MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 516: 177–185, 2014
doi: 10.3354/meps11046
Published December 3
INTRODUCTION
Animals can strongly affect the morphology and
structure of habitat-forming foundation species and
the habitats they comprise (Mopper et al. 1991).
For example, animals alter mangrove trees through
folivory, wood-boring, and feeding on propagules
(Ellison & Farnsworth 1990, Feller 2002, Sousa et al.
2003, Cannicci et al. 2008). Insects have substantial
negative effects on mangrove trees causing dramatic
and widespread alterations of growth and survivor-
ship (Piya karn chana 1981, Whitten & Damanik 1986,
Anderson & Lee 1995). However, marine wood-bor-
ing isopods may also be important structuring agents
in mangrove ecosystems. Several studies have linked
the prodigious boring by high densities of sphaero-
matid isopods to the loss of mangroves in Florida
(Rehm & Humm 1973, Rehm 1976), Kenya and Tan-
zania (Sva vars son et al. 2002), and India (Santhaku-
mari 1991).
Isopod burrowing or boring (used synonymously
herein) changes the morphology and integrity of
© Inter-Research 2014 · www.int-res.com*Corresponding author: davidsont@si.edu
Damage and alteration of mangroves inhabited by a
marine wood-borer
Timothy M. Davidson1,3,*, Catherine E. de Rivera1, Hwey-Lian Hsieh2
1Department of Environmental Science and Management, Portland State University, PO Box 751, Portland, OR 97207, USA
2Biodiversity Research Center, Academia Sinica, Nankang, Taipei, Taiwan 115, ROC
3Present address: Smithsonian Tropical Research Institute, Apartado Postal 0843-03092, Balboa, Ancon, República de Panamá
ABSTRACT: Animals can exert a strong influence on the structure and function of foundation spe-
cies such as mangroves. Because mangroves live at the interface of land and sea, both terrestrial
and marine species affect them, including numerous herbivores and boring species. These organ-
isms can affect the fecundity, performance, and morphology of mangroves. In a mangrove stand
in southwestern Taiwan, we discovered that mangroves were extensively damaged by wood-
boring isopods Sphaeroma terebrans. We examined the relationships between burrowing damage
from S. terebrans and metrics of mangrove fecundity, performance, and morphology. Individuals
of Rhizophora stylosa that were more burrowed by isopods had significantly fewer propagules,
fewer ground roots stabilizing the tree, smaller leaves, and more non-foliated twigs. Similarly,
Avicennia marina with more burrows had fewer pneumatophores and lenticels (used for gaseous
exchange), and pneumatophores with more necrotic tissue. The most heavily damaged trees were
hollowed-out with burrows (A. marina) or fell over when their supportive root system failed (R. sty-
losa). These correlations suggest that marine wood-borers can negatively influence mangroves
and alter tree morphology, although other stressors may also be involved. While studies have
examined the effects of isopods on root-level production, we provide the first quantitative evi-
dence that localized burrowing damage is correlated with tree-level effects. These results are con-
sistent with other literature demonstrating the importance of sub-lethal damage by borers in shap-
ing foundation species. Such damage may have cascading effects on the diverse assemblages of
 marine and terrestrial biota that use mangroves as habitat.
KEY WORDS:  Habitat alteration · Mangrove · Plant−animal interactions · Sphaeroma terebrans ·
Sub-lethal stress
Resale or republication not permitted without written consent of the publisher
Mar Ecol Prog Ser 516: 177–185, 2014
mangrove roots, which can have mech anical and
physiological consequences. The aerial roots of red
mangroves buttress and anchor the tree; the
destruction of aerial roots by isopods has been sug-
gested to increase the susceptibility of mangroves to
damage from storms (Rehm & Humm 1973, San-
thakumari 1991). While burrowing by sphaeromatid
isopods can initiate root branching (Simberloff et al.
1978, Perry & Brusca 1989), these sub-lethal bur-
rowing effects cause a net reduction in root growth
and productivity (Perry & Brusca 1989, Brooks &
Bell 2002) and divert resources from production to
repair (Brooks & Bell 2002). Roots are also important
for gaseous exchange and nutrient absorption, and
they are photosynthetic (Gill & Tomlinson 1977,
Tomlinson 1986, Kitaya et al. 2002). Because borer
damage diverts resources from the generation of
propagules, leaves, and other structures to repair,
we hypothesize that high levels of isopod boring
may affect unbored parts of the tree. However, most
relevant studies are anecdotal or focus on isopod
interactions at the root level (Rehm & Humm 1973,
Perry & Brusca 1989, Santhakumari 1991, Brooks &
Bell 2002); no studies have previously quantified
how isopod effects may scale up to alter and dam-
age structures other than roots.
