MGF 154
Georgia Lorenti
Madagascar Forest Terrestrial Programme
Nosy Be, North West Madagascar
MGF Phase 154 Science Report
(October-November 2015)
Georgia Lorenti (Principal Investigator)
1
MGF 154
Georgia Lorenti
Contents
1.
Introduction .................................................................................................................................... 3
2.
Survey Sites ..................................................................................................................................... 4
4.
The effect of habitat degradation on reptile communities ............................................................ 7
Introduction ........................................................................................................................................ 7
Aims and Objectives............................................................................................................................ 7
Methodology....................................................................................................................................... 8
Reptile Community ......................................................................................................................... 8
Statistical Analysis ........................................................................................................................... 8
Results ................................................................................................................................................. 9
Discussion.......................................................................................................................................... 10
5.
The effective of habitat degradation on bird communities .......................................................... 13
Introduction ...................................................................................................................................... 13
Aims and objectives .......................................................................................................................... 13
Methodology..................................................................................................................................... 13
Survey Methodology ..................................................................................................................... 13
Statistical Analysis ......................................................................................................................... 14
Results ............................................................................................................................................... 14
Discussion.......................................................................................................................................... 16
6.
The butterfly diversity of the Ambalohonko Region .................................................................... 18
Introduction ...................................................................................................................................... 18
Aims and Objectives.......................................................................................................................... 18
Methodology..................................................................................................................................... 18
Butterfly Community .................................................................................................................... 18
Statistical Analysis ......................................................................................................................... 19
Results ............................................................................................................................................... 19
Discussion.......................................................................................................................................... 20
7.
Satellite Camp: Snapshot Analysis of Potential Research Site ...................................................... 22
Introduction ...................................................................................................................................... 22
Methodology..................................................................................................................................... 22
Results ............................................................................................................................................... 23
Discussion.......................................................................................................................................... 23
References: ........................................................................................................................................... 25
2
MGF 154
1.
Georgia Lorenti
Introduction
Madagascar has been designated as a global biodiversity hotspot due to its exceptional
concentrations of endemic species (Myers et al. 2000; Ganzhorn et al. 2001 ). The island has
been heralded as 'the single highest biodiversity conservation priority in the world' (Myers et
al. 2000). Roughly 90% of Madagascar's endemic species are forest dwelling and are highly
susceptible to anthropogenic disturbance (Dufils, 2003). The island's unique biota is
currently being threatened by high levels of deforestation and environmental degradation
(Myers et al. 2000; Ganzhorn et al. 2001). There is much debate over the exact percentage of
historical forest cover still remaining (Dufils, 2003; Quéméré et al. 2012), but it is believed
that over 90% has been destroyed. Analysis of satellite imagery has shown that 40% of
contemporary forest cover was lost between 1950 and 2000 (Harper et al. 2007). Despite
increased environmental awareness in recent decades, deforestation has continued, with 0.9%
of remaining forest lost annually from 1990 to 2000 (Scott et al. 2006; Irwin et al. 2010).
Clearance for agriculture remains a major driver of forest destruction, as well as extraction of
timber for fuel and building materials (Irwin et al., 2010).
Species extinctions have substantially reduced the native vertebrate community over
the past 10,000 years (Dewar, 2003). Although it is difficult to pinpoint the mechanisms for
these extinctions, it is likely that the rate has increased since human colonisation around
2,000 years ago (Dewar, 2003). Anthropogenic habitat destruction and fragmentation are
considered to be the most important factors driving extinction in Madagascar (Andreone et
al., 2005). Deforestation resulting in habitat loss, degradation and fragmentation has a
detrimental effect on species diversity (Goodman and Rakotondravony, 2000;
Ramanamanjato and Ganzhorn, 2001). The ability of a species to tolerate or exploit human
modified landscapes will determine their persistence and future survival (Scott et al. 2006). In
order to mitigate the deleterious effects of deforestation it is essential for conservation
purposes to identify susceptible species. Such information will aid in determining
communities for conservation priority and could assist in the development of effective land
use and management practices. Madagascar National Parks (MNP, formerly ANGAP)
currently manage 46 protected areas at 44 sites, covering an area of 170,000 km2, with the
aim of preserving a broad spectrum of Madagascar’s biota. However, there is a further 40,246
km2 outside reserves (Randrianandianina et al., 2003), which contains substantial latent
biodiversity (Ingram and Dawson, 2006).
The Sambirano domain is Madagascar’s youngest biome, distributed around the
Sambirano River and the town of Ambanja, in northwest Madagascar. This humid forest
habitat developed approximately eight million years ago, when a climatic shift caused an
increase in rainfall (Wells, 2003). The domain is characterised by a climax community of
lowland forest, with a proportion of the species composition being shared with both the
eastern and, especially in degraded habitats, western domains. This is due to the close
geographical proximity and similar climates of the regions, but an additional component of
the taxa is endemic (Du Puy and Moat, 2003). The region experiences annual rainfall of
2,250 mm (Wells, 2003), with a short dry period of three to four months (Legris and Blasco,
1965) and mean annual temperatures of 28oC, ranging from 15oC to 35oC (Andreone et al.
2003). The Sambirano domain was historically a belt of unbroken forest running from North
to South Madagascar; however it is now limited to the northern end of the central mountain
range. Due to its restricted area and high levels of human disturbance, the Sambirano region
is considered one of the most endangered habitats in Madagascar (Langrand, 1990).
3
MGF 154
Georgia Lorenti
Nosy Be, Madagascar’s largest offshore island, is situated in the northwest and falls
within the Sambirano domain. It was connected to the mainland as recently as 8,000 years
ago, when rising sea levels isolated the island and a number of smaller outcrops in the Nosy
Vorona Bight. Analysis of the biodiversity of small islands, such as Nosy Be, is important for
identifying areas of species endemism. These studies also increase our understanding of
historical patterns of animal and plant dispersal and distribution, due to the fact that
extinctions occur at an accelerated rate on small islands because of their limited size and
reduced recruitment (Andreone et al. 2003). The island is an important centre for agriculture
and tourism and large parts have been deforested for agriculture. Rice paddies, coffee, ylangylang and sugar cane plantations now form the largest component of vegetation, with only a
small fragment of the original forest remaining (Andreone et al. 2003). This protected area
represents one of the last remaining relatively undisturbed areas of Sambirano forest
(Andreone et al. 2003). As a result, Nosy Be is the type locality of many of the floral and
faunal species of the historic Sambirano domain. The area represents a high research priority
with much of its unique biota in need of cataloguing.
2.
Survey Sites
The Frontier Madagascar terrestrial research programme is situated in the village of
Ambalahonko, Nosy Be (Figure 1), bordering Lokobe National Park. Twelve research sites
were selected along a gradient of human disturbance, based on the time since they were last
cleared and their current land usage. The sites range from primary forest that has never
experienced clearing and has low levels of anthropogenic disruption, to those that are cleared
annually for agricultural purposes. All research sites are located in close proximity to the
village and are accessible via the beach, except at times when the tide is high. The primary
survey sites are towards the west, near the Lokobe National Park, whilst the most disturbed
sites are those nearest Ambalahonko and neighbouring Antafondro. A distance of about 5 km
separates the survey sites. To account for the effect of human disturbance and habitat
degradation on vertebrate communities, sites were selected carefully and chosen for their
contrasting histories.
Figure 1:
Location of Nosy Be in relation to mainland Madagascar (top left and top
right) and the position of the MGF base camp within the village of
Ambalahonko (red dot), next to the Lokobe Integral Reserve.
