MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

https://doi.org/10.47193/mafis.3632023010902

ABSTRACT. This study provides information on the effects of fish grazing on a seagrass bed

that was dominated by Thalassia hemprichii in Guang-guang, Dahican, Mati City, Davao Oriental,

Philippines. We tested the assumption that herbivore exclusion keeps the growth of seagrass shoots

up to a minimum length. Fish visual census was used to count the abundance of fish in the area dur-

ing high tide at a distance of 5 m away from the treatment cages with a 15 m transect length. Among

all species, Siganus fuscescens was the most abundant (499) followed by S. guttatus (153), while

Thalassoma jansenii was the less abundant (13). The result of the weekly fish count in the seagrass

bed showed highly significant variations in the number and census of species present in the area (df

=11, MS =2.88, F=30.10, p < 0.001). Analysis of shoot length measurement in treatment cages

also exhibited highly significant differences throughout the sampling period (df = 11, MS =726.71,

F=1,633.10, p<0.001). Comparison between initial and final shoot counts also showed significant

differences, with a higher shoot count in the full cage (X2=8, df = 2, p=0.018; FC >PC >OP).

This study raises clear evidence to support the assumption that herbivore exclusion positively influ-

ences the growth of T. hemprichii in this area.

Key words: Davao Oriental, fish grazing, herbivore exclusion, plant-herbivory interactions,

Siganus fuscescens, Thalassia hemprichii.

Efectos de la herbivoría de peces sobre las praderas marinas de Guang-guang, ciudad de Mati,

Filipinas

RESUMEN. Este estudio proporciona información sobre los efectos del forrajeo de peces en un

lecho de pastos marinos dominado por Thalassia hemprichii en Guang-guang, Dahican, ciudad de

Mati, Davao Oriental, Filipinas. Se probó la suposición de que la exclusión de herbívoros mantie-

ne el crecimiento de los brotes de pastos marinos hasta una longitud mínima. Se utilizó el censo

visual de peces para contar la abundancia de peces en el área durante la marea alta a una distancia

de 5 m de las jaulas de tratamiento con una longitud de transectas de 15 m. Entre todas las espe-

cies, Siganus fuscescens fue la más abundante (499) seguida de S. guttatus (153), mientras que

Thalassoma jansenii fue la menos abundante (13). El resultado del conteo semanal de peces en el

lecho de fanerógamas marinas mostró variaciones altamente significativas en el número y censo

de especies presentes en el área (df =11, MS =2,88, F=30,10, p< 0.001). El análisis de la medi-

ción de la longitud de los brotes en las jaulas de tratamiento también mostró diferencias muy sig-

nificativas a lo largo del período de muestreo (df =11, MS =726,71, F=1.633,10, p< 0,001). La

comparación entre el conteo de brotes inicial y final también mostró diferencias significativas, con

un mayor conteo de brotes en la jaula completa (X2=8, df =2, p=0,018; FC >PC > OP). Este

219

*Correspondence:

edison.macusi@dorsu.edu.ph

Received: 21 February 2023

Accepted: 26 April 2023

ISSN 2683-7595 (print)

ISSN 2683-7951 (online)

https://ojs.inidep.edu.ar

Journal of the Instituto Nacional de

Investigación y Desarrollo Pesquero

(INIDEP)

This work is licensed under a Creative

Commons Attribution-

NonCommercial-ShareAlike 4.0

International License

Marine and

Fishery Sciences

MAFIS

ORIGINAL RESEARCH

Effects of fish herbivory on seagrass meadows of Guang-guang, Mati City,

Philippines

EDISON D. MACUSI1, 2, *, NORJAN T. MASAGNAY1and IVY M. NALLOS1, 2

1Faculty of Agriculture and Life Sciences (FALS), Davao Oriental State University (DOrSU), Mati City, Philippines. 2Fisheries Catch

Assessment Project, Davao Oriental State University (DOrSU), Mati City, Philippines.

ORCID Edison D. Macusi https://orcid.org/0000-0002-9714-1074, Ivy M. Nallos https://orcid.org/0000-0003-3752-4847

INTRODUCTION

Shallow tropical and subtropical coastal areas

contain seagrass beds considered as one of the

most biodiverse, productive, and economically

important ecosystems in the world (Fortes 2004;

Macusi et al. 2011). Seagrass beds are known to

host visitors and resident species such as sea tur-

tles, dugongs, sea cucumbers, sea urchins, clams

and shellfish, as well as adult and juvenile fish

that take advantage of the productive ecosystem

found in seagrass beds (Abreo et al. 2018). It also

provides ecological services such as filtering

debris, stabilizing sediment, reducing coastal ero-

sion, and protecting the shoreline from extreme

storm events (Koch 2001; Bjork et al. 2008a;

Macusi and Tipudan 2020). In addition, seagrass

beds also store large quantities of organic carbon,

which is crucial for fighting climate change

(Duarte and Hemminga 2000). Excess carbon

produced by seagrass is buried under sediments,

helping in biological carbon sequestration (Duarte

et al. 2005). Its proximity to other habitat engi-

neers, such as mangroves and coral reefs, allows

for the trophic transfer of energy essential to

maintaining the abundance of mangrove and reef

fish species (Valentine and Heck 2005). Seagrass

beds are also economically critical for the liveli-

hood of coastal communities as they function as

gleaning areas and ecotourism sites (Maynawang

and Macusi 2023). While these are all essential

features of seagrass beds, one neglected connec-

tion is how the trophic transfer occurs from pro-

ducers to consumers, such as herbivores.

