that
the
high
dependency
on
the
forest
by
the
local
communities
may
be
explained
by
the
high
level
of
poverty,
i.e.
per
capita
income
of
$540
in
1991.
The
figures
produced
by
Emerton
(1994)
therefore
are
indicative
of
the
importance
that
Arabuko
Sokoke
Forest
has
for
the
local
communities.
Revenues
from
tourism
and
butterfly
farming
As
well
as
offering
benefits
to
the
local
population
in
terms
of
subsistence
goods,
the
forest
has
also
brought
income
to
local
communities
through
a
butterfly
farming
project
started
in
1993
through
assistance
of
the
GEF.
The
activities
of
the
project
involve
collection
of
butterfly
eggs
and
forage
from
the
forest
by
the
forest
adjacent
communities.
The
eggs
are
kept
in
crossed
cages
and
after
hatching,
the
larvae
are
fed
with
forage
until
they
develop
into
pupae
which
are
packed
and
exported
to
the
Far
East
for
ornamental
purposes.
The
benefit
of
this
project
can
be
seen
not
only
in
income
from
export
sales,
but
also
from
visits
to
the
butterfly
farm,
sale
of
guide
books,
and
camping
fees
collected
on
the
farm
(Table
6).
Evidence
suggests
that
butterfly
farming has
no
impact
on
the
biodiversity
of
the
forest
(Ayiemba
1998).
All
the
income
from
the
butterfly
farming
is
shared
between
the
members
of
the
community.
Unlike
licensing
revenue
to
the
Forest
Department
(Table
4),
revenue
from
the
butterfly
farming
project
has
increased
to
over
US$31000
in
1998,
significantly
more
than
was
ever
shown
in
the
Forest
Depart-
ment
Annual
Report
as
income
from
other
licensed
uses
of
the
forest.
1446
Table
6.
Revenues
from
tourism
and
butterfly
farming
(US$).
Year
1994
1995
1996
1997
1998
Butterfly
export
13
224.1
15
297.1
26
893.31
30
157.04
29
454.26
Visits
to
butterfly
fanns
1241.1
1312.2
1682.3
2154.86
1939.86
Sale
of
guide
books
210
158.60
231.42
262.86
408.6
Camping
fee
53.20
48.60
60
Total
14
675.20
16
767.90
28
860.03
32
623.36
31
862.72
Source:
Kipepeo
Project
Annual
Reports
(1994-1998).
Cultural
values
Although
many
of
the
values
of
the
forest
have
been
reported
or
estimated,
some
are
not
easily
estimated
in
monetary
terms.
Within
the
Arabuko
Sokoke
Forest
there
are
sacred
groves
which
are
traditionally
referred
to
as
'Kap?,
meaning
'homestead'
in
several
local
languages
(Negussie
1997).
The
local
ethnic groups
have
historical
and
spiritual
relationships
with
the
`Kayas'.
The
shrines
are
used
for
worship
where
the
local
communities
conduct
religious
rites,
use
traditional
burial
grounds
and
offer
spiritual
connection
between
the
living
and
the
ancestors.
Although
the
cultural
and
spiritual
value
of
these
shrines
are
in
decline
due
to
erosion
of
traditional
culture,
the
traditional
institutions
play
an
important
role
in
the
protection
of
the
forest
(Negussie
1997).
Through
a
council
of
elders
the
locals
enact
customary
rules
which
regulate
access
to
use
of
the
forest
resources
through
a
mixture
of
spiritual
belief
and
superstition
(Kamuaro
1998).
Clearly,
the
Arabuko
Sokoke
Forest
has
significant,
if
declining,
cultural
value.
However,
it
was
not
possible
to
quantify
the
cultural
value
in
this
study,
since
the
survey
focused
on
visitors
to
the
forest
and
Kenya,
and
did
not
include
local
people.
However,
an
indication
of
cultural
value
can
be
illustrated
with
reference
to
a
case
study
from
Zimbabwe,
where
cultural
value
was
ranked
in
order
of
its
relative
importance
to
local
communities,
and
was
placed
above
medicines
and
crafts,
but
below
poles
and
fuel-wood
(International
Institute
of
Environment
and
Development
1997).
