the state of lemur conservation in south-eastern ...€¦ · lemurs, madagascar’s non-human...
TRANSCRIPT
Oryx Vol 39 No 2 April 2005
© 2005 FFI, Oryx, 39(2), 204–218 doi:10.1017/S0030605305000451 Printed in the United Kingdom204
Abstract The unique primates of south-easternMadagascar face threats from growing human popula-tions. The country’s extant primates already representonly a subset of the taxonomic and ecological diversityexisting a few thousand years ago. To prevent furtherlosses remaining taxa must be subjected to effectivemonitoring programmes that directly inform conserva-tion efforts. We offer a necessary first step: revision ofgeographic ranges and quantification of habitat areaand population size for diurnal and cathemeral (activeduring both day and night) lemurs. Recent satelliteimages are used to develop a forest cover geographical
information system, and censuses are used to establishrange boundaries and develop estimates of populationdensity and size. These assessments are used to identifyregions and taxa at risk, and will be a useful baselinefor future monitoring of habitat and populations. Preciseestimates are impossible for patchily-distributed taxa(especially Hapalemur aureus, H. simus and Vareciavariegata variegata); these taxa require more sophisticatedmodelling.
Keywords Conservation status, geographic range, GIS,lemurs, Madagascar, population densities, primates.
The state of lemur conservation in south-eastern Madagascar:population and habitat assessments for diurnal and cathemerallemurs using surveys, satellite imagery and GIS
Mitchell T. Irwin, Steig E. Johnson and Patricia C. Wright
Mitchell T. Irwin (Corresponding author) Department of Anthropology, SBS
Building, 5th Floor, Stony Brook University, Stony Brook, NY 11794-4364,
USA. E-mail [email protected]
Steig E. Johnson Department of Anthropology, 2500 University Drive NW,
University of Calgary, Calgary, AB T2N 1N4, Canada.
Patricia C. Wright Department of Anthropology and Institute for the
Conservation of Tropical Environments, Stony Brook University, Stony Brook,
NY 11794-4364, USA.
Received 22 February 2004. Revision requested 9 June 2004.
Accepted 2 November 2004.
Introduction
The island nation of Madagascar has recently beenclassified as both a megadiversity country and one of25 biodiversity hotspots, a classification reserved forregions combining high biodiversity with high levelsof habitat loss and extinction risk (Myers et al., 2000).Madagascar has a diverse range of habitats, created bythe interaction of an east-west rainfall gradient (mostrainfall occurs in the east) and a north-south latitudinalgradient (the 1,600 km-long island spans 12–25° S).
Native vegetation has been decimated throughoutMadagascar since the arrival of humans on the islandc. 2,000 years ago (Burney & MacPhee, 1988; Burney,1999). Archaeological evidence suggests that humanoccupation began in the south-west, followed by agradual spread northwards and eastwards. The east mayhave been the last area colonized; its humidity and lowaltitude may have contributed to a higher risk of endemic
diseases (Burney, 1999). However, once this ecoregionwas inhabited, its combination of abundant timber andnutrient-poor soil (causing a low agricultural tenuretime) led to rapid deforestation.
Green & Sussman (1990) used satellite images from1973 and 1985 and vegetation maps from 1950 to recon-struct the deforestation history of Madagascar’s easternrainforests. They found that only 3.8 million ha (34% ofthe original cover) existed in 1985, and extrapolated adeforestation rate of 111,000 ha (1.5%) per year. At thispace, the entire loss of this ecosystem is predicted tooccur by 2020.
Lemurs, Madagascar’s non-human primates, are justone threatened group (17 of the 81 lemur taxa existinga few thousand years ago are extinct; Godfrey & Jungers,2002). As an endemic radiation of charismatic mega-fauna, lemurs have received international attention,although much remains to be learned about theirdistributions and population sizes. The most recentpublications (e.g. Mittermeier et al., 1994) illustrateranges showing little congruence with the actual extentof remaining rainforest (Green & Sussman, 1990), andprovide only general population estimates. No previousstudy has combined satellite imagery and censuses togenerate more accurate figures (but see Sussman et al.,2003).
South-eastern Madagascar (including all easternrainforest from its southernmost extent to the Onive/Mangoro River in the north) contains 14 lemur taxa innine genera and six families (Table 1). Of these, five(Hapalemur simus, Hapalemur aureus, Eulemur collaris,
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
205
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
Eulemur albocollaris and Propithecus diadema edwardsi) areendemic to the region, seven (Hapalemur griseus, Eulemurrubriventer, Varecia variegata variegata, Avahi laniger,Lepilemur microdon, Cheirogaleus major and Microcebusrufus) are endemic to eastern rainforests, and only two(Eulemur fulvus rufus and Daubentonia madagascariensis)are more widely distributed. This region contains 22% ofMadagascar’s 64 lemur taxa, 29% of its 48 species, 64%of its 14 genera and all of its five endemic families (tax-onomy follows CBSG, 2002). There are seven protectedareas in the south-east (Fig. 1 & Table 2).
This paper aims to provide a quantitative assessmentof diurnal and cathemeral (active during both day andnight) lemur populations in south-eastern Madagascar.To achieve this we: (1) Use census data to delineate geo-graphic ranges, generate population density estimatesfor each taxon at each site and estimate overall densityfor each taxon across its range. (2) Create a geographicalinformation system (GIS) coverage representingrainforest cover, and subdivide this area based on rangesof individual taxa to quantify available habitat. (3) Com-bine population densities and habitat assessments toestimate the regional population size for each taxon.
Nocturnal lemurs were excluded for four reasons.Firstly, population sizes of nocturnal species as currentlyrecognized are probably high; all are found throughoutand beyond the region, are smaller-bodied, and liveat higher population densities. Secondly, they are lessaffected by hunting (although those sleeping in treeholes may be vulnerable). Thirdly, the assumptions of
line-transect surveys are harder to meet for nocturnalspecies (Duckworth, 1998), causing greater estimation er-ror. Fourthly, the taxonomy of eastern nocturnal speciesis poorly understood; revisions of nocturnal lemurtaxonomy elsewhere in Madagascar (Rasoloarison et al.,2000) found greater species richness than previouslyrecognized. Population size estimates should thereforeawait revisions of taxonomy and geographic range.
Methods
Population densities
Between 1995 and 2001 we censused 20 localitiesthroughout the south-east (Fig. 1 & Table 3). Data wascollected primarily by ourselves, but supplemented byunpublished data provided by S. Arrigo-Nelson andC. Grassi as part of an ongoing collaboration (Johnsonet al., 2003). Methodology has been described elsewhere(Johnson & Overdorff, 1999; Irwin et al., 2000, 2001) andfollowed previous studies (Struhsaker, 1981; Fashing &Cords, 2000). At each site 1–4 transects were established(length 0.8–3.8 km); existing trails were sometimesused but only when not affected by anthropogenicdisturbance. For each lemur sighting we recordedobserver-to-animal and perpendicular sighting distances(measured to the centre of the group), and group size.
Transect area was calculated using the histogram-inspection technique (Whitesides et al., 1988; Plumptre,2000). Observer-animal sighting distances were placed
Table 1 Diurnal/cathemeral and nocturnal lemur taxa occurring in south-eastern Madagascar, with their body mass, diet and Red Liststatus (IUCN, 2004)1.