The tropical and subtropical isopod Sphaeroma
terebrans is a prevalent borer in mangrove trees in
brackish areas of Australia, Africa, the Caribbean,
and Asia (Kensley & Schotte 1989, T. M. Davidson
unpubl. surveys). Sphaeromatid isopods are filter
feeders that create burrows for habitat and are often
found in the intertidal mangrove fringe. During sur-
veys of a mangrove stand in southwestern Taiwan,
we discovered that S. terebrans had extensively bored
into the mangroves Rhizophora stylosa and Avicennia
marina. The most heavily burrowed trees appeared
unhealthy, exhibiting low leaf cover, small leaves,
and few grounded aerial roots (R. stylosa) and few in -
tact pneumatophores (A. marina) compared to lightly
burrowed trees. Hence, we examined how isopod
boring was associated with tree-level metrics of
 morphology, performance (factors that affect survi -
vorship or growth, such as leaf size, tree height, etc.),
and fecundity of these 2 mangrove species. We
hypothesized that trees with more burrow damage
would be smaller in size, have smaller leaves, and
have fewer and altered root structures. We also
expected additional expressions of damage that vary
between these trees due to their different morpho -
logies (e.g. R. stylosa: percentage of broken aerial
roots, number of grounded roots; A. marina: pneuma-
tophore density, percent cover of necrotic tissue).
MATERIALS AND METHODS
Study site
We examined the relationships between boring
damage by Sphaeroma terebrans and the morpho -
logy and performance of mangrove trees in Haome i -
liao Nature Preserve (Chiayi County, Taiwan;
23° 21.661’ N, 120° 07.826’ E) in July 2010. Haomei -
liao Nature Preserve comprises a gradually sloping
intertidal mangrove stand with a muddy-sand sub-
stratum in a semi-enclosed lagoon in southwestern
Taiwan. The upland border of the stand is primarily
composed of Australian pine Casuarina sp. and few
individuals of Hibiscus tiliaceus. The upland transi-
tions to an expansive stand of small Avicennia ma -
rina (~1− 2 m tall) and then to a mixed low intertidal
zone with small A. marina, some Rhizophora stylosa
(~2−3 m tall), and scattered Kandelia obovata near
the stand edge. This seaward edge is relatively open
and has numerous gaps (Fig. S1 in the Supplement at
www. int-res. com/ articles/ suppl/ m516 p177_ supp.
pdf). Further seaward are scattered patches of the
seagrass Halophila beccarii. To augment the existing
population of A. marina, several thousand propag-
ules of R. stylosa and Lumnitzera racemosa (not ob -
served in this study) were planted in 1993−1994
(Peng et al. 1993, Chen et al. 1996), and additional
A. marina trees may have been planted in 1975, al -
though records are unclear (M. L. Xue, Taiwan
Endemic Species Research Institute, pers. comm.).
A. marina trees naturally occurred at the site before
restoration began, so their ages may vary. Both un -
damaged and damaged trees were present through-
out the study area. Hao mei liao hosts the northern-
most population of R. stylosa in Taiwan (Hsueh & Lee
2000), while the range of A. marina extends ~150 km
north to Hsinchu (T. M. Davidson pers. obs.). Natural
recruitment of R. stylosa, A. marina, and some K.
obovata was apparent throughout the stand and
along the mangrove shoreline. Salinity, air tempera-
ture, and sediment temperature were 29.7 ± 3.0 PSU,
29.3 ± 1.2°C, and 31.4 ± 0.7°C (mean ± 95% CI),
respectively, during the study period. Hao mei liao
experiences mixed semi-diurnal tides (tidal range is
96−206 cm in July).
Shoreline sampling of mangrove and abiotic
factors
We sampled 13 A. marina trees and 10 R. stylosa
trees accessible along a ~600 m transect placed
178
Davidson et al.: Mangrove damage associated with boring isopods 179
roughly parallel to the seaward edge of the man-
grove stand. Due to the inaccessibility of mangroves
on the terrestrial side, sampling occurred 11.6 ±
4.2 m (mean ± 95% CI) in from the seaward edge of
the mangrove stand. We used random numbers to
select the sampling locations along the transect. At
each of those locations, we sampled the nearest man-
grove tree orthogonal to and shoreward of the tran-
sect. Because the mangrove edge was sinuous, trees
were sampled at varying distances from the transect
but at similar tidal heights (Table S1 in the Supple-
ment at www. int-res. com/ articles/ suppl/ m516 p177 _
supp. pdf).