Forest
costs
The
forest
clearly
has
values
that
can
be
translated
into
realisable
benefits
to
the
Forest
Department
and
local
populations.
However,
these
values
should
be
weighed
against
the
costs
of
conserving
and
protecting
the
forest.
The
costs
associated
with
conservation
of
forests
or
protected
areas
are
both
direct
and
indirect
(Dixon
and
Sherman
1991).
Direct
costs
are
the
management
costs,
which
include
cost
of
staffing,
maintenance
of
vehicles
and
other
conservation
operations.
A
large
portion
of
recurrent
expenditure
is
spent
on
policing
to
stop
poaching
of
forest
produce
by
the
local
people.
Indirect
costs
are
the
opportunity
costs
of
developments
forgone
due
to
conserving
land
as
forest
(Dixon
and
Sherman
1991).
The
direct
management
costs
borne
by
the
forest
managing
authorities
are
shown
in
Table
7.
The
costs
of
forest
conservation
are
in
excess
of
US$100000
per
year.
1447
Table
7.
Forest
management
costs
(US$).
Item
1997/1998
1998/1999
%1997/1998
%1998/1999
Salaries
114
098
104
383
89.18
98.02
Transport
operating
expenses
9533.43
1071.43
7.45
1.01
Travelling
and
accommodation
-
140.14
0.13
Passage
and
leave
108
-
0.08
-
Postal
and
telegrams
142.86
21.43
0.11
0.2
Telephone
expenses
1142.86
82.71
0.9
0.08
Electricity
expenses
-
31.71
0.03
Water
charges
-
102.86
0.1
Fungicides,
insecticides
and
sprays
-
214.29
0.2
Purchase
of
uniform
and
clothing
-
101.14
0.09
Purchase
of
stationary
-
140.57
0.13
Supplies
for
production
1607.43
-
1.25
Advertising
and
publicity
-
35.14
0.03
Miscellaneous
and
other
charges
-
19.30
0.02
Purchase
of
plants
and
equipment
1135.71
12.86
0.9
0.01
Maintenance
of
machinery
-
44.57
0.04
Maintenance
of
water
supply
-
35.14
0.03
Maintenance
of
roads
171.43
60.43
0.13
0.06
Total
127
940
106
497
Source:
Government
Printed
Estimated,
Republic
of
Kenya,
1997/1998
and
1998/1999.
Much
of
the
budget
allocation
goes
to
payment
of
staff
salaries;
however,
this
and
other
aspects
of
the
budget
have
reduced
in
recent
years,
indicating
the
declining
budget
allocated
to
Arabuko
Sokoke
Forest.
This
may
be
attributed
to
the
Structural
Adjustment
Programmes
which
advocate
the
reduction
of
budget
deficit
through
reduction
of
expenditures
and
budget
rationalisation
(Republic
of
Kenya
1997).
The
opportunity
cost
of
conserving
the
Arabuko
Sokoke
Forest
is
evident
from
arable
farming
and
other
land
use
patterns
in
forest
adjacent
areas.
Estimation
of
indirect
costs
of
conserving
the
forest
is
beyond
the
scope
of
this
study
*
.
Other
costs
associated
with
conservation
of
Arabuko
Sokoke
Forest
include
the
losses
due
to
game
damage
on
neighbouring
farmland.
Abelson
(1996)
has
estimated
this
at
US$
45
000
per
year
for
farms
bordering
the
forest.
Discussion
By
combining
existing
and
new
data
relating
to
both
local
and
distant
populations,
it
has
been
possible
to
build
up
a
comprehensive
picture
of
the
costs
and
benefits
of
the
Arabuko
Sokoke
Forest
in
Kenya.
Despite
the
poor
state
of
the
forest
for
logging
purposes,
the
continued
conservation
of
Arabuko
Sokoke
Forest
clearly
yields
multiple
economic
benefits
(Wass
1995;
Abelson
1996).
