Content uploaded by Simon Hodge
Author content
All content in this area was uploaded by Simon Hodge on Apr 05, 2017
Content may be subject to copyright.
A comparison of macro-moth assemblages across three types of lowland forest
in Fiji
Si t e r i ti k o c a 1*, Si m o n Ho d g e 2,3, ma r i k a tu i w a w a 1, Sa r a H Pe n e 1, Jo H n cl a y t o n 4, gi l i a n n e Br o d i e 2
1Institute of Applied Sciences, University of the South Pacific, Suva, Fiji.
2School of Biological & Chemical Sciences, University of the South Pacific, Suva, Fiji.
3Faculty of Agriculture and Life Sciences, Lincoln University, Canterbury, New Zealand.
415 Whinny Brae, Broughty Ferry, Dundee, United Kingdom.
stikoca@gmail.com
Abstract. Although many studies have shown a relationship between forest type and quality on
resident lepidopteran assemblages, there appears to be an absence of such studies in Pacific island
countries. This study compared nocturnal macro-moth assemblages in a native rainforest, mixed
forest and a plantation of exotic trees (mahogany) near Suva, Fiji Islands. Four nightly surveys (4 h
from dusk) were performed in each forest type using a mercury vapour light. A total of 491 macro-
moths belonging to 92 species in nine families were collected. No statistically significant differences
in abundance, species richness and various diversity indices were observed across the different forest
types. Endemic species were collected in all three locations, although significantly more endemic
individuals were collected in the native forest compared to the exotic plantation. When examining
species composition, ‘analysis of similarity’ (ANOSIM) and non-metric multidimensional scaling
suggested that the faunas observed in the mixed forest and the exotic forest might be different,
with the fauna in the native forest intermediate between these two. Although we found no major
differences in the moth assemblages in these three sites, the results collected provide baseline data
for future studies and comparisons with other localities. The results also reinforce previous findings
which demonstrate that exotic plantations and semi-degraded forests may still provide useful refuges
for endemic insect species of conservation value.
Key words: Ecological monitoring, indicator species, Lepidoptera, South Pacific.
In t r o d u c t I o n
Herbivorous insects, such as Lepidoptera, have
a close functional relationship with the vegetation
they utilize, and hence often respond sensitively to
deforestation and subsequent forest regeneration
(Hilt 2005). The distribution and quality of plant
communities, in terms of host plants for larvae and
floral resources for adult insects, can have immediate
effects on the abundance and composition of the co-
occurring lepidopteran fauna (Robinson 1975; Fisher
2011). Because of their sensitivity to habitat quality,
nocturnal moths have long been considered valuable
indicators for monitoring the ecological effects of
forest change and for providing a surrogate measure
of forest ‘health’ (Willottf 1999; Schulze et al. 2000;
Beck et al. 2002; Axmacher et al. 2004, Summerville
et al. 2004).
The destruction of native forests remains a
serious threat to endemic terrestrial fauna and flora
of Pacific islands. Since 1967 an estimated 19%
(140,000 ha) of Fiji’s forests have been lost, principally
due to conversion to commercial agriculture, rural
development projects, spread of small settlements
and the development of urban growth (Evenhuis &
Bickel 2005; Prasad 2010). A comprehensive study
of the Macro-Lepidoptera of Fiji was produced by
Robinson (1975) and a later checklist provided by
Evenhuis (2013), with new species records for Fiji
regularly being reported (e.g. Clayton 2002, 2008,
2010, 2011, 2015; Tikoca et al. 2016a). However,
* Corresponding author
Volume 49: 69-79
iSSn 0022-4324 (Print)
iSSn 2156-5457 (o n l i n e )
The Journal
of Research
on the Lepidoptera
tHe lePidoP tera reSea rcH Foundation, 2 aP r i l 2017
Received: 25 January 2016
Accepted: 10 February 2017
Copyright: This work is licensed under the Creative Commons
Attribution-NonCommercial-NoDerivs 3.0 Unported License. To
view a copy of this license, visit http://creativecommons.org/
licenses/by-nc-nd/3.0/ or send a letter to Creative Commons,
171 Second Street, Suite 300, San Francisco, California, 94105,
USA.
J. Res.Lepid.70
as far as we can ascertain, in South Pacific islands
there have been few, if any, comparisons of moth
assemblages among forest types, nor any work
utilizing macro-moth communities as indicators of
habitat change, habitat degradation or restoration
success.
Fiji still contains a wide range of forest types,
from pristine native cloud and rain forests to highly
managed plantations containing exotic tree species
(Prasad 2010; Sue 2010). The aim of this study was
to examine the assemblages of nocturnal macro-
moths across three secondary lowland forests near
Suva, Viti Levu: a native forest, an exotic plantation
and a mixed forest containing regeneration of native
species after commercial use. The relationships
between forest type and macro-moth abundance,
species richness, patterns of endemism, and species
composition were assessed. In addition, two
moth-based ‘Forest Quality Indices’, as proposed
by Kitching et al. (2000), were evaluated for their
potential and applicability as conservation tools in
a Pacific island setting.