4
MGF 154
Georgia Lorenti
There are nine sites that are currently used for reptile surveying, with three being
designated as primary, three as secondary, and three as degraded. The transects within these
sites are all 200m in length and follow existing trails through the forest. The primary transects
are all located close to the border with Lokobe National Park. There is one secondary forest
transect which begins along the beach. The other two secondary forest transects are located
within in an area designated as a communal protected park by the Fokotany of Ampasipohy,
The degraded transects are all located close to the village of Ambalahonko, in forested areas
that undergo regular clear felling and disturbance. There are three transects that are presently
used for bird and butterfly surveys. Transects within these sites are all 400m in length and
utilise existing pathways. Two of the transects are within forested areas representing both
primary and secondary habitats. The degraded transect is located in open habitat, primarily
consisting of plantations. A map showing the locations of the sites is given in figure 2 and a
summary of each site in table 1.
Figure 2:
Location of survey sites in relation to Ambalahonko and Lokobe National
Park. Red indicates reptile transects, and yellow bird/butterfly transects. Image
taken from Google Earth.
5
MGF 154
Georgia Lorenti
Table 1:
An overview of each of the survey sites, describing their exact locations and
their present habitat use.
Survey Site Overview
GPS Location
South
East
13°
48°
Site
Code
BP
Survey
24’48”
20’2”
Reptile
SB
24’49”
20’5”
Reptile
GG
24’40”
20’8”
Reptile
ZB
24’31”
20’18”
Reptile
FC
23’46”
20’29”
Reptile
LKL
23’52”
20’28”
Reptile
SS
24’12”
20’30”
Reptile
PP
24’7”
20’57”
Reptile
LL
24’26”
20’59”
Reptile
ZB
24’31”
20’18”
Bird and
Butterfly
TS
24’20”
21’0”
Bird and
Butterfly
Habitat
Type
Primary
Description
Primary mature forest, running along the
border of Lokobe National Park. The trail is
kept free of vegetation to mark the
boundary. Steep in places.
Primary mature forest, located adjacent to
Primary
the beach, and close to BP
Predominantly primary mature forest, with
Primary
many Ravenala palms. Steep in places.
Secondary Secondary closed canopy forest, which has
not been heavily cleared for many years.
Accessed via beach. Some individual trees
have been recently removed for timber
Secondary Secondary closed canopy forest, located in
Ampasipohy. Sparse ground vegetation.
Some vanilla plantations. Guided tours
occur frequently.
Secondary Secondary closed canopy forest, located in
Ampasipohy. Dense ground vegetation. Part
of transect runs along border of Lokobe
National Park.
Degraded Young secondary regrowth, primarily used
for timber extraction. Located on hillside
Degraded Located along a main pathway. Recently,
clear felled recently for charcoal production
and agriculture.
Degraded Located along a main pathway. Recently,
clear felled recently for charcoal production
and agriculture.
Forest
Closed canopy forest, which has not been
heavily cleared for many years. Accessed
via beach. Some individual trees have been
recently removed for timber
Open
Pathway through mixed plantations
Habitat
6
MGF 154
4.
Georgia Lorenti
The effect of habitat degradation on reptile communities
Introduction
Reptiles are in a state of dramatic decline worldwide (Gibbons et al. 2000). In
Madagascar the primary cause of this decline is through the destruction of habitats (Andreone
et al. 2005). Much of island’s forests are still under threat, with deforestation reducing and
fragmenting available habitat. The reptiles of Madagascar are highly diverse, with 91% of the
species being endemic (Raxworthy, 2003). Despite the importance of studying and
conserving this taxa, relatively little is known and new species are routinely being described.
There are around 60 species of reptile recorded from Nosy Be, although some of these may
be ambiguous or now extirpated. Roughly 80% are considered to be found in Lokobe
National Park (LNP), whose 740ha only covers around 3% of the total area of the island
(Andreone et al. 2003).
There are several species which are also thought to exist only within the boundaries of
LNP itself (Andreone et al. 2003). For this reason, the importance of maintaining this small
protected area and its surrounding buffer zone of vegetation is high. Although there has been
previous research work conducted on the herpetofauna of Nosy Be (e.g. Glaw and Vences,
2007; Andreone et al. 2001; Andreone et al. 2003; Andreone et al. 2005), this has so far been
primarily cataloguing of the species present. There is still much work to be done from an
ecological, conservation and genetic standpoint. Irwin et al., (2010) identified the need for a
greater understanding of the patterns of anthropogenic change on Madagascan species. The
paper highlighted the importance of being able to predict the outcomes of continued
deforestation and disturbance on species richness and abundance. It is expected that diversity
will be reduced by anthropogenic disturbance, with more generalist and tolerant (and
potentially invasive non-endemic) species able to expand their populations in response to
habitat modification (Glaw and Vences, 2007).
The study area surrounding the village of Ambalahonko and the Lokobe buffer zone
is well suited to conducting studies of this nature as it comprises a variety of habitats with
varying levels of disturbance within close proximity to one another. A portion of this area
close to the border is composed of primary and secondary vegetation, which is contiguous
with the park itself, therefore is important to study and protect. Small-scale deforestation is
still ongoing in the area (pers. obs.2015), and further protective measures are required in
order to safe-guard this small fragment of intact forest. This study aims to collect data on the
abundance of reptiles, assessing species richness and diversity between forests which have
experienced various levels of anthropogenic disturbance, in order to determine the effects of
habitat loss on reptile assemblages.
Aims and Objectives
Aim: To monitor the effects of habitat disturbance on the reptile diversity and abundance
between primary, secondary and degraded habitats.
Objectives:
1.
To compare reptile abundance between different habitat types from visual encounter
surveys..
7
MGF 154
2.
Georgia Lorenti
To compare reptile species richness between different habitat types from visual
encounter surveys
Methodology
Reptile Community
An assessment of reptile abundance and diversity was conducted along 200m
transects using visual encounter surveys (VES). This technique has been described as an
effective tool for the sampling of reptiles in the tropics (Doan, 2003; Jestrzemski et al. 2013).
Nine transects were selected for reptile surveying. As forest areas are owned by local citizens,
pre-existing trails were utilised to avoid demarcating private property. Transects were
designated as either primary, secondary or degraded habitat based on the level of human
disturbance. Three transects were devoted to each habitat type. The VES searching method is
not a time-constrained technique (Doan, 2003). Researchers walked each transect at a
standard pace, but visually searched the entire transect using as much time as was needed to
thoroughly examine each area (Jestrzemski et al. 2013). Observers kept noise and other
disturbances to a minimum, in order to prevent scaring animals away before they had been
encountered. During surveys reptiles were identified at the species level. In circumstances
where species identification was not possible, differentiation was conducted at the genus
level.
Despite visual encounter surveying being a widely used and accepted method of
surveying (Andreone et al. 2003; D'Cruze et al 2008), not all species will be surveyed with
equal efficiency as their appearance or behaviour will differ. Species living in high canopy or
amongst the leaf litter will be particularly under-represented. Those which are sensitive to the
presence of researchers may be more likely to flee or hide, despite best efforts to minimise
disturbance, and is presumably more likely to occur in more pristine habitats (D'Cruze et al.
2008). Structural complexity also tends to mean species are more difficult to spot in less open
habitats. The aim of not restricting surveys to a time limit was intended to try to account for
this, as a correspondingly larger times needs to be spent searching dense vegetation.
Statistical Analysis
Analysis of reptile species richness was carried out in EstimateS, using rarefaction curves to
produce smoothed curves, and the ACE estimations of species richness were used to predict
the likely number of species if curves reach asymptote.
The Shapiro-Wilk Test has been labelled as the preferred test for normal distribution due to
its good power properties (Medes and Pala, 2003). It is recognised as the first statistical
procedure to detect departures from normality due to skewness (Razali and Wah, 2011). The
Shapiro-Wilk Test was selected to assess if variables had normal distributions as the
procedure is regarded as optimal for sample sizes less than 2000 (Razali and Wah, 2011).
Kruskal-Wallis H tests were used to determine whether a significant difference exists among
the habitat types (primary, secondary and degraded).The Kruskal-Wallis H test is a nonparametric test used to determine whether three or more independent groups are the same or
different on some variable of interest. It is used in situations where an ordinal level of data or
an interval or ratio level of data is available (Chan and Walmsley, 1997).