Table
8
shows
the
esti-
*
The
opportunity
costs
of
conserving
protected
areas
in
Kenya
as
a
whole
have
been
estimated
and
reported
by
Norton-Griffith
(1995).
1448
Table
8.
Quantified
costs
and
benefits
of
Arabuko
Sokoke
Forest
(US$).
Cost
of
forest
conservation
Forest
management
cost
in
1997
106
497
Cost
of
wildlife
damage
to
farmlands
in
1997
45
000
Actual
direct
benefits
Forest
Department
revenues
in
1997
7536
Tourism
revenues
in
1997
469
Revenues
from
butterfly
farming
in
1997
31
394
Illegal
subsistence
use
of
forest
in
1994
21
184
Potential
income
from
entry
fee
in
1999
2676-9324
Potential
donation
for
forest
conservation
(one
off
donation)'
22
730
000
'Note
the
discussion
on
calibration
in
the
text,
which
means
that
these
figures
should
be
treated
with
some
caution.
mated
values
of
the
different
uses
of
the
forest,
and
an
estimate
of
the
non-use
value
derived
from
the
contingent
valuation
survey.
Clearly,
it
would
be
wrong
to
aggregate
these
figures,
as
they
relate
to
different
years,
and
the
methods
of
data
collection
are
not
compatible.
However,
they
do
illustrate
the
asymmetry
between
costs
of
conserving
the
forest
and
the
revenue
the
forest
is
currently
bringing
into
the
Kenyan
Forestry
Department.
Even
when
the
benefits
to
local
communities
of
illegal
uses
of
the
forest
are
taken
into
account,
the
costs
appear
to
be
significantly
greater
than
the
benefits
of
conserving
the
forest.
Currently
no
entry
fee
is
charged
to
visitors
of
the
forest,
however,
a
significant
increase
in
income
could
come
from
making
some
charge.
Even
if
this
was
to
take
place
however,
the
evidence
indicates
that
income
from
the
forest
would
not
outweigh
the
costs
incurred
to
the
Forest
Department.
It
is
only
when
non-use
values
are
taken
into
account
that
it
becomes
economically
sound
to
conserve
the
forest.
This
highlights
the
problem
that,
although
there
is
clearly
additional
value
for
the
forest
both
within
Kenya
and
the
rest
of
the
world,
the
means
to
capture
this
value
in
terms
of
benefits
to
local
populations
and
the
Forest
Department
are
not
currently
available.
The
butterfly
farm
project
is
of
interest
and
may
indicate
means
by
which
such
global
non-use
values
can
be
captured.
Although
the
project
itself
captures
use-
value
of
the
forest,
the
funding
comes
from
the
GEF,
and
represents
a
transfer
of
funds
from
international
populations
to
local
populations.
The
GEF
is
a
financial
mechanism
that
provides
funds
to
recipient
countries
for
projects
and
activities
that
aim
to
protect
the
global
environment.
It
is
jointly
implemented
by
the
United
Nations
Development
Programme
(UNDP),
the
United
Nations
Environment
Pro-
gramme
(UNEP)
and
the
World
Bank.
The
funding
which
facilitates
the
butterfly
farming
project
in
the
forest
could
therefore
be
argued
to
relate
to
international
non-use
values
for
the
forest.
This
GEF
funding
can
be
seen as
a
means
by
which
some
of
the
global
value
such
as
that
estimated
from
the
contingent
valuation
survey
can
be
captured.
This
facility
not
only
allows
such
value
to
be
transferred,
but
allows
additional
use-value
to
be
captured
which
then
becomes
a
sustainable
source
of
income
from
the
forest
for
local
communities.
1449
Conclusions
This
study
shows
that
even
when
a
forest
has
been
degraded
to
such
an
extent
that
commercial
logging
is
no
longer
viable,
it
may
still
hold
significant
value,
which
suggests
that
conservation
of
the
forest
should
take
place.
Although
much of
this
value
is
captured
by
local
people
in
terms
of
legal
and
illegal
use
of
the
forest,
other
values
do
not
accrue
to
local
people.
