Content uploaded by Roy Lewis
Author content
All content in this area was uploaded by Roy Lewis on Apr 23, 2015
Content may be subject to copyright.
BULLETIN OF MARINE SCIENC E, 80(3): 823–837, 2007
823
Bulletin of Marine Science
© 2007 Rosenstiel School of Marine and Atmospheric Science
of the University of Miami
IMPORTANT CONSIDERATIONS TO ACHIEVE
SUCCESSFUL MANGROVE FOREST RESTORATION
WITH OPTIMUM FISH HABITAT
Roy R. Lewis III and R. Grant Gilmore
ABSTRACT
Mangrove forest restoration projects commonly fail to achieve significant plant
cover for two reasons: because there is a misunderstanding of mangrove forest hy-
drology, or, acceptance of the false assumption that simply planting mangroves is
all that is required to establish a fully-functional mangrove ecosystem. Even res-
toration projects that meet a restoration goal within 3–5 yrs often fail to provide
adequate habitat for fish and invertebrates. Here we discuss how fish and mangrove
ecosystems are coupled in time and space, offer several restoration strategies that
match these couplings, and provide simple sequential checklist of design tasks to
use to prevent most failures. Tidal hydrology must be carefully designed to incor-
porate fish habitat, including tidal creeks, to provide access and low tide refuge for
mobile nekton because the mangrove forest floor is generally flooded by tidal waters
less than 30 percent of the time. A fully successful restoration design must mimic
tidal stream morphology and hydrology along an estuarine gradient across a hetero-
geneous mixture of mangrove ecosystem communities.
Mangrove ecosystems currently cover 146,530 km² of the tropical shorelines of the
world (FAO, 2003). is represents a decline from 198,000 km² of mangroves in 1980,
and 157,630 km² in 1990 (FAO, 2003). ese losses represent about 2.0% per year
between 1980–1990, and 0.7% per year between 1990–2000. ese figures show the
magnitude of mangrove loss, and hence the magnitude of mangrove restoration op-
portunity, presented by areas like mosquito control impoundments in Florida, USA
(Brockmeyer et al., 1997), and abandoned shrimp aquaculture ponds in ailand and
the Philippines (Stevenson et al., 1999).
ere are a multitude of opportunities to restore mangrove forests, and reasons to
do so. e reasons are often based on fish habitat value, but include others such as
direct wildlife use and dependence on fisheries production as a source of food. Here
we discuss the basis for the fish—mangrove coupling and restoration efforts meant
to restore these habitats. We offer several restoration strategies that match these
couplings, and provide a simple sequential checklist of design tasks to be used to
prevent most failures.
C F M H
We have previously written about the value of mangroves as habitat for fish and
invertebrates (Gilmore et al., 1983; Lewis et al., 1985) and have noted the pioneering
work of Heald (1971), Odum (1971), and Odum and Heald (1975) in Florida. Wolanski
and Boto (1990) looked at similar issues with regard to specific mangrove forests in
ailand, Brazil, Malaysia, Japan, West Africa, and northern Australia and noted
that “mangrove forests are a highly productive and valuable marine ecosystem.” Dan-
iel and Robertson (1990) state that “[C]omparisons of the densities of fish and Old
World mangrove sites have generally supported the contention of a nursery ground
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
824
function for mangrove habitats…” but note that “…there is often high variance in the
nursery ground value of mangrove habitats in any one region.”
While the controversy about the role of mangrove forests as “nursery habitat” for
fish and invertebrates continues (Sheridan and Hays, 2003; Manson et al., 2005), it is
generally accepted that mangroves and adjacent nearshore habitats provide impor-
tant coastal habitats for fish and invertebrates, as well as coastal wildlife species that
depend on mangrove forests for both habitat and food resources, such as wading and
seabirds, crocodilians, monkeys, and even large mammals such as tigers (Saenger,
2002). A good example of this is the Roseate spoonbill in Florida (Ajaia ajaja Lin-
naeus, 1758) about which Lorenz et al. (2002) state “…small fishes of the orders Cy-
prinodontidae and Poecilidae appear to be the primary dietary items.”