8
MGF 154
Georgia Lorenti
Both the Shapiro-Wilk Test Kruskal-Wallis conducted using IBM SPSS, version 17.0. All
tests were two-tailed, and statistical significance was set at p <0.05.
Results
A total of 59 visual encounter surveys were conducted over the course of the phase,
encountering 641 individuals across 28 different species.
Figure 3:
Mean number of reptiles in three habitat types (n = 641). Error bars
represent the standard error (SE).
Primary, secondary and degraded habitat types showed a significant difference in mean
abundance of reptile individuals (Kruskal-Wallis H = 8.679, df = 2, p < 0.05). Mean reptile
abundance was highest in the primary habitat 7.57 ± 3.61 (mean ± SE). Secondary habitat
showed a mean abundance of 6.68 ± 3.27 (mean ± SE). Reptile mean abundance was lowest
in the degraded habitat 3.08 ± 1.01 (mean ± SE).
Analysis of species richness using rarefaction between the three habitat types (two transects
showed that there was little difference between primary and secondary sites. At the
equivalence point of 19 (lowest sampling intensity - in degraded habitat), secondary (17.7)
had a species richness only slightly higher than primary (19.6 species). The degraded site
type was lower than the other two habitat types (13 species).
9
MGF 154
Georgia Lorenti
Figure 4:
Rarefaction curve showing how species richness increases with sampling
intensity in primary, secondary and degraded habitats. Dashed lines indicate
mean ACE richness estimator, and how quickly they reach asymptote.
As none of the curves reached asymptote, it can be assumed that not all the species within the
habitats were observed. ACE estimates of total species richness from the collected data
predicted primary habitat to be the most diverse site (species = 28.82, SD = 0.00). Species
richness for secondary habitat was estimated to be 25.41 (SD = 0.00). The least diverse site is
likely to be the degraded habitat (species = 15.41, SD = 0.00). The difference between habitat
types is likely to be significant as standard deviations did not overlap.
Table 2:
Total number of reptile species recorded in each habitat type. List of species
unique to either primary, secondary or degraded habitat.
Species
Observed
Unique
Species
Primary
Secondary
Degraded
22
15
13
Brookesia minima
Zonosaurus subunicolor
Lygodactylus madagascariensis
Uroplatus henkeli
Bibilava stumpffi
Liophidium torquatum
Mimophis mahfalensis
Hemidactylus spp.
Phelsuma abbotti
Langaha madagascariensis
Discussion
The high level of diversity and endemicity of animals and plants in Madagascar is
well documented (Myers et al. 2000; Ganzhorn et al. 2001). The comprehension that
understanding of biogeographic patterns is crucial to assess conservation priorities has led to
an increased interest in floral and faunal inventories. However, most studies have been
conducted in protected areas on the mainland (Raxworthy and Nussbaum, 1995; Raxworthy
et al. 1998; Raselimanana et al. 2000), while relatively few data are available for non10
MGF 154
Georgia Lorenti
protected and secondary habitats in Madagascar (Andreone et al. 2000). Even less is known
about the herpetofauna of the Malagasy offshore islands. These may be important centres of
endemicity and whose study could aid in the understanding of historical patterns of reptile
distribution and dispersal (Andreone et al. 2003). Moreover, islands represent a particular
conservation priority as their limited size may accelerate extinction. This has previously been
described for lemurs and birds by Goodman (1993), where human disturbance and intensive
hunting has caused the extinction of the snail-eating coua bird and the disappearance of
native lemurs at Sainte Marie. Only limited data exists in regards to the effect of habitat
degradation on reptile abundance and species richness on Malagasy offshore islands
(Andreone et al. 2003). A lack of information on how reptilian responses to habitat
disturbance is a major impediment to effective conservation in Madagascar.
This phase revealed that there was a significant difference between mean reptile
abundances between primary, secondary and degraded habitats. The highest mean abundance
of herpetofauna was recorded in primary forest 7.57 ± 3.61 (mean ± SE). Secondary habitat
showed a mean abundance of 6.68 ± 3.27 (mean ± SE). Overlapping standard deviations
might reduce the significance of the difference between secondary and primary habitats. This
could have an important impact on potential reptile management practices as despite the
higher levels of human disturbance, secondary forests are often able to support important
populations. Recently, forests that occur outside the protected areas in Madagascar have been
highlighted as sites of conservation value (Jenkins et al. 2003).Whilst the conservation of
pristine tropical habitats remains a principal objective for conservation biologists, the
inclusion of disturbed habitats in tropical forest management is becoming increasingly
important (Bawa and Seidler, 1998). Reptile mean abundance was lowest in the degraded
habitat 3.08 ± 1.01 (mean ± SE). This result is supported by previous studies that have
attributed the loss of reptile abundance and diversity to anthropogenic disturbance (Irwin et
al. 2010; Gibbons et al. 2000). Degraded forests sites are often unsuitable for species that are
restricted to primary vegetation (Stephenson, 1995). Scientists consider the loss of suitable
habitat to be the largest contributor to the global decline of reptile species (Gibbons et al.
2000).
A comparison of species richness between the sites using rarefaction, to control for
relative density, showed that there was a significant difference habitat types. ACE estimates
of total species richness, obtained by extrapolating the data, indicate that degraded sites are
likely to be the least diverse (species = 15.41). This is congruent with the literature as human
disturbances, such as logging or the planting of non-native tree plantations, have been shown
to influence reptile communities (Vonesh, 2001). Disturbed sites are frequently characterised
by lower species richness and evenness compared to similar undisturbed sites (Inger, 1980;
Heinen, 1992). However, as none of the curves reached asymptote, it can be assumed that not
all the species within the habitats were observed. Simply comparing sites by species richness,
or other diversity measures, does not account for differences in community composition
which are likely to occur between habitat types. Different habitat types may contain a similar
number of species, but they are not necessarily the same ones. This is an important point
when considering the effects of disturbance upon ecological communities. It is likely that
species more sensitive to disturbance, or with more specialised niches, are unable to complete
with more generalist species which essentially replace their role within the ecosystem
(Andreone et al. 2003). It is therefore important for studies to examine the presence or
absence of each species for each research site.
11
MGF 154
Georgia Lorenti
There were six species found solely in the primary habitat: Brookesia minima,
Zonosaurus subunicolor, Lygodactylus madagascariensis, Uroplatus henkeli, Bibilava
stumpffi and Liophidium torquatum (See Table 2).There were three species found solely in
degraded sites, and only one solely in secondary, although all at very low encounter rates.
Many of these species have been encountered in other habitat types, either outside of
surveying periods (pers. obs.), or in previous phases (Burger, 2015; Burger 2014). Short
survey periods the low detectability of certain species, snakes especially, makes them prone
to misinterpreted as specialising in only one particular habitat. For example, Liophidium
torquatum was only recorded in primary habitat, but has also been seen in secondary forest.
recorded solely in the degraded habitat, but is generally observed in primary and secondary
forest. Given a longer time frame, such a result might mark this one reporting of Langaha
madagascariensis as an outlier. It is also surprising that the Hognose snake, Leioheterodon
madagascariensis, was not encountered in any of the habitat types, when it is commonly
encountered in all three habitat types (pers. obs.) and known to be tolerant of disturbance
(Glaw and Vences, 2007). Six species were found in all three habitat types; Brookesia
stumpffi, Calumma boettgeri, Furcifer pardalis, Zonosaurus madagascariensis, Phelsuma
grandis and Phelsuma seippi. All of these species are known to have a wide variety of niches
and highly adaptable (Glaw and Vences, 2007). Future studies should include a larger survey
effort data that will increase the chances of encountering those species with very low levels
of detectability.
12
MGF 154
5.