Capturing
some
of
these
values
may
be
a
relatively
simple
matter,
such
as
charging
an
entry
fee
to
the
forest,
and
to
attempt
to
increase
visitors
in
the
area
through
the
establishment
of
a
visitor
centre,
nature
trails
and
trained
guides.
However,
it
may
be
more
difficult
to
realise
other
benefits,
such
as
international
non-use
value.
The
Arabuko
Sokoke
Forest
has
benefited
from
the
transfer
of
some
international
funds,
which
may
be
seen
as
a
realisation
of
international
non-use
value.
However,
a
large
proportion
of
non-use
value
of
the
forest
remains
unavailable
to
local
populations
and
the
Forest
Department.
The
case
of
the
Arabuko
Sokoke
Forest
illustrates
that,
whilst
it
is
possible
to
quantify
many
aspects
of
value
of
a
forest,
according
to
the
economic
approach
to
saving
the
tropical
rainforests,
further
efforts
are
required
to
design
capture
mechanisms
which
allow
benefits
to
be
realised.
Such
mechanisms
are
particularly
vital
for
the
capture
of
global
non-use
values,
which
in
the
case
of
the
Arabuko
Sokoke
Forest
are
captured
to
a
limited
extent
only.
It
is
only
when
practical
capture
mechanisms
are
implemented
that
the
economic
approach
provides
a
robust
rationale
for
conserving
rainforests
such
as
the
Arabuko
Sokoke
Forest.
References
Abelson
P.
1996.
Project
Appraisal
and
Valuation
of
the
Environment.
General
Principles
and
Six
Case-Studies
in
Developing
Countries.
Macmillan
Press
Ltd,
Basingstoke,
UK.
Alfarez-Farizo
B.,
Hanley
N.,
Wright
R.E.
and
Macmillan
D.C.
1999.
Estimating
the
benefits
of
agri-environment
policy
econometric
issues
in
open-ended
contingent
valuation
studies.
Journal
of
Environmental
Planning
and
Management
42:
23-43.
Arabuko
Sokoke
Forest
Annual
Report
(1999)
Kenyan
Forest
Department,
Nairobi,
Kenya.
Arrow
K.,
Solow
R.,
Portney
P.R.,
Learner
E.E.,
Radner
R.
and
Schuman
H.
1993.
Report
of
the
NOAA
panel
on
contingent
valuation.
Federal
Register
58:
4601-4614.
Ayiemba
W.O.
1998.
Biodiversity
Utilisation
by
the
Local
Communities.
Impact
of
Butterfly
Farming
on
Wild
Population
of
Butterfly
to
Forest
Conservation.
National
Museums
of
Kenya,
Nairobi,
Kenya.
Bann
C.
2000.
A
Contingent
Valuation
of
the
Mangroves
of
Benut.
Paper
presented
at
the
European
Association
of
Environmental
and
Resource
Economists
Conference,
Crete.
Johor
State,
Malaysia
(unpublished).
Boyle
K.J.,
Johnson
F.R.,
McCollum
DW.,
Desvouges
W.H.,
Dunford
R.W.
and
Hudson
S.P.
1996.
Valuing
public
goods:
discrete
versus
continuous valuation
responses.
Land
Economics
72:
381-396.
Carson
R.
1994.
A
Bibliography
of
Contingent
Valuation
Studies
and
Papers.
University
of
California,
San
Diego,
California,
pp.
1-104.
Dixon
J.A.
and
Sherman
P.B.
1991.
Economics
of
Protected
Areas.
A
New
Look
at
Benefits
and
Costs.
Earthscan
Publications,
London.
Edwards-Jones
G.
and
Kenyon
W.
1998.
What
level
of
information
enables
the
public
to
act
like
experts
when
evaluating
ecological
goods.
Journal
of
Environmental
Planning
and
Management
41:
463
475.
1450
Emerton
L.
1994.
Summary
of
the
Current
Value
Use
of
Arabuko
Sokoke.
Forest
Department,
Nairobi,
Kenya.
