8.2 Introduction Because of their isolation and relatively recent human occupation, Pacific islands are
especially vulnerable to invasive species, to such an extent that invasive species are
the primary cause of the extinction of island native species (Tye, 2009). Previous
work from numerous authors has focussed on documenting their presence and to
some extent their distribution and abundance within Fiji.
Pernetta and Watling (1978) compiled a list of native and introduced vertebrates
which included reptiles. Since then monitoring of some of the major invasives have
revealed new additions for example, a second mongoose species (Morley et al., 2007).
Some species (rats, mongooses and goats) have also been the target of concerted
eradication efforts on some smaller islands to protect native biodiversity.
In terms of invasive plants, 52 species have been identified as being present in Fiji
(Meyer, 2000). These have been classified under three groups according to their
degree of invasiveness, namely: thirteen dominant invaders, seventeen medium
invaders and 22 potential invaders. For some of these individual species, such as
Mikania micrantha (Macanawai et al. 2010) and Spathodea campanulata (Auld and
Nagatalevu-Seniloli, 2003), research has focussed on their ecology and control.
Invasive species surveys have been a component of wider biodiversity assessments
done in eastern Viti Levu (Morrison and Nawadra, 2009, Morrison et al., 2010,
Morrison, 2003). This is however, the first survey of invasive species in the Emalu
8.3 Methodology Rodent Survey Traps were laid for rodents (rats and mice) on three consecutive nights during the
July 2012 survey in the Nasa River catchment. Traps were baited with roasted
coconuts and positioned in protected spots under hanging boulders, large tree bases
and below fallen logs. The traps were laid in pairs along a transect, according to the
established methodology of Cunningham and Moors (2006). The location of the three
trapping transects is shown in Map 11.
A total of 88 traps were set over the 3-day period, each for one night.The nose-to-tail
length and the weight of captured animals were measured using vernier calipers and
a 1kg spring balance. The species and sex of each captured animal was recorded,
along with an estimate of its age (based on body size).
Other Mammals Opportunistic surveys were conducted to identify the presence of other invasive
mammal species such mongooses, feral cats, cows, dogs, horses, cows and goats. This
included simple visual surveys for individuals, or for traces such as footprints, scat,
and feeding evidence. Information was also obtained from other teams conducting
surveys in other parts of the forest study site. One of the guides used dogs to hunt
feral pigs, and these captures were also recorded.
Invasive Plants A checklist of all invasive plants sighted was compiled during the survey with notes
taken as to their relative abundance and habitat preferences.
8.4 Results A full checklist of all invasive and potentially invasive animals documented during
the survey is provided in Appendix 20.
Rodent Trapping A summary of the trapping results is provided in Appendix 21. The trapping
transects only successfully captured one rat; a juvenile male black rat (Rattus rattus)
that weighed 135g. This rat was captured on the first trap night in Transect 1.
Another individual of the same species (Rattus rattus) was caught opportunistically
by one of the guides without a trap during the survey of the Waikarakarawa River
catchment (Fig. 62). The other two species of rats, Rattus exulans, Rattus norvegicus and the mouse Mus musculus were not caught or observed during this trip, but it is
highly likely that they are present in the area.
Other Animals Pigs
Six pigs were caught with the use of hunting dogs, including one pregnant female.
Descriptions of the pigs caught are given in Appendix 22. Numerous wild pig
wallows were observed in the forest including resting areas such as large tree
hollows. Plantations near the village of Navitilevu showed some evidence of pig
damage, including the uprooting of root crops such as cassava, taro and giant taro.
A juvenile male feral cat (Felis cattus) was caught at the Tovatova base camp. The cat
managed to escape but was seen around the camp several times looking for food. Cat
scat was also found along one of the tracks. No cats or evidence of cats were found
around the Waikarakarawa Creek base camp.
Due to the remoteness of the area and villagers depend greatly on their horses (Equus caballus) for transportation to their plantations or into the forest. Horses were used
for transportation of equipment and supplies to the campsites in the Mavuvu and
Waikarakarawa Creek catchments during this study (Fig. 63). There was no evidence
of the presence of feral horses in the Emalu area.