Sea urchins and fish have replaced historically

dominant mega-grazers like sirenians and turtles

in tropical areas and waterfowl in temperate areas

as primary grazers in current seagrass beds (Heck

and Valentine 2006). Moreover, sea urchins are

crucial to managing damaged aquatic vegetation

and recycling nutrients (Shears and Babcock

2002; Vonk et al. 2008). One commonly found

seagrass ecosystem resident is the rabbitfish of

Genus Siganus with 28 species (Randall and Kul-

bicki 2005). These are found in the Red Sea and

the Mediterranean in great abundance causing

overgrazing along the marine ecosystem in vari-

ous locations (Pickholtz et al. 2018). Although

individual species of the Siganid family can dom-

inate the herbivore biomass within seagrass beds

and estuarine habitats (Campos et al. 1994; Soli-

man et al. 2008), siganids tend to be found in

lower abundance on reefs (Wilson et al. 2003;

Cheal et al. 2012). Herbivores usually leave

patches of seagrasses, i.e. they are not completely

overgrazed; somehow, seagrass shoots are only

clipped. Grazing by fish and invertebrates in sea-

grass beds has a long history of being studied in

Africa, South America, Australia, and Europe

(McGlathery 1995; Valentine and Heck 1999;

Smulders et al. 2022). Grazing animals usually

forage on seagrasses to gain nourishment. How-

ever, there are cases when fish and invertebrates

may be construed as directly grazing on seagrass

leaves when this may not be the case. Instead,

they are found grazing on epiphytes attached to

the seaweed or plant.

There are documented studies showing that

rabbitfish usually eat macrophytes not just for

plant leaves but also for its epiphytes and other

associated small invertebrates that are ingested

accidentally (von Westernhagen 1973; Fox and

Bellwood 2008; Hoey et al. 2013). Their diets

consist primarily of macrophytes with low lipid

220 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

estudio presenta evidencia clara para apoyar la suposición de que la exclusión de herbívoros influye positivamente en el crecimiento

de T. hemprichii en esta área.

Palabras clave: Davao Oriental, forrajeo de peces, exclusión de herbívoros, interacciones planta-herbívoro, Siganus fuscescens,

Thalassia hemprichii.

content, probably resulting in sub-optimum

growth and inconsistent production (Bariche

2006). As a matter of fact, the gut content analysis

of rabbitfish from Mediterranean coastal habitats

showed that the macroalga Dictyota spp. was the

main food item, followed by Cystoseira spp., Sar-

gassum spp., and Caulerpa racemose (Bariche

2006; Serio et al. 2006). This finding was also

similar to that of Azzurro et al. (2007), where the

most represented food items included Sargassum

spp., Cystoseira spp., and Dictyota spp. This high-

lights the wide range of herbivory by rabbitfish

and emphasizes the importance of its seagrass

habitat. Sea urchins, on the other hand, graze on

seagrasses to gain energy for their physiological

function as well as reproduction. Few studies have

been conducted in the Philippines to observe or

investigate the influence of herbivore exclusion

on the growth of marine plants like seagrasses. In

general, herbivores (such as rabbitfish and sea

urchins) are widely acknowledged as critical com-

ponents that determine the structure and resilience

of marine ecosystems (Teresa and Simone 2002;

Burkepile and Hay 2006). Herbivores facilitate

healthy seagrass meadows and nutrient recycling

by grazing (Duarte 2002) and can strongly influ-

ence ecosystem organization and functioning

(Steneck and Dethier 1994; Vinueza et al. 2006;

Macusi 2012). Plant-herbivore interactions have

substantial ecological and evolutionary conse-

quences but have traditionally been overlooked in

marine higher plants (Verges et al. 2007). Investi-

gations of plant-herbivore relationships, such as

fish grazing, are essential for understanding the

interconnectivity of different ecosystems. This

study aimed to investigate how herbivore exclu-

sion influences the growth and abundance of sea-

grass, particularly Thalassia hemprichii in

Guang-guang, Davao Oriental, Philippines. No

previous studies have been conducted on fish

grazing in this location. Therefore, it is important

to carry out this assessment as it will serve as

baseline information for further ecological studies

related to fish grazing.

MATERIALS AND METHODS

Description of the study area

The study was conducted from March to May

2018 in Guang-guang, Dahican, Mati City, Davao

Oriental, Philippines, where natural mangrove

forests and rehabilitated mangroves can be found

(Figure 1). The current mangrove area planted in

the study site covers 77 ha, with its wide intertidal

area covered by seagrass beds, sandy and rocky

substrate (Nallos and Macusi 2023). Seagrass

species found in the area include Cymodocea

rotundata, Cymodocea serrulata, Enhalus

acoroides, Halodule pinifolia, Halodule univervis,

Halophila minor, Halophila ovalis, Syringodium

isoetifolium and T. hemprichii. Seagrass species T.

hemprichii and C. rotundata are predominant in

the current study site (De Troch et al. 2008;

Jimenez 2016; Terayama et al. 2022).). This study

was established in a pre-selected seagrass bed of

T. hemprichii to better understand the ecology of

this vital ecosystem in a controlled manner in

Guang-guang mangrove sanctuary, which is part

of a Marine Protected Area (MPA) in Davao Ori-

ental (6° 54'59"N-126° 15'10"E). This area was

chosen because of its gently sloping seagrass

cover known for its abundance of siganids. More-

over, this area is undisturbed by fishermen or

gleaners and supports a rich diversity of marine

flora and fauna, including various species of man-

grove trees and seagrass beds (Abreo et al. 2020).

Exclusion experiment

This study used caging experiments to deter-

mine the effect of herbivore exclusion on sea-

grass. Cages measuring 0.5 ´0.5 m were placed

in the sampling site using a plastic chicken net

with a mesh size of 1 cm in diameter. The full

cage (FC) treatment was overlaid with a wire

mesh roof tied with a cable tie on top of the

221

MACUSI ET AL.: FISH HERBIVORY ON SEAGRASS MEADOWS

fenced area. In this treatment, it was expected that

T. hemprichii would be free from grazing effects

of large fishes, except for juvenile fish, which can

enter the mesh size (<1 cm in diameter). The par-

tial cage (PC) has an open top portion, which con-

stitutes a procedural control that allows rabbitfish

and other juvenile fishes to feed from T.

hemprichii found inside the fenced area. The

open cage (OC) was distinguished from the two

other cages by using a pegged stick with marker

flags anchored to the substrate ground floor of T.

hemprichii bed to allow all herbivores, including

sea urchins and rabbitfish, to feed on the shoots of

T. hemprichii. Four replicates of each treatment

(12 replicates in total), were randomly distributed

on the seagrass bed and arranged close to each

other with two-meter intervals (Figure 2).