Georgia Lorenti
The effective of habitat degradation on bird communities
Introduction
The avifauna of Madagascar is often considered species-poor, hosting just 258 species
(204 breeding species), when compared to other landmasses of equivalent size (e.g. Borneo
hosts 688 species) (Goodman et al. 2003). Despite its low bird diversity, nearly half of the
species found in Madagascar are endemic (115 species) (Morris and Hawkins, 1998),
meaning the fauna are still significant and have high ecological value. Another striking
feature of Madagascan avifauna is the high degree of specialisation found within endemic
lineages, most notably their dependence on forest environments, with 80 of the 115 endemic
species (representing 37 endemic genera) being restricted to forest habitats (Morris and
Hawkins, 1998).
There has been little previous research conducted on the avifauna of Nosy Be. As the
island is close to the mainland, it is relatively easy for birds to disperse to and from the island,
and as such there are no species currently recognized as being endemic to the region.
Virtually all the species known to be present on the island have very large ranges and are of
less conservation concern unless their abundances are low across the range (IUCN, 2014).
The small area of protected forest in Lokobe National Park and its surrounding buffer zone
harbours many species of bird. These are species not found in secondary or degraded habitat
and their association with forest habitat increases the likelihood that they are of high
functional value. Such information can be used to determine which species are more likely to
be locally extirpated, should deforestation of the buffer zone continue.
Aims and objectives
Aim: To examine the effect of habitat degradation on bird diversity and abundance.
Objectives:
1.
To compare avian abundance between different habitat types from point count
surveys.
2.
To compare avian species richness between different habitat types from point count
surveys.
Methodology
Survey Methodology
Bird surveying was carried out using point counts along a 400m transect, with a point
count station positioned every 100m along this transect (Gregory et al. 2004; Hutto et al.
1986). Point counts are used to explore avian habitat relationships and to determine species
diversity and abundance (Ralph et al. 1995). The transects were selected based on a gradient
of human disturbance and utilised pre-existing trails in order to minimise impact on the local
environment. Bird species were identified by species based on both auditory and visual
confirmations. Surveys in heavily vegetated habitats relied almost exclusively on auditory
detections (Faanes and Bystrak 1981; Scott et al. 1981). A sampling period of five minutes
was used at each point count station. All avian species within an estimated radius of 25m
13
MGF 154
Georgia Lorenti
from the from the observers were recorded (Hutto et al. 1986; O'Dea et al. 2004). Birds which
were heard and then subsequently seen were recorded as seen, and those located outside of
the 25m radius and subsequently entered it were recorded as being within the point count
radius. Surveys were conducted in the morning, between 0630 and 0730 in order to record
bird abundance at peak activity times.
Statistical Analysis
Analysis of bird species richness was carried out in EstimateS, using rarefaction curves to
produce smoothed curves, and the ACE estimations of species richness were used to predict
the likely number of species if curves reach asymptote.
The Shapiro-Wilk Test has been labelled as the preferred test for normal distribution due to
its good power properties (Medes and Pala, 2003). It is recognised as the first statistical
procedure to detect departures from normality due to skewness (Razali and Wah, 2011). The
Shapiro-Wilk Test was selected to assess if variables had normal distributions as the
procedure is regarded as optimal for sample sizes less than 2000 (Razali and Wah, 2011).
Kruskal-Wallis H tests were used to determine whether a significant difference exists among
the habitat types (primary, secondary and degraded).The Kruskal-Wallis H test is a nonparametric test used to determine whether three or more independent groups are the same or
different on some variable of interest. It is used in situations where an ordinal level of data or
an interval or ratio level of data is available (Chan and Walmsley, 1997).
Both the Shapiro-Wilk Test Kruskal-Wallis conducted using IBM SPSS, version 17.0. All
tests were two-tailed, and statistical significance was set at p <0.05.
Results
Figure 5:
Mean number of birds in two habitat types (n =140). Error bars
represent the standard error (SE).
Closed forest and open habitat types showed a highly significant difference in bird abundance
(Kruskal-Wallis H = 12.851, df = 1, p < 0.001). The mean number of birds was highest in the
14
MGF 154
Georgia Lorenti
open habitat 2.45 ± 0.83 (mean ± SE). The closed forest habitat showed a mean abundance of
0.88 ± 0.47 (mean ± SE).
Figure 6:
Rarefaction curve showing how species richness increases with sampling
intensity. Dashed lines indicate mean ACE richness estimator, and how
quickly they reach asymptote.
Analysis of species richness using rarefaction between the three habitat types (Figure 6)
showed that there was little difference between sites. At the equivalent point of 4 (lowest
sampling intensity - open habitat) closed forest (8) had a species richness significantly lower
than open habitat (15). As none of the curves reached asymptote, it can be assumed that not
all species within the habitats were observed.
ACE estimates of total species richness from the collected data showed that the open habitat
was likely to be the most diverse site (species = 36.00, SD = 0.00). Should the data have
reached asymptote, ACE predicts a significantly lower species richness for closed forest
(species = 11.00, SD = 0.00).
15
MGF 154
Table 3:
Georgia Lorenti
Total number of bird species recorded in each habitat type. List of species
unique to closed forest and open habitat.
Closed Forest
Species Observed
Unique Species
8
Common Newtonia
Madagascar Green
Pigeon
Madagascar Kestrel
Paradise Flycatcher
Open Habitat
15
African Palm Swift
Broad-billed Roller
Common Myna
Crested Drongo
Frances' Sparrowhawk
Long-billed Tetraka
Madagascar Buzzard
Madagascar Red Fody
Madagascar Turtle Dove
Squacco Heron
Discussion
Quantifying the species diversity of bird communities has gained increasing
importance in environmental impact assessments, conservation planning (Bibby et al. 1992;
Stotz et al. 1996), and ecological research (Huston, 1994). The precise mechanisms for the
creation and maintenance of avian species diversity are still being debated, although there is a
general consensus that variables related to habitat heterogeneity play a prominent role
(Kissling et al. 2008). Different groups of organisms that show similar direct or indirect
responses to environmental factors are expected to display species richness patterns that are
spatially congruent (Kissling et al. 2008). Such patterns could have profound implications for
biodiversity conservation and for assessing the effects of habitat modification on birds.
During this survey phase, the trend for high mean abundance in open habitat was
found to be significant. The total number of individuals recorded in open habitat was 103,
while closed forest was significantly lower with only 37 individuals. A comparison of species
richness between the sites using rarefaction, to control for relative density, showed that there
was a difference between habitat types. ACE estimates of total species richness, obtained by
extrapolating the data, indicate that closed forest sites are likely to be the least diverse
(species = 11.00). The resulting differences between closed forest and degraded areas appear
to contradict previous studies, which generally suggest loss of diversity and abundance with
anthropogenic disturbance (Irwin et al. 2010; Watson et al. 2004). The reasons for this
disparity are difficult to ascertain. It might be attributed to the majority of species found
within the Ambalahonko area being generalists, and existing in a wide variety of habitats
over large parts of Madagascar (Sinclair and Langrand, 2013). The species accumulation
curve for both habitat types was not complete, therefore, there is the potential for more
species to be observed in these sites with greater survey effort. Biases are unavoidable when
using the point count methods, there will be differences in detectability in closed forest as
opposed to open habitat. Birds may remain undetected by researchers in dense vegetation and
there is the potential for vocalisations to be too faint, as sound attenuation in closed forest is
reduced (Schiek, 1997). It may be the case that true forest specialists, such as the Vasa
Parrots (Coracopsis spp.) (Morris et al. 1998), existed historically on the island of Nosy Be,
16
MGF 154
Georgia Lorenti
but have been extirpated, leaving only more tolerant species. Surveying more pristine core
forest areas on the edge of Lokobe National Park would be desirable for further examination
of the effect of habitat degradation on avian communities. It is also worth mentioning that
during the study period (October-December), large numbers of cicadas were present
throughout the area, their sound reducing the detection of bird calls. This could have
drastically affected bird sampling in closed forest where calls play a significant role in bird
encounters.