As noted in Lewis (1992), declines in both the commercial and recreational har-
vests of fishes have been attributed to overharvesting, water pollution, and habitat
loss (Stroud, 1983; Royce, 1987). In response to these declines, various fishery man-
agement tools have been proposed and implemented. ey include legally limiting
harvests (by size and number), closing seasons, limiting entry, stocking from hatch-
eries, and enhancing fisheries, the latter most often defined as the creation of artifi-
cial reefs (Royce, 1987).
Fish and invertebrate habitat protection is often discussed in relation to declining
fishery harvests, but restoration of degraded or lost coastal wetland habitats has only
recently been recognized as having great potential as a cost-effective fishery mainte-
nance and restoration strategy (Lewis, 1992; Benaka, 1999).
Mangrove communities worldwide often contain a suite of fish with widely vary-
ing life history strategies. A few “resident” species spend their entire lives reproduc-
ing and feeding within mangrove habitats. ese fish often small and numerous,
numerically dominating the mangrove fish fauna (Blaber, 1997). Ephemeral visitors
to mangrove communities, or “transients” typically utilize mangrove systems to for-
age, or to seek refuge from predation. Mangrove ecosystems are sought as nursery
sites for many transient species, such as the elopids, megalopids, centropomids, and
lutjanids, which are major tropical estuarine predators as adults. ese species en-
ter the mangrove system as small larvae or early juveniles, often grow rapidly, and
leave the system before maturing as top predators in the adjacent estuary. e most
common transient omnivorous primary consumers are represented by mugiliids,
clupeids, sparids, and gerreids (Gilmore and Snedaker, 1993). e omnivorous/detri-
tivorous transient fishes may use the mangrove system as a nursery ground, but they
also enter these systems in large numbers as adults to feed on the abundant primary
producers, algal/bacterial conglomerates, and detritus that is abundant in tropical
mangrove habitats.
Resident species that reproduce and feed primarily in mangrove forest ecosystems
are typically micro/macro-consumers, herbivores or omnivores (Table 1). is group
typically consists of small fishes with highly variable diets, often capable of consum-
ing algae, cyanobacteria, and micro-invertebrates. In the tropical western Atlantic,
resident mangrove species typically consist of cyprinodonts, fundulids, rivulins,
poeciliids, eleotrids, and gobiids. To remain within mangrove ecosystems through-
out the year, a resident species must have eco-physiological capabilities that allow
it to survive the extreme physical conditions characteristic of mangrove habitats:
periodic dewatering, low dissolved oxygen, high sulfide levels, extremes of high tem-
perature, and salinity variation. Resident mangrove forest species can typically with-
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 825
stand major variations in physical parameters: water temperature, salinity, pH and
dissolved oxygen (Harrington and Harrington, 1961; Gilmore, et al., 1982). Transient
species such as tarpon, Megalops atlanticus Valenciennes, 1847, and common snook,
Centropomus undecimalis (Bloch, 1792), have adaptations that allow them to with-
stand low dissolved oxygen conditions and variable salinity regimes characteristic of
mangrove ecosystems (Peterson and Gilmore, 1991).