Fish visual census

In order to quantify the abundance of rabbitfish

and other fish species in the study site, fish visual

census (FVC) was used. A visual census was per-

formed only when the water was clear and with a

minimum distance of 5 m from the treatment

cages with a 15 m transect length. To do this,

researchers counted and identified the fish abun-

dance over the 15 m transect line to determine the

density of rabbitfish and other fish species found

in the area during high tide. By using a slate,

names and sizes (cm) of fish species were record-

ed, including the time of starting and finishing

diving. This was performed during high tide near

T. hemprichii bed of the study area. The FVC last-

ed from 15-20 min per transect. There were two

perpendicular transect lines during each sampling

period.

Shoots of T. hemprichii were initially tagged

during the first sampling using a colored thread

and then measured using a tape measure for

counting and length measurement. A total of 120

shoots of T. hemprichii were monitored and

measured in all cages. To quantify the effects of

herbivore exclusion on seagrass, the number of

shoots was counted, and shoot length was meas-

ured separately for the tagged shoots of T.

222 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

Figure 1. Study area in Guang-guang, Dahican, Mati City, Philippines, indicating the location of caging experiments.

hemprichii in the full cage, partial cage, and open

cage, respectively. Initial shoot counts for T.

hemprichii were determined during the first sam-

pling, and final cover was counted during the last

sampling. These data were later used to compare

different treatments in terms of the number of

shoots, but no attempt to statistically compare the

data was done because of the sparse number of

unit observations. Tagged individual shoots of T.

hemprichii in each treatment were measured

repeatedly during the weekly monitoring to the

study site. Initial and final shoot length measure-

ments were noted and compared in all treatments.

Seagrass counting and measurement of shoots

were done during the low tide period. During data

collection, observations on the abundance and

density of rabbitfish and other fish species found

in the sampling site were performed during high

tide using FVC. On the other hand, counting and

measuring of leaf shoots happened during low

tide. The observation started during the daytime

cycles of high and low tide, usually from 6:00 am

to 5:00 pm.

Taxonomic identification of fishes

Fishing nets were used to collect individual

fish samples, which were subsequently pho-

tographed and identified based on their morpho-

logical characteristics, such as color, size or

length, eyes, fin size, fin color, tail size, tail color,

and other bodily markings according to Lieske

and Myers (2002) and (Froese and Pauly 2020).

Data analysis

The standard formula to calculate the relative

abundance of taxa was applied to rabbitfish and

other herbivorous fishes in the sampling site. Rel-

ative abundance (%RA) was calculated as:

223

MACUSI ET AL.: FISH HERBIVORY ON SEAGRASS MEADOWS

Figure 2. Different caging treatments in the study site (A), full cage (B), partial cage (C), and open cage (D).

Number of individuals per species

%RA =´ 100

Number of individuals of all species

Fish density was calculated by getting the

number of individuals per unit area =ni/A, where

niis the total number of individuals of a species

found in the transect, and A is the total length of

two transects times (15 ´15 m) the width of

vision (5 m) of the researcher during the FVC.

The area was calculated based on the transect

used, and the width of the area swam by the

researcher from one end to the other.

The relative density was quantified using the

number of individuals of a given species over the

total density of all species. To analyze the relative

density of rabbitfish and other herbivorous fish in

the study area, the formula Di/ƩD was applied,

where Diis the density of a species A and ƩD is

the density of all species

Minimum and maximum ﬁsh lengths for each

species were estimated through observation dur-

ing the FVC, with a minimum size set to 1 cm

total length. Average length was used to estimate

a biomass index (kg) of all species according to

Bi=LiNi/1,000 g (Taquet et al. 2007), where Li is

the mean length of the species i(in cm), Ni is the

number of fish counted for the species i, and

1,000 is for the conversion of g to kg.

ANOVA was used to find differences in total

weekly abundance during the 12 weeks of FVC

and to compare shoot number length among treat-

ments (full, partial, and open cages). Data were

checked for the assumptions of ANOVA. Non-

transformed data were checked for normality and

Kolmogorov-Smirnov (KS) test for equal vari-

ance. When this violated the assumptions of

ANOVA, data were fourth-root transformed and

again checked for its normality. The FVC count

data were also analyzed using one-way ANOVA

to compare the abundance of the species during

the 12 weeks of observation. When data showed

significant differences, this was further explored

by using post hoc Tukey test and Fisher’s LSD.

Means that differed and those grouped together

according to their means were recorded. A one-

way ANOVA was also used to compare treat-

ments and determine if there was variation

between shoot length growths. Shoot length data

were first explored and then check for violations

of ANOVA assumptions. Then, if this did not sat-

isfy the assumptions of ANOVA, a data transfor-

mation was used (e.g. fourth root transformation,

log10 transformation, or square root transforma-

tion). Furthermore, when data transformation did

not work out, untransformed data were used

because ANOVA can be robust to violations of

the assumptions of ANOVA given large sample

sizes (Quinn and Keough 2002).

In addition, initial and final shoot length meas-

urements from treatments were also compared to

determine differences using one-way ANOVA.

Mood’s median test was also used owing to the

small number of replicates per treatment and two

instances of counting, initial versus final. All sta-

tistical analyses were performed using Minitab

version 17 (Minitab Inc., Pennsylvania, USA).

RESULTS

Species composition

A total of 13 fish species were found in the

sampling site. These species were categorized into

6 families, namely: Siganidae (2), Labridae (4),

Acanthuridae (1), Pomacentridae (3), Chaetodon-

tidae (2), and Apogonidae (1).

First-level carnivores consuming benthic inverte-

brates

The feeding levels of the 13 species were clas-

sified based on their feeding habit. Siganus

fuscescens (Siganidae), Siganus guttatus (Sigan-

idae), and Acanthurus triostegus (Acanthuridae)

grazed mainly on algal turfs and seagrasses.