There were only four species recorded solely in closed forest habitat: Common
Newtonia, Madagascar Green Pigeon, Madagascar Kestrel and Paradise flycatcher, all of
which were recorded at low encounter rates (See Table 3). There were 10 species reported
solely in open habitat: African Palm Swift, Broad-billed Roller, Common Myna, Crested
Drongo, Frances' Sparrowhawk, Long-billed Tetraka, Madagascar Buzzard, Madagascar
Turtle Dove, Madagascar Red Fody and Squacco Heron. It was surprising that the Frances's
Sparrowhawk and Broad-billed Roller were reported exclusively in open habitat as both
these species are considered forest specialists (Morris et al. 1998). A study conducted by Peh
et al. (2006) suggested that nearby primary forest could act as a source habitat, resulting in a
steady influx of forest birds to degraded areas. Thus, the bird community we have observed
may not be representative of the forest avifauna in degraded lands that have no primary
forests nearby. There were only three species of birds found in both habitat types:
Madagascar Coucal, Madagscar Cuckoo and Souimanga Sunbird. These results may be
biased due to the inherent difficulties sampling in closed forest. Future studies should include
a larger survey effort data that will increase the chances of encountering those species with
very low levels of detectability.
17
MGF 154
6.
Georgia Lorenti
The butterfly diversity of the Ambalohonko Region
Introduction
Butterflies are often used as indicator species as they are conspicuous and relatively
easy to identify and are far better documented than other invertebrate taxa. Madagascar has
been used previously in case studies for sampling tropical butterflies which provides a useful
means of assessing overall invertebrate diversity (Kremen, 1992; Kremen, 1994). There are
currently around 297 species of butterfly described from Madagascar, with 210 (70%) of
these being endemic (Sáfián et al., 2009).
Previous studies into Lepidopteran diversity and abundance have shown that habitat
modification and loss have significant detrimental effects (Lawton et al., 1998). Butterflies
are known to be good indicator species as they are extremely sensitive to minor changes
within their micro-habitats, especially with regards to altered light levels (Kremen, 1992).
They are therefore useful tools in assessing and monitoring the levels of habitat degradation
through forest clearance, reacting rapidly to any changes in light concentrations associated
with the loss of canopy cover (Steer et al., 2009). On the nearby Comoros Islands, a study
conducted on the effects of habitat loss on the Lepidopteran communities showed that more
geographically widespread species replaced species which were more specialised and had
smaller distributions, in some cases restricting their distributions further and causing some
species to become endangered (Lewis et al., 1998). Monitoring over a long period of time can
allow for us to understand better how the butterfly assemblage changes over the course of the
seasons (Kunte, 1997).
Aims and Objectives
Aim: To examine the effect of habitat degradation on butterfly diversity and abundance.
Objectives:
1.
To compare avian abundance between different habitat types from surveys.
2.
To compare avian species richness between different habitat types from surveys.
Methodology
Butterfly Community
Surveys were carried out along three 400m transects (ZB, MA, TS; see figure 2 for location
on map, and table 1 for transect descriptions) between the hours of 1000 and 1400 at the time
of peak butterfly activity (Steer et al. 2009). Butterflies were surveyed using a modified
version of the Pollard walk (Pollard, 1977), similar to that used by Sparrow et al. (1994),
where researchers would walk slowly along a transect, counting the numbers of each species
observed and identifying them visually. Many of the species found in the area are difficult to
identify on the wing (notably the families Hesperidae and Lycaenidae), any individuals that
could not be positively identified in this way were captured using a sweep net and identified
using a set of photographic plates. Nomenclature primarily followed that of Safian et al.
(2009).
18
MGF 154
Georgia Lorenti
Statistical Analysis
Analysis of bird species richness was carried out in EstimateS, using rarefaction curves to
produce smoothed curves, and the ACE estimations of species richness were used to predict
the likely number of species if curves reach asymptote.
The Shapiro-Wilk Test has been labelled as the preferred test for normal distribution due to
its good power properties (Medes and Pala, 2003). It is recognised as the first statistical
procedure to detect departures from normality due to skewness (Razali and Wah, 2011). The
Shapiro-Wilk Test was selected to assess if variables had normal distributions as the
procedure is regarded as optimal for sample sizes less than 2000 (Razali and Wah, 2011).
Kruskal-Wallis H tests were used to determine whether a significant difference exists among
the habitat types (primary, secondary and degraded).The Kruskal-Wallis H test is a nonparametric test used to determine whether three or more independent groups are the same or
different on some variable of interest. It is used in situations where an ordinal level of data or
an interval or ratio level of data is available (Chan and Walmsley, 1997).
Both the Shapiro-Wilk Test Kruskal-Wallis conducted using IBM SPSS, version 17.0. All
tests were two-tailed, and statistical significance was set at p <0.05.
Results
Figure 7:
Mean number of butterflies in two habitat types (n = 412). Error bars
represent the standard error (SE).
A Kruskal-Wallis H carried out in SPSS showed that there were significant differences in the
abundance of butterflies between habitat types (Kruskal-Wallis H = 12.851, df = 1, p <0.001).
Open habitat had the highest mean number of butterflies 8.95 ± 5.38 (mean ± SE). Butterflies
in closed forest had a mean abundance of 0.59 ± 0.24 (mean ± SE).
19
MGF 154
Georgia Lorenti
Analysis of species richness in the two habitat types (Figure 8) showed that at an equivalent
point of sampling intensity, open habitat was more diverse (13 species). Closed forest was
significantly less diverse with a species richness of 8.
Figure 8:
Rarefaction curve showing how species richness increases with
sampling intensity in forest, forest edge and open habitats.
ACE estimates of species richness, should the data have reached asymptote, predict a species
richness for closed forest and open habitat to be 11.27 and 13.69, respectively. Open habitat
is therefore likely to be the most diverse site, being significantly higher than closed forest.
Table 4:
Total number of reptile species recorded in each habitat type. List of species
unique to closed forest and open habitat.
Total Species
Unique Species
Closed Forest
8
Neptis kikideli
Open Habitat
13
Papilio grosesmithi
Papilio demodocus
Colotis evanthe
Acraea ranavalona
Acraea serena
(=eponina)
Junonia goudotii
Discussion
Butterfly abundance was highest in the open habitat with a mean value of 8.95 ± 5.38
(mean ± SE), while closed forest habitat showed a mean abundance of 0.59 ± 0.24 (mean ±
SE). Additionally, a comparison of species richness between research areas using rarefaction
showed that at an equivalent point of sampling intensity, open habitat was more diverse (13)
while closed forest was significantly less diverse (8 species). As none of the curves reached
asymptote, it can be assumed that not all the species within the habitats were observed.
20
MGF 154
Georgia Lorenti
ACE estimates of total species richness, obtained by extrapolating the data, indicate that
closed forest sites are likely to be the least diverse site (11.27) in comparison to open habitat
(species = 13.69). Changes in species abundance and diversity are likely to be linked to both
the environmental and physical effects that human disturbance has on the ecological and life
history traits of species (Cleary et al. 2009). Habitat degradation increases the frequency and
size of gaps in the canopy which are liable to favour some groups of species while hindering
others. The loss of biodiversity in anthropogenically modified habitats reduces the number of
species with a restricted area distribution. These specialised species are of conservation
importance because they are vulnerable and prone to extinction (Spitzer et al. 1997). For
example, Neptis kikideli was observed exclusively in closed forest (See Table 4). This is
congruent with the literature as it is known to be a forest specialist and thus unlikely to
survive in disturbed environments (Henning et al. 1997). Papilio demodocus, Colotis
evanthe, Acraea serena and Junonia goudotii were observed solely in open habitat and are
known to thrive in forest margins as well as transformed grasslands (Henning et al. 1997).
More interesting is the sole observation of Acraea ranavalona and Papilio grosesmithi in
open habitat. Both these species are endemic to Madagascar and generally considered to be a
forest specialists (Henning et al. 1997). Their appearance in open habitat might suggest that
these species are more robust than previously considered.
