OCEANOGRAPHIC PROCESSES OF CORAL REEFS: Physical and Biological Links in the Great Barrier Reef - Chapter 7 ppsx

94 Effects of Siltation on Seagrass Depth Distribution and Abundance.. It is hypothesized that the reduction in light availability brought about by siltation or sedimentation is the most

The Effects of Siltation

on Tropical Coastal

Ecosystems

Miguel Fortes

CONTENTS

Introduction 93

Responses of Seagrass to Siltation 94

Effects of Siltation on Seagrass Depth Distribution and Abundance 94

Effects of Siltation on Diversity, Biomass, and Survival 95

Effects of Siltation on Seagrass Growth and Primary Productivity 96

Effects of Siltation on Seagrass Morphology 98

Responses of Corals to Siltation/Sedimentation 100

Effects of Sedimentation on Coral Abundance, Diversity, and Distribution 100

Effects of Sedimentation on Coral Growth and Productivity 101

Modeling Reef Status and Sedimentation 101

Responses of Mangroves to Siltation 102

Conclusion 103

Effects of Siltation on Seagrass 103

Effects of Sedimentation on Coral Reefs 104

Effects of Siltation on Mangroves 105

References 105

INTRODUCTION

Coral reefs, seagrass beds, and mangroves are the major ecosystems in coastal Southeast Asia They are experiencing widespread deterioration, largely as a result of siltation (Fortes, 1988) During the past 25 years rates of siltation in the region have increased substantially and are among the highest in the world (Milliman & Meade, 1983; Milliman & Syvitski, 1992) These have been caused largely by human distur-bances such as land reclamation or changes in land use (Fortes, 1988 and 1995; Short

& Wyllie-Echeverria, 1996) The rapid progression of coastal development, near and offshore mining, agricultural land use, and deforestation have led to increasing silt load and eutrophication These brought about dramatic changes in the development 7

93

of coastal plant and animal communities in both tropical and temperate waters (Orth

& Moore, 1983; Cambridge & McComb, 1984; Onuf, 1994; Terrados et al., 1998) The aim of this chapter is to describe the changes in depth distribution, abun-dance, growth and photosynthetic performance, and morphological changes in sea-grasses and corals along siltation gradients In addition, the effect of siltation on the demography of mangrove seedlings is briefly discussed It is hypothesized that the reduction in light availability brought about by siltation or sedimentation is the most operationally significant factor forcing changes in species composition and commu-nity distribution along gradients of siltation Hence, at less perturbed sites, a change

in species composition along a gradient should parallel a similar change with increas-ing depth

RESPONSES OF SEAGRASS TO SILTATION

Seagrasses are submerged angiosperms that can fulfil their entire life cycle under water, forming extensive meadows on sandy to muddy sediments in shallow coastal waters (den Hartog, 1970; Valiela, 1984) Among the most productive components of coastal ecosystems (Hillman et al., 1989), these meadows are an important link between land and ocean (Holligan & de Boois, 1993; Hemminga et al., 1994) and support a high primary production (Valiela, 1984; Hillman et al., 1989; Duarte, 1989) Seagrass leaves and stems add considerable three-dimensional structure to the seabed, providing habitat, feeding, and breeding grounds as well as nurseries for a diverse array of fauna (e.g., sirenians, birds, fish, and invertebrates: Jacobs et al., 1981; Bell & Pollard, 1989; Howard et al., 1989; Klumpp et al., 1993) Seagrass meadows also act as sediment traps (Bulthuis et al., 1984; Ward et al., 1984; Fonseca

& Fisher, 1986; Fonseca, 1989) and as breakwaters offering natural shoreline protec-tion (Fonseca et al., 1982; Hemminga & Nieuwenhuize, 1990)

The effects of siltation on seagrasses are manifested in their depth distribution, abundance, species composition, growth, primary productivity, and changes in mor-phology These changes are briefly discussed below