The 2 species of mangroves differed in morphol-
ogy, thus we sampled slightly different tree-level
effects and different numbers of roots of each species
to test our hypotheses. R. stylosa has aerial roots that
grow down from the tree to form a flying buttress. In
contrast, A. marina exhibits a morphology similar to
terrestrial trees (single trunk, broad crown) but is
surrounded by numerous pencil-sized pneumato-
phores (breathing roots) (Tomlinson 1986).
For each tree of R. stylosa, we measured how 2
metrics of isopod damage, the percentage of aerial
roots burrowed and mean number of burrows per cm
of aerial root, were associated with tree-level effects,
including the number of grounded aerial roots and
the percentage of broken roots, the number of
propagules per tree, tree height, the percentage of
first-order branches without leaves (hereinafter: non-
foliated twigs), and mean leaf area (cm2). We consid-
ered the latter 3 measurements as indicators of per-
formance in R. stylosa because those factors could
affect or indicate growth and survivorship. For each
tree, we counted the number of broken, burrowed,
free- hanging, and grounded aerial roots present.
We also haphazardly collected 5 free-hanging un -
branched aerial root tips by cutting the aerial root
tips off at their initiation point (when it was acces -
sible) or approximately 5 cm above the burrowed
area of the root (near the high tide mark on the root).
The roots were then dissected in the laboratory to
determine the number of isopods per root.
On each tree of A. marina, we examined if the total
number of burrows per tree was associated with
tree height, percentage of broken pneumatophores,
pneumatophore length and weight, pneumatophore
density and mass, lenticel density, and mean leaf size
(cm2). Because these factors could affect or indicate
survivorship and growth, we consider these factors to
be metrics of performance in A. marina; however,
many of these factors (e.g. pneumatophore size and
density) are also metrics of morphology. We also
examined how the mean numbers of burrows per cm
of pneumatophore were related to pneumatophore
length and weight, percent cover of necrotic tissue
on pneumatophores, and the number of lenticels per
pneumatophore. At each sampled tree, we counted
the numbers of burrows present, and assessed tree
height and the characteristics of the pneumato-
phores. We randomly collected 10 pneumatophores
(pencil-sized roots) from each of 13 A. marina trees
by cutting the pneumatophores off at the surface of
the sediment at random distances (within 1 m) and
directions from the bole of the tree. We measured the
weight and length, counted the number of lenticels,
isopods, and burrows, and visually estimated the per-
cent cover of necrotic or damaged tissue on each
pneumatophore. In addition, to measure the density
of unburrowed, burrowed, or broken pneumato-
phores, we randomly placed 3 quadrats (25 × 25 cm)
within 1 m of each A. marina tree. We characterized
pneumatophores as broken if their tips were missing
or hollowed out. The mean mass of pneumatophores
per quadrat was estimated by multiplying pneumato-
phore density by the mean weight of the 10 sampled
pneumatophores. Similarly, we calculated the mean
density of lenticels per quadrat by multiplying the
mean number of lenticels per sampled pneumato-
phore by the mean number of pneumatophores in the
quadrats around the trees. Fecundity could not be
assessed for A. marina because propagules were not
present at the time of sampling.
Approximately 20 leaves per tree of both species of
mangrove were haphazardly collected (without look-
ing) then digitally scanned. The leaf area (cm2), per-
cent herbivore damage per leaf, such as discoloration
and holes, and percentage of leaves damaged were
measured using digital image analysis software
(ImageJ, version 1.46r) with a calibrated scale bar.
The latter 2 metrics were used to assess the level of
leaf herbivory on the trees.
To determine if other factors could be responsible
for the inter-tree variation in structure and perform-
ance, we measured several other biotic and environ-
mental variables. We examined trees for the pres-
ence of other boring organisms (insects, pholads,
shipworms, etc.) and conspicuous pathogens (galls,
cankers, etc.; Wier et al. 2000). We also measured the
potential environmental covariates of tree perform-
ance and isopod prevalence including salinity, air
and sediment temperature, distance from the sea-
ward edge of the mangrove stand, and relative tidal
height. We estimated relative tidal height by compar-
ing the height of the high tide mark on the bole of
sampled trees and by using PVC poles coated with a
Mar Ecol Prog Ser 516: 177–185, 2014
marker composed of water-soluble glue and crystal
violet stain (dissolvable when tide water contacts it)
and placed adjacent to some trees.
Statistical analyses
We tested the associations between isopod damage
and tree morphology, performance, and fecundity
using Pearson’s correlations. Square-root or log
trans formations were used to meet the statistical as -
sumptions of Pearson’s correlation,
re duce the influence of outliers, and
im prove linearity between variables.