The data collection method is somewhat biased. Although the Pollard method
(Pollard, 1977) reduces the chance of pseudoreplication by moving continuously along a
transect, observer bias has to be taken into account. This is because many of the species that
inhabit Nosy Be are small and difficult or impossible to identify on the wing, particularly
Heteropsis spp., and the families Lycaenidae and Hesperiidae. Heteropsis need to be closely
examined as many are dull coloured with differing numbers and positions of eye spots.
Lycaenidae and Hesperiidae consist of mostly very small, fast moving species, and also need
to be examined closely in the hand to be able to differentiate them. The method attempted to
account for this is through capturing species not identifiable on the wing, however, to catch
all individuals by sweep netting is difficult, even by a very experienced researcher.
Lycaedinae and Hesperiidae tend to stay close to vegetation, so it is often difficult to capture
them without damaging the nets. This clearly represents an undersampling bias for these
groups. The more enclosed environment within forested transects also restricts the ability to
sweep net effectively in dense vegetation, so for species which cannot be identified on the
wing there is likely to be further disparity in sampling between forested habitats and the more
open scrubland and plantations of the degraded habitats. As mentioned above, it is also
probable that forest specialist species spend more time in the canopy, close to sunlight and
available food, and are thus under sampled in closed forest habitat (Dumbrell and Hill, 2005).
21
MGF 154
7.
Georgia Lorenti
Satellite Camp: Snapshot Analysis of Potential Research Site
Introduction
During this phase, the MGF team held a satellite camp to assess the potential of a new
research site. A combination of active searches and explorative walks were utilised to
investigate both the scientific value and logistical implications of the potential site. The
prospective site was roughly located at 13°22'44.7"S 48°19'56.2"E, between the villages of
Ampasipohy and Ambatazavavy. The area was highlighted as a locale of potential interest by
Richard Burger, four months prior to this satellite camp. He described an area of intact forest,
bursting with wildlife and very little anthropogenic disturbance. Sadly, this is no longer the
case as practically all of the forest area has been decimated. The current landscape is
composed of large expanses of cleared forest and a small cluster of trees. Such large scale
habitat degradation meant that the new potential site was determined to be not appropriate for
future research. A short analysis of the fauna present in the area is discussed in this report.
Methodology
Explorative walks were used to locate small clusters of trees that could be used for
surveying. Establishing transect lines in private owned forests was inappropriate and could
potentially assist in furthering logging activities. For this reason, active searches were
conducted using reconnaissance walks. Such walks followed pre-existing trails and allowed
for movement around obstacles rather than cutting through them. In some instances,
reconnaissance walks simply followed the river bed as the cluster of remaining trees was so
small that pre-existing trails did not exist. Although biased in some respects, several studies
have shown a good correlation between encounter rates from reconnaissance walks and
population density. Active searches were used to sample reptile, bird, butterfly and mammal
communities. Most survey periods took place between the hours of 0800 and 1400. Night
time surveys were also conducted to ensure that both nocturnal and diurnal species were
equally represented in study. The sum total area sampled was 15,000m2.
Species diversity was then assessed using the Shannon diversity index (H’).
Where: pi is the proportion of the sample that belongs to the ith species (Stirling, 1999). It
generally varies from 1 (low diversity) to 5 (high diversity) (Gering et al. 2002). The index
takes into account both the number and evenness of species. The diversity value is increased
through a greater number of species combined with a more even distribution (Stirling, 1999).
22
MGF 154
Georgia Lorenti
Results
Figure 9:
Percentage of communities (reptile, bird, butterfly and mammal)
recorded during surveys (n = 49).
The graph depicts the percentage composition of each community type over the entire survey
period. Birds represented the largest proportion of the total species observed (40.82%).
Mammals only constituted the smallest percentage of the overall species composition
(10.2%). It is also important to note that over the five day sample period a total of only 49
individuals were observed. The Shannon diversity index (H') value was 2.80.
Discussion
The severe lack of forest area made surveys difficult to accomplish with the scientific
rigour associated with statistically valid studies. Reconnaissance walks were purposely
situated to run through the small cluster of trees or along the riverbed. As a result, the data
gathered during surveys was inevitably biased. The small number of individuals observed
during the five days of sampling is evident of the highly degraded nature of the area.
Deforestation threatens species survival and diminishes biodiversity by creating forest
fragments that struggle to maintain viable populations. Species diversity at the site was higher
than expected, but still comparatively low in contrast to other parts of Nosy Be (Andreone et
al. 2003). This could easily be explained by the extreme level of deforestation that has
occurred over a short time period (less than four months). Under such conditions, the
appearance of higher species diversity could be an artefact of resident communities being
‘packed’ in at artificial densities. Reduced habitat area and crowding of individuals can have
a negative effect on the overall health of populations (Gillespie et al. 2008). Eventually the
competition for limited resources will result in the mortality of many conspecific and
reducing species diversity.
Birds represented the largest proportion of the total species observed (40.82%).
Several studies have addressed the conservation value of agricultural landscapes using
community data from various taxonomic bird groups. In several cases, studies showed that a
relatively high number of individuals and species can still be found in anthropogenically
23
MGF 154
Georgia Lorenti
modified landscapes, and that a considerable proportion of these species are part of the
natural forest fauna (Estrada et al. 1993, Daily et al. 2001, Hughes et al. 2002). However,
thus is a cautious interpretation of abundance and species richness data as the intensification
of agricultural land is still ongoing and long-term stability of bird populations is uncertain.
Mammals constituted the smallest percentage of the overall species composition
(10.2%). Lemur sightings were restricted to only three individuals. Habitat conversion and
fragmentation are the main reasons for the global decline of primate populations (ArroyoRodriguez and Dias, 2010) and thus more than half of the world’s primate species are
currently threatened by extinction (Chapman and Peres, 2001). The severity of deforestation
in the area means that new site is not conducive to the continued survival of lemur
populations.
The extremely degraded nature of the area means it of little scientific value and has
little to none chance of recovery. Roughly 90% of the Madagascar's endemic species are
forest dwelling and are highly susceptible to anthropogenic disturbance (Dufils, 2003).
Clearance for agriculture remains a major driver of forest destruction, severely threatening
the island's unique biota (Irwin et al., 2010). From a conservation perspective, the site serves
as a warning of the desperate need for immediate proactive preservation activities in
Madagascar. The results of this satellite camp should come as a call to action. A reminder of
the work that needs to be done. The future of Madagascan wildlife depends critically on the
interactive management of both ecological and anthropogenic requirements. As the American
scientist Aldo Leopold once wrote "conservation is a state of harmony between men and
land."
24
MGF 154
Georgia Lorenti
References:
Abe, AS (1983). Observations on dormancy in tegu lizards: Tupinambis teguixin (Reptilia,
Teiidae). Naturalia. 8: 235–239.
Andreone, F, Guarino, FM and Randrianirina, JE (2005). Life history traits, age profile, and
conservation of the panther chameleon, Furcifer pardalis (Cuvier 1829), at Nosy Be, NW
Madagascar. Tropical Zoology. 18: 209–225.
Andreone, F,Vences, M and Randrianirina, JE (2001). Patterns of amphibian and reptile
diversity at Berara Forest (Sahamalaza Peninsula), NW Madagascar. Italian Journal of
Zoology. 68: 235-241.
Andreone, F, Randrianirina, JE, Jenkins, PD and Aprea, G (2000). Species diversity of
Amphibia, Reptilia and Lypotyphla (Mammalia) at Ambolokopatrika, a rainforest between
Anjanaharibe-Sud and Marojejy massifs, NE Madagascar. Biodiversity and Conservation. 9:
1587–1622.
Andreone, F, Glaw. F, Nussbaum, RA, Raxworthy, CJ, Vences, M and Randrianirina, JE
(2003). The amphibians and reptiles of Nosy Be (NW Madagascar) and nearby islands: a case
study of diversity and conservation of an insular fauna. Journal of Natural History. 37: 21192149.