E FFECTS OF S ILTATION ON S EAGRASS D EPTH D ISTRIBUTION

Seagrass beds are subject to both direct and indirect influences of man’s interference

in the coastal zone Urbanization, large-scale reclamation and shore protection works, increased sediment delivery by rivers draining watersheds with changing land-use practices, eutrophication, and increased fishing pressure have severely affected the depth distribution, density, and areal extent of seagrass meadows (Cambridge et al., 1986; Fortes, 1988; Shepherd et al., 1989; Giesen et al., 1990; Holligan & de Boois, 1993; Lundin & Linden, 1993)

Distribution and abundance of seagrasses are controlled by a range of environ-mental conditions including light availability (Dennison & Alberte, 1985; Dennison, 1987), nutrient availability (Short, 1987), water motion (Fonseca & Kenworthy, 1987), and grazing (Lanyon et al., 1989) Of these, light availability is considered one of the

more important environmental parameters, controlling the depth to which seagrasses can grow and excluding seagrasses from areas with low light conditions (Dennison

et al., 1993; Abal & Dennison, 1996; Bach, 1997; Bach et al., 1998; Duarte et al., 1997) Siltation is a major factor that limits light availability for benthic organisms The relationships between light conditions and depth distribution of temperate seagrasses clearly document that light availability is the prime regulating factor for plant performance (e.g., Bulthuis, 1983; Dennison, 1987; Olesen, 1996) Silt from rivers and land reduces underwater light penetration by increasing both light absorp-tion and scattering (Kirk, 1983; Onuf, 1994) Increase in nutrient load, similarly asso-ciated with an increase in silt load (Malmer & Grip, 1994), favors the growth of microalgae and epiphytes (Sand-Jensen & Borum, 1991; Duarte, 1995), thereby reducing light availability to seagrass In turn, reduced seagrass abundance decreases the ability of the plants to protect surface sediments (Fonseca et al., 1982), enhanc-ing sediment resuspension (Bulthuis et al., 1984) Deterioration of the underwater light climate for the remaining seagrass stands results

At Cape Bolinao, northwestern Philippines, the depth penetration of the mixed seagrass beds declined systematically with increasing siltation (Bach, 1997 and

1998; Terrados et al., 1998) At the control site, leaf growth of Thalassia hemprichii,

Cymodocea rotunda, and Cymodocea serrulata responded clearly to artificial

reduc-tion of light However, in natural stands of T hemprichii, C serrulata, and Enhalus

acoroides growing along the siltation gradient, there was no differential leaf growth

to variations in light regime They responded only moderately to reduced light with increasing depth

E FFECTS OF S ILTATION ON D IVERSITY , B IOMASS , AND S URVIVAL

While siltation smothers and buries benthic organisms (Duarte et al., 1997), at the same time it increases the nutrient load in both water and the sediments (Malmer & Grip, 1994; Mitchel et al., 1997) These changes in the water and sediment conditions are particularly detrimental for seagrasses (Giesen et al., 1990; Duarte, 1991; Sand-Jensen

& Borum, 1991; Duarte, 1995) At Cape Bolinao, the diversity of the mixed seagrass beds was reduced with increasing silt load (Bach et al., 1998) From the most to the least tolerant, the seagrass species could be ranked after their tolerance to siltation as:

Enhalus acoroides  Cymodocea serrulata  Halodule uninervis  Thalassia

hemprichii  Halophila ovalis  Cymodocea rotundata  Syringodium isoetifolium.

This sequential loss of species agrees well with that found in a related study among sea-grass beds along siltation gradients in the Philippines and Thailand (Terrados et al., 1998), suggesting that the sequence may represent a general pattern of tolerance to sil-tation among Southeast Asian seagrass species

At the initial phase under conditions of severe light reduction some seagrasses

exhibit a rapid loss of biomass Leaf densities of Heterozostera tasmanica (Bulthuis, 1983) and Posidonia sinuosa (Gordon et al., 1994) decreased by 70% during the first month of exposure to 2 and 1% of ambient light, respectively H pinifolia, on the

other hand, can survive long periods of light deprivation, a feature of great impor-tance for the species especially in the Southeast Gulf of Carpentaria (Australia)

which periodically receives monsoonal rains that result in highly turbid floodwaters covering over the seagrass beds (Shepherd et al., 1989)