Body Red ListFamily Taxon Common name mass (kg)2 Diet status3
Diurnal/ Indriidae Propithecus diadema edwardsi Milne-Edwards’ sifaka 6.0 leaves, seeds, fruits ENcathemeral Lemuridae Eulemur fulvus rufus4 Red-fronted brown lemur 2.2 fruit LR/nt
Eulemur albocollaris White-collared brown lemur 2.2 fruit CREulemur collaris Collared brown lemur 2.5 fruit VUEulemur rubriventer4 Red-bellied lemur 2.0 fruit VUHapalemur griseus griseus4,5 Lesser bamboo lemur 0.7 bamboo, fruit LR/ntHapalemur aureus Golden bamboo lemur 1.5 bamboo CRHapalemur simus Greater bamboo lemur 2.0 bamboo CRVarecia variegata variegata4 Black-and-white ruffed lemur 3.6 fruit EN
Nocturnal Cheirogaleidae Cheirogaleus major Greater dwarf lemur 0.4 fruit, nectar, insects LR/ntMicrocebus rufus Brown mouse lemur 0.04 fruit, insects LR/nt
Daubentoniidae Daubentonia madagascariensis Aye-aye 2.6 insect larvae, seeds ENIndriidae Avahi laniger Eastern woolly lemur 1.2 leaves LR/ntLepilemuridae Lepilemur microdon Small-toothed sportive lemur 1.0 leaves LR/nt
1General references: Petter et al. (1977), Tattersall (1982), Harcourt & Thornback (1990), Mittermeier et al. (1994)2Average of male and female body mass from Smith & Jungers (1997)3IUCN (2004), CBSG (2002): LR/nt, Lower Risk/near threatened; VU, Vulnerable; EN, Endangered; CR, Critically Endangered4Taxa whose range extends beyond area considered in this study5H. g. griseus is considered here to include the range of the putative taxon H. (g.) meridionalis and other unnamed variants that may exist inthe south-east (Fausser et al., 2002). As it was impossible in our surveys to distinguish between H. g. griseus and H. g. meridionalis, we cannotcomment on the validity or geographic range of these taxa.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
206 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
Fig. 1 Rainforest of south-eastern Madagascar (green) based on Landsat 7 satellite images for 1999–2001, with east-flowing rivers andprotected areas (SR, Special Reserve; NP, National Park) including rainforest (outlined in red; data courtesy of ANGAP). Rainforestcorridor extends north past study area (indicated in grey). Red dots indicate the 20 survey sites (see Table 3). Inset: position of study areawithin Madagascar, with extent of satellite images used in this study. Image ID numbers and acquisition dates are:1, E1SC:L70RWRS.002:2000973352, 19/04/2000; 2, E1SC:L70RWRS.002:2000319961, 17/10/1999; 3, E1SC:L70RWRS.002:2000443464, 11/11/1999; 4, E1SC:L70RWRS.002:2000319964, 17/10/1999; 5, E1SC:L70RWRS.002:2001961696, 29/11/2000; 6, E1SC:L70RWRS.002:2004077212,13/09/2001.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
207
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
into 5 m categories, and the boundary at which observa-tion frequencies dropped to f50% of the previousinterval was the ‘fall-off distance’. We used observer-animal distance because perpendicular distances werenot available for some sites, and, although the methodwas formulated for perpendicular distances, it can beapplied to observer-animal distances (Fashing & Cords,2000). Considering all species for which both typesof data were available, and for which observationsexceeded 30 sightings, density estimates differed by<15% (n= 3).
Effective distance E was calculated as:
EN
FDf
=Nt
where Nt is the number of sightings, Nf is the number ofsightings at less than the fall-off distance, and FD is thefall-off distance (Whitesides et al., 1988). Effective dis-tance was calculated individually for each species, exceptfor the brown lemur group, within which sightings werepooled.
Population density D was calculated as:
DELt
=I
2
where I=numbers of individuals observed, E= effectivedistance, and Lt= total transect length (transect length *number of repetitions).
Geographic ranges
In assessing geographic ranges we considered the mostrecent published data as null hypotheses to be tested byour data. Our survey locations were distributed through-out the study area, except for the extreme south; toconfirm species incidence in this region we relied onO’Connor et al. (1986) and Feistner & Schmid (1999).
Because of the close spacing of survey sites, we believeour data provide a fine-grained test of range limits.
Habitat assessment
We analyzed six Landsat 7 multispectral satellite imagesacquired between 1999 and 2001 (USGS, 2004). Imageswere chosen to be in as narrow a time window aspossible whilst minimizing cloud cover. Using ErdasImagine (Leica Geosystems, Atlanta, USA) we performeda 2-class supervised classification to identify forestedareas. For each image, a spectral signature was acquiredfrom a ground-truthed rainforest area, and this signature(incorporating visible, infrared, and near-infrared wave-lengths), was used to classify the image into forest andnon-forest. The resulting images were imported intoArcGIS 8.0 (ESRI, Redlands, USA) and smoothed usingmajority neighbourhood statistics. All fragments <1 hawere removed.
The six images were then examined to judge theamount of offset between them in areas of overlap (dueto small georectification errors). Merging offset imagescreates a ghosting effect, whereby forest in mosaic habi-tats can be counted twice. When offset was found, imageswere manually shifted to match one another (accuracyin area calculations was given priority over small geo-graphic shifts). The images were then joined to create asingle image; in overlap areas, pixels were judged to beforest if one or more of the contributing images showedforest at that location (this additive approach was criticalin one overlap area, where images had cloud cover at dif-ferent locations). The resulting image was then convertedto a polygon coverage and clipped to correspond to theobserved geographic ranges of individual lemur taxa,thus producing an ‘available habitat’ layer for eachtaxon.
To investigate the forest’s altitudinal distribution, weused a GTOPO30 digital elevation model, (USGS, 2003)
Table 2 Protected areas of south-eastern Madagascar (see Fig. 1 for locations), with date gazetted, area, approximate rainforest area, andaltitudinal range1.
Name Date gazetted Area (ha) Approximate rainforest area (ha) Altitudinal range (m)
Ranomafana National Park 1991 43,500 43,500 400–1,534Andringitra National Park 19272 31,160 15,000 650–2,658Pic d’Ivohibe Special Reserve 1964 3,453 3,453 775–2,060Kalambatritra Special Reserve 1959 28,250 13,0603 740–1,680Manombo Special Reserve 1962 5,320 2,6604 0–137Midongy du Sud National Park 1997 197,900 197,900 850–1,357Andohahela National Park 19395 76,020 63,1006 90–1,972Totals 385,603 338,673
1From Nicoll & Langrand (1989), Ramarokoto et al. (1999)2Upgraded from Réserve Naturelle Intégrale (Integral Natural Reserve) to National Park in 19983From Intercoopération Suisse & Marie ERTA (1999)4Approximately half of Manombo Special Reserve remains forested (Steig E. Johnson, pers. obs.; J. Ratsimbazafy, pers. comm.)5Upgraded from Réserve Naturelle Intégrale to National Park in 19976Parcel 1 (rainforest) only; Parcels 2 and 3 (dry and transitional forest) excluded
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
208 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
Tab
le 3
The
20
surv
ey s
ites
(see
Fig
. 1 fo
r lo
cati
ons)
, wit
h th
eir
alti
tud
e, in
vest
igat
ors
and
ref
eren
ces
(whe
re a
ppro
pria
te),
and
num
ber
of tr
anse
cts
and
tota
l tra
nsec
t len
gth
of s
urve
ys.