There were five other fish species classified as

224 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

omnivores: Chaetodon auriga (Chaetodontidae),

C. kleinii (Chaetodontidae), Plectroglyphidodon

leucozonus (Pomacentridae), P. dickii (Pomacen-

tridae), and Abudefduf vaigiensis (Pomacentri-

dae) which fed on algal turfs, benthic seaweeds,

crustaceans, zooplanktons and finfish.

Second-level carnivores consuming fish

Halichoeres margaritaceus (Labridae), H. hor-

tulanus (Labridae), Thalassoma jansenii (Labri-

dae), Stethojulis bandanensis (Labridae), and

Pristiapogon kallopterus (Apogonidae) were

classified as zooplanktivores mainly feeding on

crustaceans, zooplanktons and zoobenthos.

Relative abundance and density of fishes

Among different fish species observed in the

study site, S. fuscescens was the most abundant,

with 499 individuals (mean =41.76), followed by

S. guttatus with 153 individuals (mean =12.8),

then A. triostegus with 127 individuals (mean =

10.63). Compared to all other fish species found

in the study site, T. jansenii was the least abun-

dant fish species with only 13 individuals. These

species were recorded for their feeding levels,

whether an herbivore, carnivore or omnivore

(Table 1).

The ANOVA result of the total weekly abun-

dance of fish density in the study area showed no

significant differences. However, a comparison of

fish count in terms of the abundance of various

fish species showed highly significant differences

(df =12, MS =2.88, F =30.10, p <0.001). The

post hoc analysis showed different means: S.

fuscescens had the highest mean count (40.97)

compared to all the other species, followed by S.

guttatus (12.49), A. triostegus,and P. kallopterus

(10.27 and 7.41), A. vaigiensis (5.92), H. margar-

itaceus (5.34), P. leucozonus (4.30), C. auriga

(2.22), P. dickii (1.57), and S. bandanensis (1.31).

Three species had similar fish count means: H.

hortulanus (0.63), C. kleinii (0.52), and T.

jansenii (0.43).

Fish length and biomass index of fishes

The estimated fish length ranged from 5 to

12 cm. H. hortulanus and S. bandanensis had the

highest estimated length (12 cm), followed by S.

fuscescens,S. guttatus,H. margaritaceus, and T.

jansenii (10 cm) while C. kleinii,P. leucozonus,

P. dickii, and P. kallopterus were the smallest (5

cm). The mean length of fish species was derived

from the computation of minimum and maximum

fish length with a range of 3.0 to 6.5 cm. In terms

of their biomass index, S. fuscescens recorded the

highest biomass with 2.47 kg, followed by S. gut-

tatus with 0.84 kg, while C. kleinii had the lowest

biomass obtained from their feeding activity with

0.05 kg.

Seagrass (Thalassia hemprichii) cover and shoot

data

In the full cage, all shoot counts in each repli-

cate increased in number from five to ten individ-

ual shoots. Some of these new individual shoots

were shorter than the others in this treatment. In

the second treatment, the partial cage, replicates

one and two also increased in the number of

shoots in the final counts, while replicates three

and four decreased in the number of individual

shoots in the final counts. However, in the open

cage treatment, all individual shoot counts

decreased in their number from five to ten indi-

viduals in the final count. The test showed that

initial counts were not different between treat-

ments as shown by Mood’s median test: X2=2, df

=2, p =0.368; FC =PC =OP, but the final shoot

counts showed significant differences (X2=8, df

=2, p = 0.018; FC >PC >OP). Median values

showed highest final shoot counts in the full cage

(170 shoots) compared to the partial cage (156

shoots) and the open cage (38 shoots) (Figure 3

A). By using these median values, it was estimat-

ed that fish predation damage comparing full and

partial cages was about 8%, while using average

values this figure could be 10%. Although full and

225

MACUSI ET AL.: FISH HERBIVORY ON SEAGRASS MEADOWS

open cages can be compared, there is a high prob-

ability that predation in the open cage is not only

carried out by fish, but also by other invertebrates.

Nonetheless, comparison values between the full

cage and open cage using the median count will

put that to 77%, while if average values are used,

that number could be 22%.

Among mean shoot length treatments, the full

cage recorded the highest mean length (16.75 cm),

followed by the partial cage treatment (11.81 cm),

and the open cage with the shortest length

(8.86 cm). In terms of replicates, full cage repli-

cate two registered the longest length at 16.86 cm,

followed by replicate one with a length of 16.82

cm, and then replicate four with a length of 16.76

cm and followed by replicate three with the short-

est length measured at 16.54 cm. In the partial

cage, replicate two was the longest, measured at

12.13 cm, followed by replicate four at 11.88 cm,

then replicate one with a length of 11.68 cm, while

replicate three recorded the shortest length, meas-

ured at 11.57 cm. In the open cage, replicate four

was the longest (8.96 cm), followed by replicate

one (8.92 cm), then replicate three (8.88 cm), and

last was replicate four (8.71 cm) (Figure 3 B).

Highly significant (p <0.001) results from

weekly shoot length of T. hemprichii from the

full, partial and open cages were observed. The

shoot length measured in the full and partial

cages exhibited a weekly increased since the ini-

tial week. The open cage recorded the opposite,

and the plotted chart showed a decreasing trend in

growth.

Comparison of the initial shoot length of T.

hemprichii among treatments showed differ-

ences, with the full cage recording initial mean

226 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

Table 1. Relative abundance, density, relative density, fish length, biomass and trophic and feeding levels of fish species found

in the study area.