In contrast to the high tolerance of H pinifolia, H ovalis has a low tolerance to

darkness, death occurring after only 38 days in the dark A similar intolerance to light

deprivation has also been demonstrated for monospecific H ovalis plants growing in

sub-tropical waters (Longstaff et al., 1999) This long-term survival strategy of

Halophila species to perturbations has also been suggested to occur elsewhere

(Kenworthy, 1992) The explanation is that seagrasses growing under reduced light conditions allocate a lower fraction of photosynthetic products to underground tis-sues (Madsen & Sand-Jensen, 1994) and formation of new shoots This results in low shoot density For the remaining shoots, however, light becomes more available because of a concomitant reduction in self-shading among them Prolonged condi-tions of improved (Williams, 1987) or reduced light availability (Zieman et al., 1989) induce changes in shoot density and biomass

Species loss may also result indirectly from the effect of siltation on sediment grain size, resuspension, and resistance to oxygen diffusion (Duarte et al., 1997) Fine-grained sediments are more readily resuspended, and therefore seagrass beds in

silted areas more often experience partial burial The large Enhalus acoroides and species which grow profusely via their vertical rhizomes (e.g., Cymodocea serrulata)

can comparatively tolerate both silt and burial (Vermaat et al., 1997), while smaller

species (e.g., Halodule uninervis and Syringodium isoetifolium) cannot survive

bur-ial (Duarte et al., 1997) Seagrasses also respond differently to changes in redox potential of the sediment, depending on their morphology and ability to maintain root oxygen supply (Smith et al., 1988) Hence, the integrated response of mixed seagrass beds to siltation is likely to be caused by changes in both water and sediment condi-tions it brings about

E FFECTS OF S ILTATION ON S EAGRASS G ROWTH

The relationships between siltation, the light conditions it brings about, and growth and photosynthesis of seagrasses clearly demonstrate that light is the prime factor regulating plant performance (e.g., Bulthuis, 1983; Dennison, 1987; Olesen, 1996) Seagrasses generally require a higher quantity of light in comparison to other marine and terrestrial flora (Dennison et al., 1993; Duarte, 1991; Abal et al., 1994) However,

as to the actual amount of light required for long-term survival, more studies have yet

to be done Estimates of light requirements of seagrasses differ between species (e.g., 4.4 to 29% of surface light) and within a species (e.g., 5 to 20% of surface light) (Dennison et al., 1993), while an average requirement of seagrasses as a group of plants has been calculated to be 11% of surface light (Duarte, 1991)

In tropical seas, productivity of shallow seagrass stands seems to be limited largely by the availability of nutrients (e.g., Agawin et al., 1996) However, nutrient availability is low in fine-grained carbonate sediments (Short et al., 1985; Short, 1987) but increases in coarse-grained carbonate and terrigenous sediments (Erftemeijer, 1994; Erftemeijer & Middelburg, 1995) These findings suggest that not all tropical seagrass meadows might be nutrient limited (Erftemeijer et al., 1994) The

nutrient status of seagrasses, however, may be reduced by a decrease in the availabil-ity of light (Abal et al., 1994), thereby reducing the nutrient requirements of some tropical seagrasses

Three photosynthetic parameters have been found to respond strongly to both the gradient in natural light and light deprivation, conditions which are associated with

siltation: chlorophyll a-to-b ratio, leaf amino acid concentration, and leaf 13C value

Decreasing chlorophyll a:b with depth has been observed in a number of seagrass species including Zostera marina (Dennison & Alberte, 1985), H ovalis (Longstaff

et al., 1999), Halophila spp., Halodule spp., Syringodium filiforme, and Thalassia

testudinum (Wigington & McMillan, 1979; Lee & Dunton, 1997) A decrease in the

chlorophyll a:b has been considered an adaptive response that increases the light

absorption efficiency of seagrass (Abal, 1996; Lee & Dunton, 1997)

Changes in amino acid concentrations in seagrasses are brought about by a num-ber of environmental variables Two of these which are associated with siltation are water depth and nutrient addition Depth has been shown to affect amino acid

con-centrations in Posidonia oceanica (Pirc, 1984), although this was not in the case of