Arroyo-Rodrigues, V and Dias, PAD (2010). Effects of fragmentation and disturbance on
howler monkeys: A review. American Journal of Primatology. 72: 1-16.
Bawa, KS, Seidler, R. (1988). Natural forest management and conservation of biodiversity in
tropical forests. Conservation Biology. 12: 46–55.
Bibby, CJ, Collar, NJ, Crosby, MJ, Heath, MF, Imboden, C, Johnson, TH, Long, AJ,
Stattersfield, AJ, Thiergood, SJ (1992). Putting biodiversity on the map: Priority areas for
global conservation. ICBP, Cambridge, United Kingdom.
Burger, R. (2014). Frontier Madagascar Terrestrial Research Project Science Report Phase
142. Unpublished.
Burger, R. (2015). Frontier Madagascar Terrestrial Research Project Science Report Phase
151. Unpublished.
Chapman CA, Peres CA (2001). Primate conservation in the new millennium: the role of
scientists. Evolutionary Anthropology. 10:16–33.
Chan, Y and Walmsley, RP. (1997). Learning and understanding the Kruskal-Wallis one-way
analysis-of-variance-by-ranks test for differences among three or more independent groups.
Physical therapy. 77(12): 1755-1761.
Cleary, DFR, Genner, MJ, Koh, LP, Boyle, TS, de Jong, R and Menken, SBJ (2009).
Butterfly species and traits associated with selectively logged forest in Borneo. Basic and
Applied Ecology. 10: 237-245.
25
MGF 154
Georgia Lorenti
Daily, GC, Ehrlich, PR and Sanchez-Azofeifa, GA (2001). Countryside biogeography: use of
human-dominated habitats by the avifauna of southern Costa Rica. Ecological Applications
11:1–13.
Dewar, RE (2003). Relationship between human ecological pressure and the vertebrate
extinctions. In: Goodman, SM and Benstead, JP (Eds). The Natural History of Madagascar.
University of Chicago Press, Chicago, pp 119-122.
D'Cruze, N, Kohler, J, Franzen, M, and Glaw, F (2008). A conservation assessment of the
amphibians and reptiles of the Foret d'Ambre Special Reserve, north Madagascar.
Madagascar Conservation and Management. 3: 44-54.
Doan, TM (2003). Which methods are most effect for surveying rain forest herpetofauna?
Journal of Herpetology. 37(1): 72-81.
Dufils, JM (2003). Forest Ecology. In: Goodman, SM and Benstead, JP (Eds). The Natural
History of Madagascar. University of Chicago Press: Chicago, pp 88-111.
Du Puy, DJ and Moat J (2003). Using Geological substrate to identify and map primary
vegetation types in Madagascar and the implications for planning biodiversity conservation.
In: Goodman, SM and Benstead, JP (Eds). The Natural History of Madagascar. University of
Chicago Press: Chicago, pp 51-67.
Dumbrell, AJ and Hill, KJ (2005). Impacts of selective logging on canopy and ground
assemblages of tropical forest butterflies: implications for sampling. Biological Conservation.
125: 123-131.
Estrada, A, Coates-Estrada, R and Meritt, DJ (1993). Bat species richness and abundance in
tropical rain forest fragments and in agricultural habitats at Los-Tuxtlas, Mexico. Ecography
16:309–318.
Fannes, CA and D. Bystrak. (1981). The role of observer bias in the North American
breeding bird survey. In: Ralph, CJ and Scott, JM (Eds). Estimating numbers of terrestrial
Birds: Studies in Avian Biology, pp 353-359.
Ganzhorn, JU, Lowry, PP, Schatz, GE and Sommer, S (2001). The biodiversity of
Madagascar: One of the world’s hottest hotspots on its way out. Oryx. 35(4): 346-348.
Gering.JC, Crist, TO and Veech, JA (2003). Additive partitioning of species diversity across
multiple spatial scales: implications for regional conservation of biodiversity. Conservation
Biology. 17(2): 488-499.
Gibbons, JW, Scott, DE, Ryan, TJ, Buhlmann, KA, Tuberville TD, Metts, BS, Greene, JL,
Mills, T, Leiden, Y, Poppy, S and Winne, CT (2000). The global decline of reptiles, deja vu
amphibians. BioScience. 50: 653-666.
Gillespie, TR and Chapman, CA (2006) Prediction of parasite infection dynamics in primate
metapopulations based on attributes of forest fragmentation. Conservation Biology. 20(2):
441-448.
26
MGF 154
Georgia Lorenti
Glaw, F and Vences, M (2007). A field guide to the amphibians and reptiles of Madagascar.
3rd edition. Vences and Glaw Verlag. Koln.
Goodman, SM (1993). A reconnaissance of Ile Sainte Marie, Madagascar: the status of the
forest, avifauna, lemurs and fruit bats. Biological Conservation. 65: 205–212.
Goodman, SM, Rakotondravony, D (2003). The effects of forest fragmentation and isolation
on insectivorous small mammals (Lipotyphla) on the Central High Plateau of Madagascar.
Journal of Zoology. 250: 193–200.
Gregory, RD, Gibbons, DW and Donald, PF (2004). Bird census and survey techniques. Bird
Ecology and Conservation. pp. 17-56.
Hailey, A and Loveridge, JP (1997). Metabolic depression during dormancy in the African
tortoise Kinixy spekii. Canadian Journal of Zoology. 75: 1328 - 1335.
Harper, GJ, Steininger, MK, Tucker, CJ, Juhn, D and Hawkins F (2007). Fifty years of
deforestation and forest fragmentation in Madagascar. Environmental Conservation. 34: 325333.
Heinen, JT (1992). Comparisons of the leaf litter herpetofauna in abandoned cacao
plantations and primary rain forest in Costa Rica: Some implications for faunal restoration.
Biotropica. 24: 431-439.
Henning, GA, Henning, SF, Joannou, JG and Woodall, SE (1997). Living butterflies of
Southern Africa: biology, ecology and conservation. In: Dixon, R and Steffens, SE (Eds).
Umdaus Press, Hatfield.
Hughes, JB, Daily, GC and Ehrlich, PR (2002). Conservation of tropical forest birds in
countryside habitats. Ecology Letters 5:121–129
Huston, MA (1994). Biological Diversity. In: Huston, MA (Ed). Cambridge University Press,
Cambridge.
Hutto, RL, Pletschet, SM and Hendricks, P (1986). A Fixed-Radius Point Count Method for
Nonbreeding and Breeding-Season Use. The Auk. 103: 593–602.
Ingram. JC and Dawson TP (2006). Forest Cover, Condition, and Ecology in HumanImpacted Forests, South-Eastern Madagascar. Conservation and Society. 4(2): 194-230.
Irwin, MT, Wright, PC, Birkinshaw, C, Fisher, BL, Gardener, CJ, Glos, J, Goodman, SL,
Loiselle, P, Rabeson, P, Raharison, JL, Raharelalao, MJ, Rakotondravony, D, Raselimanana,
A, Ratsimbafany, J, Sparks, JS, Wilme, L and Ganzhorn, J (2010). Patterns of Species
Change in Anthropogenically changed Forests in Madagascar. Biological Conservation,
143(10): 2351-2362.
Jenkins, RKB, Brady, LD, Bisoa, M, Rabearivony, J and Griffiths, RA (2003). Forest
disturbance and river proximity influence chameleon abundance in Madagascar. Biological
Conservation. 109: 407-415.
27
MGF 154
Georgia Lorenti
Jestrzemski, D, Schutz, S, Nguyen, TQ and Ziegler, T (2013). A survey of amphibians and
reptiles in Chu Mom Ray National Park, Vietnam, with implications for herpetofaunal
conservation. Asian Journal of Conservation Biology. 2(2): 88-110.
Kissling WD, Field, R and Bohing-Gaese, K (2008). Spatial patterns of woody plant and bird
diversity: functional relationships or environmental effects? Global Ecology and
biogeography. DOI: 10.1111/j.1466-8238.2007.00379.