Fish length

Species Relative Density Relative Mean Max. Min. Biomass Trophic Feeding

abundance (ni/A) density index (kg) level* level*

(Di/ƩD)

Siganus fuscescens 41.76 83.17 0.42 5.5 10 1 2.47 2 Herbivore

Siganus guttatus 12.80 25.5 0.13 5.5 10 1 0.84 2 Herbivore

Halichoeres margaritaceus 5.61 11.17 0.06 5.5 10 1 0.37 >3 Omnivore

Halichoeres hortulanus 1.51 3.00 0.02 6.5 12 1 0.12 >3 Omnivore

Thalassoma jansenii 1.09 2.17 0.01 5.5 10 1 0.07 3.1 Omnivore

Stethojulis bandanensis 1.84 3.67 0.02 6.5 12 1 0.14 3.2 Omnivore

Acanthurus triostegus 10.63 21.17 0.11 3.5 6 1 0.44 2 Herbivore

Chaetodon auriga 2.34 4.67 0.02 3.5 6 1 0.10 3.7 Omnivore

Chaetodon kleinii 1.34 2.67 0.01 3.0 5 1 0.05 3.5 Omnivore

Plectroglyphidodon 4.52 9.00 0.05 3.0 5 1 0.16 2 Omnivore

leucozonus

Plectroglyphidodon dickii 2.18 4.33 0.02 3.0 5 1 0.08 3.7 Omnivore

Abudefduf vaigiensis 6.78 13.50 0.07 4.5 8 1 0.36 2.6 Omnivore

Pristiapogon kallopterus 7.62 15.17 0.08 3.0 5 1 0.27 3.6 Planktivore

*Taken from Froese and Pauly 2020.

values of 9.97 cm, followed by the open cage

with a value of 9.93 cm and the partial cage with

9.74 cm. The post hoc Fisher’s pairwise compar-

ison revealed that shoot lengths in the full and

open cages were similar but the partial cage was

lower, e.g. OP =FC >PC (df =2, MS =0.6, F =

3.41, p =0.036) (Table 2).

Results of the final assessment of shoot length

of T. hemprichii in all cages were highly signifi-

cant. The post hoc analysis using Fisher’s pair-

wise comparison showed that the full cage treat-

ment obtained the highest shoot length (22.24

cm) compared to the partial (13.49 cm) and open

cages (7.24 cm), e.g. FC >PC >OP (df =2, MS

=2271.54, F =3476.46, p <0.001) (Table 2).

DISCUSSION

A total of 13 marine fish species belonging to

six families were observed in the area. Fish

species commonly occur in seagrass beds in the

Philippines (Salita et al. 2003). Many of these

fish species use the seagrass beds either through-

out their life (residents) or only as a nursery for

younger individuals of the species (Nagelkerken

et al. 2000). Members of the Family Labridae and

Siganidae are frequently observed in seagrass

beds not only in the local area but also in other

parts of the country (Salita et al. 2003). The

Genus Siganus is herbivorous and considered res-

ident of seagrass beds (Randall and Kulbicki

227

MACUSI ET AL.: FISH HERBIVORY ON SEAGRASS MEADOWS

Figure 3. Initial and final mean shoot counts of seagrass (A)

and mean shoot length measurements of Thalassia

hemprichii under different treatments (B).

Table 2. ANOVA results on weekly shoot length increase of Thalassia heprichii in full cage, partial cage, open cage, and for all

treatments.

Full cage Partial cage Open cage All treatments

df MS F* MS F* MS F* df MS F** MS F*

Weeks 11 726.71 1,633.1 96.85 140.28 42.80 181.99 2 0.60 3.41 2,271.54 3,476.46

Error 468 0.45 0.69 0.24 117 0.18 0.65

Total 479 119

*p <0.01; **p <0.5.

100

120

140

160

180

200

Full cage Partial cage Open cage

Count

Treatments

Initial Final

Full cage Partial cage Open cage

Shoot length (cm)

Treatments

2005; Jumawan-Nanual and Metillo 2008; Metil-

lo et al. 2016) which explain its high density. Her-

bivores may play an essential role in mitigating

impacts of increased N2as nutrient by interacting

with primary productivity (Gruner et al. 2008).

Furthermore, herbivores can positively impact on

seagrass by consuming nutrient-rich algae

(McSkimming et al. 2015). A frequent natural

disturbance in seagrass meadows is herbivory,

which shows that grazing generally reduces sea-

grass growth (Hughes et al. 2004). Intensive graz-

ers can severely reduce seagrass production and

distribution to the point of degradation in mead-

ows, often known as overgrazing event (Peterson

et al. 2002).

Rabbitfish in the area is a crucial species in

determining the structure and resilience of sea-

grass aquatic ecosystems (Burkepile and Hay

2007). Understanding and quantifying the impact

of herbivores on the ecosystem is, therefore, fun-

damental to our ability to devise strategies for

protecting these ecosystems (Mumby et al. 2006).

The most abundant fish species in this study

were S. guttatus and S. fuscescens. Significant

differences in length and number of T. hemprichii

shoots in the full cage and open cage set-ups were

attributable to lesser grazing pressure due to pro-

tection offered by the cage from herbivores.

Restricting the access of herbivorous fish in the

full cage allowed the seagrass to increase in shoot

number and length. Alcoverro and Mariani

(2000) observed lesser fish grazing in seagrass

beds with fewer herbivorous fish, a scenario in

which the full cage treatment emulates. Signifi-

cant differences in T. hemprichii shoot count and

length measurement from week 1 to 12 showed

evidence of high grazing intensity of rabbitfish in

all replicates of treatment cages in the sampling

area. This grazing was observed closely in the

slow growth of T. hemprichii particularly in par-

tial and open cages.

While the importance of grazing in algal-domi-

nated communities is well documented (Macusi

2010), the premise that ingesting living seagrass

228 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

biomass is infrequent and inconsequential remains

a central tenet of current food web theory (Cebrian

2002). This idea has been so widely accepted that

it is found in most marine ecology textbooks in

North America. As an example, Levinton (2001)

asserts in his widely used marine biology textbook

that ‘one of the most interesting aspects of sea-

grasses is their apparent unpalatability to grazers’.

Our observation of the impact of rabbitfish graz-

ing, which affected the growth and cover of T.

hemprichii in the control treatment (partial cage)

and the open cage (marked grazing plot) compared

to the full cage, contradicts this statement. The

result of higher shoot count and longer shoots

found in the full cage supports the hypothesis that

the higher density of rabbitfish in the area poten-

tially reduced the chance for higher growth of T.

hemprichii in the partial and open cages.

In Inner Ambon Bay, white-spotted rabbitfish

Siganus canaliculatus correlates with available

seagrass meadows (Latuconsina et al. 2013;

2020a). This fish matures at six months of age, has

a high exploitation and mortality rate (Latuconsina

et al. 2020b), and gives a high grazing pressure on

seagrasses (Irawan and Nganro 2016; Noya et al.