Ma t e r I a l s & M e t h o d s
Study sites
Macro-moth assemblages were compared in
three secondary lowland forest types, namely: (i)
native forest (Savura), (ii) exotic plantation forest
(Mt. Korobaba) and (iii) mixed forest (Colo-i-Suva).
The three sites are located on the south-eastern part
of Fiji’s largest Island, Viti Levu, at elevations < 300
m above sea level (a.s.l). Savura (-18.070, 178.448)
consists of 397 ha of native forest located in the
province of Naitasiri, 14 km west of Nausori. The
site was established as a forest reserve in 1963 and has
not been logged since that time. A total of 587 plant
species have been recorded from the area, of which
560 (96%) were considered native to Fiji, with 29%
considered endemic. The dominant plant families
present include Myristicaceae, Cyatheaceae and
Clusiaceae (Keppel et al. 2005).
Mahogany (Swietenia macrophylla) plantations
cover a considerable area of the south-eastern parts
of Viti Levu (Tuiwawa et al. 2013). Mt. Korobaba is
located 8 km west of Suva (-18.097, 178.388), and was
cleared and systematically planted and managed for
the mahogany timber trade from the late 1950’s to
1970’s (Kirkpatrick & Hassall 1985). The sampling
sites within Mt. Korobaba were in elevations <200
m a.s.l., in areas which contained mature unlogged
mahogany forest with a 90% relative dominance of
mahogany.
Colo-i-Suva is located in the province of Naitasiri,
7 km north-west of Suva, (-18.328, 178.274). Sampling
sites were within a two and a half square kilometre
of tropical rainforest that was set up as a reserve in
1964 (soon after mahogany stands were planted in
the area) and established as the Colo-i-Suva forest
park in 1970 (Paine 1991). The vegetation at Colo-
i-Suva contains a mixture of both exotic timber
species and native species at various growth stages in
the understorey, and approximately 70 native plant
species have been recorded from the site (Tuiwawa
& Keppel 2013).
Light trapping and insect identification
Moths were collected using a manual light
trapping system, consisting of a 125W mercury vapour
lamp powered by a portable generator and a 2 x 2 m
white sheet positioned in front of the light source
which was spread out and secured onto nearby trees
or branches. All moths that landed on the white sheet
were collected and placed into jars charged with ethyl
acetate as a killing agent. Each sample consisted of
the individuals collected in one night in the four
hours after dusk.
Previous research comparing light trap efficiency at
one of the field sites (Colo-i-Suva) indicated that four
nights of sampling would obtain a good proportion (c.
90%) of the estimated moth species present (Tikoca
2016c). Therefore, sampling was carried out on four
nights within each site, performed over six nights in
October 2012 with two sites being randomly selected
for sampling on each occasion.
Specimens were assigned to species level by
reference to keys, images and nomenclature provided
by Robinson (1975), Holloway (1998), Clayton
(2004), CSIRO (2011) and Evenhuis (2013), with
family designations as revised by Zahiri et al. (2011).
Species were classified as being ‘endemic’ if they have
only been recorded from Fiji. This classification is
based on taxa at the species level, and no account
is taken of possible endemic sub-species. We accept
that any designation of a species as endemic has to
be made with some reservations given the incomplete
knowledge of occurrences in different island groups
in the South Pacific, and the relatively unstable
taxonomy in some groups. Individuals of the large
genus Cleora were not identified to species level and
‘Cleora sp.’ was treated as a single taxon. Of the ten
species belonging to the genus Cleora in Fiji only C.
injectaria a nd C. samoana are not considered endemic.
However, as neither of these species was recorded in
this study, ‘Cleora sp.’ was considered as an endemic
taxonomic unit in our analyses.
71
Measures of community structure
For each sample, macro-moth abundance (N),
species richness (S) and rate of endemism were
obtained. Species diversity was defined using the
Shannon-Weiner index [H’ = -Σpi.log(pi)] and
ev enne ss ind ex [ J’= H’/ l o g (s)], where pi = proportion
of individuals consisting of the ith species.
Kitching et al. (2000) proposed an index to
measure forest quality in terms of the abundance
of certain moth families, calculated as: 100 ×
[Geometridae / (Arctiidae + Noctuidae)]. However,
due to taxonomic revisions at family level, Arctiidae
and some Noctuidae are now placed in the family
Erebidae (Zahiri et al. 2011). Therefore we calculated
the Forest Quality Index (FQI) proposed by Kitching
(2000) using previous taxonomy, and then a second
FQI (‘Tikoca FQI’) based on current family-wise
designations calculated as: 100 × [Geometridae /
(Erebidae + Noctuidae)]. Each FQI was calculated
for each of the twelve samples separately and also
based on the overall catch from each forest using
pooled data.
Statistical analysis
All statistical analyses were performed using
Minitab (v17, Minitab Inc, USA) and Community
Analysis Package (v4, Pisces Conservation Ltd, UK).