During the 1950s and 1960s the majority of wetland mangrove habitat on the cen-
tral east coast of Florida between latitude 27°00´ and 29°00´N, approximately 225
coastal km, was impounded to control mosquito populations. Impounding destroyed
thousands of hectares of mangrove forest habitat and caused significant perturba-
tion to the mangrove ecosystem and its dependent biota.
e first studies implemented to determine the influence of impounding man-
grove ecosystems on associated fish communities were conducted in the 1950s by
Harrington and Harrington (1961, 1982). As the first study to document and suggest
positive restoration measures Gilmore et al. (1982) demonstrated that even though
impoundments destroyed thousands of hectares of fish habitat, these habitats could
be restored for fish use through simple tidal reconnection (hydrologic restoration;
Brockmeyer et al., 1997). In fact, tidally connected ditches dug to produce impound-
ment dikes were found to offer new habitat that mimicked tidal creeks, thus allowing
Table 1. Typical east Florida restored mangrove ecosystem sh community: common species of a
total of 93 captured in 5,731 collections at 326 sites in the Indian River Lagoon, Florida. Species
are divided into residential and trophic groups from data collected simultaneously in three seperate
barrier island mangrove systems for 24 hrs every 2 wks over 2 yrs.
Biomass
Wet weight (g) %No. Individuals %
Omnivores grazing on detritus, algae, cyanobacteria, micro and macro invertebrates
Cyprinodon. variegatus 253,700.16 57 405,944 41
Poecilia latipinna 146,412.94 32 234,495 24
Gambusia holbrooki 50,993.30 11 350,769 35
Total residents 451,106.40 49.3 991,208 95
Mangove transient species
Top predators feeding on sh and macro invertebrates
Centropomus undecimalis 5,1487.77 44 12,431 53.0
Elops saurus 25,442.41 22 10,279 44.0
Megalops atlanticus 22,649.31 20 408 2.0
Lutjanus griseus 16,544.94 14 193 0.8
Total transient predators 116,124.43 12.7 23,311 2.2
Omnivores grazing on detritus, algae, cyanobacteria as well as micro and macro invertebrates
Mugil cephalus 283,754.38 81.0 21,184 74.1
Mugil curema 564,66.10 16.0 3,064 10.7
Brevoortia smithi 935.22 0.3 1,544 5.4
Diapterus auratus 1,596.04 0.5 1,612 5.6
Archosargus probatocephalus 4,934.83 1.4 169 0.6
Gerres cinereus 806.74 0.2 1,033 0.4
Total transient omnivores 348,493.31 38.1 28,606 2.7
Total sh 915,724.14 1,043,125
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
826
water to remain in the wetland during low water periods, offering more habitat to
transient species such as common snook and striped mullet, Mugil cephalus Lin-
naeus, 1758.
As an example, fish fauna data from pre- and post-impoundment of a restored
mangrove habitat show that hydrologic restoration restored transient fish commu-
nities, previously precluded from the mangrove wetland by impoundment dikes
(Table 2). Invertebrate and vegetative communities were also restored through the
restoration of tidal access and natural plant succession. Tidal connections to adja-
cent estuary restored biological transport of biomass from the mangrove wetland to
the adjacent estuary and allowed recruitment to the mangrove wetland of juvenile
transient species that normally utilize this habitat as nursery habitat (centropomids,
elopids, megalopids, lutjanids, and mugiliids; Gilmore 1987a,b).
Extensive impoundment and subsequent restoration of mangrove forest wetlands
throughout coastal east Florida allowed researchers to demonstrate the direct value
of mangrove restoration by hydrologic restoration on fish communities, particularly
Table 2. Comparison of the east Florida mangrove sh fauna before and after restoration of a 26
ha mangrove wetland ecosystem, Round Island, Indian River County, Florida. These data were
collected from 1955–1968 by Harrington and Harrington (1961, 1982) and from 1978–1986 by
Gilmore et al. (1982) and Gilmore (1987a). Pre-impoundment collections were from 10 Septem-
ber 1956 to 17 October 1956 and post-impoundment collections from September 1956 to October
1965 and from September 1984 to August 1986.