Although no mongooses were observed within Emalu forest, according to the guides
they are present in and around their villages. Since no individuals were sighted it is
not known if both species of mongoose (Herpestes auropunctatus, Herpestes fuscus) are
present in the area, or just one.
Cane toads (Bufo marinus) were present in the area, and several were caught and
photographed (Fig. 64). They tended to be larger in size than those in urban areas.
Invasive Plants A total of 26 invasive plant species were observed to be present within the area.
Seven of these are currently listed in the top 100 of the World’s Worst Invasive Alien
Species (Lowe et al., 2000). The complete list of all invasive and potentially invasive
plant species found in the Emalu forest is in Appendix 23.
The majority of species and the highest abundances were recorded in proximity to
human habitation and to roads and agricultural land, with a much reduced number
penetrating into the forest. Stream flats and embankments often had multiple
invasive species in close proximity to each other (Fig. 65).
African tulip (Spathodea campanulata)
Spathodea campanulata, although one of the most problematic invasive plant species in
Fiji, was recorded as only two individual trees within the forested area of the Nasa
catchment. One other tree was found near the basecamp in the Waikarakarawa
catchment. In and around villages and roadsides however this species was more
False kava (Piper aduncum)
Large monotypic stands of Piper aduncum were observed outside the Emalu
boundary especially around Navitilevu Village, but within the Emalu site it is
generally restricted to creek banks and disturbed open areas.
Mikania micrantha and Clidemia hirta Although generally occurring at low densities within the forest, these two species are
the most pervasive. Both were observed in higher altitudes, even in closed forest,
above 700m and Clidemia hirta was also recorded in the cloud forest of Mt Vonolevu.
Although both species are capable of growing in the low light conditions beneath
canopy, they are found in much higher density in open and disturbed areas
especially around tracks and stream banks.
Several ornamentals plants introduced deliberately, probably as aboriginal
introductions, were found in the area. These included species such as Brugmansia suaveolens, Musa x paradisiaca, Saccharum edule, Citrus grandis, Bambusa vulgaris, Artocarpus altilis, Citrus limon, Derris malaccensis and Bischofia javanica. Large terraces
were observed within the Nasa and Waikarakarawa river catchments indicating that
at one time the area was intensively farmed.
8.5 Discussion In general the occurrence and abundance of invasive species in the Emalu forest was
associated with proximity to human habitation and to disturbed areas such as tracks
and past or present campsites, fallow land, old village sites, burnt out forest and
A total of 26 invasive plants species were recorded during the trip including some
ornamental plant species that have the potential to become invasive. The majority of
the invasive plant species were found along stream banks, abandoned plantations
and old village sites. Continuous clearing of forest for plantation will certainly
facilitate the incursion of invasive plants into the forest. Disturbance and both human
and animal traffic along tracks will also contribute to the spread of invasive plants
into the interior of the forest.
Compared to invasive plants, invasive animals tend to have a more negative effect on
the native fauna. Even though there is a high possibility that all rat species are
present in the Nasa catchment of Emalu forest, the low trapping rate indicated either
a low density of rodents or the need to improve trapping methods. It is possible that
a grid system (Weihong et al., 1999) would have had a higher trapping success than
the transect system. Trapping data suggest the rat abundances are not very
significant but the presence of the black rat (Rattus rattus) is worrisome as they are a
very agile and frequent climber and therefore can easily access nesting birds.
The presence of feral cats could have also impacted the number of rats caught. Some
farms are located at a significant distance from villages and farmers prefer to build
temporary shelter where they camp and tend to their farms during the week and
return to the village during the weekend. Domestic animals such as cats and dogs
can aggregate around such makeshift camps and in this way become feral. Feral cats
in particular are a major threat to native birdlife.
Other invasive mammals observed during the trip included one juvenile feral cat and
six wild pigs caught by guide Aporosa Maya. Wild pig wallows are common
throughout the Emalu forest and according to the guides the wild pigs are a
significant cause of crop damage. Pigs can also cause damage to native biodiversity,
in particular through destroying seedlings, and contributing to soil erosion.
Although mongooses were not directly observed within the Emalu Forest according
to guides they can be seen around the villages, and it is likely there is incursion into
the forest where, like cats they pose a significant threat to native birdlife.