Forest types were compared using a one way analysis
of variance (ANOVA) test with post-hoc Tukey’s tests
used for pairwise comparisons after a significant
result. Abundance of moths and species richness
data were log10 transformed prior to analysis to help
reduce the effects of the relationship between mean
and variance (Tikoca 2016c), and prior to ANOVA
being performed, homogeneity of variance was
verified for all variables examined using Levene’s test.
Due to a prevalence of zero scores, the abundances
of each family at the three sites were compared using
a non-parametric Kruskal-Wallis test.
The species-sample matrix obtained was extremely
sparse, with 79% of cells equal to zero. Legendre
and Gallagher (2001) indicated that, with sparse
matrices such as this, principle components analysis
on raw data might be inappropriate because samples
that actually contain no common species may appear
similar due to a prevalence of shared absences.
Therefore we compared the compositions of the
moth faunas among the three forest types using
non-metric multidimensional scaling (NMDS) and
‘analysis of similarity’ (ANOSIM) using square root
transformed data (Community Analysis Package;
Henderson & Seaby 2008). For the NMDS, a
Bray-Curtis similarity measure was employed and
principal components analysis used to give initial
positions of the samples. The ANOSIM procedure
examines whether samples from within pre-defined
groups are more similar in composition than samples
from different groups, again using a Bray-Curtis
Individuals Species
Savura Colo-i-Suva Mt Korobaba Savura Colo-i-Suva Mt Korobaba
Family Native Mixed Exotic Total Native Mixed Exotic Total
Cossidae 1 0 0 1 1 0 0 1
Erebidae 99 50 63 212 20 19 21 32
Geometridae 81 39 30 150 9 10 11 21
Limacodidae 7 6 3 16 4 4 2 7
Noctuidae 22 29 13 64 10 15 16 19
Nolidae 2 7 6 15 1 3 4 5
Sphingidae 3 1 1 5 2 2 1 3
Thyrididae 5 3 7 15 2 2 2 3
Uranidae 4 4 5 13 1 1 1 1
Grand Total 224 139 128 491 50 55 48 92
No. of singletons 21 27 24 34
Proportion of
Singletons (%) 42.0 49.1 50.0 37.0
Table 1. Abundance of individuals and number of species in macro-moth families collected at three forests near Suva, Viti Levu,
Fiji, produced by four hours trapping on four separate nights using an MV light.
49: 69-79, 2016
J. Res.Lepid.72
measure of similarity. The test statistic produced, R,
ranges from -1 to +1, with +1 indicating all the most
similar samples are within groups, and -1 indicating
that all the most similar samples are never in the
same group. Both of these multivariate procedures
were performed three times: on a matrix including
the abundance of all species, a matrix including only
species with total abundance ≥ 3, and on a sample-
by-family matrix.
results
Moth abundance and diversity
A total of 491 macro-moth individuals belonging
to 9 families and 92 species were collected. Three
families - Erebidae, Geometridae and Noctuidae
- made up the majority of individuals (87%) and
species (78%) collected (Table 1; Appendix).
The total number of species collected at each site
was similar: 55 species were recorded at Colo-i-Suva,
50 species at Savura and 48 species at Mt. Korobaba.
There were no significant differences in abundance
among the three sites for any of the families recorded
(Kruskal-Wallis tests, P > 0.180 in all cases) (Table 1).
There were also no statistically significant differences
among the three forests in terms of total moth
abundance, species richness, species diversity and
evenness of moth assemblages (Table 2).
A considerable proportion (c. 35%) of the total
catch in each forest type consisted of endemic
species, although there were no statistically significant
differences among the three forests in terms of
numbers of species or proportions of endemic species
in the individual collections (Table 2). However,
there were clear differences in the abundances of
endemic species among the three forests. The exotic
forest at Mt Korobaba had significantly fewer endemic
individuals than the native forest at Savura, with
the mixed forest at Colo-i-Suva being intermediate
between these two extremes (Table 2).
Comparison of macro-moth assemblage
composition
When comparing the three forests in a pairwise
fashion, the ANOSIM procedure identified no
significant differences among the moth faunas in
the three locations when considering family-level
identifications (R < -0.10; P > 0.35). However, the
ANOSIM procedure indicated there was moderate
evidence that the moth assemblages in the mixed
and exotic forests exhibited some differences when
considering all species (R = 0.19; P = 0.07) and when
considering only those species with abundances ≥ 3
(R = 0.18; P = 0.10). The findings from the ANOSIM
were supported by the results of the NMDS (Figure
1), where no obvious clustering of the samples from
the three forests occurred when the analysis was based
Site Savura Colo-i-Suva Mt Korobaba
Forest type Native Mixed Exotic F2,11 P
Abundance (N) 56.0 ± 13.8 34.8 ± 10.4 32.0 ± 9.9 0.86*0.457
Species richness (S) 20.5 ± 4.2 20.8 ± 6.0 17.2 ± 4.3 0.13*0.881
Species diversity (H’) 2.36 ± 0.23 2.61 ± 0.35 2.54 ± 0.25 0.21 0.815
Evenness (J’) 0.80 ± 0.06 0.91 ± 0.01 0.93 ± 0.02 3.39 0.080
Endemic abundance (EN) 37.2 ± 8.0 a14.5 ± 4.4 ab 10.0 ± 1.4 b5.06*0.034
Endemic abundance (EN %) 69.4 ± 6.5 42.5 ± 6.6 40.5 ± 10.4 4.03 0.056
Endemic richness (ES) 7.8 ± 1.1 7.5 ± 2.4 5.2 ± 0.9 0.36*0.708
Endemic richness (ES %) 40.2 ± 6.1 33.5 ± 7.1 35.2 ± 8.5 0.23 0.799
Site total Kitching FQI 118.6 57.4 40.5 - -
Site total Tikoca FQI 68.6 49.4 39.5 - -
Mean sample Kitching FQI 156.8 ± 64.2 54.9 ± 16.0 95.7 ± 68.1 0.87 0.450
Mean sample Tikoca FQI 65.9 ± 6.2 46.3 ± 15.6 94.3 ± 68.6 0.95 0.423
Table 2. Abundance and species richness of macro-moths and levels of endemism at three forests near Suva, Viti Levu, Fiji, produced
by four hours light trapping (mean ± se; n = 4). Samples with different letter codes (a or b) were separated by Tukey test at P < 0.05.