Fitz-Gibbon
C.D.,
Mogaka
H.
and
Fanshawe
J.H.
1995.
Subsistence
hunting
in
Arabuko
Sokoke
Forest
and
its
effect
on
mammal
population.
Conservation
Biology
9:
1116-1126.
Hanley
N.
and
Ruffell
R.J.
1993.
The
contingent
valuation
of
forest
characteristics:
two
experiments.
Journal
of
Agricultural
Economics
44:
217-229.
International
Institute
of
Environment
and
Development
1997.
Valuing
the
Hidden
Harvest:
Methodo-
logical
Approaches
for
Local-level
Economic
Analysis
of
Wild
Resources.
BED,
London.
Kamuaro
0.
1998.
State
and
community
conflicts
in
natural
resource
management
in
Kenya.
In:
Veit
P.
(ed.),
Africa's
Valuable
Assets.
A
Reader
in
Natural
Resource
Management.
World
Resources
Institute,
Washington,
DC.
Kramer
R.A.
and
Mercer
D.E.
1997.
Valuing
a
global
environmental
good:
US
residents'
willingness
to
pay
to
protect
tropical
rain-forest.
Land
Economics
73:
196-210.
Kingdon
J.
1997.
The
Kingdon
Field
Guide
to
African
Mammals.
Academic
Press,
San
Diego,
California.
KIFCON
1991.
Arabuko
Sokoke
Forest
Ornithological
Survey.
Forest
Department,
Nairobi,
Kenya.
KIFCON
1992.
Arabuko
Sokoke
Forest
Vegetation
Survey.
Forest
Department,
Nairobi,
Kenya.
KIFCON
1993.
Arabuko
Sokoke
Forest
Mammal
Survey.
Forest
Department,
Nairobi,
Kenya.
KIFCON
1995.
Arabuko
Sokoke
Forest
and
Mida
Creek,
The
Official
Guide.
Forest
Department,
Nairobi,
Kenya.
Mitchell
R.C.
and
Carson
R.T.
1989.
Using
Surveys
to
Value
Public
Goods:
The
Contingent
Valuation
Method.
Resources
for
the
Future,
Washington,
DC.
Negussie
G.
1997.
Use
of
traditional
values
in
search
for
conservation
goals:
the
Kaya
forests
of
the
Kenyan
coast.
In:
Doolan
S.
(ed.),
African
Rainforest
and
the
Conservation
of
Biodiversity.
Proceedings
of
the
Limbe
Conference,
17-24
January
1997,
Limbe
Botanic
Garden,
Cameroon.
Earthwatch
Europe,
Oxford.
Norton-Griffith
M.
1995.
The
opportunity
cost
of
biodiversity
conservation
in
Kenya.
Ecological
Economics
12:
125-139.
Pearce
D.W.
1991.
An
economic
approach
to
saving
the
tropical
forests.
In:
Helm
D.
(ed.),
Economic
Policy
Towards
the
Environment.
Blackwell,
Oxford,
pp.
239-262.
Pearce
D.W.
and
Moran
D.
1994.
The
Economic
Value
of
Biodiversity.
Earthscan,
London.
Pearce
D.,
Georgiou
W.S.
and
Moran
D.
1994.
Economic
Values
and
the
Environment
in
Developing
Countries.
A
Report
to
the
United
Nations
Environmental
Programme.
UNEP,
Nairobi,
Kenya.
Ready
R.,
Buzby
J.C.
and
Hu
D.
1996.
Difference
between
continuous
and
discrete
contingent
valuation
estimates.
Land
Economics
72:
397-411.
Republic
of
Kenya
1968.
Sessional
Paper
No.
1
of
1968,
A
Forest
Policy
for
Kenya.
Government
Printer,
Nairobi,
Kenya.
Republic
of
Kenya
1997.
National
Development
Plan
1997-2001.
Government
Printer,
Nairobi,
Kenya.
Wass
P.
1995.
Kenya
Indigenous
Forests,
Status,
Management
and
Conservation.
The
World
Conserva-
tion
Union,
Gland,
Switzerland.