Figure 26: Status and distribution of macroinvertebrate taxa across all sites
Figure 27 shows the total number of taxa recorded at each sampling station and their
status/distribution shown as a proportion of total taxa richness within each
community. The number of endemic and endemic/native taxa recorded at sampling
stations ranged between seven taxa at Upper Doguru stream (VL3) to 22 at
Nasealevu village stream (VL1). This amounted to 78% and 85% of the total taxa per
sites respectively, highlighting that endemic or native species are the dominant taxa
at all sites. The majority of endemic/native taxa recorded were insects; inclusive of
both qualitative and quantitative collections (35 taxa in total).
Other endemic taxa recorded were the small (<4 mm) spring snails (Fluviopupa spp.).
All the crustaceans (shrimps and prawns) are native but also found throughout the
Indo-Pacific; the exception was the first record of two atyid shrimps (Caridina sp. A
and Caridina sp. B) which have a very high chance of being new to science as these
were compared to shrimp keys from Fiji, PNG, the Philippines and New Caledonia.
There were also two new prawn records (Macrobrachium sp. A and Macrobrachium sp.
Status & distribution
B) which did not match the taxonomic keys stated previously. These specimens
were placed under unknown origin. Two commonly introduced taxa found were the
mosquitoe larvae (Culicidae) and the Thiarid snail Melanoides tuberculata
common introduced mosquitoe larvae (Culicidae) was found at Nasealevu village
(VL1) while M. tuberculata
was found at several sites. These species are common
throughout streams in Fiji and the Melanoides
snail is known to be a hardy species
that can successfully make its way to highland streams.
Figure 27: Status and distribution of taxa across individual sites
A lower number of endemic-endemic/native taxa were observed as part of the
quantitative survey at upper Doguru (VL3) (7 taxa) and Vunidogoloa village stream
(VL13) (8 taxa). This is probably due to the absence of a stable aquatic habitat
(natural riffle-run-pool sequence coupled with stream sides trees providing mass
fibrous roots extended into the channel) for aquatic insect fauna such as mayflies
damselflies, shrimps, whirligig beetles and caddisfly species which generally
contributes to the highest proportion of endemic-endemic/native fauna in Fiji inland
streams. Another possibility could be the removal of stream site trees that would
have contributed to food availability of the macroinvertebrate community. The
streamside trees provide leaf matter and indirectly maintain algal biofilms (prevent
washing away of sediments that would otherwise smother the algal film on
submerged rocks), both of which are food sources for aquatic invertebrates.
Functional Feeding Groups (FFG)
Functional feeding groups include collector-filterers, filter/gatherers, scrapers,
grazers, shredders and predators. An overview of macroinvertebrates and their FFG
categories is presented in Table 5 with the relative proportions of each group at each
site shown in Figure 28.
Table 5: Functional feeding groups for freshwater macroinvertebrate taxa
(minute moss beetle)
Collector-filterers were diverse and ubiquitous across the waterways sampled but
low in relative abundance compared to the scrapers. The collector-filterer feeding
group was represented by nineteen taxa while the scraper functional feeding group
was represented by ten taxa. The scrapers were the most abundant group and made
up between 16% (Nasealevu village site -VL1) to 93% (Waisali village site-VL9) of
total community abundance at stony streambed sites. Scrapers recorded included
mayflies, caddisflies, oligochaetes, moths, beetles and snails. The most abundant
scraper taxon recorded across sites surveyed was the mayfly Pseudocloeon
which grazes on thin biofilms growing on stable in-stream substrates (e.g., cobbles,
boulders, leaf litter). Other widely distributed scrapers included Cloeon
sp. (mayflies), Odontoceridae
(caddisfly), Anisocentropus fijianus
caddisfly) and Nymphula
Figure 28: Proportion of total abundance that each functional feeding group made up at
Filterer/gatherers included caddisfly larvae and dipterans and also represented a
major component of the macroinvertebrate communities recorded. Only three
filterer/gatherers were recorded within this functional feeding group, but they made
up between 2% (Vunidogoloa) to 56% (Doguru village) of total abundance. The
most abundant filterer/gatherer taxon was Abacaria fijiana
(caddisfly), whilst other
widely distributed collector/filterer taxa included A. ruficeps
and Muscidae (stable
fly). Collector-filterers were represented by shrimps, true-flies and beetles and
highly diverse (19 taxa) but of low relative abundance making up between 1%
(Waisali village-VL9) to 25 % (Sorolevu-VL6) of community abundance at the sites.