Pre-impound. Post-impound. Restored
Harrington and
Harrington, 1961
Harrington and
Harrington, 1982
Gilmore
et al., 1985
No. Individuals %No. Individuals %No. Individuals %
Mangrove resident species
Omnivores grazing on detritus, algae, cyanobacteria, micro and macro invertebrates
Cyprinodon variegatus 1,986 40.0 235 25 258,597 41.0
Poecilia latipinna 1,215 25.0 400 43 160,317 26.0
Gambusia holbrooki 1,769 36.0 300 32 210,080 33.0
Total residents 4,970 76.0 935 100 628,994 94.0
Mangrove transient species
Top predators feeding on sh and macro invertebrates
Centropomus undecimalis 172 38.0 0 0.0 4,005 53.0
Elops saurus 33 7.0 0 0.0 8,849 44.0
Megalops atlanticus 254 55.0 0 0.0 316 2.0
Lutjanus griseus 0 0.0 0 0.0 34 0.8
Total transient predators 459 7.0 0 13,204 2.0
Omnivores grazing on detritus, algae, cyanobacteria as well as micro and macro invertebrates
Mugil cephalus 1,124 100.0 0 0.0 17,856 74.1
Mugil curema 0 0.0 0 0.0 3,064 10.7
Brevoortia smithi 0 0.0 0 0.0 78 5.4
Diapterus auratus 0 0.0 0 0.0 102 5.6
Archosargus probatocephalus 0 0.0 0 0.0 15 0.6
Gerres cinereus 0 0.0 0 0.0 328 0.4
Total transient omnivores 1,124 17.0 21,443 3.0
Total sh 6,553 900 670,194
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 827
transient fish species which supported regional sport and commercial fisheries. For
example, over 1500 juvenile common snook were captured during a single 3-hr cul-
vert trap set in a restored, previously impounded, mangrove wetland, Jack Island
State Park, on the barrier island in St. Lucie County, Florida. is is the single larg-
est capture of this species as juveniles recorded in Florida waters and demonstrated
the value of mangrove habitat for this species and its utilization of a restored tidal
connection between the mangrove forest and the estuary. e trophic work of Lucz-
kovich et al. (1995) included this species, verifying the major role mangrove residents
play in the diet of predaceous transients in restored mangrove wetlands first demon-
strated by Harrington and Harrington (1961, 1982) and Gilmore et al. (1983).
One of the first definitive research efforts in the United States to target early juve-
nile fish habitat modification as a major limiting factor to adult fishery stock consid-
ered of commercial importance was that of Gilmore et al. (1983) documenting the
life history of the common snook in the Indian River on the east coast of Florida. is
species is a prized game fish that had been primarily managed by closure of its fishery
to commercial harvest in 1957, and subsequent recreational harvest limits, neither of
which has been shown to maintain or increase the adult stock. Gilmore et al. (1983)
determined that while adult common snook spawn in higher salinity coastal passes,
the early juvenile snook (mean 27.5 mm standard length, SL) are found in shallow
freshwater tributary streams entering estuarine waters and in polyhaline to marine
barrier island mangrove forests. Juvenile common snook typically leave peripheral
mangrove and freshwater habitats as they reach 100–150 mm SL, moving into sea-
grass meadows as they mature, at a mean of 240 mm SL. However, larger juvenile
and mature adults will periodically visit tidal fringe mangrove forest habitats, rivers
and creeks where adequate prey and water depth allow. Important food items for
juveniles included fish, shrimp, and microcrustaceans, with the eastern mosquito-
fish, Gambusia holbrooki Girard, 1859, being the dominant identifiable fish species
in specimens < 100 mm SL (Gilmore et al., 1983; Luczkovitch et al., 1995).