Thalassodendron ciliatum (Parnik et al., 1992) Ambient sediment nutrient

concen-tration and sediment nutrient addition can also have a significant effect on amino acid concentrations (Udy & Dennison, 1997a and b) The increase in concentration at depth is linked to a response to reduced light availability and could be related to the balance of nutrient against light limitation of seagrass growth, the light condition bringing about the elevated amino acid content in the plants

In response to shading and increased water depth, the carbon isotope ratio (13C)

of H pinifolia leaves became more negative (Abal, 1996; Grice et al., 1996;

Longstaff et al., 1999) This may be due to a more rapid uptake of 12C in relation to 13

C, the preferential rate occurring because 12C uptake requires less energy in com-parison to 13C (Abal & Dennison, in press; Grice et al., 1996; Longstaff et al., 1999) Whether reductions in light availability have significant effects on seagrass growth and survival depends primarily on the efficiency with which light energy is used in the autotrophic accumulation of plant biomass These are often described using photosynthesis-irradiance (or P-1) curves (Drew, 1979) Species that are able

to physiologically acclimate to reduced light by adjusting their P-1 curves will have

a better chance to survive severe siltation events

Measured P-1 curves of different species revealed that in the Philippine Enhalus

acoroides and Thalassia hemprichii, variation in the compensation depth (i.e., the

depth at which daily respiratory demand and photosynthetic oxygen supply are just

in balance) with water depth and turbidity correlates well with predicted maximum colonization depth An important observation may then be derived from the colo-nization depth-turbidity curve which would suggest that small reduction in water clarity may dramatically affect seagrass performance in relatively clear waters of the Mediterranean and the Philippines (K0.5 m1

) Furthermore, it would suggest that moderately eutrophicated waters subjected to increases in turbidity may not allow seagrasses to colonize deeper parts

Recent shading studies have shown that the survival period of a seagrass below the minimum light required may be altered by adaptations in photosynthetic

parame-ters (e.g., increased chlorophyll content, changes in the chlorophyll a:b ratio,

increased canopy height and shoot thinning: Abal et al., 1994; Lee & Dunton, 1997).

This was the case with H pinifolia which demonstrated an increased chlorophyll con-tent, decreased chlorophyll a:b ratio, and an increased canopy height under

condi-tions of shading (Longstaff et al., 1999)

Traditionally changes in the morphology of seagrasses have been used as an

indica-tor of an adverse environmental effect on a seagrass community (e.g., Posidonia

sin-uosa, shoot density and leaf length) (Gordon et al., 1994) In the present study, the

morphological responses to siltation under consideration include decreases in bio-mass, shoot density, and canopy height It should be noted, however, that physiolog-ical responses can detect declining seagrass health and impending seagrass die-off before substantial morphological changes occur

Sediment dynamics over a seagrass bed may range from a gradual, continuous deposition to a sudden storm-related event (Marba et al., 1994a), and from a homo-geneous rate over large areas to small-scale variability associated with sand ripples

or dunes (Marba et al., 1994b) Seagrasses may respond to the latter via horizontal rhizome growth On the other hand, the species respond to homogeneous sedimenta-tion rates only via vertical stem elongasedimenta-tion or re-establishment from seeds

Vertical stem growth, even in Philippine seagrasses, has been shown to be sea-sonal (Duarte et al., 1994; Vermaat et al., 1995): during the growing season, longer internodes are formed and this often also occurs at a higher rate than at other, less favorable times of the year (Duarte et al., 1994) It is probable that the capacity of sea-grasses to respond to burial may also be seasonal, and off-season siltation may have more dramatic effects than expected Genera without differentiated vertical stems may respond with a redirection of the horizontal rhizome to survive excessive silta-tion and burial

Shoot size is an obvious determinant of the chance to survive a burial event:

larger shoots are simply less easily buried The largest Philippine species Enhalus

acoroides, for example, has horizontal rhizome branches that curve upward to

posi-tion the apical meristems at an average distance of 10 cm from the main rhizome, which is generally several centimeters above the sediment With full-grown leaves measuring about 80 cm, the leaf canopy reaches considerably further upward