No.
Site
1L
ocal
ity
Alt
itud
e (m
)In
vest
igat
or(s
)R
efer
ence
No.
of t
rans
ects
Tot
al tr
anse
ct le
ngth
(km
)
1K
iris
iasy
20d1
7.4p
S, 4
7d41
.4p E
1,20
0–1,
400
M. I
rwin
, T. S
mit
hIr
win
et a
l., 2
000
240
2M
arof
otsy
21d0
0.0p
S, 4
7d28
.0p E
1,00
0–1,
200
M. I
rwin
, T. S
mit
hIr
win
et a
l., 2
000
252
3R
anom
afan
a N
P: N
amah
oaka
21d0
7.5p
S, 4
7d32
.3p E
1,10
0–1,
200
M. I
rwin
, T. S
mit
hIr
win
et a
l., 2
000
235
4R
anom
afan
a N
P: B
evoa
hazo
21d1
0.1p
S, 4
7d30
.5p E
1,05
0–1,
250
P. W
righ
t1
405
Ran
omaf
ana
NP:
Mia
rano
ny21d1
0.9p
S, 4
7d32
.8p E
800
–1,1
00S.
Arr
igo-
Nel
son
240
6R
anom
afan
a N
P: M
anid
ika
21d1
6.9p
S, 4
7d23
.9p E
1,10
0–1,
300
S. A
rrig
o-N
elso
n2
487
Ran
omaf
ana
NP:
Vat
ohar
anan
a21d1
7.4p
S, 4
7d26
.0p E
1,02
5C
. Gra
ssi,
S. Jo
hnso
n,Jo
hnso
n &
Ove
rdor
ff,
411
4P.
Wri
ght
1999
; Joh
nson
, 200
28
Tol
ongo
ina:
Man
dri
and
ry21d3
5.5p
S, 4
7d29
.1p E
750
P. W
righ
t1
409
Am
bant
ofot
sy: A
mba
haka
21d4
4.2p
S, 4
7d24
.5p E
750
P. W
righ
t1
4010
Ikon
go: A
mba
tam
be21d4
9.3p
S, 4
7d21
.5p E
625
P. W
righ
t1
4011
Ikon
go: A
nkop
akop
aka
21d4
9.7p
S, 4
7d20
.3p E
645
P. W
righ
t1
4012
And
ring
itra
NP:
Imai
tso
22d0
8.0p
S, 4
6d56
.0p E
1,50
0S.
John
son,
S.
John
son,
200
23
52R
azaf
iman
dim
by13
And
ring
itra
NP:
Am
baro
ngy
22d1
3.3p
S, 4
7d01
.2p E
725
–1,1
00S.
John
son
John
son
& W
yner
, 200
0;3
106
John
son,
200
214
And
ring
itra
NP:
‘Par
c’22d1
3.3p
S, 4
7d01
.1p E
725
–900
S. Jo
hnso
nJo
hnso
n &
Wyn
er, 2
000;
147
John
son,
200
215
Eve
ndra
222d2
6.1p
S, 4
7d13
.8p E
425
S. Jo
hnso
nJo
hnso
n &
Wyn
er, 2
000
––
16V
evem
be22d4
7.7p
S, 4
7d11
.3p E
525
S. Jo
hnso
nJo
hnso
n, 2
002
187
17L
ambo
hazo
222d5
2.9p
S, 4
7d11
.3p E
c. 3
00S.
John
son
John
son
& O
verd
orff
, 199
9–
–18
Man
ombo
SR
23d0
1.5p
S, 4
7d42
.0p E
25
S. Jo
hnso
n, P
. Wri
ght
John
son
& O
verd
orff
, 199
92
3219
Kal
amba
trit
ra S
R23d2
2.4p
S, 4
6d28
.2p E
1,40
0–1,
680
M. I
rwin
, K. S
amon
ds
Irw
in e
t al.,
200
12
2420
Mid
ongy
du
Sud
NP
23d4
6.0p
S, 4
7d01
.0p E
1,05
0P
. Wri
ght
347
Tot
als
3492
4
1 NP,
Nat
iona
l Par
k; S
R, S
peci
al R
eser
ve2 S
ite
used
for
det
erm
inin
g ra
nge
boun
dar
ies
only
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
209
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
The model was classified into 200 m intervals andintersected with the forest coverage, dividing the forestby altitudinal category.
Population estimates
We estimated the regional population size of each taxonby multiplying estimated habitat area (km2) by averagepopulation density (individuals km−2). Population den-sity values were the arithmetic means of densities at allsites within the taxon’s range, including sites wherethey are known to exist but were not detected duringcensuses (in these cases, zero values were used). We alsocalculated estimates using only habitat within protectedareas.
Sources of error
Spatial heterogeneity in population densityPopulation density was estimated using the mean ofdensities at different sites; we did not model spatialheterogeneity. This undoubtedly leads to error, as lemuroccurrence and densities vary non-randomly withhabitat type, disturbance and altitude (Ganzhorn, 1995;Sterling & Ramaroson, 1996; Ganzhorn et al., 1997a;Goodman & Rasolonandrasana, 2001). However, thesampling effort necessary to adequately model thisheterogeneity is beyond the scope of this study. As oursurvey sites are evenly distributed throughout thesouth-east (geographically and by altitude), and includeprotected and unprotected forests, we believe that ourestimates are reasonable reflections of overall densities.Estimates will be least accurate for patchily-distributedtaxa.
Interspecific differences in detectionTaxa differ in habitat use, locomotion, group size,anti-predator strategies and response to observers; thesefactors cause differences in detectability. Our methodsunderestimate the population size of less easily detectedtaxa.
Misclassification of forest in satellite imageryWe believe that misclassification was rare in our analysis.Ground-truthing upheld the sharp distinction betweenforest and non-forest visible even in raw images.However, it is possible that our forest coverage containssome non-habitat areas (e.g. degraded forest, planta-tions), leading to overestimation. Further ground-truthing would be necessary to determine whether theseareas can be reliably distinguished.
Under-representation of forest due to cloud coverAlthough we attempted to avoid cloud cover in theimages certain forest areas were obscured, and there are
no reliable techniques of countering this shortcoming.We considered cloud-covered areas as non-forest; under-estimation of habitat area due to clouds was estimated tobe <5%, and is primarily found between the Mananaraand Manampatrana Rivers.
Results
Habitat assessment
We found that 11,433 km2 of rainforest remains in thesouth-east (Fig. 1), of which 3,386 km2 (29.6%) is withinprotected areas. Visual inspection reveals that thenorth-south corridor has become discontinuous in twolocations. The first is just north of Ranomafana (approxi-mately 21° S); the images used had no cloud cover in thisregion. The second location is between the Manam-patrana and Mananara Rivers (continuing south of theMananara). The images had significant cloud cover inthis area, making finer quantification of deforestationdifficult.
Remaining rainforest is concentrated at mid-altitudes(70% is between 600 and 1,400 m), straddling the escarp-ment dividing the eastern lowlands from the centralplateau (Fig. 2). Lowland rainforest (<500–600 m;Ganzhorn et al., 1997b) appears to be rare, especiallyrelative to the north-east (Green & Sussman, 1990). Thisincludes the unique coastal forest, which has beenreduced to disconnected fragments (Ganzhorn et al.,2001).