Predators were represented by caddisflies
, damselflies, dragonflies, water bugs, true-
flies and prawns. The predator functional feeding group was diverse (13 taxa) but of
low relative abundance and made up between 0% (Vunidogoloa-VL13) and 15%
(Waicacuru-VL7) of community abundance at the sites. The shredders were
represented by only five taxa making up between 0 and 4% of total community
abundance across stony streambed sites. Shredders recorded included Trianodes
(caddisfly larvae), beetles and cranefly larvae (Tipula
sp.). The highest
proportion of shredders occurred at upland forested sites (Sorolevu mountain forests
(VL6 and Waisali Forest reserve-VL9), where leaf litter was abundant and retained
within the waterways long enough to be assimilated. The shredders are known to
contribute only a minor component of macroinvertebrate community biomass in Fiji
and tropical Pacific Island riverine systems (Bright, 1982; Resh et al.
, 1990; Haynes,
1999). The low proportions of shredder community is due to absence of stoneflies
from Fiji ecosystems and the nature of leaves (food) entering streams from
surrounding native forests, which tend to be tough with thick cuticles that are
broken down slowly (Haynes, 1999).
Taxa of interest
Certain macroinvertebrate taxa that were recorded during the freshwater
macroinvertebrate surveys may be of potential ecological interest (pictured in
Appendix 17). Some of these taxa, such as Fluviopupa
spp. and Nesobasis
spp. have a
very high chance of being new to science and either catchment endemic or endemic
to Vanua Levu.
These taxa are very good bioindicators for state of streams and the catchment it
drains; ranging from highly sensitive to resilient species. The densities
) of these species reflect the state of streams. These species have
previously being surveyed and found to be varying in abundance in slightly
degraded to intact streams in Viti Levu.
A major finding of this survey was a prawn species, Macrobrachium spinosum
29) which is a new record for Fiji. This species was first discovered in Halmahera,
Indonesia in 2001 (Cai and Ng, 2001) and also recently collected and identified in
Vanuatu (Keith et al.
, 2011). The official documentation of this species is still in
Figure 29: Macrobrachium spinosum
The freshwater macroinvertebrate community (in total 70 taxa) of Vanua Levu
survey areas showed that the endemic/native taxa were the most dominant with
insects making up the majority of the taxa. This is typical of inland tropical riverine
system headwaters. In comparison with other studies in Fiji, Viti Levu catchment
headwaters (by the author), 76 taxa were identified from waterways in the Emalu
area (Navosa highlands), 27 from Wainavadu creek, the headwaters of the Waidina
river and 32 taxa were identified from the Wainibuka river headwaters in the
Nakauvadra range. Waterways in the Emalu area supported much higher taxa
richness than other stream/river headwaters that have been surveyed in Fiji as the
headwaters drained intact catchments. The Vanua Levu survey areas mostly
supported secondary forest except some part of Waisali reserve and Savusa reserve.
A total of 12 macroinvertebrate taxa collected as part of the survey may be of
potential ecological interest (Appendix 17). These include four species of mayfly
nymphs (Ephemeroptera: one Pseudocloeon sp. and two Cloeon spp. and one Caenis
sp.), four species of damselfly nymphs (Odonata: Nesobasis spp.), two species of
caddisfly larvae (Trichoptera: Apsilochorema
one cranefly larvae
sp.) and one snail (Fluviopupa
spp. (<4 mm). These taxa are very
good bioindicators, ranging from highly sensitive to resilient species. Some of them,
for example the Pseudocloeon
sp. and the Cloeon
sp A, are typical of pristine streams
draining intact watersheds. In addition special taxa such as the spring snails
spp.) are very likely to be catchment endemic or area endemic species.
Ten species of spring snails are already known to be endemic to Fiji, have restricted
distribution and are usually catchment endemic, inhabiting springs and small creeks
or riffles (Haase et al.
They almost exclusively live in springs and in the
headwater of streams. The presence of these spring snails is indicative of very clean
water. These snails are specialists with very low ecological amplitude; reacting to the
slightest difference in environmental conditions. They are mostly threatened by
human activities that lead to sedimentation and eutrophication such as logging,
mining, intensive agriculture, forest burning and removal of riparian vegetation
which results in the springs snail density decreasing or the population disappearing
altogether (Great Basin EF, 2012)
The damselfly nymphs collected (Nesobasis spp
. W, X, Y, Z) were morphologically
different from those commonly found in Viti Levu streams and have a high chance
of being endemic to Vanua Levu. Further scientific research is needed to confirm
this. Additionally these larval stages will need to be matched to an adult stage before
it can be confirmed if they are a new species or not. In addition this survey
documented for the first time two atyid shrimps (Caridina
sp. A and Caridina
which have a very high chance of being new to science as these were compared to
shrimp keys from Fiji
, PNG, Philippines, New Caledonia, Japan, Malaysia
and Indo-West Pacific. There were also two new records of prawn specimens
sp. A and Macrobrachium
sp. B). These specimens seem to have
partial resemblance to Macrobrachium placidulum
(Holthuis, 1952; Chace, 1997; Short,
2004; Cai and Shokita, 2006; Cai et al.