(Vermaat et al., 1995) Halophila ovalis, the smallest Philippine species, also lacks

vertical stems, but its oval leaf blades have petioles that may reach a length of 2 cm,

a height that allows a substantial short-term sediment deposition rate over the short shoot life span of this species (1 to 2 weeks) (Duarte, 1991; Vermaat et al., 1995) For species that do have vertical stems, considerable variation exists in annual mean vertical growth rates, particularly among the Philippine species: 1.5 to 13 cm shoot1yr1(for Cymodocea rotundata and C serrulata, respectively) Additionally,

species differ in the height of their vertical stems Stem lengths range between 1 and 8 cm These vertical stems are partly buried in the sediment, but particularly in

C serrulata, also reach above the sediment surface Although mean annual vertical

stem growth is strictly not comparable to an instantaneous response to a short-term

sedimentation event, the former will set the order of magnitude of the short-term response of the seagrass shoot Short-term responses, however, have been quantified

in a few species only

In contrast to vertical stem growth, horizontal rhizome expansion is closely cor-related with seagrass size and longevity Rhizome growth is slowest (2 to 5 cm yr1)

in larger and longer-lived species (Duarte et al., 1994; Vermaat et al., 1995; Duarte, 1991) This capacity may allow shorter-lived species to migrate away from or into newly deposited sediment forms Horizontal expansion rates at patch edges, i.e., into newly available bare-ground, are often considerably higher than those measured in established beds Whereas the difference between species in mean annual vertical stem growth amounts to about a factor of 10, horizontal rhizome growth differs by a factor of 30, a difference present among both Mediterranean and Philippine species

Annual horizontal growth in northern temperate Zostera species is limited, though

these species have relatively short life spans and small shoots This is mainly caused

by the reduced length of the growing season (Marba et al., 1994a; Vermaat & Verhagen, 1995), since rhizome internodal lengths and growth rates during the grow-ing season are comparable to those of other small species

The slowest-growing and longest-lived Philippine species, Enhalus acoroides,

as well as the oldest Mediterranean species, Posidonia oceanica, also have the largest

shoots and rhizomes (Duarte, 1991; Vermaat et al., 1995) In the Mediterranean, the larger and longer-lived species showed less annual variation in photosynthetic para-meters than the shorter-lived species, supporting the suggestion of increased seasonal buffering with increased size and age (Duarte, 1991) This pattern, however, was not confirmed for the three studied Philippine species, which are all rather long-lived Morphological adjustments may also improve light availability considerably Longer leaves or stems raise the photosynthetic tissue closer to the water surface, an investment which will probably pay off in turbid, shallow waters where light is

atten-uated exponentially The tallest tropical seagrass Enhalus acoroides is able to lift its

leaves much closer to the water surface, growing in turbid water on shallow (1 to

2 in.) mudflats close to river mouths (Nienhuis et al., 1989; Brouns & Heijs, 1991; Erftemeijer & Herman, 1994)

In mixed meadows, form and size could be decisive and one would expect that the smallest species in the lower leaf canopies would suffer most the impact of light

deprivation, e.g., Halophila ovalis, Halodule uninervis, and Syringodium

isoeti-folium (Vermaat et al., 1995) However, in clear waters, Halophila species have been

found to grow considerably deeper than most other seagrass species (Duarte, 1991), and for one species Drew (1979) found a comparatively low compensation point (9E m2

s1for Halophila stipulacea) Hence, species from the genus Halophila

may survive longer under reduced light regimes

In some areas, seagrasses have to cope with burial through sediment deposition and resuspension Burial affects seagrasses adversely by reducing light availability to affected photosynthetic tissue, reducing diffusion of O2to roots and rhizomes; and mechanically counteracting the production of new leaves by deeply buried meristems (Duarte et al., 1997) Seagrass responses to increased sedimentation include adjust-ments in vertical stem elongation or horizontal rhizome expansion (Duarte et al.,