We included trendlines and slopes (b)
from linear regression to aid in the
interpretation of relationships be -
tween variables. To account for pos-
sible multiple testing issues, we cal-
culated the false discovery rate (FDR)
(Benjamini & Hochberg 1995) and
report adjusted p-values (p(adj)) for
each family of tests (raw and adjusted
p-values are provided in Table S2 in
the Supplement). The design included
4 families of tests respective of both
metrics of burrowing damage for both
mangrove species (Table S2). We set
our significance level at 0.05 but also
report marginally significant values
(between 0.05 and 0.10). We present
values as mean ± 95% CI.
RESULTS
Prevalence, density, and patterns of
isopod  burrowing damage
Both Rhizophora stylosa and Avi-
cennia marina trees exhibited exten-
sive burrowing damage from Sphae -
roma terebrans (Fig. 1). The most
heavily burrowed individuals of R.
sty losa had discolored leaves, numer-
ous necrotic and broken aerial roots,
and few supportive ground roots
(some trees had fallen over or were
tipping over; Fig. 1b). Approximately
77.1% of the collected, lab-inspected
free-hanging aerial roots (pooled)
were burrowed, and 70.8% harbored
iso pods. The sampled free-hanging
aerial roots had mean burrow densities of 10.3 ±
5.5 per root (0.7 ± 0.3 per cm of root) and isopod
 densities of 17.4 ± 10.2 per root (1.3 ± 0.7 per cm of
root). The mean number of burrows in roots was
strongly positively correlated to the mean number of
isopods in roots (r2 = 0.89, b = 0.511, t = 7.92, df = 8,
p(adj) < 0.001).
Isopods were also found in dense aggregations
in A. marina. The mean number of burrows in the
sampled A. marina trees was 194 ± 90 per tree. In
contrast to the root-dominated occurrence in R. sty-
180
Fig. 1. Prodigious burrowing by Sphaeroma terebrans damaged the man-
groves in Haomeiliao Nature Preserve, including (a−c) the free-hanging aer-
ial roots of Rhizophora stylosa (used to anchor and support the tree), and (d)
the pneumatophores (roots used in gaseous exchange) and (e,f) the trunk and
branches of Avicennia marina. Tree height: (b) ~2 m, (e) ~1.5 m. Scale bar: 
(a) 5 cm, (d) 3 cm
Davidson et al.: Mangrove damage associated with boring isopods
losa, more burrows were in the trunk (53.1%) and
branches (30.1%) of A. marina than in the exposed
areas of roots anchoring the tree in the sediment
(16.8%). The trees with the most visible damage by
burrows had truncated, broken, or necrotic pneuma-
tophores, discolored leaves, and perforated sections
or hollowed-out trunks (Fig. 1d−f). Very little force
was necessary to break off sections from those dam-
aged trees.
In the collected pneumatophores of A. marina,
mean burrow and isopod densities were 1.8 ± 0.5 and
0.5 ± 0.2 per root (0.4 ± 0.2 and 0.1 ± 0.1 per cm of
root), respectively. Burrows were present in pneuma-
tophores of all sampled trees and in 65.8% of all sam-
pled pneumatophores. Isopods were present in pneu-
matophores of 12 out of 13 sampled trees and in
32.5% of sampled pneumatophores. Approximately
78.3% of collected pneumatophores exhibited discol-
ored, necrotic tissue. The tissue had been completely
removed in the most heavily burrowed, rotten, and
discolored pneumatophores and fell apart when
 handled.
Associations between isopod  burrowing and
mangrove  morphology, performance, and
 fecundity
Both measures of isopod burrowing damage
(percent of the roots that were burrowed or
number of burrows per root) in R. stylosa were
related to in creased root breakage and fewer
grounded aerial roots. The percentage of aerial
roots burrowed by isopods (as measured in the
field) was positively correlated with the percentage
of roots that were broken (Fig. 2a) and negatively
related with the number of grounded roots present
and anchoring the tree in the ground (Fig. 2b).
Similarly, the mean number of burrows per cm in
aerial roots was negatively associated with the
number of ground roots (r2 = 0.44, b = −38.0, t =
−2.49, df = 8, p(adj) = 0.084) and positively associ-
ated with the percentage of broken roots (r2 = 0.44,
b = 0.517, t = 2.51, df = 8, p(adj) = 0.084), al though
these weak relationships were not significant
after FDR corrections were applied (Table S2 in
the Supplement at www. int-res. com/
articles/ suppl/ m516 p177_ supp. pdf).