Lemur populations
Geographic ranges of lemur taxa as determined by thisstudy are shown in Fig. 3; Table 4 summarizes differ-ences between previous accounts of range boundariesand those found by this study. A revised incidence oflemurs in south-eastern protected areas is shown inTable 5.
Habitat area and population size estimates are summa-rized in Table 6. We did not calculate density for H. simus(due to low sample size), or estimate population size forthree patchily-distributed species (H. aureus, H. simusand V. variegata variegata).
Discussion
South-eastern Madagascar still contains viable foresthabitat, but continual monitoring and intervention willbe necessary for preservation. Specifically, two majorrecommendations arise from this study. Firstly, specialattention should be paid to the two regions in which theforest corridor has become discontinuous. These discon-tinuities will pose barriers to gene flow and reduceeffective population size of subdivided populations.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
210 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
Fig. 2 Areas of rainforest in south-eastern Madagascar by altitude.
Secondly, special attention should be paid to both low-altitude and high-altitude forests, as examples of bothare rare and isolated. Low-altitude forests exist as smallfragments, whereas high-altitude forests are part ofthe corridor but patchily-distributed (i.e. connected bylower-altitude forests). This combination of scarcity andfragmentation indicates disproportionate extinction riskfor taxa restricted to these forest types; i.e. local extinc-tions due to low effective population size, and lack ofrecolonization due to isolation. Efforts should be made toinvestigate the biodiversity at risk, and extend habitatprotection to include more lowland and montane areas.
Our surveys provided small but important clarifica-tions of range boundaries for the diurnal and cathemerallemur species of south-eastern Madagascar, and revealedthe large extent of the E. f. rufus x E. albocollaris hybridzone in and around Andringitra (Johnson & Wyner, 2000;Johnson, 2002; Wyner et al., 2002). We were able to esti-mate regional population size for seven of 10 taxa consid-ered. These estimates reveal a wide range of populationsizes (from <8,000 to almost 100,000 individuals) andindicate clear conservation priorities. Similarly, althoughthe overall proportion of protected habitat is relativelyhigh, species vary widely in the size of their protectedpopulation.
E. albocollaris appears to be the most seriously threat-ened, with by far the smallest overall and protectedpopulations. Its range was the area most affected bycloud cover; however, even assuming a generous under-estimate of area (e.g. 25%), this species is still morethreatened than other species. P. d. edwardsi also has asmall protected population, due to low density and thefact that the northern part of the study area, where theyare found, has less protected area than the south.E. rubriventer has similarly low numbers but is wellrepresented in protected areas in the north-east. Othertaxa are in less immediate danger, having large rangesand/or good representation in protected areas. How-ever, all taxa should be monitored both within andoutside protected areas.
Various characteristics of particular taxa affect the reli-ability of our results, and precluded estimations for threetaxa. Problems with estimation always involved patchyor incomplete presence within the geographic range,arising from the following processes: true ecologicalpatchiness due to habitat requirements, altitudinalrestrictions, and hunting. Specific cases are as follows:
Eulemur rubriventer is apparently confined to medium-and high-altitude forests (Mittermeier et al., 1994), butwe were unable to test this assertion because lowland
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
211
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
Fig. 3 Ranges of individual lemur taxa discussed in text: i, Propithecus diadema edwardsi; ii, Eulemur fulvus rufus (red), Eulemur fulvus rufus xEulemur albocollaris (green), Eulemur albocollaris (yellow), Eulemur collaris (purple); iii, Eulemur rubriventer; iv, Hapalemur griseus griseus;v, Hapalemur aureus and Hapalemur simus (same range); vi, Varecia variegata variegata.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
212 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
Tab
le 4
Ran
ge b
ound
arie
s of
diu
rnal
/ca
them
eral
lem
ur ta
xa in
sou
th-e
aste
rn M
adag
asca
r as
rep
orte
d b
y pr
evio
us s
tud
ies
and
this
stu
dy.
Tax
onPr
evio
us a
ccou
nts
of r
ange
bou
ndar
ies1
Ran
ge b
ound
arie
s in
dic
ated
by
this
stu
dy1
Eul
emur
fulv
us r
ufus
Man
goro
/O
nive
R. (
N) t
o vi
cini
ty o
f Man
ampa
tran
a R
. (S)
2,3
Man
goro
/O
nive
(N) t
o Pi
c d
’Ivo
hibe
SR
(S);
rang
e in
terr
upte
d b
y hy
brid
zon
e ar
ound
Man
goro
/O
nive
R. (
N) t
o M
anam
patr
ana
R. (
S)4,
5A
ndri
ngit
ra N
PE
. f. r
ufus
x E
. alb
ocol
lari
sN
one
Hea
dw
ater
s of
Man
ampa
tran
a R
. inc
lud
ing
And
ring
itra
NP,
& e
xten
din
g al
ong
the
corr
idor
s S
(tow
ard
s Iv
ohib
e) &
SE
(tow
ard
s th
e lo
wer
Man
ampa
tran
a R
.)E
ulem
ur a
lboc
olla
ris
Fara
ony
R. (
N) t
o M
anan
ara
R. (
S)2,
3 (t
ext n
ames
Far
aony
as
NFr
om h
ybri
d z
one
(wit
h E
. f. r
ufus
) SE
of A
ndri
ngit
ra N
P (N
) to
Man
anar
a R
. (S)
boun
dar
y bu
t fig
ure
ind
icat
es M
anam
patr
ana
R.)
Man
ampa
tran
a R
. (N
) to
Man
anar
a R
. (S)
4,5
Eul
emur
col
lari
sM
anan
ara
R. (
N) t
o S
coas
t2,3,
4,5
Man
anar
a R
. (N
) to
S co
ast
Eul
emur
rub
rive
nter
6B
eyon
d s
tud
y ar
ea (N
) to
Man
ampa
tran
a R
. and
Pic
d’I
vohi
be S
R2,
3,4,
5B
eyon
d s
tud
y ar
ea (N
) to
Man
ampa
tran
a R
. & P
ic d
’Ivo
hibe
SR
Hap
alem
ur g
rise
us g
rise
usB
eyon
d s
tud
y ar
ea (N
) to
S co
ast2,
3,4,
5B
eyon
d s
tud
y ar
ea (N
) to
S co
ast
Hap
alem
ur a
ureu
sV
icin
ity
of R
anom
afan
a N
P3V
icin
ity
of R
anom
afan
a N
P (N
) to
Man
ampa
tran
a R
. (S)
Vic
init
y of
Ran
omaf
ana
NP
(N) t
o vi
cini
ty o
f And
ring
itra
NP
(S)4
Vic
init
y of
Ran
omaf
ana
NP
and
vic
init
y of
And
ring
itra
NP
5
Hap
alem
ur s
imus
Vic
init
y of
Ran
omaf
ana
NP3
Vic
init
y of
Ran
omaf
ana
NP
(N) t
o M
anam
patr
ana
R. (
S)V
icin
ity
of R
anom
afan
a N
P (N
) to
vici
nity
of A
ndri
ngit
ra N
P (S
)4
Vic
init
y of
Ran
omaf
ana
NP
& v
icin
ity
of A
ndri
ngit
ra N
P5
Pro
pith
ecus
dia
dem
a ed
war
dsi7,
8M
ango
ro/
Oni
ve R
. (N
) to
‘nea
r M
anak
ara’
(S)2,
3M
ango
ro/
Oni
ve (N
) to
Man
ampa
tran
a R
. (S)
, not
incl
udin
g Iv
ohib
e SR
Man
goro
/O
nive
R. (
N) t
o M
anan
ara
R. (
S)4,
5
Var
ecia
var
iega
ta v
arie
gata
Bey
ond
stu
dy
area
(N) t
o ‘S
of F
araf
anga
na b
ut N
of t
heB
eyon
d s
tud
y ar
ea (N
) to
Man
ampa
tran
a R
. (S)
, wit
h ad
dit
iona
l iso
late
d p
opul
atio
n at
Man
anar
a R
.’ (S
)2,3
Man
ombo
SR
Bey
ond
stu
dy
area
(N) t
o M
anan
ara
R. (
S)4,
5
1 NP,
Nat
iona
l Par
k; S
R, S
peci
al R
eser
veSo
urce
s: 2T
atte
rsal
l (19
82);
3 Har
cour
t & T
horn
back
(199
0); 4 M
itte
rmei
er e
t al.