, 2006; Cai et al.
Another interesting observation during the survey was the absence of the fingernet
caddisflies of the genus Chimarra. These caddisfly larvae (Figure 30) have been
observed in slightly disturbed to intact streams in Viti Levu and has been highly
abundant (average= 66 individual/m²) in intact (primary forested) catchment such as
Emalu in Navosa highlands. Their absence in the areas surveyed could be due to the
species not being able to reach the areas as the water quality recorded supported
their usual habitat water physicochemistry.
Figure 30: Fingernet caddisfly Chimarra sp. (Philopotamodae)
Invasive alien species are described in the context of the Convention on Biological
Diversity as "alien species whose introduction and/or spread threaten biological
diversity" (CBD, 2002). The Millennium Ecosystem Assessment (UNEP, 2005)
confirms that invasive alien species have been a significant driver of biodiversity loss
over the last century, and forecasts that this trend will continue or increase in all
biomes across the globe. Island ecosystems like those in the Pacific are particular
vulnerable to the impact of invasive alien species (CBD, 2003).
The list of plant invasives in Fiji (Meyer, 2000) is currently composed of 52 species,
classified under three groups according to their degree of invasiveness, namely: 13
dominant invaders, 17 medium invaders and 22 potential invaders).
Pernetta and Watling (1978) compiled a list of introduced vertebrates in Fiji which
includes most of the globally common invasive species such as rats, mongooses and
the Indian mynah. Fiji has, however, successfully prevented the entry of the giant
African snail and the brown tree snake, which have had devastating impacts on
other islands in the Pacific (Sherley, 2000).
Invasive species management in Fiji has focused for the most part on control
methods; physical, biological and chemical. A few eradication programmes have
been implemented on small islands, for example Vatu-i-Ra, where the Pacific rat
) was successfully eradicated to protect seabirds (Seniloli et al.
Whilst eradication programmes are feasible for small isolated islands, it is not a
realistic approach for widespread plant and animal invasives in larger areas on the
This invasive species survey was conducted as part of a rapid biodiversity
assessment of sites in inland Vanua Levu that are being considered for designation
as protected areas.
A checklist of invasive plant species was compiled based on observations at all areas
surveyed, which included the Mt Delaikoro summit road, the Navakuro to Mt
Sorolevu road, and part of the Waisali reserve. A more detailed assessment of
invasive plant species was made on the Navakuro to Mt Sorolevu road. This logging
road has been made within the last 10 years and ascends close to the peak of Mt
Sorolevu, Vanua Levu’s highest mountain. The survey team followed this road as
close as possible to the summit of Mt Sorolevu, making records of invasive plant
species encountered along the way that were visible from the road. These points
were georeferenced and aligned to the corresponding elevation profile of the track.
A checklist of the invasive animal species was compiled based on reports from the
vertebrate fauna specialists. Both direct sightings as well as indirect observations
(scat, chewing marks etc.) were recorded. Where reports were based on indirect
observations identification to species level could not be reliably made, the list
indicates the possible species (“cf.”). Invertebrate invasive species (such as
agricultural insect pests) were not recorded.
Invasive plant species were readily observed in all areas surveyed, and as
anticipated were most abundant in disturbed habitats such as roads, tracks,
waterways, agricultural areas and near human habitation. The checklist comprised
21 species (Appendix 18), including most of the dominant and moderate invaders
listed by Meyer (2000).
The distribution of some of the most common invasive species along the altitudinal
gradient on Mt Sorolevu is shown in Figure 31. A greater variety of invasive species
were observed in the lowland areas nearer to human habitation and agricultural
The giant reed, Arundo donax
, was very common sight, not only along the many
streams and rivulets on the Mt Sorolevu track (Figure 32), but also along the track
itself. In areas where there was still or slow-moving water, such as ponds and
ditches, the presence of water hyacinth (Eichornia crassipes
) was noted (Figure 33).
Some species, such as Mimosa invisa
, and Stachytarpheta urticifolia
were very common
along most of the track, forming thickets or large stands of groundcover along the
was one of the most highly visible invasive species and dominated,
not just as a blanketing climber over large shrubs and trees, but also spreading out
over the road itself (Figure 34). Clidemia hirta
, a very common shrub species, was less
noticeable at the lower altitudes but became more visible as Merremia
dominant at higher altitudes (Figure 35).
, classified as potentially invasive (Meyer, 2000), was recorded in
great abundance along most of the track, even at higher altitudes. Since it was
flowering, the African tulip was visible at long distances, and was observed not just
near the roadside but also penetrating into forest. The individual recorded at the
highest altitude was at 500 m, over 5 km away from the village of Navakuro (Figure
In areas of intact forest (such as at the Waisali reserve), the only invasive species
generally observed were Clidemia hirta
and the climber Mikania micrantha