1994; Marba et al., 1994 a and b), or by recolonization from seeds (Duarte et al., 1997) Architectural differences among species result in considerable ecological advantages for survival

RESPONSES OF CORALS TO

SILTATION/SEDIMENTATION

Sediment deposition and suspended sediments affect coral community structure dif-ferently The inability of coral planulae to settle in areas where soft sediments con-tinually cover the bottom support the observation that sediment deposition has generally an adverse effect on living coral (Ruitenbeek et al., 1999) Adult coral colonies of some species may survive silt cover for short periods (e.g., hours to days) However, coverage for longer periods is lethal to virtually all species (Ruitenbeek

et al., 1999)

On the other hand, greater coral abundance may be found in many reefs with high suspended sediment loads Species composition in these areas may differ sub-stantially from that in areas with low suspended sediment This is in part due to the differential ability of the polyps to eject sediment Hence, coral reefs may exhibit wide variations in species composition in areas of differing suspended sediment loads, but coral cover may not vary significantly with suspended sediment loading (Ruitenbeek et al., 1999)

Sedimentation patterns exert a significant control on reef development via their influence on both sediment deposition and suspended sediment In St Croix, U.S Virgin Islands, lower transport rates of sediments permit faster reef growth (Hubbard, 1986) Annual storms (wave height 3 to 5 m), however, result in order-of-magni-tude increases in sediment transport They periodically flush sediments and offset the usual imbalance between sediment import and export

E FFECTS OF S EDIMENTATION ON C ORAL A BUNDANCE , D IVERSITY ,

Sedimentation is among the important factors that determine coral abundance, growth, and distribution (Hodgson, 1990; Babcock & Davies, 1991) High turbidity and sedimentation decrease coral abundance, alter coral growth forms to a more branching habit, and decrease species diversity (Dodge & Vaisnys, 1977) The diver-sity of corals on all intertidal flats in the vicinity of tin dredging and smelting activi-ties around Laem Pan Qah peninsula, Phuket, was low (six genera), the dominant

genera being Porites, Montipora, Acropora, and Platygyra (Brown & Holley, 1981).

Dodge and Vaisnys (1977) likewise reported that analysis of coral growth patterns and populations in Bermuda reveals that living coral abundance on the reefs of Castle Harbor, a location where extensive dredging occurred during 1941 to 1943, is much reduced in comparison to external North–South reefs

In Bolinao (NW Philippines), Wesseling et al (1997) further found that

Acropora completely buried with littoral sediment (16% silt, 38% fine sand, and 38%

coarse sand) experienced high mortality This finding suggests a reduction in coral

composition in reefs subjected to intense sedimentation Less sensitive taxa (e.g.,

Porites), however, were found to recover within a month of exposure.

The probable causes of these events include turbidity, physical tissue damage, reduced larval recruitment and mortality, and their effects on coral survival Turbidity reduces underwater light due to scattering from sediment particles in the water col-umn Hence, a source of energy is virtually lost In addition, time and energy that could be used to capture food, grow, metabolize, and reproduce are likewise lost (Dodge & Vaisnys, 1977)

Experimental application of sediments onto living coralline tissues has demon-strated detrimental effects including expulsion of zooxanthellae, cellular damage, and after complete burial, death (Babcock & Davies, 1991) On the other hand, they found that while higher sedimentation rates reduced the number of larvae settling on upper surfaces, total numbers of settled larvae were not significantly affected by sed-imentation regime

E FFECTS OF S EDIMENTATION ON C ORAL G ROWTH

AND P RODUCTIVITY

At 13 sites with varying siltation levels in the Philippines, studies were conducted on the responses of corals to sedimentation At the level of the colony, the comparatively

fewer number of white and dark bands observed in Porites at a more silted site indicated

slower growth rate when compared to colonies with a greater number of bands observed

at a less silted site (Mamaril-Villanoy et al., 1997) Barnes and Lough (1993) found that coral growth over a year is represented by adjacent dense and less dense bands which may be caused by different factors, among which are turbidity and sedimentation