(199
4); 5 G
arbu
tt (1
999)
6 Our
sur
veys
con
firm
ed th
e so
uthe
rn li
mit
of E
. rub
rive
nter
at P
ic d
’Ivo
hibe
Spe
cial
Res
erve
, and
faile
d to
find
this
taxo
n so
uth
of th
e M
anam
patr
ana
Riv
er (c
ontr
a Pe
tter
et a
l., 1
977)
.7 T
here
has
bee
n so
me
ques
tion
abo
ut w
heth
er th
is ta
xon
is fo
und
wit
hin
Pic
d’I
vohi
be S
peci
al R
eser
ve. B
oth
Nic
oll &
Lan
gran
d (1
989)
and
Mit
term
eier
et a
l. (1
994)
sta
te th
at it
is p
rese
nt o
rre
port
ed, w
here
as th
e re
cent
sys
tem
atic
sur
vey
of R
asol
oari
son
& R
asol
onan
dra
sana
(199
9) fa
iled
to fi
nd it
in th
e re
serv
e, b
ut r
epor
t tha
t vill
ager
s at
test
to it
s pr
esen
ce in
the
corr
idor
link
ing
Pic
d’I
vohi
be S
peci
al R
eser
ve to
And
ring
itra
Nat
iona
l Par
k. W
e ha
ve c
onst
ruct
ed P
. d. e
dwar
dsi’s
ran
ge s
o as
to e
xten
d h
alfw
ay a
long
this
fore
st c
orri
dor
.8 O
’Con
nor
et a
l. (1
986)
des
crib
e a
brie
f sig
htin
g of
a d
ark
form
of P
. dia
dem
a at
And
ohah
ela
Nat
iona
l Par
k; h
owev
er, s
ubse
quen
t sur
veys
hav
e fa
iled
to lo
cate
it (F
eist
ner
& S
chm
id, 1
999)
.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
213
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
Tab
le 5
Occ
urre
nce
of d
iurn
al/
cath
emer
al r
ainf
ores
t lem
ur ta
xa in
the
prot
ecte
d a
reas
of s
outh
-eas
tern
Mad
agas
car
(see
Fig
. 1)1 .
Prot
ecte
d a
rea
Ran
omaf
ana
And
ring
itra
Pic
d’I
vohi
beM
anom
boK
alam
batr
itra
Mid
ongy
du
Sud
And
ohah
ela
Tax
onN
atio
nal P
ark
Nat
iona
l Par
kSp
ecia
l Res
erve
Spec
ial R
eser
veSp
ecia
l Res
erve
Nat
iona
l Par
kN
atio
nal P
ark
Eul
emur
fulv
us r
ufus
2+
(+)3
+E
ulem
ur a
lboc
olla
ris
(+)3
+E
ulem
ur c
olla
ris
++
+E
ulem
ur r
ubri
vent
er2
++
+H
apal
emur
gri
seus
gri
seus
2+
++
++
++
Hap
alem
ur a
ureu
s+
+H
apal
emur
sim
us+
+P
ropi
thec
us d
iade
ma
edw
ards
i+
+V
arec
ia v
arie
gata
var
iega
ta2
+?
4+
1 Sou
rces
oth
er th
an p
rese
nt s
urve
ys: A
ndri
ngit
ra N
atio
nal P
ark,
Ste
rlin
g &
Ram
aros
on (1
996)
; Pic
d’I
vohi
be S
peci
al R
eser
ve, R
asol
oari
son
& R
asol
onan
dra
sana
(199
9); M
anom
bo S
peci
al R
eser
ve,
J. R
atsi
mba
zafy
(per
s. c
omm
.); K
alam
batr
itra
Spe
cial
Res
erve
, Irw
in e
t al.
(200
1); M
idon
gy d
u Su
d N
atio
nal P
ark,
P. W
righ
t (un
publ
. dat
a); A
ndoh
ahel
a N
atio
nal P
ark,
Fei
stne
r &
Sch
mid
(199
9).
2 Tax
a w
hose
ran
ge e
xten
ds
beyo
nd a
rea
cons
ider
ed in
this
stu
dy
3 (+
) Hyb
rid
zon
e; o
nly
hybr
id g
roup
s ar
e kn
own
from
wit
hin
And
ring
itra
Nat
iona
l Par
k4 V
. v. v
arie
gata
app
ears
to b
e ra
re o
r sp
orad
ical
ly p
rese
nt w
ithi
n A
ndri
ngit
ra N
atio
nal P
ark;
it h
as b
een
repo
rted
by
Nic
oll &
Lan
gran
d (1
989)
and
Mit
term
eier
et a
l. (1
994)
but
was
not
foun
d in
mor
e re
cent
sur
veys
by
us o
r ot
hers
(Goo
dm
an &
Ras
olon
and
rasa
na, 2
001)
.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
214 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
Tab
le 6
Prev
ious
pop
ulat
ion
esti
mat
es o
f the
diu
rnal
/ca
them
eral
lem
ur ta
xa in
sou
th-e
aste
rn M
adag
asca
r an
d th
e nu
mbe
r of
sig
htin
gs, a
nd e
stim
ates
of a
vaila
ble
habi
tat,
popu
lati
on d
ensi
tyan
d m
axim
um p
opul
atio
n si
ze b
oth
in th
e st
udy
area
and
wit
hin
prot
ecte
d a
reas
.
Ove
rall
popu
lati
on in
stu
dy
area
Popu
lati
on in
pro
tect
ed a
reas
Est
imat
edM
axim
umE
stim
ated
Prev
ious
Est
imat
edpo
pula
tion
esti
mat
edav
aila
ble
habi
tat
Est
imat
ed p
opul
atio
nSa
mpl
e si
zepo
pula
tion
avai
labl
ed
ensi
typo
pula
tion
in p
rote
cted
size
in p
rote
cted
Tax
on(n
o. s
ight
ings
)es
tim
ate1
habi
tat (
km2 )
(km
−2)P
SEsi
zeP
SEar
eas
(km
2 )ar
easP
SE
Eul
emur
fulv
us r
ufus
212
0>
100,
000
4,31
822
.72P
2.02
98,0
92P
8,72
547
010
,677P
950
E. f
. ruf
us x
E. a
lboc
olla
ris
81–
670
57.4
6P
8.21
38,4
95P
5,49
815
0 8
,618P
1,23
1E
ulem
ur a
lboc
olla
ris
141,
000–
10,0
0070
110
.37P
3.23
7,26
5P
2,26
827
2
80P
87E
ulem
ur c
olla
ris
131,
000–
100,
000
5,74
514
.14P
3.98
81,2
41P
22,8
532,
741
38,7
61P
10,9
03E
ulem
ur r
ubri
vent
er2
8610
,000
–100
,000
5,27
95.