Along the coasts of central Florida, the role of mangroves and the mangrove-marsh
mixture in the production of small forage fish species [eastern mosquitofish; rainwa-
ter killifish: Lucania parva Baird and Girard, 1855; sailfin molly: Poeicilia latipinna
(Leseur, 1821); sheepshead minnow: Cyprinodon variegates Lacépède, 1803; Fundu-
lus spp.; diamond killifish: Adinia xenica (Jordan and Gilbert, 1882)] essential as di-
etary items for larger and more commercially important fishes (as well as wading
birds and seabirds) is underappreciated and rarely documented (Fore and Schmidt,
1973; Gilmore et al., 1983; Luczkovitch et al., 1995). e few studies focusing on this
function of mangroves and mangrove-marsh communities include Harrington and
Harrington (1961), Gilmore et al. (1982, 1983), Rey et al. (1990), Ley et al. (1994),
and Whitman and Gilmore (1993). Robertson and Duke (1990) make a similar point
about the valuable Australian fishery for barramundi (Lates calcarifer Alexander,
1845): “[O]nly a small number of fish species examined in this study are of direct
economic importance…(H)owever, the valuable commercial gill net and recreation-
al line fisheries for barramundi…are intimately linked to mangrove habitats…our
data…show that more than 50% of the diet of subadult L. calcarifer is composed of
the small fish which dominate the mangrove habitats…”
In spite of this, the recent review of Manson et al. (2005), in discussing groups of
mangrove dependent fish species states “(A) final grouping is the true estuarine spe-
cies that complete their entire life cycle within estuaries. ese species are clearly es-
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
828
tuarine-dependent but many are small and short lived…and few contribute directly
to fisheries,. . . (and). . . will not be discussed further. Instead, the focus will be on the
marine-estuarine category that includes a number of economically important spe-
cies.” We believe ignoring the role of small mangrove-dependent fish in the families
Gobiidae, Atherinidae, Cyprinodontidae, and Poecilidae, for example, as important
mangrove-dependent species is neither an ecologically sound approach to mangrove
forest management and restoration, nor estuarine fisheries management.
Gilmore et al. (1983) concluded that, “loss of habitat and general degradation of
water quality has undoubtedly had a more permanent and, therefore, far greater
effect on reducing snook populations than the fishery. Removal of juvenile habitat
curtails recruitment from the largest portion of the snook population and that por-
tion which is most susceptible to natural mortality.” Similar life history implications
are reported for other key commercial and recreationally important estuarine fish
species, such as redfish (Sciaenops ocellatus Linnaeus, 1766), spotted seatrout [Cy-
noscion nebulosus (Cuvier, 1830)], and tarpon (Lewis et al., 1985; Lewis, 1990).
Similar restoration is reported by Barimo and Serafy (2003) where fish utilization
of a 30 ha mangrove restoration site in Biscayne Bay, Florida, with 28 ponds con-
nected to adjacent waters by a series of tidal creeks, supported 25 fish species 1–3 yrs
after restoration. e most common of these species were the gold spotted killifish
[Floridichthys carpio (Gunther, 1866)]; yellowfin mojarra [Gerres cinerus (Walbaum,
1792)]; and rainwater killifish. Sampling of ponds of various ages revealed a general
trend in increasing fish community diversity over time.
Finally, fragmented or hydrologically restricted or isolated mangrove habitat in the
Bahamas has shown reduced numbers of fish species (Layman et al., 2004) similar to
the data of Harrington and Harrington (1961, 1982), Gilmore et al. (1982), and Gilm-
ore (1987a). Community-based hydrologic restoration of these areas led to increases
in fish and invertebrate utilization after restoration (Layman and Arringon, 2005).
Layman et al. (2004) proposed that hydrologic restrictions influence fish and inver-
tebrate abundance through: (1) reduced tidal exchange producing decreased habitat
quality (e.g., greater salinity extremes), (2) reduced tidal exchange lowering the influx
of planktonic larvae and juveniles of estuarine dependent species, and (3) non-per-
meable landscape features (e.g., roads without culverts) halting upstream movement
by transient marine species [e.g., bonefish, Albula vulpes (Linnaeus, 1758)].