* - ANOVA performed on log10 transformed data
73
on families (Figure 3c). However, some separation
of the groups was observed along NMDS Axis 1 when
the analysis was based on species-level identifications,
especially between the moth samples taken from the
mixed and the exotic forests (Figure 1a,b).
Twenty percent of the total species recorded in
this study were found in all three forest types (Figure
2; Appendix). In addition, a further 27% of species
were shared by at least two of the sites (Figure 3).
However, this indicates that over half of the species
recorded (53%) were only found at a single site,
and thus may have potential as indicators of certain
habitat types. Unfortunately 34 of these 49 site-
unique species were represented by singletons and
thus could not be considered as indicator species.
Similarly, a further six of the site-unique species
were only recorded in a single night’s trapping and
thus exhibited no consistency of capture within that
location. Indeed, no species were found that were
unique to a single site and occurred in all of the
samples taken from that site.
However, based on the results of the NMDS
analysis, there appeared some tendency for the
abundances of Ericaea leichardtii and Ericaea inangulata
(Erebidae), and less so Sasunaga oenistis (Noctuidae)
and Rusicada nigritasis (Erebidae), to be correlated
(rank correlation) with the NMDS Axis 1 score,
indicating a positive association with the exotic forest.
Also, by examining the raw data, it was found that two
endemic taxa, Cleora sp and Calliteara fidjiensis, made
up approximately 50% of the individuals in the native
forest at Savura, and so it might be speculated that
a high abundance of these taxa may indicate high
quality forest in a Fijian setting.
Forest quality indices
The values of the FQI proposed by Kitching et al.
(2000) and the alternative ‘Tikoca FQI’ proposed
here were highly correlated across the 12 moth
samples (rs = 0.944, P < 0.001). When considering
the FQIs based on the total catch at each site, both
FQIs exhibited a similar pattern: the FQIs for the
native forest at Savura were considerabley higher
that that seen in the exotic forest at Mt Korobaba,
with the mixed forest at Colo-i-Suva intermediate
between these two (Table 2).
However, there were some discrepancies, and a
difference in the ranking of sites, when using FQIs
based on the total catch and those based on the
sample means (Table 2). These anomolies resulted
primarly because some samples consisting of small
numbers of individuals (e.g. 10, 14 and 16) produced
some extreme FQI values, such as an FQI of 300 for
one sample from the exotic forest at Mt Korobaba.
This value was given equal weighting when the mean
FQI values for Mt Korobaba was calculated (95.7 &
94 .3), but t he smal l nu mber of i ndividu als me ant the
effect of this sample on the pooled FQI estimates
(40.5 & 39.5) was much reduced.
dIscussIon
Moth abundance, diversity and composition
Many previous studies have examined macro-
moth community patterns in forests as a response
to the state of the forests studied, including logging
and recovery regimes (Fisher 2011; Hilt 2005;
Willott 1999), natural disturbance (Chaundy 1999),
reforestation age (Taki et al. 2010), native and
agricultural habitats (e.g. Ricketts et al. 2001), and
plantations (Hawes et al. 2009). Often there are clear
negative relationships between moth abundance and
species richness with increasing habitat degradation
and forest disturbance (Hawes et al., 2009; Ricketts
et al. 2001). However, only slight differences were
found in the abundance and species diversity of the
moth faunas in the three different Fijian forests
investigated here. The lack of distinctiveness
may be due to the forests we examined all being
geographically close to each other and in recent
years they have all suffered similarly low amounts of
disturbance in terms of logging management. Also,
this study was performed over a relatively narrow
time p eriod . Rece nt wor k has id ent ified con sider abl e
seasonal fluctuations in Fijian moth populations,
and it is possible that clearer differences do occur
between forest types at different times of the year
(Tikoca et al 2016b).
