3.6 Recommendations Considering that baseline survey within the Emalu forest has now been conducted,
the best option available will be to build on this by conducting subsequent surveys
and standardising the survey techniques especially for the sticky traps and frog
surveys, carrying them out over different seasons and assessing species densities.
Any changes in terms of presence/absence and density over time will indicate the
status of the forest. It is recommended that these intensive and dedicated surveys
focus on a particular area or along standard transects. It is also recommended that
tree climbing techniques be used to enable better capture rates of cryptic skinks and
CHAPTER 4: AVIFAUNA Alivereti Naikatini
4.1 Summary The main objectives of the study were to compile a checklist of the birds and bats
species present and observed, and determine the presence of species of high
conservation importance (focal species) for monitoring in the future. The assessment
methods used during the survey were the Point Count Method with a fixed radius of
50m; evening (dusk) bat counts using a Bat Detector device to detect presence of
micro-bats; interviewing of local guides, and opportunistic surveys. About 4000
minutes of avifauna studies were conducted during the two surveys where 59 points
were assessed in 2012 and an additional 37 points in 2013. A total of 35 species of
birds were recorded during the two surveys which included 25 endemic and one
exotic species. Two species of bats were also recorded during the surveys. Ten focal
species were identified (eight bird species and two bat species). The bird diversity of
Emalu is comparable to the four largest Important Bird Areas (IBAs FJ07, FJ08, FJ09,
and FJ10) on Viti Levu and ranks even higher in terms of bird density.
4.2 Introduction Fiji’s bats play an essential role as seed dispersing agents, major pollinators, and
insect control agents in the rainforest and other terrestrial ecosystems (Palmeirim et al., 2007). However, bats are understudied in Fiji in terms of ecological research and
there is little public awareness of their role and importance. Bats are the only native
terrestrial mammals of Fiji and six species occur in Fiji, four of which are native and
two endemic (Flannery, 1995, Palmeirim et al., 2007). Four bat species are listed as
threatened (Palmeirim et al., 2007).
Like bats, birds are also very important indicators of the forest health. They are
important seed dispersers, pollinators and insect control agents. In a pristine forest
system, one would expect to find more native and endemic species. There are 68
species of land birds found in Fiji, eleven of which are introduced species.
No previous bird or bat surveys have been carried out in the Emalu area. A few
recent studies were carried to areas close to Emalu, including a bat survey of the
Tatuba caves in the vicinity of Saweni in the Namataku District (Palmeirim et al.,
2007). The most recent bird survey close to the study area was carried out by Birdlife
International (Fiji) in the Southern Viti Levu Highlands (IBA FJ10), which is to the
south of Emalu, Sovi Basin (IBA FJ08) to the east and the Rairaimatuku Plateau (IBA
FJ09) to the north (Masibalavu and Dutson, 2006).
The main objectives of this survey were to:
Provide a checklist of all avifauna species (birds and bats) present in the site,
Highlight species that are of conservation importance (focal species),
Provide preliminary abundances of species present, and to
Develop a methodology for avifauna monitoring work in the future.
4.3 Methodology The survey methods used in the survey were the:
Point count method (for both bats and birds),
Evening counts for bats,
Bat detector surveys in the evenings,
Interviews with local communities.
The point count method was the most commonly used method to survey for the bats
and birds. It was only carried out in the morning and afternoons when birds are
more active. Counts in a point were restricted within a 50m radius for a period of ten
minutes according to an established methodology (Naikatini, 2009). Stations were not
randomly located, due to the rugged terrain of the area, but were placed along tracks
and accessible areas. To maximise the size of the area covered, points were placed at
least 200 – 400m apart. This was also done to minimise the likelihood of double
counts. Each morning or afternoon session would last two to four hours depending
on the weather. All birds detected within the 50m radius area were recorded and
GPS locations noted. The inclusion of as many sub-habitats as possible – riparian,
flat, slope, ridge and ridge top - in disturbed and undisturbed areas was attempted.
The total number of points, birds and species recorded were tabulated to give the
relative abundance or density of each species.
Bat surveys were also carried out by conducting bat counts in the early evenings
(from about one hour before sunset – 17:00 to 18:00) from a good lookout or open
area to determine what bat species were flying over and their direction of flight. The
total number of bats counted in an hour would give an idea of the bat activity and
abundance in the study area. Bat detectors were also used in the evenings near the
camp site by walking along the trail and stopping at various points where there was
an opening or gap in the canopy and pointing the bat detector into the direction of
the sky. The bat detector enabled us to tune to the frequencies at which the two
micro-bat species (present in Fiji) would be detected if they flew over or were feeding
nearby. These surveys were only carried out for about an hour between 1900 and
2200 hours, and also when weather conditions were favorable for such surveys.