25P
0.64
27,7
15P
3,38
662
03,
255P
398
Hap
alem
ur g
rise
us g
rise
us2
33>
100,
000
11,4
331.
65P
0.36
18,8
72P
4,07
03,
387
5,59
1P
1,20
6H
apal
emur
aur
eus*
141,
000+
3,12
52.
04P
0.57
*58
5*
Hap
alem
ur s
imus
*1
1,00
0+3,
125
**
585
*P
ropi
thec
us d
iade
ma
edw
ards
i60
1,00
0–10
,000
5,16
97.
65P
1.19
39,5
28P
6,13
658
54,
474P
694
Var
ecia
var
iega
ta v
arie
gata
2 *81
1,00
0–10
,000
5,37
011
.12P
1.11
*46
2*
1 Mit
term
eier
et a
l. (1
994)
; est
imat
es a
re n
ot r
estr
icte
d to
stu
dy
area
but
app
ly to
ent
ire
popu
lati
on2 T
axa
who
se r
ange
ext
end
s be
yond
are
a co
nsid
ered
in th
is s
tud
y; p
opul
atio
n es
tim
ates
app
ly to
stu
dy
area
onl
y*I
ndic
ates
a p
atch
ily-d
istr
ibut
ed s
peci
es fo
r w
hich
rel
iabl
e es
tim
ates
can
not y
et b
e m
ade
(see
text
for
det
ails
)
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
215
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
rainforests are extremely rare in the south-east, andsurveys were thus concentrated in medium to high-altitude forests. This does mean, however, that our popu-lation size estimate is unlikely to be greatly inflated dueto altitudinal preference.
Hapalemur aureus was discovered in 1986 atRanomafana (Meier et al., 1987; Wright et al., 1987), and in1993 at Andringitra (Sterling & Ramaroson, 1996). Recentsurveys in the corridor between these parks foundH. aureus at four of eight sites (Goodman et al., 2001). Weare aware of no sightings north of Ranomafana or southof Andringitra, but tentatively place its southern limit atthe Manampatrana River, the closest major barrier. Asthis species is patchily-distributed due to habitat require-ments (it is a bamboo specialist), simple estimation ofpopulation size is impossible. The population size ifuniformly distributed would be 5,916 individuals, butwe believe that actual population size is much smaller,perhaps 25% of this estimate. Further surveys are neces-sary to understand the microhabitat preferences ofthis Critically Endangered species. If these could bequantified, and preferred habitats could be detected insatellite images, incidence models could be developed togenerate more accurate estimates.
Hapalemur simus was first described in 1870, but duringthe 1980s was almost unknown from contemporaryaccounts and thought to be extinct. Museum recordsindicate a large range in historic times throughout mostof the eastern forests, and subfossil remains from north-western and central Madagascar (beyond the rainforest’spresent extent) suggest a wider distribution in previouscenturies (Tattersall, 1982; Godfrey et al., 2004). Presently,H. simus occurs only within Ranomafana, Andringitra,corridor and isolated forests between and to the east ofthese localities, and one locality (Evendra) south-east ofAndringitra but north of the Manampatrana River.Goodman et al. (2001) found H. simus at only one of eightsites in the Ranomafana-Andringitra corridor. We there-fore tentatively present a geographic range between theManampatrana River in the south, and Ranomafana inthe north. As this species is furtive and restricted bymicrohabitat, and was rarely detected during surveys,we cannot reliably estimate population size. Given itsobserved range and extreme patchiness, we believeH. simus to be substantially less abundant than H. aureus.
Propithecus diadema edwardsi is patchily distributed dueto hunting pressure. Evidence from outside protectedareas indicates that this taxon is heavily hunted through-out much of its range (especially north of Ranomafana;Irwin et al., 2000; Lehman & Wright, 2000). Our estimateis probably too large, but without more detailed informa-tion the degree of overestimation is unclear. We conser-vatively estimate population size to be approximatelyhalf of the calculated value, or 20,000 individuals.
Varecia variegata variegata is patchily distributed, bothaccording to previous accounts and this study. Threefactors contribute to this patchiness. Firstly, this large-bodied frugivore is dependent on primary rainforest andis absent from areas with even mild disturbance (Whiteet al., 1995). Secondly, V. v. variegata is confined to low-and mid-altitude forests (<1,200 m; Mittermeier et al.,1994). This is particularly critical in south-eastern Mada-gascar, where low-altitude forest is rare; 25% of forest isabove 1,200 m (Fig. 2). Thirdly, V. v. variegata is especiallysusceptible to hunting due to its large size and loudalarm calls. This species’ complex, patchy distributionprecludes simple estimation of population size. Furthermodels incorporating altitude, habitat quality andhunting should be applied to develop more accurateestimates for this Endangered species.
Lemurs in south-eastern Madagascar face four maintypes of threat: (1) habitat loss, (2) fragmentation, (3)habitat disturbance, and (4) hunting. Forest continues tobe lost throughout the south-east, mostly due to small-scale shifting agriculture. The practice of tavy, cuttingforest and then burning the dried vegetation, is usedto create arable land (Gade, 1996). After a brief periodof cultivation (usually <10 years), the land becomesunusable and new plots are cleared.
Because deforestation usually occurs at small spatialscales, remaining forest becomes increasingly frag-mented. Living in fragments poses several dangers toforest-adapted species (Turner, 1996). Small, isolatedpopulations face extinction due to genetic, demographicand/or environmental stochasticity (Lande, 1998), andreduction in migration means that local extinctions maynot be easily reversed by recolonization. In addition,edge effects change habitat suitability (Murcia, 1995)through abiotic effects (e.g. wind and sun penetration),consequent biological effects, and anthropogenic effectssuch as hunting and disturbance.
Human populations also affect plant and animal com-munities through resource extraction. The use of treesand other forest products for food, construction or textilemanufacture affects vegetative structure and speciescomposition, thus affecting lemur populations. Onesevere example of this has been observed in the northernpart of the study region. The bark of Syzygium trees(Myrtaceae) is used as an additive in rum-making;bark-stripping causes high mortality, possibly affectinglemur populations by reducing food availability (Irwin &Ravelomanantsoa, 2004).
Hunting displays great geographic and culturalvariability, but appears capable of locally decimatinglemur populations (Irwin et al., 2000; Goodman &Raselimanana, 2003). Two types of hunting are known:projectile hunting (spears, slings, blowguns and, rarely,firearms) and trapping. Trapping usually involves snaresset along horizontal supports in deforested strips, and is
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
216 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
thought to mostly target Eulemur, Varecia and Hapalemurspecies, which move quadrupedally across the supports.
Extensive, quantitative monitoring programmes willbe essential to the survival of these unique primates. Weoffer a necessary first step: revision of species ranges andquantitative assessments of habitat area and populationsize. These assessments are used to identify regions andtaxa at risk, and can be used as a baseline for future moni-toring of forest cover and population densities. However,methodologies should be refined for patchily-distributedtaxa (especially H. aureus, H. simus and V. v. variegata),which require more sophisticated incidence models. Wetherefore recommend: (1) continued monitoring of forestcover changes through remote sensing, and (2) accumu-lating survey data at finer scales and across broaderhabitat gradients (location, elevation and disturbancelevel). The data generated through these approaches willbe indispensable in future efforts to identify and managelemur populations at risk of extinction.