More heavily burrowed trees of R.
stylosa also had fewer twigs bearing
leaves and propagules and harbored
smaller leaves. The total percent age
of roots burrowed by isopods was pos-
itively associated with the percentage
of non-foliated twigs in R. stylosa
(Fig. 2c) and negatively asso ciated with
the number of propagules (Fig. 2d).
Trees of R. stylosa with a higher per-
centage of burrowed roots had signifi-
cantly smaller leaves (r2 = 0.48, b =
−15.1, t = −2.53, df = 7, p(adj) = 0.047).
The mean number of burrows per
cm in aerial roots had a slight
 posi tive association with the percent-
age of non-foliated twigs (r2 = 0.41,
b = 0.162, t = 2.39, df = 8, p(adj) =
0.084) and a slight negative associa-
tion with leaf size (r2 = 0.41, b = −11.4,
t = −2.23, df = 7, p(adj) = 0.092), but
these relationships were not signifi-
cant. We did not detect a significant
association be tween the percentage
of roots burrowed by isopods and tree
height or between the mean number
of burrows per cm in aerial roots and
tree height or the number of propag-
ules on trees (p(adj) > 0.10).
181
Fig. 2. Relationships between percentage of roots burrowed by Sphaeroma
terebrans in Rhizophora stylosa and (a) root breakage, (b) abundance of
grounded roots, (c) percentage of non-foliated twigs, and (d) abundance of
propagules (all data given ‘per tree’). Sample size (n = 10) differed due to miss-
ing data on propagules for 2 trees. The y-axes of (a) and (c) are back-trans-
formed (from square-root transformations). Note the log-scaling in (b) and (d)
Mar Ecol Prog Ser 516: 177–185, 2014
In A. marina, more heavily burrowed trees had
fewer pneumatophores, a lower mass of pneumato-
phores, and fewer lenticels. The total number of bur-
rows in A. marina was negatively associated with the
mean density of pneumatophores (Fig. 3a). The rela-
tionship remained significant even after an influen-
tial outlier (from the most heavily burrowed tree in
the data collected) was excluded from the analysis
(r2 = 0.47, df = 10, p(adj) = 0.014). The number of bur-
rows per tree was also negatively associated with the
mass of pneumatophores per m2 (Fig. 3b) and mean
lenticel density (Fig. 3c), but not pneu mato phore
length or weight or percent of broken pneumato-
phores (p(adj) > 0.10). The mean number of burrows
per cm in pneumatophores was positively related to
the percentage of pneumatophore surface that was
damaged, dis colored, and necrotic (r2 = 0.75, b = 49.1,
t = 5.7, df = 11, p(adj) < 0.001), and negatively related
to the numbers of lenticels per pneu ma to phore
(although the association was weak and not signifi-
cant after FDR correction; r2 = 0.37, b = −48.9, t =
−2.54, df = 11, p(adj) = 0.054). There were no signifi-
cant associations de tected be tween the numbers of
burrows in pneu mato phores and the pneumatophore
length or weight (p(adj) > 0.10), likely because evi-
dence of burrowing became ob scured or removed as
the heavily damaged roots  further decomposed or
broke off. More heavily burrowed trees also had
smaller leaves in A. marina (r2 = 0.39, b = −0.0053, t =
−2.41, df = 9, p(adj) = 0.079), al though the association
was weak and not significant. We did not detect a
significant association between the number of bur-
rows per tree and tree height (p(adj) > 0.10).
Influences of environmental factors and other
organisms
None of the environmental factors (salinity, air
temperature, sediment temperature, distance from
the seaward edge of the stand, relative tidal height)
explained the variability in the percentage of bur-
rowed aerial roots, numbers of burrows per tree,
mean density of burrows per root, or the metrics of
tree performance (p(adj) > 0.10). However, a mar -
ginally significant negative association was found
between the percentage of burrowed roots and the
distance from the edge of the stand in R. stylosa (r2 =
0.52, b = −0.017, t = −2.97, df = 8, p(adj) = 0.072), indi-
cating that isopods were present many meters in
from the edge. Herbivory (percent cover of leaf dam-
age, percentage of leaves damaged) was not signifi-
cantly associated with any measurement of burrow-
ing damage, morphology, or performance in either
mangrove species (p(adj) > 0.10). No other boring
animals (insects, shipworms, etc.) were found in the
roots of R. stylosa or in A. marina. We did not detect
any conspicuous pathogens on trees.