2016). S. canaliculatus also affects seagrass vege-

tation, as it is necessary for its growth and repro-

ductive biology (Madduppa et al. 2019). Rabbit-

fishes are tropical herbivorous fish species and

highly dependent on benthic algae and seagrasses

(Simora et al. 2015).

CONCLUSIONS

This study contributes to the rising evidence

that grazing activity of rabbitfish on seagrass beds

strongly influences the growth of T. hemprichii.

Of the various fish species found in the study area,

two species of rabbitfish (S. fuscescence and S.

gutatus) were the most abundant. These species

were observed engaging in intense grazing activi-

ty, which strongly influenced the growth of sea-

grass. The hypothesis that rabbitfish grazing on T.

hemprichii beds is a significant cause of

decreased shoot cover and length in the study

area is strongly supported by our findings.

ACKNOWLEDGMENTS

This research has been possible thanks to the

joint efforts of La Nina Ubal, Darshell Estor, Nit-

cel Aymie Albarido, Dolly Pera, Christian Dave

Tipudan for help in the field set-up and during the

sampling time. We also extend our gratitude to Dr

Girley S. Gumanao of the Davao Del Norte State

College for his study’s taxonomic validation of

the fish species. The map was also possible

thanks to the help of Remie S. Aurelio of UP Min-

danao.

REFERENCES

ABREO NAS, MACUSI ED, JIMENEZ LA. 2018. A

survey of subtidal anthropogenic marine

debris (AMD) in Mayo Bay, Mati City, Davao

Oriental, Philippines. Philipp J Sci. 147: 597-

600.

ABREO NAS, SIBLOS SKV, MACUSI ED. 2020.

Anthropogenic marine debris (AMD) in man-

grove forests of Pujada Bay, Davao Oriental,

Philippines. J Mar Isl Cult. 9: 38-53.

ALCOVERRO T, MARIANI S. 2000. The effects of

the sea urchin grazing on the Thalassodendron

ciliatum seagrass beds of a Kenyan lagoon.

Mar Ecol Prog Ser. 226: 255-263.

AZZURRO E, FANELLI E, MOSTARDA E, CATRA M,

ANDALORO F. 2007. Resource partitioning

among early colonizing Siganus luridus and

native herbivorous fish in the Mediterranean:

an integrated study based on gut-content

analysis and stable isotope signatures. J Mar

Biol Assoc UK. 87: 991-998.

BARICHE M. 2006. Diet of the Lessepsian ﬁshes,

Siganus rivulatu, and S. luridus (Siganidae) in

the eastern Mediterranean: a bibliographic

analysis. Cybium. 30: 41-49.

BJORK M, SHORT FT, MCLEOD E, BEER S. 2008a.

Managing seagrasses for resilience to climate

change. Gland: IUCN Resilience Science

Group Working Paper Series. 3. 56 p.

BURKEPILE DE, HAY ME. 2006. Herbivore vs.

nutrient control of marine primary producers:

context-dependent effects. Ecology. 87: 3128-

3139.

BURKEPILE DE, HAY ME. 2007. Predator release

of the gastropod Cyphoma gibbosum increases

predation on gorgonian corals. Oecologia.

154: 167-173.

CAMPOS WL, DEL NORTE A, CAMPOS GC, MCMA-

NUS JW. 1994. Yield estimates, catch, effort

and fishery potential of the reef flat in Cape

Bolinao, Philippines. J Appl Itchthyol. 10: 82-

95.

CEBRIAN J. 2002. Variability and control of car-

bon consumption, export, and accumulation in

marine communities. Limnol Oceanogr. 47:

11-22.

CHEAL A, EMSLIE M, MILLER I, SWEATMAN H.

2012. The distribution of herbivorous fishes

on the Great Barrier Reef. Mar Biol. 159:

1143-1154.

DETROCH M, MELGO JL, JIMENEZ L, GHEE-

RARDYN H, VINCX M. 2008. Diversity and

habitat selectivity of harpacticoid copepods

from seagrass beds in Pujada Bay, The Philip-

pines. J Mar Biol Assoc UK. 88: 515-526

DUARTE CM. 2002. The future of seagrass mead-

ows. Environ Conserv. 29: 192-206.

DUARTE CM, HEMMINGA MA. 2000. Seagrass

ecology. Cambridge: Cambridge University

Press. 312 p.

DUARTE CM, MIDDELBURG J, CARACO N. 2005.

Major role of marine vegetation on the ocean-

ic carbon cycle. Biogeosciences. 2: 1-8.

DRIFMEYER JE. 1981. Urchin Lytechinus variega-

tus grazing on Eelgrass, Zostera marina.

229

MACUSI ET AL.: FISH HERBIVORY ON SEAGRASS MEADOWS

Estuaries. 4: 374-375.

FORTES MD. 1995. Seagrass of East Asia: envi-

ronmental and management perspectives.

RCU/EAS Technical Report Series No. 6.

Bangkok: UNDEP. 79 p.

FORTES MD. 2004. Wetland conservation and

management in the Philippines: where are we

now? the case of seagrass and mangrove. In:

WONG MH, editor. Wetlands ecosystems in

Asia. Amsterdam: Elsevier. p. 233-262.

FOX RJ, BELLWOOD DR. 2008. Remote video

bioassays reveal the potential feeding impact

of the rabbitfish Siganus canaliculatus (f:

Siganidae) on an inner-shelf reef of the Great

Barrier Reef. Coral Reefs. 27 (3): 605-615.

FROESE R, PAULY D, editors. 2020. FishBase.

[accessed 2020 Feb]. https://www.fishbase.org.

GRUNER DS, SMITH JE, SEABLOOM EW, SANDIN

SA, NGAI JT, HILLEBRAND H, BOLKER BM.

2008. A cross‐system synthesis of consumer

and nutrient resource control on producer bio-

mass. Ecol Lett. 11 (7): 740-755.