Acknowledgements
We thank the Government of the Democratic Republic ofMadagascar, the Association Nationale pour la Gestiondes Aires Protégées, the Direction des Eaux et Forêts, andCAFF/CORE (Commission Tripartite) for permission toconduct research. Thanks to Benjamin Andriamihajaand MICET/ICTE for research facilitation. Funding wasprovided by Margot Marsh Biodiversity Foundation,Primate Conservation, Inc., University of Texas atAustin, Wenner-Gren Foundation for AnthropologicalResearch (6414), National Science Foundation (BCS-9910257), IIE Fulbright, United Nations Foundation, andUNESCO. We thank our students (Vaniah Andrianjaka,Tantely Andrianjazalahatra and Jean-Luc Raharison),and research assistants (François Monja, Emile Rajeriari-son, Jean-Claude Rakotoniaiana, Georges Rakotonirina,Joseph Rakotozafy II, Richard Randriamampionona,Jean-Francois Randriatsilanisoa, Loret Rasabo, FrançoisRatalata, Raymond Ratsimbazafy, Sylvain Razafiman-dimby, Berthin Razafy-Jerome, Justin Solo and PierreTalata). We are greatly appreciative for unpublished datasupplied by Summer Arrigo-Nelson and ChristinaGrassi. Additional thanks to Karen Samonds, Avri Beard,Rickie van Berkum, David Burney, Lauren Donovan,Jukka Jernvall, Rachel Kitko, Bill Konstant, StephenNash, Deborah Overdorff, Aimée Razafiarimalala, NoelRowe, Tanya Smith, Elizabeth Stone and RebeccaStumpf.
References
Burney, D.A. (1999) Rates, patterns, and processes of landscapetransformation and extinction in Madagascar. In Extinctionsin Near Time: Causes, Contexts, and Consequences (ed. R.D.E.MacPhee), pp. 145–164. Plenum/Kluwer, New York, USA.
Burney, D.A. & MacPhee, R.D.E. (1988) Mysterious island:what killed Madagascar’s large native animals? NaturalHistory, 97, 46–55.
CBSG (Conservation Biology Specialist Group) (2002)L’Evaluation et Plans de Gestion pour la conservation (CAMP) dela faune de Madagascar: Lémuriens, autres mammifères, reptiles etamphibiens, poisons d’eau douce et evaluation de la viabilité despopulations et des habitat de Hypogeomys antimena (Vositse).Conservation Biology Specialist Group, Apple Valley, USA.
Duckworth, J.W. (1998) The difficulty of estimating populationdensities of nocturnal forest mammals from transect countsof animals. Journal of Zoology (London), 246, 466–468.
Fashing, P.J. & Cords, M. (2000) Diurnal primate densities andbiomass in the Kakamega forest: an evaluation of censusmethods and a comparison with other forests. AmericanJournal of Primatology, 50, 139–152.
Fausser, J.-L., Prosper, P., Donati, G., Ramanamanjato, B. &Rumpler, Y. (2002) Phylogenetic relationships betweenHapalemur species and subspecies based on mitochondrialDNA sequences. BMC Evolutionary Biology, 2, 4.
Feistner, A.T.C. & Schmid, J. (1999) Lemurs of the RéserveNaturelle Intégrale d’Andohahela, Madagascar. In A Floraland Faunal Inventory of the Réserve Naturelle Intégraled’Andohahela, Madagascar: With Reference to ElevationalVariation (ed. S.M. Goodman), pp. 269–283. Fieldiana Zoology,n.s. 94. Field Museum of Natural History, Chicago, USA.
Gade, D. (1996) Deforestation and its effects in highlandMadagascar. Mountain Research and Development, 16, 101–116.
Ganzhorn, J.U. (1995) Low-level forest disturbance effects onprimary production, leaf chemistry, and lemur populations.Ecology, 76, 2084–2096.
Ganzhorn, J.U., Lowry II, P.P., Schatz, G.E. & Sommer, S. (2001)The biodiversity of Madagascar: one of the world’s hottesthotspots on its way out. Oryx, 35, 346–348.
Ganzhorn, J.U., Malcomber, S., Andrinantoanina, O. &Goodman, S.M. (1997a) Habitat characteristics and lemurspecies richness in Madagascar. Biotropica, 29, 331–343.
Ganzhorn, J.U., Rakotosamimanana, B., Hannah, L., Hough, J.,Iyer, L., Olivieri, S., Rajaobelina, S., Rodstrom, C. & Tilkin,G. (1997b) Priorities for biodiversity conservation inMadagascar. Primate Report (Special Issue), 48–1, 1–81.
Garbutt, N. (1999) Mammals of Madagascar. Yale UniversityPress, New Haven, USA.
Godfrey, L.R. & Jungers, W.L. (2002) Quaternary fossil lemurs.In The Primate Fossil Record (ed. W. Hartwig), pp. 97–122.Cambridge University Press, Cambridge, UK.
Godfrey, L.R., Simons, E.L., Jungers, W.L., DeBlieux, D.D. &Chatrath, P.S. (2004) New discovery of subfossil Hapalemursimus, the greater bamboo lemur, in western Madagascar.Lemur News, 9, 9–11.
Goodman, S.M. & Raselimanana, A. (2003) Hunting of wildanimals by Sakalava of the Menabe region: a field reportfrom Kirindy-Mite. Lemur News, 8, 4–6.
Goodman, S.M. & Rasolonandrasana, B.P.N. (2001) Elevationalzonation of birds, insectivores, rodents and primates on theslopes of the Andringitra Massif, Madagascar. Journal ofNatural History, 35, 285–305.
Goodman, S.M., Razafindratsita, V., Schütz, H. &Ratsimbazafy, R. (2001) Les Lémuriens. In InventaireBiologique du Parc National de Ranomafana et du couloir forestierqui la relie au Parc National d’Andringitra (eds S.M. Goodman& V.R. Razafindratsita), pp. 231–243. CIDST, Antananarivo,Madagascar.
Green, G.M. & Sussman, R.W. (1990) Deforestation history ofthe eastern rain forests of Madagascar from satellite images.Science, 248, 212–215.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
217
© 2005 FFI, Oryx, 39(2), 204–218
Lemur populations in south-eastern Madagascar
Harcourt, C. & Thornback, J. (1990) Lemurs of Madagascar andthe Comoros: The IUCN Red Data Book. IUCN, Gland,Switzerland.
Intercoopération Suisse (Organisation Suisse pour leDéveloppement et la Coopération) & Marie ERTA (Etude etRéalization de Travaux d’Amenagement) (1999) Rapport desEtudes Physico-bio-ecologiques de la Réserve Spéciale deKalambatritra. Unpublished Report to ANGAP,Antananarivo, Madagascar.
Irwin, M.T. & Ravelomanantsoa, H.V. (2004) Illegal rumproduction threatens health of lemur populations atTsinjoarivo, eastern central Madagascar: brief report andrequest for information. Lemur News, 9, 16–17.
Irwin, M.T., Samonds, K.E. & Raharison, J.-L. (2001) Abiological inventory of the lemur community of RéserveSpéciale de Kalambatritra, South-Central Madagascar. LemurNews, 6, 24–28.
Irwin, M.T., Smith, T.M. & Wright, P.C. (2000) Census of threeeastern rainforest sites north of Ranomafana National Park:preliminary results and implications for lemur conservation.Lemur News, 5, 20–22.