182
Fig. 3. Association between number of burrows created by
Sphaeroma terebrans per Avicennia marina and (a) mean
pneumatophore density, (b) mean mass of pneumatophores,
and (c) mean lenticel density. Sample size varied due to a
missing sample of pneumatophores for 1 tree needed for
 estimations of mean mass and lenticel density. Note the 
log-scaling
Davidson et al.: Mangrove damage associated with boring isopods
DISCUSSION
High levels of isopod boring were related to de -
pressed performance and fecundity and alterations
in the morphology of mangrove trees. Previous
experiments have quantified the effects of isopod
boring on production at the root level (Perry 1988,
Ellison & Farnsworth 1990); however, the present
study provides the first quantified evidence that the
cumulative effect of localized isopod boring damage
can scale up to affect the whole tree. Consistent with
our hypotheses, heavily burrowed Rhizophora sty-
losa had altered root architectures (fewer intact aer-
ial roots to anchor the tree), smaller and fewer leaves,
and fewer propagules. Similarly, more burrowed
trees of Avicennia marina had smaller leaves and
fewer pneumatophores; remaining pneumatophores
tended to be more necrotic and damaged than in less
burrowed trees.
Trees more affected by isopods could suffer lower
performance and survivorship, as photosynthetic
capacity, gaseous exchange, and nutrient uptake
would cease or be reduced compared to unburrowed
trees. The accumulation of minor damage can nega-
tively affect trees by diverting resources to repairing
damage (Kulman 1971, Brooks & Bell 2002) or by
causing mortality (Kulman 1971, Ozaki et al. 1999).
Many of these trees also exhibited changes in root
and canopy architecture related to isopod damage,
including fewer supportive aerial roots and smaller
and fewer pneumatophores. However, manipulative
studies are necessary to definitively assign causality
between isopod boring and mangrove damage.
Numerous studies have documented boring spha -
eromatid isopod damage in brackish mangroves
around the world (Rehm & Humm 1973, Perry &
Brusca 1989, Santhakumari 1991, Svavarsson et al.
2002), but the extent and intensity of the damage
varies among sites. In mangroves experiencing a lim-
ited tidal range (<0.5 m; e.g. Caribbean and Florida),
isopod attack is mostly limited to the first few meters
of aerial roots of fringing red mangroves that are sub-
merged during high tide (Estevez & Simon 1975,
Simberloff et al. 1978, T. M. Davidson pers. obs.).
However, in mangrove sites exhibiting a larger tidal
range (e.g. East Africa, Svavarsson et al. 2002; the
present study; some sites in Pacific Panama, T. M.
Davidson pers. obs.), and in sloping sites with many
openings and gaps along the shoreline, isopods can
attack trees many meters inside the mangrove stand
(Fig. S1 in the Supplement at www. int-res. com/
articles/ suppl/ m516 p177_ supp. pdf) because abun-
dant root habitat is available for colonization at high
tide. Environmental stressors also affect tree mor-
phology and performance (Tomlinson 1986, Kathire-
san & Bingham 2001) and may exacerbate the effects
of borer damage. For example, in our study site, R.
stylosa and A. marina appear stunted and may be
experiencing physiological stresses associated with
living near their range edge (Hsueh & Lee 2000), or
they could be stressed by subsidence that has appar-
ently affected the region (Wang 2012). In addition,
some of the sampled trees (R. stylosa and perhaps A.
marina) might be the result of a mangrove restoration
decades ago, and it is possible that some unknown
stressor is affecting these small trees in concert with
isopod damage. With respect to such environmental
stressors, we do not rule out the possibility that indi-
vidual trees could vary in their susceptibility to a
potential unknown stressor that is correlated with
borer damage.
The changes documented in this study in tree
architecture (through root alterations), leaf cover,
and habitat complexity (through the extirpation of
trees) may more broadly affect mangrove and other
forest ecosystems as well. For example, alterations in
the root structure of mangroves and terrestrial trees
alter erosion and sedimentation regimes (Spenceley
1977, Kathiresan 2003, Krauss et al. 2003, Reubens et
al. 2007). Herbivory mediates primary productivity
(Mattson & Addy 1975, Feller & Mathis 1997), alters
the distribution of canopy gaps (Feller & McKee
1999), and changes nutrient flux by facilitating the
breakdown and export of plant materials (Lightfoot &
Whitford 1990, Feller 2002, Chapman et al. 2003).
Furthermore, the complex structure of mangroves
provides important nursery habitats (Primavera 1998,
Kathiresan & Bingham 2001, Nagelkerken et al.
2008); isopod-associated changes could alter the
quality and quantity of habitat available to numerous
other organisms.
This study suggests that boring marine isopods are
important structuring agents of mangroves, although
additional manipulative studies are necessary. While
the structuring role of herbivory is well documented
(Mopper et al. 1991), our study posits that the non-
consumptive boring damage of isopods may also
have important effects on foundation species. The
present work also contributes to the literature of sub-
lethal effects that identifies systemic effects from
cumulative small-scale stressors (Mopper et al. 1991,
Feller 2002). Through cumulative boring damage,
isopods may affect the performance, fitness, sur -
vivor ship, and morphology of mangroves — all of
which can alter the structure and function of this
important ecosystem and the biota living therein.