The proportion of singletons was high for each
location (> 40%), and for the total catch (37%), which
suggests that, even though the moth sampling regime
was based on previous appraisals of suitable sampling
effort (Tikoca 2016c), the number of samples used
per site was insufficient in this case (Coddington
et al. 2009). However, given the various summary
statistics used to compare abundance and diversity
of the moth assemblages in each forest type, we do
not believe that the lack of statistically significant
results occurred due to a lack of statistical power,
and was more a reflection of the small differences
that actually occurred between sites.
Although there are few data on host plant
specificity for Fijian moth species, we can assume
that endemic moth species primarily utilize native
host plants, and that diversity of endemic plants
should be associated with high incidence of endemic
49: 69-79, 2016
J. Res.Lepid.74
Figure 1. Scatter plots of NMDS Axis 2 versus NMDS Axis 1 scores of twelve macro-moth samples. NMDS was performed on
square-root transformed data of moths collected during one nights sampling in a native (Na; Savura), a mixed (Mi; Cool-i-Suva)
and an exotic (Ex; Mt Korobaba) forest on Viti Levu, Fiji Islands: (a) based on NMDS on data for all species, (b) based on NMDS
on data for species represented by ≥ 3 individuals and (c) based on the abundance of each family in each sample.
moths (Miller & Hammond 2007). This turned out
to be the case: Savura had much higher abundance
(threefold) of endemic individuals than the other
two forests studied, and Savura is also the most
floristically diverse forest with the highest degree of
indigenous plant species and composition (Keppel
et al. 2005). Conspicuous numbers of endemic
moths were also recorded at Colo-i-Suva and Mt.
Korobaba which suggests that these moth species
are finding adequate resources in these habitats.
The value of plantations of exotic tree species for
providing habitats for native invertebrates has been
observed previously (e.g. Pawson et al. 2010, 2011),
but until more is known of the specific life history
requirements of Fijian endemic moth species no
explicit management actions can be taken in order
to increase their numbers at these, and other,
locations.
In ter ms of species composition, only 20%
of species were found at all three sites, and the
multivariate methods suggested there could be
dissimilarities in the compositions of the moth
assemblages at the different sites. However, this
separation was not between the exotic and native
forest as might have been predicted, and thus
did not indicate any gradient of change in moth
assemblages from high quality native forest to low
quality exotic plantation, with mixed forest having
a fauna intermediate between these two. Although
the data obtained suggested that a few species might
show some weak associations with certain forest
types, we could not identify any strong candidates
as indicators of habitat quality or class.
The moth-based Forest Quality Index proposed
by Kitching et al. (2000) utilized family-level
ident ification s, and in upland rainfore st s in
Queensland, Australia, FQI values of 98.7 for
uncleared remnants, 68.2 for regrowth remnants
and 18.6 for ‘scramberland’ remnants were obtained.
The values we obtained using the pooled catches
for each site were of similar magnitude: 118.6 for
the native forest at Savura, 57.4 for the mixed forest
75
at Colo-i-Suva and 40.5 for the exotic plantation at
Mt. Korobaba. The ranking of the sites by these
pooled-data FQI scores appeared sensible, in that the
FQI ranks matched the sequence of habitat quality
we had notionally pre-determined: exotic → mixed
→ native forest. However, the values we obtained
when using the sample averages suggest that these
indices can produce some highly anomalous values
when sample sizes are small, and might only be of
value when a large number of individuals (> 100)
has been recorded at each location.
Changes of macro-moth community composition
with forest structure over time
The abundance and diversity of macro-moths
in Colo-i-Suva revealed an interesting development
in this community over the last 50 years or so. One
of Robinson’s (1975) primary collecting sites in the
1960s and 1970s was Colo-i-Suva, where he identified
the dominant species as Nola fijiensis (Nolidae),
Progonia micrastis (Noctuidae), and Hypenagonnia spp.
(Noctuidae). In the current study, none of these
species were encountered at the Colo-i-Suva site at all
(although additional collections made by the authors
at Colo-i-Suva have since revealed a few occurrences
of Hypenagonnia spp. but neither of the other two
species). The dominant taxa in our collections from
Colo-i-Suva were Cleora sp. (Geometridae), Calliteara
fidjiensis (Lymantridae) and Spodoptera mauritia
(Noct uid ae) (Appendix). T he Colo -i-Suva area w as
cleared to aid the mahogany trade from 1950–1960
(Tuiwawa et al. 2012), which means Robinson’s
collecting was conducted during the late 1960s/70s
on a disturbed forest system, and consequently Nola
fijiensis was described by Robinson (1975) as a species
typical of a disturbed lowland forest. The forest at
Colo-i-Suva has changed considerably over the last
50 years: the understory of largely native forest plant
species has been allowed to develop and remnant
mahogany plants have matured. Robinson (1975)
suggested that moth faunal composition develops
with forest structure and age, and the absence of
N. fijiensis in Colo-i-Suva during the current study
suggests that the recovery of the forest may have
caused a shift in macro-moth species composition
and the loss of this species.
conclusIons
Although this study showed that the three forest
types sustain similar macro-moth communities in
terms of abundance and species richness, the major
difference identified was the ability of the native
forest to sustain higher populations of endemic
species than the exotic forest. Over two thirds of
the total macro-moths collected in the native forest
belonged to endemic species, and this relatively
large population of endemic moths in the native
forest may be explained by the high frequency of
indigenous plant species that presumably support
them, although this hypothesis requires further
research on host plant usage for confirmation.