Opportunistic surveys were also conducted whilst travelling from one point station
to another, or whilst travelling within the area from one base camp to another.
Interviews with the local guides were carried out on some evenings. Local guides
knew the area well, including where the main bat roosts are located, as well as the
species of birds they may have encountered in the area previously.
4.4 Results and discussion In total approximately 4000 minutes were spent actively conducting bat and bird
surveys, and over 70 hectares were covered using the point count method. A total of
35 species of land birds and two species of bats were recorded in the study site, and
these are listed in Appendix 7. Identifications were verified using a published field
guide (Watling, 2001). A total of 96 point stations were surveyed during the 20 days
of survey. These point stations (shown on Map 6) were located in the different sub-
habitat types found with the main vegetation systems; lowland rainforest (<600m
elevation), upland rainforest (600-800m elevation) and cloud forest (>800m
elevation). A table of the location and habitat of each station and a summary of the
species diversity and bird abundance is provided in Appendix 8
Of the 35 species of land birds recorded, one is an exotic species and 25 are endemic
to Fiji. The exotic species, commonly known as the red-vented bulbul (Pycnotus cafer)
on the IUCN Red List as being a species of Least Concern (Birdlife International,
2012a) and is more common on the western edges of the Emalu site. Eight species of
birds recorded are listed as focal bird species for conservation in Fiji because of their
status (Appendix 9). Stations where bird and bat focal species were recorded are
marked on (Map 7).
The long-leggd warbler (Fig. 18), classified as Endangered on the IUCN Red List
(Birdlife International, 2012b) was found to be common in the upland and
undisturbed riparian vegetation; an example of this habit is shown in Fig. 17.
Sightings of the collared lorry, Phigys solitarius (Fig. 19), and the golden dove,
Ptilinopus luteovirens (Fig. 20), were also made during the survey.
Only two species of bats were recorded throughout the survey; Pteropus samoensis, the Samoan flying-fox and P. tonganus the Pacific flying-fox. Pteropus samoensis (Fig.
21) is listed in the IUCN Red List as Near Threatened (Brooke and Wiles, 2008). P. tonganus (Fig. 22) was not commonly encountered in the study area in 2012 however
it was common in the areas surveyed in 2013 and seemed to be more common in the
upper Mavuvu catchment. Here, two of the guides were able to catch seven bats one
evening in just one hour, with sticks. The guides also mentioned that the upper
Mavuvu area was well known for bats. No bat roost for P. tonganus was sighted in
the Emalu REDD+ site. The closest roost was located outside the study area, and
consisted of over 1000 bats. There could be roosts located in the forested areas on the
Namosi side of Emalu however time constraints did not allow for a confirmation of
this. No micro-bats were detected using the bat detectors. However this should not
imply that there are no micro-bats foraging for food in Emalu as there needs to be
more follow up studies to confirm this.
Table 1. Comparison of Emalu to the four largest Important Bird Areas (IBAs) of Viti
Emalu & IBAs
Viti Levu Southern Highlands
Table 1 shows that native bird species diversity in Emalu is comparable to Viti Levu’s
four largest Important Bird Areas (IBAs), and has a slightly higher number of
endemic species. In terms of species density it is the highest ever recorded for
anywhere in Fiji to date.
4.5 Recommendations To better understand the ecology and abundance of the avifauna of Emalu there is a
need to carry out further monitoring work. To monitor the bird and Pteropus samoensis populations, we recommend the use of the point count method with a fixed
50m radius and 8-10 minute counts per station. For best practice, future monitoring
surveys should include approximately 70 point count stations spread out over the
various vegetation systems present; cloud forest (10 stations), upland rainforest (20
stations), lowland rainforest (20 stations), grassland (10 stations), secondary forest(10
stations), and ensuring within these that there is coverage across the different sub-
habitats (riparian, flat, slope, ridge, and ridge-top).
To monitor for the other bat species a further survey of the area is needed to locate
the roosts, both in the area and the surrounding forest systems as it is most likely that
bats roosting outside the Emalu site will be flying in to forage for food, e.g. from the P. tonganus roost at Vurunamasima near Navitilevu Village and the Notopteris macdonaldi roost in Saweni (Navosa) and Nabukelevu (Serua). These roosts are both
about 10 km from the edge of the Emalu site. When the roosts are located, population
counts will be performed for monitoring purposes.