183
Mar Ecol Prog Ser 516: 177–185, 2014
Acknowledgements. We thank C.P. Chen, Hsiao-Hang
(Stacy) Tao, Joe Huang, and the rest of the Hsieh and Chen
laboratories for logistical and field support. Ernie Estevez
and anonymous reviewers greatly improved previous ver-
sions of this manuscript. Heejun Chang, Elise Granek, Gre-
gory Ruiz, and Mark Sytsma provided helpful advice. The
staff of the Summer Institute in Taiwan and Academia Sinica
provided key logistical support. We are grateful for the
funding provided by the National Science Foundation East
Asia and South Pacific Summer Institute fellowship program
in Taiwan (OISE-101514 to T.M.D.) and the Taiwan National
Science Council (NSC-99-2911-I-007-014).
LITERATURE CITED
Anderson C, Lee SY (1995) Defoliation of the mangrove Avi-
cennia marina in Hong Kong: cause and consequences.
Biotropica 27: 218−226
Benjamini Y, Hochberg Y (1995) Controlling the false dis-
covery rate: a practical and powerful approach to multi-
ple testing. J R Stat Soc B 57: 289−300
Brooks RA, Bell S (2002) Mangrove response to attack by a
root boring isopod: root repair versus architectural modi-
fication. Mar Ecol Prog Ser 231: 85−90
Cannicci S, Burrows D, Fratini S, Smith TJ III, Offenberg J,
Dahdouh-Guebas F (2008) Faunal impact on vegetation
structure and ecosystem function in mangrove forests: a
review. Aquat Bot 89: 186−200
Chapman S, Hart S, Cobb N (2003) Insect herbivory
increases litter quality and decomposition: an extension
of the acceleration hypothesis. Ecology 84: 2867−2876
Chen TS, Lai GX, Xue ML (1996) Preliminary restoration
results of Rhizophora stylosa. In: Peng GD (ed) Proceed-
ings of Mangrove ecosystem conference. 29 December
1995, Taichung, Taiwan. Taiwan Endemic Species Re -
search Institute, Nantou County, p 81–91
Ellison AM, Farnsworth EJ (1990) The ecology of Belizean
mangrove-root fouling communities. I. Epibenthic fauna
are barriers to isopod attack of red mangrove roots. J Exp
Mar Biol Ecol 142: 91−104
Estevez E, Simon J (1975) Systematics and ecology of
Sphaeroma (Crustacea: Isopoda) in the mangrove habi-
tats of Florida. In: Walsh G, Snedaker S, Teas H (eds) Pro-
ceedings of the international symposium of biology and
management of mangroves. Institute of Food and Agri-
cultural Sciences. University of Florida, Gainesville, FL,
p 286−304
Feller IC (2002) The role of herbivory by wood-boring insects
in mangrove ecosystems in Belize. Oikos 97: 167−176
Feller IC, Mathis WN (1997) Primary herbivory by wood-
boring insects along an architectural gradient of Rhi-
zophora mangle. Biotropica 29: 440−451
Feller IC, McKee KL (1999) Small gap creation in Belizean
mangrove forests by a wood-boring insect. Biotropica 31: 
607−617
Gill A, Tomlinson P (1977) Studies on the growth of red man-
grove (Rhizophora mangle L.) 4. The adult root system.