With a caveat that sample numbers must be large
to avoid anomalous values, the use of forest quality
indices appears to have some potential as a tool to
compare the quality of Fijian forests in terms of their
moth assemblages. Endemic moth species were also
found in considerable numbers in both the mixed
and exotic forest, and management of these sites to
enhance their ability to sustain native invertebrate
species should be further explored.
ac k n o w l e d g e M e n t s
This work was funded with grants from The University of the
South Pacific. Site access and collection permission for the work
in Colo-i-Suva Forest Reserve was granted by The Department of
Fisheries and Forestry and the Water Authority of Fiji. We gratefully
acknowledge staff of the South Pacific Regional Herbarium for
their logistical and fieldwork assistance, in particular Alifereti
Naikatini, Hilda Waqa Sakiti, Tokasaya Cakacaka, Apaitia Liga,
Manoa Maiwaqa, and Ratu Filimoni Rokotunaceva. Konrad
Fiedler provided insightful comments on an earlier draft of this
paper and advice on appropriate multivariate statistics.
Figure 2. Venn diagram illustrating the number of moth
species (total = 92) collected in a native (Savura), a mixed
(Cool-i-Suva) and an exotic (Mt Korobaba) forest on Viti
Levu, Fiji Islands. The numbers are based on total catches
obtained on four sample nights in each forest using a MV-
lamp combined with hand collecting of specimens.
49: 69-79, 2016
J. Res.Lepid.76
lI t e r a t u r e c I t e d
ax m a c H e r , J.c., g. Ho lt m a n n , l. Sc H e u e r m a n n , g. Br e H m , k. mü l l e r -
Ho H e n S t e i n & k. Fi e d l e r . 2004. Diversity of geometrid moths
(Lepidoptera: Geometridae) along an Afrotropical elevational
rainforest transect. Diversity and Distributions 10: 293-302.
Be c k , J., c.H. Sc H u l z e , k.e. li n S e n m a i r & k. Fi e d l e r . 2002. From
forest to farmland: diversity of geometrid moths along two
habitat gradients on Borneo. Journal of Tropical Ecology 18:
33-51.
cH a u n d y , r.F.c. 1999. Moth diversity in young Jack Pine-deciduous
forests after disturbance by wildfire or clear-cutting. MSC thesis.
University of Toronto. 95 p.
clayton, J. 2002. A new species of Uraniidae (Lepidoptera)
and a new species of Limacodidae (Lepidoptera) from Fiji.
Entomologists’ Record and Journal of Variation 114: 193-197.
cl a y t o n , J. 2004. Moths in Fiji. Retrieved 2nd February, 2015,
from http://www.usp.ac.fj/index.php?id=8504
cl a y t o n , J . 2008. Notes on the genus Palpita Hübner (Lepidoptera,
Pyralidae, Pyraustinae) from Fiji, with descriptions of two new
species. Entomologist’s Record and Journal of Variation 120:
199-203.
cl a y t o n , J. 2010. Two new species of Noctuidae (Lepidoptera),
subfamilies Hypenodinae and Hypeninae, from Fiji.
Entomologists’ Record and Journal of Variation 122: 219-223.
cl a y t o n , J. 2011. Two new species of Noctuidae (Lepidoptera),
subfamily Herminiinae, from Fiji. Entomologists’ Record and
Journal of Variation 123: 185-189.
cl a y t o n , J.a. 2015. The Lophocoleus group of genera (Lepidoptera:
Erebidae: Herminiinae) in Fiji, with the description of a new
genus and species. Australian Entomologist 42: 1-12.
coddington, J.a., i. agnarSSon, J.a. miller, m. kuntner & g.
Ho r m i g a . 2009. Undersampling bias: the null hypothesis for
singleton species in tropical arthropod surveys. Journal of
Animal Ecology 78: 573-584.
cSiro. 2011. Australian moths online: a photo gallery Retrieved
02/02/2012. http://www.csiro.au/outcomes/environment/
biodiversity/australian-moths
ev e n H u i S , n.l. 2013. Checklist of Fijian Lepidoptera. Retrieved
02/01/2016, from http://hbs.bishopmuseum.org/fiji/
checklists/lepidoptera.html
ev e n H u i S , n.l. & d.J. Bi c k e l . 2005. Fiji arthropods. Bishop Musuem
Occassional Papers 82: 3-25.
Fi S H e r , m.r. 2011. Changes in macromoth community structure
following deforestation in Western Washington State. Western
Washington University.
Ha w e S , J., c.d.S. mo t t a , w.l. ov e r a l , J. Ba r l o w , t.a. ga r d n e r &
c.a. Pe r e S . 2009. Diversity and composition of Amazonian
moths in primary, secondary and plantation forests. Journal
of Tropical Ecology 25: 281-300.