Kremen, C (1992). Assessing the indicator properties of species assemblages for natural areas
monitoring. Ecological applications. 2(2): 203-217.
Kremen, C (1994). Biological inventory using target taxa: a case study of the butterflies of
Madagascar. Ecological applications. 4(3): 407-422.
Kunte, KJ (1997). Seasonal patterns in butterfly abundance and species diversity in four
tropical habitats in northern Western Ghats. Journal of Bioscience 22, 593-693.
Langrand, O (1990). Guide to the Birds of Madagascar. Yale University Publication.
Lawton, JH, Bignell, D E, Bolton, B, Bloemers, GF, Eggleton, P, Hammond, PM, Hodda, M,
Holt, RD, Larsen, TB, Mawdsley, Stork, NE, Srivastava, DS and Watt, AD (1998).
Biodiversity inventories, indicator taxa and effects of habitat modification in tropical forest.
Nature. 391(6662): 72-76.
Legris, P and Blasco, F (1965). Carton des bioclimats. In: Humbert, H and Cours-Darne, G.
(Eds). Notice de la carte de Madagascar. Travaux de la Section Scientifique et Technique de
l’Institut Français de Pondichéry, pp. 31–45.
Lewis, OT, Wilson, RJ, and Harper, MC (1998). Endemic butterflies on Grande Comore:
habitat preferences and conservation priorities. Biological Conservation. 85(1): 113-121.
Lieberman, SS (1986). Ecology of the leaf litter herpetofauna of a Neotropical rainforest: La
Selva, Costa Rica. ActaZool. Mex. Nueva Ser. 15: 1-72.
McEachern, MA, Adams, AAY, Klug, PE, Fitzgerald, LA and Reed, RN (2015). Brumation
of introduced black and white tegus, Tupinambis merianae (Squamata: Teiidae), in southern
Florida. Southeastern Naturalist. 14(2): 319-328.
Miyata, KI (1980). Patterns of diversity in tropical herpetofaunas. Ph.D. dissertation. Harvard
University, Cambridge, Massachusetts.
Morris, P, Hawkins, F and Andrews, M (1998). Birds of Madagáscar. A photographic guide.
Pica Press.
Myers, N, Mittermeier, RA, Mittermeier, CG, da Fonseca, GAB and Kent, J (2000).
Biodiversity hotspots for conservation priorities. Nature. 403: 853-868.
O'Dea, N, Watson, JEM, WHittaker, RJ (2004). Rapid assessment in conservation research: a
critique of avifaunal assessment techniques illustrated by Ecudorian and Madagascan case
study data. Diversity and Distributions. 10: 55-63.
28
MGF 154
Georgia Lorenti
Peh, KSH, Sodhi, NS, De Jong, J, Sekercioglu, CH, Yap, CAM and Lim, SLH. (2006).
Conservation value of degraded habitats for forest birds in southern Peninsular Malaysia.
Diversity and Distributions. 12(5): 572-581.
Pollard, E (1977). A method for assessing changes in the abundance of butterflies. Biological
conservation. 12: 115-134.
Ralph, CJ, Droege, S and Sauer, JR (1995). Managing and monitoring birds using point
counts: Standards and applications. In: Ralph, CJ, Sauer, JR and Droege, S (Eds). Monitoring
bird population by point counts. U.S. Department of Agriculture, Forest Service General
Technical Report, pp 161-168.
Ramanamanjato, JB, Ganzhorn, JU ( 2001). Effects of forest fragmentation, introduced
Rattus rattus and the role of exotic tree plantations and secondary vegetation for the
conservation of an endemic rodent and a small lemur in littoral forests of southeastern
Madagascar. Animal Conservation. 4: 175–183.
Randrianandianina, BN, Andriamahaly, LR, Harisoa, FM, and Nicoll, ME (2003). The role of
protected areas in the management of the island’s biodiversity. In: Goodman, S.M., and
Benstead, JP (Eds). The Natural History of Madagascar. University of Chicago Press,
Chicago, pp 1423-1432.
Raselimanana, AP, Raxworthy, CJ and Nussbaum, RA (2000). Herpetofaunal species
diversity and elevational distribution within the Parc National de Marojejy, Madagascar. In
Goodman, SM (Ed.) A Floral and Faunal Inventory of the Parc National de Marojejy: with
Reference to Elevational Variation. Fieldiana: Zoology. 92: 157–174.
Raxworthy, CJ and Nussbaum, RA (1994). A rainforest survey of amphibians, reptiles and
small mammals at Montagne d’Ambre, Madagascar. Biological Conservation. 69: 65–73.
Raxworthy, CJ, Andreone, F, Nussbaum, RA, Rabibisoa, N and Randriamahazo, H (1998).
Amphibians and reptiles of the Anjanaharibe-Sud Massif: elevational distributions and
regional endemicity. In Goodman, SM (Ed). A Floral and Faunal Inventory of the Reserve
Speciale d’Anjanaharibe-Sud, Madagascar: with Reference to Elevational Variation.
Fieldiana: Zoology. 90: 79–92.
Razali, NM and Wah, YB (2011). Power comparisons of Shapiro-wilk, Kolmogorovsmirnov, Lilliefors and Anderson-darling tests. Journal of Statistical Modelling and
Analytiscs. 2(1): 21-33.
Sáfián, SZ, Collins, SC, Kormos, B and Siklósi, A (2009). African Butterfly Database.
Version 1.0. Retrieved from: (www.adbd-africa.org): 26 September 2015.
Schieck, J (1997). Biased detection of bird vocalizations affects comparisons of bird
abundance amongst forest habitats. The Condor.. 99: 179-190.
Scott, DM, Brown, D, Mahood, S, Denton, B, Silburn, A and Rakotonraparany, F (2006).
The impacts of forest clearance on lizard, small mammal and bird communities in the arid
spiny forest, southern Madagascar. Biological conservation. 127: 72-87.
29
MGF 154
Georgia Lorenti
Scott, JM, Ramsey, FL and Kepler, CB. (1981). Distance estimation as a variable in
estimating bird numbers from vocalizations. In: Ralph, CJ and Scott, JM (Eds). Estimating
numbers of terrestrial Birds: Studies in Avian Biology, pp 334-340.
Spitzer, K, Jaros, J, Havelka, J and Leps, J (1997). Effect of small-scale disturbannce on
butterfly communities of an Indochinese montane rainforest. Biological conservation. 80: 915.
Steer, MD, Vater, A and McCann-Wood, S (2009). The Effect of Forest Degradation on the
Species Richness and Diversity of a Diurnal Lepidoptera Community in Northern
Madagascar. Frontier Madagascar Environmental Research Report. 22: 1-20.
Stephenson, PJ (1995). Small mammal microhabitat use in lowland rain forest of north-east
Madagascar. Acta Theriologica 40, 425– 438.
Stiling, P (1999). Ecology: Theories and Applications (3rd edition). New Jersey, Prentice
Hall.
Stotz, DF, Fitzpatrick, JW, Parker, TA and Moskovits, DK (1996). Neotropical Birds:
Ecology and Conservation. In: Stotz, DF, Fitzpatrick, JW, Parker, TA and Moskovits (Eds).
University of Chicago Press, Chicago.
Quéméré, E. Amelot, X. Pierson, J. Crouau-Roy, B and Chikhi, L (2012). Genetic data
suggest a natural prehuman origin of open habitats in northern Madagascar and question the
deforestation narrative in this region. Proceedings of the National Academy of Sciences. 109
(32): 13,028-13,033.
Vonesh, JR (2001). Patterns of richness and abundance in tropical African leaf-litter
herpetofauna. Biotropica. 33(3): 501-510.
Wells, NA (2003). Some Hypotheses on the Mesozoic and Cenozoic paleoenvironment
history of Madagascar. In: Goodman, SM and Benstead, JP (Eds). The Natural History of
Madagascar. University of Chicago Press: Chicago, pp 16-34.
30