IUCN (2004) 2004 IUCN Red List of Threatened Species. IUCN,Gland, Switzerland [http://www.redlist.org, accessed 23November 2004].
Johnson, S.E. (2002) Ecology and speciation in brown lemurs:white-collared lemurs (Eulemur albocollaris) and hybrids(Eulemur albocollaris x Eulemur fulvus rufus) in southeasternMadagascar. PhD thesis, University of Texas at Austin,Austin, USA.
Johnson, S.E., Irwin, M.T., Wright, P.C., Arrigo-Nelson, S.J.,Grassi, C., Samonds, K.E. & Smith, T.M. (2003) Natural andanthropogenic influences on lemur population structure insoutheastern Madagascar. American Journal of PhysicalAnthropology, 120(S36),123.
Johnson, S.E. & Overdorff, D.J. (1999) A census of brownlemurs (Eulemur fulvus sspp.) in southeastern Madagascar:methods-testing and conservation implications. AmericanJournal of Primatology, 47, 51–60.
Johnson, S.E. & Wyner, Y. (2000) Notes on the biogeography ofEulemur fulvus albocollaris. Lemur News, 5, 25–28.
Lande, R. (1998) Anthropogenic, ecological and genetic factorsin extinction and conservation. Research in Population Ecology,40, 259–269.
Lehman, S.M. & Wright, P.C. (2000) Preliminary study of theconservation status of lemur communities in theBetsakafandrika region of eastern Madagascar. Lemur News,5, 23–25.
Meier, B., Albignac, R., Peyriéras, A., Rumpler, Y. & Wright, P.(1987) A new species of Hapalemur (Primates) fromsouth-east Madagascar. Folia Primatologica, 48, 211–215.
Mittermeier, R.A., Tattersall, I., Konstant, W.R., Meyers, D.M.& Mast, R.B. (1994) Lemurs of Madagascar. ConservationInternational, Washington, DC, USA
Murcia, C. (1995) Edge effects in fragmented forests:implications for conservation. Trends in Ecology and Evolution,10, 58–62.
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca,G.A.B. & Kent, J. (2000) Biodiversity hotspots forconservation priorities. Nature, 403, 853–858.
Nicoll, M.E. & Langrand, O. (1989) Madagascar: Revue de laConservation et des Aires Protégées. WWF, Gland, Switzerland.
O’Connor, S., Pidgeon, M. & Randria, Z. (1986) Conservationprogram for the Andohahela Reserve, Madagascar. PrimateConservation, 7, 48–52.
Petter, J.-J., Albignac, R. & Rumpler, Y. (1977) Faune deMadagascar 44: Mammifères Lémuriens (Primates Prosimiens).ORSTROM/CNRS, Paris, France.
Plumptre, A.J. (2000) Monitoring mammal populations withline transect techniques in African forests. Journal of AppliedEcology, 37, 356–368.
Ramarokoto, S., Rakotosamimanana, B. & Raharivololona, B.M.(1999) Situation actuelle des aires protégées à Madagascar:plan stratégique de l’ANGAP (Association Nationale pour laGestion des Aires Protégées) de 1998 à 2000. Lemur News, 4,5–7.
Rasoloarison, R.M., Goodman, S.M. & Ganzhorn, J.U. (2000)Taxonomic revision of mouse lemurs (Microcebus) in thewestern portions of Madagascar. International Journal ofPrimatology, 21, 963–1019.
Rasoloarison, R.M. & Rasolonandrasana, B.P.N. (1999) LesLémuriens. In Inventaire Biologique de la Réserve Spéciale du Picd’Ivohibe et du Couloir Forestier qui la relie au Parc Nationald’Andringitra (eds S.M. Goodman & B.P.N.Rasolonandrasana), pp. 159–169. CIDST, Antananarivo,Madagascar.
Smith, R.J. & Jungers, W.L. (1997) Body mass in comparativeprimatology. Journal of Human Evolution, 32, 423–559.
Sterling, E.J. & Ramaroson, M.G. (1996) Rapid assessment ofthe primate fauna of the Eastern slopes of the RéserveNaturelle Intégrale d’Andringitra, Madagascar. In A Floraland Faunal Inventory of the Eastern Slopes of the RéserveNaturelle Intégrale d’Andringitra, Madagascar: With Reference toElevational Variation (ed. S.M. Goodman), pp. 293–305.Fieldiana Zoology, n.s. 85. Field Museum of Natural History,Chicago, USA.
Struhsaker, T.R. (1981) Census methods for estimatingdensities. In Techniques for the Study of Primate PopulationEcology (ed. Subcommittee on Conservation of NaturalPopulations), pp. 36–80. National Academy Press,Washington, DC, USA.
Sussman, R.W., Green, G.M., Porton, I., Andrianasolondraibe,O.L. & Ratsirarson, J. (2003) A survey of the habitat of Lemurcatta in south-western and southern Madagascar. PrimateConservation, 19, 32–57.
Tattersall, I. (1982) The Primates of Madagascar. ColumbiaUniversity Press, New York, USA.
Turner, I.M. (1996) Species loss in fragments of tropical rainforest: a review of the evidence. Journal of Applied Ecology, 33,200–209.
USGS (2003) United States Geological Survey.Http://edc.usgs.gov/products/elevation/gtopo30.html[accessed 23 November 2004].
USGS (2004) United States Geological Survey.Http://edc.usgs.gov/products/satellite/landsat7.html[accessed 23 November 2004].
White, F.J., Overdorff, D.J., Balko, E.A. & Wright, P.C. (1995)Distribution of ruffed lemurs (Varecia variegata) inRanomafana National Park, Madagascar. Folia Primatologica,64, 124–131.
Whitesides, G.H., Oates, J.F., Green, S.M. & Kluberdanz, R.P.(1988) Estimating primate densities from transects in a WestAfrican rain forest: a comparison of techniques. Journal ofAnimal Ecology, 57, 345–367.
Wyner, Y.M., Johnson, S.E., Stumpf, R.M. & DeSalle, R. (2002)Genetic assessment of a white-collared x red-fronted lemurhybrid zone at Andringitra, Madagascar. American Journal ofPrimatology, 67, 51–66.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at
218 M. T. Irwin et al.
© 2005 FFI, Oryx, 39(2), 204–218
Biographical sketches
Mitchell Irwin’s research examines the effects of forestfragmentation on lemur ecology and behaviour. He hasexamined patterns of lemur extinctions in a network of forestfragments, and recently completed a year-long study of thefeeding ecology of the diademed sifaka Propithecus diadema inforest fragments and continuous forest.
Steig Johnson’s research interests include behaviouralecology, biogeography and speciation in strepsirhineprimates, with a focus on the brown lemur Eulemur fulvusspecies complex. He has recently conducted fieldworkon disturbance effects on forest structure and animalpopulations as part of an interdisciplinary project modellingdeforestation in south-eastern Madagascar.
Patricia Wright co-discovered the golden bamboo lemurHapalemur aureus in 1986 at Ranomafana, and was instrumen-tal in the creation of Ranomafana National Park in 1991.Recently she has spearheaded Centre ValBio, a research andtraining center at Ranomafana. She has received two medalsof honor from the Malagasy government.
https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0030605305000451Downloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 14 Sep 2020 at 16:44:29, subject to the Cambridge Core terms of use, available at