HenderSon, P. & r. Se a B y . 2008. A practical handbook for
multivariate methods. Pisces Conservation Ltd., UK.
pp224.
Hi l t , n. 2005. Diversity and species composition of two different
moth families (Lepidoptera: Arctiidae vs. Geometridae) along
a successional gradient in the Ecuadorian Andes. MSc thesis.
University of Bayreuth.
Ho l l o w a y , J.d. 1983. The moths of Borneo, Part 14. Noctuidae:
Euteliinae, Stictopterinae, Plusiinae, Pantheinae. Kuala
Lumpur: Southdene Sdn Bhd.
ke P P e l , g., J.c. na v u S o , a. na i k a t i n i , n.t. tH o m a S , i.a. ro u n d S , t.a.
oS B o r n e , n. Ba t i n a m u & e . Se n i v a S a . 2005. Botanical diversity at
Savura, a lowland rain forest site along the PABITRA Gateway
Transect, Viti Levu, Fiji. Pacific Science 59: 175-191.
ki r k P a t r i c k , J.B. & d.c. Ha S S a l l . 1985. The vegetation and flora
along an altitudinal transect through tropical forest at Mount
Korobaba, Fiji. New Zealand Journal of Botany 23: 33-46.
kitcHing, r.l., a.g. or r , l. tHaliB, H. mitcHell, m.S. HoPkinS
& a.w. graHam. 2000. Moth assemblages as indicators of
environmental quality in remnants of upland Australian rain
forest. Journal of Applied Ecology 37: 284-297.
legendre, P. & e.d. gallagHer. 2001. Ecologically meaningful
transformations for ordination of species data. Oecologia
129: 271-280.
miller, J.c. & P.c. Hammond. 2007. Butterflies and moths of
Pacific Northwest forests and woodlands: rare, endangered, and
management sensitive species. USA: USDA, USFS & FHTET.
Pa i n e , J.r. 1991. lUCN Directory of protected areas in Oceania.
UK: lUCN.
Pa w S o n , S.m., e.g. Br o c k e r H o F F , m.S. wa t t & r.k. di d H a m . 2011.
Maximising biodiversity in plantation forests: Insights from
long-term changes in clearfell-sensitive beetles in a Pinus radiata
plantation. Biological Conservation 144: 2842-2850.
Pa w S o n , S.m., c.e. ec r o y d , r. Se a t o n , w.B. SH a w & e.g. Br o c k e r H o F F .
2010. New Zealand’s exotic plantation forests as habitats for
threatened indigenous species. New Zealand Journal of Ecology
34: 342-355.
Pr a S a d , B.c. 2010. Natural resource inventory report of the Fiji
Islands (Vol. 3): Land resources inventory of the Fiji Islands.
Fiji: The University of the South Pacific.
ri c k e t t S , t.H., g.c. da i l y , P.r. eH r l i c H & J.P. Fa y . 2001. Countryside
biogeography of moths in a fragmented landscape: biodiversity
in native and agricultural habitats. Conservation Biology 15:
378-388.
ro B i n S o n , g.S. 1975. Macrolepidoptera of Fiji and Rotuma. Oxford:
E. W. Classey.
Sc H u l z e , c., c. Hä u S e r & m. ma r y a t i . 2000. A checklist of the
Hawkmoths (Lepidoptera: Sphingidae) of Kinabalu park,
Sabah, Borneo. Malayan Nature Journal 54: 1-20.
Su e , d. 2010. Facilitating financing for sustainable forest
management in small islands developing states and low forest
cover countries. An analytical report prepared for the United
Nations Forum on Forests Country case study: Fiji.
Su m m e r v i l l e , k.S., l.m. ri t t e r & t.o. cr i S t . 2004. Forest moth taxa
as indicators of Lepidopteran richness and habitat disturbance:
a preliminary assessment. Biological Conservation 116: 9-18.
ta k i , H., y. ya m a u r a , i. ok o c H i , t. in o u e , k. ok a B e & S.i. ma k i n o .
2010. Effects of reforestation age on moth assemblages
in plantations and naturally regenerated forests. Insect
Conservation and Diversity 3: 257-265.
ti k o c a , S., J. cl a y t o n , S. Ho d g e , m. tu i w a w a , S. Pe n e & g. Br o d i e .
2016a. A record of Dactyloplusia impulsa (Walker, 1865)
(Lepidoptera: Noctuidae: Plusiinae) new to Fiji. Australian
Entomologist 43:35-37.
Ti k o c a , S., S. Ho d g e , S. Pe n e , J. cl a y t o n , m. tu i w a w a & g. Br o d i e .
2016b. Temporal variation in macro-moth abundance and
species richness in a lowland Fijian forest. Pacific Science
70:447-461.
ti k o c a , S., S. Ho d g e , m. tu i w a w a , S. Pe n e , J. cl a y t o n & g. Br o d i e .