Biotropica 9: 145−155
Hsueh ML, Lee HH (2000) Diversity and distribution of the
mangrove forests in Taiwan. Wetlands Ecol Manage 8: 
233−242
Kathiresan K (2003) How do mangrove forests induce sedi-
mentation? Rev Biol Trop 51: 355−359
Kathiresan K, Bingham B (2001) Biology of mangroves and
mangrove ecosystems. Adv Mar Biol 40: 81−251
Kensley B, Schotte M (1989) Guide to the marine isopod
crustaceans of the Caribbean. Smithsonian Institution
Press, Washington, DC
Kitaya Y, Yabuki K, Kiyota M, Tani A (2002) Gas exchange
and oxygen concentration in pneumatophores and prop
roots of four mangrove species. Trees (Berl) 16: 155−158
Krauss K, Allen J, Cahoon D (2003) Differential rates of ver-
tical accretion and elevation change among aerial root
types in Micronesian mangrove forests. Estuar Coast
Mar Sci 56: 251−259
Kulman H (1971) Effects of insect defoliation on growth and
mortality of trees. Annu Rev Entomol 16: 289−324
Lightfoot D, Whitford W (1990) Phytophagous insects
enhance nitrogen flux in a desert creosotebush commu-
nity. Oecologia 82: 18−25
Mattson W, Addy N (1975) Phytophagous insects as regula-
tors of forest primary production. Science 190: 515−522
Mopper S, Maschinski J, Cobb N, Whitham T (1991) A new
look at habitat structure: consequences of herbivore-
modified plant architecture. In: Bell SS, McCoy ED,
Mushinsky HR (eds) Habitat structure: the physical
arrangement of objects in space. Chapman & Hall, New
York, NY, p 260−280
Nagelkerken I, Blaber SJM, Bouillon S, Green P and others
(2008) The habitat function of mangroves for terrestrial
and marine fauna: a review. Aquat Bot 89: 155−185
Ozaki K, Kitamura S, Subiandoro E, Taketani A (1999) Life
history of Aulacaspis marina Takagi and Williams (Hom.,
Coccoidea), a new pest of mangrove plantations in
Indonesia, and its damage to mangrove seedlings. J Appl
Entomol 123: 281−284
Peng GD, Lai GX, Huang ZQ, Liu KY, Xue ML (1993) Resto-
ration results of Rhizophora stylosa and Lumnitzera
 racemosa. Research program implementation results of
Taiwan Endemic Species Research Institute. Project no.
3.7. Taiwan Endemic Species Research Institute,  Nantou
County
Perry D (1988) Effects of associated fauna on growth and
productivity in the red mangrove. Ecology 69: 1064−1075
Perry D, Brusca R (1989) Effects of the root-boring isopod
Sphaeroma peruvianum on red mangrove forests. Mar
Ecol Prog Ser 57: 287−292
Piyakarnchana T (1981) Severe defoliation of Avicennia
alba BL. by larvae of Cleora injectaria Walker. J Sci Soc
Thailand 7: 33−36
Primavera JH (1998) Mangroves as nurseries: shrimp popu-
lations in mangrove and non-mangrove habitats. Estuar
Coast Mar Sci 46: 457−464
Rehm A (1976) The effects of the wood-boring isopod
Sphaeroma terebrans on the mangrove communities of
Florida. Environ Conserv 3: 47−57
Rehm A, Humm HJ (1973) Sphaeroma terebrans: a threat to
the mangroves of Southwestern Florida. Science 182: 
173−174
Reubens B, Poesen J, Danjon F, Geudens G, Muys B (2007)
The role of fine and coarse roots in shallow slope stability
and soil erosion control with a focus on root system archi-
tecture: a review. Trees (Berl) 21: 385−402
Santhakumari V (1991) Destruction of mangrove vegetation
by Sphaeroma terebrans along Kerala India coast. Fish
Technol 28: 29−32
Simberloff D, Brown BJ, Lowrie S (1978) Isopod and insect
184
Davidson et al.: Mangrove damage associated with boring isopods
root borers may benefit Florida mangroves. Science 201: 
630−632
Sousa WP, Quek SP, Mitchell BJ (2003) Regeneration of Rhi-
zophora mangle in a Caribbean mangrove forest: inter-
acting effects of canopy disturbance and a stem-boring
beetle. Oecologia 137: 436−445
Spenceley A (1977) The role of pneumatophores in sedimen-
tary processes. Mar Geol 24: M31−M37
Svavarsson J, Osore MKW, Olafsson E (2002) Does the
wood-borer Sphaeroma terebrans (Crustacea) shape the
distribution of the mangrove Rhizophora mucronata?
Ambio 31: 574−579
Tomlinson PB (1986) The botany of mangroves. Cambridge
University Press, New York, NY
Wang HW (2012) National Important Wetland Conservation
Action Plan: rehabilitating the hydrologic and ecological
environment at Haomeiliao wetland and Budai Salt Pan
Wetland (Part 2). Final report to Construction and Plan-
ning Agency, Ministry of Interior, Taiwan and Chiayi
County Government. Construction and Planning Agency,
Ministry of Interior, Taipei
Whitten A, Damanik S (1986) Mass defoliation of mangroves
in Sumatra, Indonesia. Biotropica 18: 176
Wier AM, Tattar TA, Klekowski EJ Jr (2000) Disease of red
mangrove (Rhizophora mangle) in southwest Puerto Rico
caused by Cytospora rhizophorae. Biotropica 32: 299−306
185
Editorial responsibility: Christine Paetzold,
Oldendorf/Luhe, Germany
Submitted: February 12, 2014; Accepted: September 12, 2014
Proofs received from author(s): November 22, 2014