8.6 Recommendations In terms of the potential for further study of invasive species, the following have
been identified as areas for further action:
Long-term monitoring of Spathodea campanulata. Emalu is an ideal site for
long-term monitoring as this species is considered highly invasive, but
currently has very low abundance in the area. Assessing its spread in relation
to disturbances or other ecological factors over time would elucidate further
information as to its invasiveness potential in Fiji.
A dedicated mongoose-trapping study over a longer time period is needed to
definitively establish if one or both of the two mongoose species in Fiji are
present in the Emalus area.
Some general control measures which would help lessen the damage done by
invasive species, and on which there needs to be awareness-raising in the villages
Control of feral pig populations through de-sexing of alpha male pigs, and a
bounty system for pig hunters.
Limitation of the establishment of new farms to reduce encroachment on
grassland and secondary forest.
CHAPTER 9: ARCHAEOLOGICAL SURVEY Elia Nakoro and Sakiusa Kataiwai
9.1 Summary The land belonging to the mataqaliEmalu is rich in historical and cultural material
remains that have not been documented until this survey. The historical remains are
scattered all throughout the mataqali land, a widespread distribution of elaborate
hilltop and lowland settlement and fortifications some of which are associated with
sophisticated irrigation systems for terrace agriculture.
The general physical setup of settlements depicts various forms of insecurity in the
past, a time of great rivalry and competition. Supporting evidence for this can be
found in some of the structures of the hill fortifications that were encountered.
Constructing on high elevation is a survival strategy whereby communities used
their natural environment and rugged terrain to provide security.
Further evidence to support the notion that the area was densely populated was
given by the series of large intricate irrigation systems discovered during this survey.
The discovery of these elaborate channels suggests larger populations to implement
and maintain this agricultural system. The study of the cultural footprints within the
Emalu study area is vital in understanding why the people of Emalu chose to live in
such remoteness and rugged terrain, as well as their socio-cultural relations and their
responses to altering natural and climatic conditions.
Generally, the archaeological finds during this survey have considerable cultural
value to the local community as well as at the national level. The significance of these
sites can be determined and derived by deconstructing the value of the individual
sites into the following components; aesthetic, symbolic, social, historic, authenticity
and spiritual values.
9.2 Introduction A collaborative archaeological survey was undertaken to outline the cultural
connection the land has to the people of the mataqali Emalu and surrounding
communities with an emphasis on identifying and describing cultural sites of
significance for which there is tangible evidence.
The mataqali Emalu, from the village of Draubuta, possesses a rich historical
background with ancestral ties and links connected to the Emalu forest in which their
generational history and cultural livelihood have been strongly maintained. The
Emalu forest plays a primary role in the cultural identity and history of the mataqali, as their forefathers inhabited the area utilizing its resources and settling extensively
throughout the land as highlighted in this extract by A. Brewster, the Commissioner
of Colo North and East provinces in the late 1800s:
At the time when the ancestral gods were sent forth, Qicatabua went to Qamo, which is near to Serua. Having arrived there, he disliked being on the sea coast, and made up his mind to proceed into the hills in the interior, and so departing thence he went upwards to Vautabu and came to the Wailevu, which he followed down to Navua and then he arrived at Busa Levu. (Note: The great plateau of moorland and swamp in which rises the Wainisavulevu or River of the great falls, a feeder of the Wainimala head of the Rewa. North of the Busa Levu is the valley of the Sigatoka River and south west of the Navua River. This plateau is very nearly in the centre of Viti Levu. A.B.B.) He followed the plateau up until he came to a hill, which he called Emalu and thought that he would build his village there. He did build houses and made clearings for gardens but he took a dislike to it, and arising he went on his way. He followed the Mavuvu River (a tributary of the Navua) until he came to Veinuqa, afterwards arriving at a place called Nasaqaruku. Some of his men who were with him said they would like to remain there and Qicatabua gave them permission to do so, and they built some houses there at Nasaqaruku. Now the name of the clan who remained at that place is Nataritale. Then he went upstream to Toluga, and then some more of his followers said they would stop there. Then he went on to another place called Nasakikisaqora. There stayed his priest, Siliece. Then he went on to Sirowala, and there stopped another of his followers whose name was Vadra. Then he went along the ridge at Naonou and descending followed another ridge, Naraiyawa, then he got down to the river (the Sigatoka River) and got to a place called Nakavu, and there stayed another of his young men by the name Quna. Then he came to another stream, the Wainimosi, which he followed up to a marsh called Nabudoi. Then he ascended a ridge and he said he would rest there, and called the place Emalu… (Brewster, 1921)
9.3 Methodology With the assistance of village guides through collaboration of oral history and
correspondence, areas of interest were identified and located in the study area.