HECK KL, VALENTINE JF. 2006. Plant-herbivore

interactions in seagrass meadows. J Exp Mar

Biol Ecol. 330: 420-436.

HOEY AS, BRANDL SJ, BELLWOOD DR. 2013. Diet

and cross-shelf distribution of rabbitfishes (f.

Siganidae) on the northern Great Barrier Reef:

implications for ecosystem function. Coral

Reefs. 32 (4): 973-984.

HUGHES AR, STACHOWICZ JJ, 2004. Genetic

diversity enhances the resistance of a seagrass

ecosystem to disturbance. Proc Natl Acad Sci

USA. 101: 8998-9002.

IRAWAN A, NGANRO NR. 2016. Distribution of

seagrass in Inner Ambon Bay. J Trop Mar Sci

Technol. 8 (1): 99-114.

JIMENEZ LA. 2016. Guide to common seagrass in

Pujada Bay. Davao Oriental State College of

Science and Technology. Mati City, Davao

Oriental, Philippines, Regional Integrated

Coastal Resource Management Center (RIC-

XI). [accessed 2023 Mar]. http://kulturalink.

nlp.gov.ph/cgi-bin/koha/opac-detail.pl?biblio

number=6860.

JUMAWAN-NANUAL B, METILLO EB. 2008. Popu-

lation structure and reproductive biology of

Siganus fuscescens Houttuyn 1782 (Perci-

formes, Siganidae) in Pujada Bay, Southeast-

ern Mindanao, Philippines. Philipp Sci. 45:

62-79.

KOCH E. 2001. Beyond light: physical, geological

and geochemical parameters as possible sub-

mersed aquatic vegetation habitat require-

ments. Estuaries. 24: 1-17.

KOIKE I, MUKAI H, NOJIMA S. 1987. The role of

the sea urchin, Tripneustes gratilla (Lin-

naeus), in decomposition and nutrient cycling

in a tropical seagrass bed. Ecol Res. 2: 19-29.

LATUCONSINA H, AFFANDI R, KAMAL MM, BUTET

NA. 2020a. Spatial distribution of white-spot-

ted rabbitfish Siganus canaliculatus Parak,

1797 on different seagrass beds habitat of the

Inner Ambon Bay. J Trop Mar Sci Technol. 12

(1): 89-106. Indonesian. DOI: https://doi.org/

10.29244/jitkt.v12i1.27908

LATUCONSINA H, AFFANDI R, KAMAL MM, BUTET

NA. 2020b. On the assessment of white-spot-

ted rabbitfish (Siganus canaliculatus Park,

1797) stock in the Inner Ambon Bay, Indone-

sia. AACL Bioflux. 13 (4): 1827-1835.

LATUCONSINA H, AMBO-RAPPE R. 2013. Daily

variability of fish community in seagrass beds

of Tanjung Tiram-Inner Ambon Bay. J Indones

Ichthyol. 13 (1): 35-53. Indonesian.

LEVINTON JS. 2001. Marine biology: function,

biodiversity, ecology. Oxford: Oxford Univer-

sity Press.

LIESKE E, MYERS R. 2002. Coral reef fishes:

Indo-Pacific and Caribbean. Princeton:

Princeton University Press.

MACUSI ED. 2010. Effect of different herbivore

guilds on the abundance and distribution pat-

terns of autotrophs on a semi-exposed tropical

rocky shore, Hong Kong. Philipp Sci. 47: 173-

203.

MACUSI ED. 2012. Grazing drives spatial varia-

tion in the abundance and distribution patterns

230 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)

of autotrophs in tropical rocky shore. Philipp J

Sci. 141: 141-156.

MACUSI ED, KATIKIRO RE, DEEPANANDA KHMA,

JIMENEZ LA, CONTE AR, FADLI N. 2011.

Human induced degradation of coastal

resources in Asia-Pacific and implications on

management and food resources. J Nat Stud.

9/10: 13-28.

MACUSI ED, TIPUDAN CD. 2020. Effects of bio-

turbation of fiddler crabs in relation to the

growth of mangrove saplings (Rhizophora

apiculata) in a mangrove reforested area. J

Mar Isl Cul. 9 (2): 1-10.

MCGLATHERY KJ. 1995. Nutrient and grazing

influences on a subtropical seagrass commu-

nity. Mar Ecol Prog Ser. 122: 239-252.

MADDUPPA H, MARGARETHA MK, BRAMANDITO

A, PRARTONO T, SUBHAN B, ARAFAT D, ANG-

GRAINI MP. 2019. Intraspecific genetic diversi-

ty and population subdivision of rabbitfish

(Siganidae: Siganus canaliculatus) in urban-

ized reefs of Jakarta Bay, Indonesia. Biodiver-

sitas. 20 (10): 2897-2902. DOI: https://doi.

org/10.13057/biodiv/d201017

MAYNAWANG IS, MACUSI ED. 2023. Catch assess-

ment of commercially important gastropods in

Guang-Guang, Mati City, Davao Oriental,

Philippines. Acad Biol. 1 (1). DOI: https://doi.

org/10.20935/AcadBiol6029

MCSKIMMING C, TANNER JE, RUSSELL BD, CON-

NELL SD. 2015. Compensation of nutrient pol-

lution by herbivores in seagrass meadows. J

Exp Mar Biol Ecol. 471: 112-118.

METILLO EB, CADELINIA EE, HAYASHIZAKI KI,

TSUNODA T, NISHIDA S. 2016. Feeding ecology

of two sympatric species of Acetes (Decapoda:

Sergestidae) in Panguil Bay, the Philippines.

Mar Freshwater Res. 67: 1420-1433.

MUMBY PJ, DAHLGREN CP, HARBORNE AR, KAP-

PEL CV, MICHELI F, BRUMBAUGH DR, HOLMES

KE, MENDES JM, BROAD K, SANCHIRICO JN, et

al. 2006. Fishing, trophic cascades, and the

process of grazing on coral reefs. Science.

311: 98-101.