2016c. An appraisal of sampling method and effort for
investigating moth assemblages in a Fijian forest. Austral
Entomology 55:455-462.
tu i w a w a , S.H. & g. ke P P e l . 2013. Species diversity, composition and
the regeneration potential of native plants at the Wainiveiota
Mahogany Plantation, Viti Levu, Fiji Islands. The South Pacific
Journal of Natural and Applied Sciences 30: 51-57.
willott, S.J. 1999. The effects of selective logging on the
distribution of moths in a Bornean rainforest. Philosophical
Transactions of the Royal Society of London Series B: Biological
Sciences 354(1391): 1783-1790.
za H i r i , r., i.J. kit c H i n g , J.d. la F o n t a i n e , m. mu t a n e n , l. ka i l a , J.d.
Ho l l o w a y & n. wa H l B e r g . 2011. A new molecular phylogeny
offers hope for a stable family level classification of the
Noctuoidea (Lepidoptera). Zoologica Scripta 40: 158-173.
77
APPENDIX. Species of moths recorded using a mercury vapour lamp for four nights at each of three forest sites near
Suva, Fiji Islands, October 2012.
Family Species Savura Colo-i-Suva Mt. Korobaba
Cossidae Acritocera negligens 1 0 0
Erebidae Achaea robinsoni 1 0 1
Adetoneura lentiginosa 0 0 1
Aedia sericea 0 2 0
Argina astraea 0 3 0
Asota woodfordi 0 1 0
Bocana manifestalis 2 0 1
Calliteara fidjiensis 52 11 2
Dysgonia duplicata 2 1 0
Dysgonia prisca 1 2 1
Ericaea inangulata 3 2 7
Ericaea leichardtii 10 3 10
Euchromia vitiensis 1 0 0
Eudocima fullonia 1 0 0
Hydrillodes surata 2 7 2
Hypenagonia emma 3 0 0
Hypocala deflorata 0 0 4
Mecodina variata 2 0 1
Mocis trifasciata 0 0 1
Neogabara plagiola 0 1 0
Nyctemera baulus 1 0 0
Oeonistis delia 6 2 3
Oxyodes scrobiculata 4 1 13
Palaeocoleus sypnoides 2 2 1
Polydesma boarmoides 0 3 5
Rhesalides curvata 0 2 2
Rusicada nigritasis 1 0 3
Rusicada vulpina 0 1 0
Serodes mediopallens 0 1 2
Serrodes campana 1 0 0
Simplicia cornicalis 1 2 1
Thyas coronate 0 0 1
Thyas miniacea 3 3 1
Geometridae Agathia pisina 1 0 0
Anisodes gloria 0 1 0
Anisodes monetara 5 6 4
Anisodes niveopuncta 0 0 1
Anisodes oblivaria 2 1 0
Bulonga philipsi 6 2 1
Chlorochaeta cheromata 0 0 1
Chloroclystis encteta 0 2 0
Cleora sp. 59 21 14
49: 69-79, 2016
J. Res.Lepid.78
Family Species Savura Colo-i-Suva Mt. Korobaba
Gelasma stuhlmanii 0 0 1
Gymnoscelis sara 0 1 0
Horisme chlorodesma 0 3 0
Mnesiloba eupitheciata 0 0 2
Nadagara irretracta 0 1 0
Polyclysta gonycrota 0 0 2
Pseuderythrolopus bipunctatus 0 1 1
Pyrrhorachis pyrrhogona 1 0 2
Thalassodes pilaria 3 0 0
Thalassodes chloropis 0 0 1
Thalassodes figurate 1 0 0
Thalassodes liquescens 3 0 0
Limacodidae Beggina albafascia 0 1 2
Beggina bicornis 0 1 0
Beggina mediopunctata 2 0 0
Beggina minima 2 1 0
Beggina unicornis 1 0 0
Beggina zena 0 3 1
Beggina sp. 2 0 0
Noctuidae Aegilia vitiscribens 0 2 0
Agrotis ipsilon 0 1 0
Athetis thoraicica 0 0 3
Chasmania tibialis 1 1 0
Chrysodeixis eriosoma 8 1 3
Condica conducta 0 1 0
Condica illecta 1 3 2
Dactyloplusia impulse 0 1 0
Gyrtonia purpurea 0 1 0
Leucania venalba 1 0 0
Leucania yu 1 1 0
Penicillaria jocosatrix 0 1 0
Sasunaga oenistis 2 0 3
Spodoptera litura 0 1 0
Spodoptera mauritia 4 10 0
Stictoptera stygia 0 1 1
Stictoptera vitiensis 1 2 1
Targalla delatrix 1 0 0
Tiracola plagiata 2 2 0
Nolidae Austrocarea albipicta 2 0 1
Barasa triangularis 0 2 1
Earias flavida 0 4 0
Maceda mansueta 0 0 1
Maceda savura 0 1 3
APPENDIX. Cont.
79
Family Species Savura Colo-i-Suva Mt. Korobaba
Sphingidae Macroglossum godeffroyi 2 1 0
Theretra nessus 1 0 0
Theretra silhetensis 0 0 1
Thyrididae Banisia anthina 0 1 0
Banisia messoria 4 0 6
Striglina navigatorum 1 2 1
Uraniidae Urapteroides anerces 4 4 5
APPENDIX. Cont.
49: 69-79, 2016