The Emalu REDD+ site should be an area of conservation priority for the
Government of Fiji. As yet Fiji has no dedicated bird reserve and it is recommended
that, given the species diversity and high endemism levels as well as its ideal
location, the Emalu area be designated an established protected bird area.
Conservation should be a priority and logging should not be permitted in this area if
you take into account the true value of the site ecosystem function, rich biodiversity,
cultural and spiritual importance, all of which are invaluable monetarily.
CHAPTER 5: TERRESTRIAL INSECTS Hilda Waqa-Sakiti
5.1 Summary A total of 26 families of the target taxa Coleoptera (beetles) was recorded in the
Emalu areas, as well as a high abundance of the family Formicidae (ants). These taxa
provide critical ecosystem services in forests systems such as soil processing,
decomposition, herbivory, pollination and seed dispersal. Insects of conservation
value recorded from Emalu included: Hypolimnas inopinata (a rare and endemic
butterfly), Nysirus spinulosus and Cotylosoma dipneusticum (rare and endemic stick
insects) and Raiateana knowlesi (the rare and endemic cicada). These findings suggest
that the Emalu area is pristine and an important site for rare insects on Viti Levu.
5.2 Introduction This was the first entomological survey to be conducted within the Emalu forest. A
baseline survey was carried out with the primary aim of determining the general
diversity of insects in the area. The survey targeted a diversity of habitats (slopes,
flats, ridges and riparian areas) and vegetation types (grassland, lowland, upland
and cloud forest). A variety of collection techniques (light traps, leaf litter sampling,
pitfall trapping, 1km transect counts, active and opportunistic surveys) was
employed. The general diversity of insects and those species of higher conservation
value (i.e. focal species) were sampled as an indicator of the status or health of the
forest in Emalu.
5.3 Methodology Site selection and habitat considerations A number of key habitat types (shown on Map 3 and Map 4) were surveyed to
maximise the chance of encountering individuals of focal species as well as to
adequately sample the diversity of insects;
Lowland forest areas: targeted specifically to find Fiji’s rare endemic
butterflies Papilio schmeltzi and Hypolimnas inopinata.
Upland forest areas: leaf litter sampling, pitfall traps and light traps on slopes
mainly targeted the general diversity of insects within this specific habitat.
Active searches for the endemic phasmids (stick insects) were also conducted.
Ridges: leaf litter sampling and light traps on ridges targeting the general
diversity of insects found within this specific habitat. A high diversity of
insects (and in particular the focal order Coleoptera) is indicative of intact
Riparian surveys in all vegetation types: These sruveys specifically targeted
butterflies (namely Fiji’s rare endemic butterfly, H. inopinata) and damselflies
(namely those of the endemic genus Nesobasis). These often fly out to open
areas on a fine day in search for sunlight and food, and usually aggregate
along the streams in forested areas. Their presence, abundance and richness
are excellent indicators of forest and stream systems in good health.
Survey methods and sites Nocturnal surveys Nocturnal surveys were conducted using ultra violet (UV) light traps. These were set
up and left to run for 12hour periods from 6pm-6am. Insect specimens were sorted to
Order and then to Family level. Specimens are currently being curated, catalogued
and stored at the South Pacific Regional Herbarium, USP.
Leaf Litter surveys Leaf litter surveys were conducted targeting different habitat types (i.e. river flats,
slopes and ridges) in the lowland and upland vegetation types. 1m
laid at 5m intervals along a 50m transect. Leaf litter from each quadrat was sieved
through 12mm mesh sieves and transferred into Winkler bags (Fig. 24 and Fig. 25).
The Winkler bags were hung out for at least 48 hours to allow drying of the leaf litter.
Insect specimens were stored in ethanol for further sorting and identification.
Pitfall Traps Pitfall traps were set in varous habitat types (i.e. river flats, slopes and ridges) in the
lowland and upland forest areas. Pitfall traps were placed at 5m intervals along a
50m transect within the vegetation plots used by the botany team. Specimens were
collected and transferred into ethanol after 48 hours.
Active sampling- Lepidoptera (butterflies) and Odonates (damselflies) Butterflies and damselflies were also actively sampled in open grassland and
riparian areas along creeks and streams using handheld nets. Voucher specimens
were taken for identification.
1km Transect Count Method 1 km transect counts were conducted for the indicator taxa Hypolimnas inopinata (for
abundance) and Odonata (damselfly) diversity along streams within the Mavuvu
and Waikarakarawa catchments.