At the population level, Wesseling et al (1997) differentiated two types of lesions in corals found along siltation gradients: Type I lesions, surrounded with liv-ing tissue, and Type II lesions, at the edge of colonies Colony size and density of lesions varied among reefs, with smaller colonies and more lesions observed in more exploited and silted areas A relation with sedimentation rate, however, was found only for Type II lesions where it increased significantly above a sedimentation thresh-old rate of about 25 mg/cm2/day

Sediment affects coral metabolism by decreasing photosynthetic production, increasing relative respiration, and increasing carbon loss through greater mucus out-put (Riegl & Brance, 1995) In nine coral species investigated under simulations of natural sedimentation levels and light conditions, a severe reduction in productivity and respiration was recorded under sedimented conditions P/R ratios of all species were above 1 in no-silt conditions In silted conditions, on the other hand, the ratios dropped below 1 In relation to mucus secretion, it averaged 35% of daily respiration under the unsilted condition; the value rose to 65% under silt treatment (Riegl

& Brance, 1995)

M ODELING R EEF S TATUS AND S EDIMENTATION

Two recent procedures are used to generate a surface dose-response model of the rela-tionship among coral abundance and various inputs including sedimentation These

are fuzzy logic procedures and watershed-based modeling The first is linked to a non-linear economic structure incorporating technical intervention (e.g., pollution treatment) and policy interventions (e.g., taxation) (Ruitenbeek et al., 1999) The result of the optimization process gives insights into the most cost-effective means to protect reefs under different reef quality targets In Montego Bay, Jamaica, for exam-ple, appropriate policy measures costing (US) $12 million are estimated to improve coral abundance by 10% in 25 years At a cost of (US) $153 million, these are expected to provide up to 20% increase

Watershed-based modeling of sedimentation and inland pollution is a part of a global analysis, involving 3000 watersheds in the world It integrates data on slope, precipitation, and land cover type to estimate “relative erosion potential” (REP) by roughly a 2-km grid cell (Burke, L., personal communication) The results are sum-marized by watershed to develop criteria for watersheds of low, medium, or high mean REP The zone of effect for sediment discharge is estimated based upon an esti-mate of flow (discharge) for the peak rainfall month

RESPONSES OF MANGROVES TO SILTATION

Siltation is of primary importance in the development of mangroves In deltas along the coasts of Southeast Asia, mangroves cover large areas This is largely because of high rainfall and rivers with high silt loads which combine to provide favorable con-ditions for their development (Milliman & Meade, 1983; Milliman & Syvitski, 1992) Highest productivity values are usually reported in mangroves associated with rivers (Twilley et al., 1986) River flow and tides transport a large fraction of mangrove pro-duction (on average 29.5%: Duarte & Cebrian, 1996) to nearby habitats in the form

of leaf litter and propagules (e.g., Twilley et al., 1986; Hemminga et al., 1994; Panapitukkul et al., 1998) In addition a substantial fraction of mangrove production

is buried in the sediments (10.4% on average: Duarte & Cebrian, 1996), causing a large fraction of the mangrove production (therefore, a large quantity of nutrients) to

be lost from the ecosystem (Boto & Bunt, 1981; Twilley et al., 1986) Primary pro-duction of mangrove habitats therefore tends to depend on continuous nutrient sup-ply from land or sea (Duarte et al., 1998) This nutrient dependence led to the hypothesis that mangrove growth may be nutrient-limited, as has been shown by Boto and Wellington (1983) and Feller (1995)

Growth of Rhizophora apiculata seedlings living at the edge of progressing

man-grove forests at the study sites in the Philippines and Thailand is directly correlated

to the nutrient and silt contents within the sediments (Duarte et al., 1998) Sites with low nutrients and coarse sediments yielded seedlings with very low growth rates On the other hand, nutrient-rich, silty sediments yielded seedlings with much faster growth rates

The size of the watersheds drained by the rivers where mangroves grow has a strong linkage with, among others, sediment composition and mangrove growth (Duarte et al., 1998), while autochthonous substances are received by the mangrove itself (Boto & Bunt, 1981; Boto, 1984; Twilley et al., 1986) However, substantial