I S R
Great potential exists to reverse the worldwide loss of mangrove forests, and thus
enhance fish and invertebrate habitats, through application of basic principles of
ecological restoration using ecological engineering approaches. Previously docu-
mented attempts to restore mangroves, where successful, have largely concentrated
on creation of plantations consisting of a few mangrove species targeted for harvest
as wood products or temporarily used to collect eroded soil and raise intertidal areas
to elevations usable for terrestrial agriculture [see review by Lewis (2005)]. But many
expensive efforts to restore mangroves have failed (deLeon and White, 1999; Erfte-
meijer and Lewis, 2000), and most attempts to restore mangroves never even achieve
significant cover by mangrove species, or meet their stated goals (Erftemeijer and
Lewis, 2000; Lewis and Streever, 2000; Lewis, 2000, 2005).
We suggest there should be three restoration strategies for fish habitat optimization:
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 829
Strategy 1: Achieve plant cover similar to that in an adjacent relatively undisturbed
and mature mangrove ecosystem.—Due to the length of time it may take to achieve
full maturity in a restored mangrove forest (25–50 yrs minimum), the likely actual
measurable criterion will be the establishment of a trajectory (based upon regular
monitoring of multiple parameters, such as plant cover by species and density and
height of mangroves) that, if continued, would establish such mature forest cover [see
Lewis et al. (2005) for a typical trajectory curve and successful forest cover estab-
lishment at 36 mo post restoration]. Such vegetation development trends are rarely
documented or reported, however, in part due to the high failure rate of mangrove
restoration.
Strategy 2: Establish a network of channels that mimics the shape and form of a nat-
ural tidal creek system.—Tidal creeks are important as pathways for the movement of
mangrove detritus and for their function in maintaining water quality (particularly
with regard to the dissolved oxygen levels required to support fish and invertebrates),
as well as for their role as entry points on flood tides and exit points on ebb tides for
mobile nekton. Although this seems obvious from a fisheries perspective, mangrove
restoration projects are frequently completed without concern for reconnection or
construction of tidal creeks. Even when a network of channels is incorporated into
the design, it too often consists of just rough-cut ditches dug without consideration
of the resulting tidal prism, the desired exchange of tidal waters with adjacent es-
tuarine or oceanic waters, or whether velocities will be high enough to keep the new
channels open over time (Turner and Streever, 2002). Instead, channels should be
designed to serve natural functions and be as self-maintaining as possible.
Strategy 3: Establish a heterogeneous landscape similar to that exhibited by the lo-
cal mangrove ecosystem.—Mangrove ecosystems include the mangrove forest itself,
its associated animal community, and adjacent and hydrologically linked wetlands.
Important contiguous plant communities include terrestrial plant communities in
the watershed, and shallow marine communities including tidal ponds, tidal flats,
seagrass meadows, and coral reefs. e landward portion of the mangrove ecosys-
tem includes salt flats or salinas, as well as freshwater streams and wetlands (both
marshes and forested wetlands) that ultimately drain to the sea via both surficial
and underground waterways through mangrove forests. e seaward edge of the
mangrove ecosystem is the outer edge of the influence of water flows coming from
drainage through the mangrove forests and may extend several kilometers offshore.
Mangroves have also been ecologically linked with fish movements from adjacent
coral reefs and seagrass meadows (Ley and McIvor, 2002; Mumby et al., 2004).
Lewis (2005) reviews the most likely points of failure in successful mangrove forest
restoration and offers a simple sequential checklist of design tasks, that, if followed
rigorously, would prevent most of these failures. Callaway (2001) recommends seven
similar steps be taken in designing for hydrologic and geomorphologic development
of tidal marshes in California. Both Vivian-Smith (2001) and Sullivan (2001) suggest
addressing these factors through the use of a reference tidal marsh for restoration plan-
ning and design, which applies equally well here for mangrove forest restoration.
Five sequential design tasks have been suggested by Stevenson et al. (1999) as essen-
tial to implementing the strategies listed above, and successfully restoring mangrove
forests. Following these design tasks in sequence will prevent most of the common
errors in mangrove forest restoration design.