Opportunistic Encounters In addition to the survey methods described above, collections were made during the
course of the survey period in response to opportunistic encounters of interesting
Identification and curation Identification of specimens was carried out with the aid of available taxonomic
references for each of the main groups; butterflies and moths (Waterhouse, 1920,
Robinson, 1975, Prasad and Waqa-Sakiti, 2007), dragonflies and damselflies
(Donnelly, 1990, Van Gossum et al., 2006), ants (Folgarait, 1998), beetles (Lawrence
and Britton, 1994) and spiders (McGavin, 2000). The specimens are currently being
curated and catalogued at the South Pacific Regional Herbarium.
5.4 Results and discussion Insect Diversity The results of the insect survey of each catchment are provided in Appendix 10,
Appendix 11 and Appendix 12. A total of 26 Coleopteran (beetle) families were
sampled from within the entire study area. The most abundant taxa sampled
included the beetle families Curculionidae (weevils) and Scolytidae (bark beetles)
and from the Order Hymenoptera, Family Formicidae (ants). Rare beetle families:
Cerambycidae (long-horn beetles), Eucnemidae, Cantharidae, Lathrididae and
Passalidae were also encountered in the surveys. The great diversity of the target
taxa Coleoptera and the Hymenopteran family Formicidae are a good indication that
ecosystem services such as soil processing, decomposition, herbivory, pollination
and seed dispersal within the study area of the lowland, upland and cloud forests in
Emalu are well intact.
Another interesting find was in the order Odonata (i.e. damselflies). The endemic
genus Nesobasiswere abundantly found along tributaries, creeks, stream and rivers
especially for the species Nesobasis angolicolis (Fig. 26), N. erythrops and N. heteroneura.
Their diversity along streams is an excellent indicator of good water quality and
intact status of neighbouring ecosystems. Moths sampled from light traps (nocturnal
surveys) were also significant especially for a few species which are native and
known to be restricted to primary forested areas i.e. Cleora diversa, Agathia pisina,
Pyrrhorachis pyrrhogona, Thallasodes figurate and Mecodina variata. Focal Species Hypolimnas inopinata (Order Lepidoptera) Hypolimnas inopinata (Fig. 27 and Fig. 28) is a rare butterfly, endemic to the Fiji
Islands. It is a montane species and lives in rainforests. It is often found in or near
pristine mountain areas, usually in semi-open areas along streams leading up to the
mountains. Its presence and abundance has also proven to be a very good indicator
of the pristine nature of the rainforest system. H. inopinata was sampled along the
Nasa Creek, adjacent tributaries including the Wairovurovu stream (Tovatova
catchment), Waikutukutu stream (Waikarakarawa catchment) and the Wainasiga
stream and Wainasoba Creek (Mavuvu catchment) suggesting that these catchment
areas in Emalu are intact and pristine (i.e. sites P4, P7, P11 & P16, P26, P30, P31, P32,
P33, P39 & P40 on Map 8). Extent populations have only been located on Viti Levu in
the forests of Navai and Nasoqo (Ra Province) and Waisoi, Wainavadu and Saliadrau
(Namosi Province) and Naikorokoro (Rewa Province). This find is a first record for
the Navosa Province and the study area has a healthy population of this species.
Nysirus (syn. Cotylosoma) spinulosus and Cotylosoma dipneusticum (Order Phasmida) Nysirus spinulosus (Fig. 29), a rare endemic stick insect was first described in 1877,
and previously recorded from Viti Levu, Fiji and only recently (i.e. 2008 & 2009) from
Nakauvadra and Nakorotubu ranges in the Ra Province. Cotylosoma dipneusticum is
another rare endemic stick insect and has been previously recorded from Taveuni
and Viti Levu (Nakorotubu range and Savura Forest Reserve). Both were sampled
from intact upland rainforests near Tovatova Creek, a tributary of the Nasa Creek
and upland forest within the Waikarakarawa catchment. From previous
observations, these two species of stick insects have been known to be closely
associated with such pristine forest systems (P13, 14, 15, 20, 21 on Map 8).
Raiateana knowlesi (Order Hemiptera: Family Cicadidae) Raiateana knowlesi (Fig. 23) is an endemic and rare cicada with a unique life cycle in
which adults emerge every eight years (periodic emergence). The last appearance of
the adults was in 2009 from within this vicinity. It is locally known as nanai and has
been previously recorded from parts of the Serua and Navosa provinces. It is of great
cultural significance to the mataqali Emalu, being one of their ‘totem’ species. The
chiefly daughters of the mataqali are usually accorded the title Rokonai