Location data of each site was captured utilising a GPS unit (Garmin GPSmap
76CSx). Site notation was carried out and photographic documentation of sites was
made with a camera (Practica Luxmedia 14-Z4).
9.4 Results During the field survey, a total of 77 sites were documented. Their locations are
shown in Map 12 and Map 13 and a brief description of each site is given in
Appendix 24. Photographs of some of the sites are shown in Fig. 68 - Fig. 97.
9.4.1 Sites of interest Seven of the most interesting sites are described in detail below.
M28-0008 Extending along a flatland situated in lowland forest, the site displayed identified
features approximately within a 20m length zone from the first identified feature to
the furthest. Altogether, three evident house mound features were identified: the first
of which exuded significance in structure as the mound was more elevated than
usual, at a height of about 3m, containing a stone alignment that was demonstrated
along a portion of the mound surface. According to locals, this significant mound
may be the remains of a temple mound or burekalou indicated by the elevation in the
structure. The remaining identified features were two house mounds which were
sufficiently preserved, displaying a vague structure that demonstrated an adequate
appearance of its original formation.
M28-0016 This is a significant site of sentimental value to the mataqali Emalu as it represents an
aspect linked directly to the ancestral relations, background, oral accounts and values
that define and verify the mataqali and its cultural affiliations. The site is demarcated
by a pool, in which flows the Nasa Creek. According to the local oral narrative, the
pool is the final resting place for the endemic cicada, locally known as nanai (Raiateana knowlesi). In the final stages of their life cycle, the cicadas flock to this pool
to perish, an event that occurs every eight years. The nanaiis the traditional
manumanu or animal totem of the mataqali Emalu and through this site
identification, a considerable part of the historical link the mataqali Emalu has with
the land or vanua of Emalu, was established.
M28-0055 The site is quite extensive, covering a large area along the ridgeline with a total of
nine house mounds identified among two platforms that are conjoined, forming a
terrace-like construction over an extensive distance, as the landscape descends
towards the west. The first identified platform is situated at the initial area of
inspection on the east side of the site area. This platform reached a total length of
50m from either end with an identified width of 30m accommodating much of the
identified cultural features belonging to this site.
The first identified mound is rectangular in structure, 8.5 x 6.5m and is highly raised
compared to associated mounds-at least a meter above the ground. Due to its
elevated structure, this mound feature may represent a rank or status associated in
cultural communities. The second identified mound has a diameter of 8m and is
raised at 60cm from the ground displaying a well-preserved structure of its original
form. The third mound is circular and has a diameter of 8m with height raised at
60cm. The fourth mound is the largest mound feature identified in the area and is
centrally located. The mound is rectangular structured, having a length of 9m with a
width of 7.5m. The fifth house mound is identified as circular with a diameter of 7m
and a height of 150cm, well preserved. The sixth identified house mound is situated
on the western edge as the landscape descends to the second identified platform. The
mound is circular structured with a diameter of 7.5m and raised at 50cm, however,
closer inspection revealed that this mound has undergone disturbance through
erosion processes. At the edge of the northern wall of the platform, the team
identified the seventh house mound that was conjoined to the platform unlike the
associated mounds which were situated upon the platform. This mound is circular
structured and extends outwards from the platform. The platform is elevated 40cm
higher than the mound feature, creating a terrace-like structure. The mound has a
diameter of 7m and is raised 60cm from ground level.
As inspection continued towards the west, the team descended onto a second
platform that accommodated two house mounds. The initial identified mound on
this platform is the eighth house mound. This mound is circular structured with a
diameter of 7m and raised at 50cm. Along with this mound is the ninth identified
feature of another circular mound with a diameter of 6.5m and raised at 60cm. Both
mounds are situated on this second platform on the west section of the site area. This
platform extends 12 x 9.5m and is thickly vegetated with little undergrowth.