NAGELKERKEN I, DORENBOSCH M, VERBERK

WCEP, DE LA MORINIERE CE, VAN DER VELDE

G. 2000. Day-night shifts of fishes between

shallow-water biotopes of a Caribbean Bay,

with emphasis on the nocturnal feeding of

Haemulidae and Lutjanidae. Mar Ecol Prog

Ser. 194: 55-64.

NALLOS IM, MACUSI ED. 2023. Behavior and diet

composition of fiddler crabs in Guang-guang,

Dahican, Mati City, Davao Oriental, Philip-

pines. Mar Fish Sci. 36 (2): 137-147. DOI:

https://doi.org/10.47193/mafis.3622023010506

NOYA YA, PURBA M, KOROPITAS AF, PRARTONO T.

2016. Cohesive sediment transport modeling

on Inner Ambon Bay. J Trop Mar Sci Technol.

8 (2): 671-687. Indonesian. DOI: https://doi.

org/10.29244/jitkt.v8i2.15834

PETERSON BJ, ROSE CD, RUTTEN LM, FOURQURE-

AN JW. 2002. Disturbance and recovery fol-

lowing catastrophic grazing: studies of a suc-

cessional chronosequence in a seagrass bed.

Oikos. 97: 361-370.

PICKHOLTZ SM, KIFLAWI M, FRIEDLANDER AM,

BELMAKER J. 2018. Habitat utilization by an

invasive herbivorous fish (Siganus rivulatus)

in its native and invaded range. Biol Inva-

sions. 20: 3499-3512.

QUINN GP, KEOUGH MJ. 2002. Experimental

design and data analysis for biologists. Cam-

bridge: Cambridge University Press.

RANDALL JE, KULBICKI M. 2005. Siganus wood-

land, new species of rabbitfish (Siganidae)

From New Caledonia. Cybium. 29: 185-189.

SALITA JT, EKAU W, SAINT-PAUL U. 2003. Field

evidence on the influence of seagrass land-

scapes on fish abundance in Bolinao, northern

Philippines Mar Ecol Prog Ser. 247: 183-195.

SERIO D, ALONGI G, CATRA M, CORMACI M, FUR-

NARI G. 2006. Changes in the benthic algal

flora of Linosa island (Straits of Sicily,

Mediterranean Sea). Bot Mar. 49: 135-144.

SHEARS NT, BABCOCK RC. 2002. Marine reserves

demonstrate top-down control of community

structure on temperate reefs. Oecologia. 132:

231

MACUSI ET AL.: FISH HERBIVORY ON SEAGRASS MEADOWS

131-142.

SIMORA RMC, TRAFALGAR RFM, LEGARIO FS.

2015. Characterization of extracellular

enzymes from culturable autochthonous gut

bacteria in rabbitfish (Siganus guttatus).

ELBA Bioflux. 7: 67-76.

SMULDERS FOH, BECKER ST, CAMPBELL JE, BAK-

KER ES, BOÄSSON MJ, BOUWMEESTER MM,

VONK JA, CHRISTIANEN MJA. 2022. Fish graz-

ing enhanced by nutrient enrichment may

limit invasive seagrass expansion. Aquat Bot.

176: 103464.

SOLIMAN VS, MENDOZA JR. AB, YAMAOKA K.

2008. Seaweed-associated fishes of Lagonoy

Gulf of Bicol, the Philippines, with emphasis

on siganids (Teleostei: Siganidae). Kuroshio

Science. 2: 67-72.

STENECK RS, HACKER SD, DETHIER MN. 1991.

Mechanism determining competitive domi-

nance between crustose coralline algae: a her-

bivore-mediated reversal. Ecology. 72: 938-

950.

TAQUET M, DAGORN L, GAERTNER JC, GIRARD C,

AUMERRUDDY R, SANCHO G, ITANO D. 2007.

Behavior of dolphinfish (Coryphaena hippu-

rus) around drifting FADs as observed from

automated acoustic receivers. Aquat Living

Resour. 20: 323-330.

TERAYAMA K, MIZUNO K, TABETA S, SAKAMOTO S,

SUGIMOTO Y, S UGIMOTO K, FUKAMI H, SAKA-

GAMI M, JIMENEZ LA. 2022. Cost-effective

seafloor habitat mapping using a portable

speedy sea scanner and deep-learning-based

segmentation: a sea trial at Pujada Bay, Philip-

pines. Methods Ecol Evol. 13 (2): 339-345.

TERESA A, SIMONE M. 2002. Effects of sea urchin

grazing on seagrass (Thalassodendron cilia-

tum) beds of a Kenyan lagoon. Mar Ecol Prog

Ser. 226: 255-263.

VALENTINE JF, HECK KL. 2005. Perspective

review of the impacts of overﬁshing on coral

reef food web linkages. Coral Reefs. 24: 209-

213.

VALENTINE JFV, HECK JRKL. 1999. Seagrass her-

bivory: evidence for the continued grazing of

marine grasses. Mar Ecol Prog Ser. 176: 291-

302.

VERGES A, BECERRO MA, ALCOVERRO T, ROMERO

J. 2007. Variation in multiple traits of vegeta-

tive and reproductive seagrass tissues influ-

ences plant-herbivore interactions. Oecologia.

151: 675-686.

VINUEZA LR, BRANCH GM, BRANCH ML, BUSTA-

MANTE RH. 2006. Top-down herbivory and

bottom-up El niño effects on Galapagos rocky

shore communities. Ecol Monogr. 76: 111-

131.

VONK JA, PIJNAPPELS MHJ, STAPEL J. 2008. In

situ quantification of Tripneustes gratilla

grazing and influences on a mixed-species

tropical seagrass meadow. Mar Ecol Prog Ser.

360: 107-114.

VON WESTERNHAGEN H. 1973. The natural food of

the rabbitfish Siganus oramin and S. striolata.

Mar Biol. 22 (4): 367-370.

WILSON SK, BELLWOOD DR, CHOAT JH, FURNAS

MJ. 2003. Detritus in the epilithic algal matrix

and its use by coral reef fishes. Oceanogr Mar

Biol Annu Rev. 41: 279-309.

232 MARINE AND FISHERY SCIENCES 36 (3): 219-232 (2023)