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
830
1. Understand the autecology (individual species ecology) of the mangrove species
in the vicinity of the restoration site, paying particular attention to patterns of
reproduction, zonation, propagule distribution, and seedling establishment.
2. Understand the normal hydrologic patterns controlling the distribution and
successful establishment and growth of the targeted mangrove species.
3. Determine what modifications and stresses of the previous mangrove environ-
ment are currently preventing natural secondary succession.
4. Design the restoration program to first reestablish the appropriate hydrology
at an appropriate restoration site, and then utilize natural volunteer mangrove
propagule recruitment for plant establishment.
5. Only plant propagules or seedlings after determining through steps 1–4 that
natural recruitment will not provide the quantity of successfully established
seedlings, rate of stabilization, or rate of growth of saplings set as goals for the
restoration project.
e more common alternative approach to mangrove restoration is to first build a
nursery to grow mangrove seedlings, then produce large numbers of potted mangroves
for planting, and finally identify a location to install the nursery-grown plant materi-
als. is is referred to by Lewis (2005) as the “gardening” approach. Often unvegetated
mudflats that have never supported mangroves are chosen for such garden sites, and
frequently most or all of the plants installed die in a short time, or remain stunted due
to excessive flooding and stress due to extended soil saturation (Erftemeijer and Lewis,
2000). In reality, actual planting of mangroves is rarely needed to establish plant cover
unless a site has been identified as “propagule limited” (Lewis, 2005).
Previous research has documented the general principle that mangrove forests
worldwide occur on raised and sloped platforms situated above mean sea level and
inundated less than thirty percent of the time by tidal waters (Lewis, 2005). More
frequent flooding can cause stress and death of these tree species (Turner and Lewis,
1997). Preventing such damage requires a detailed understanding of local mangrove
hydrology. Again the inclusion of correctly designed tidal creeks in a mangrove res-
toration design provides access to this platform on high tides, and refuge within the
creeks on low tides if properly designed.
Because degraded mangrove forests may recover without active restoration efforts,
restoration projects should begin with a search for hydrologic sources of stress (such
as blocked tidal inundation), or roadways within the forest without adequate cross
connections both upstream and downstream of the road, that might be preventing
secondary succession, then focus on removing that stress (Hamilton and Snedak-
er, 1984; Cintrón-Molero, 1992; Layman et al., 2004; Layman and Arringon, 2005).
Once normal hydrology has been reestablished, the next step is to observe whether
natural seedling recruitment occurs. Planting should be considered only if this does
not happen.
It is important to understand that mangrove forests occur in a wide variety of hy-
drologic and climatic conditions that result in a broad array of mangrove community
types. In Florida, Lewis et al. (1985) identified at least four variations on the original
classic mangrove zonation pattern described by Davis (1940), all of which include a
tidal marsh component dominated by such species as smooth cordgrass (Spartina
alterniflora Loisel.) or saltwort (Batis maritima Linnaeus). Lewis (1982a,b) describes
the role that smooth cordgrass plays as a “nurse species” that initially colonizes bare
soil and facilitates primary or secondary succession to a climax community domi-
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 831
nated by mangroves, but with some remnants of the original tidal marsh vegetation
typically remaining. Crewz and Lewis (1991) further discuss the tidal marsh compo-
nent typical of Florida mangrove forests.
e typical “mangrove restoration” project aimed at establishing a monospecific
stand of planted Rhizophora spp. fails to restore the freshwater-marine salinity gra-
dient and plant biodiversity characteristic of intact mangrove ecosystems. Establish-
ing this salinity gradient and a related variety of habitat types must be the goal of any
mangrove restoration project intended to provide fish habitat and enhance biodiver-
sity. Besides providing food and habitat for a broad spectrum of fish, invertebrates,
and associated wading birds, seabirds, and other wildlife, incorporating this estua-
rine gradient maximizes both above- and belowground productivity of mangroves
and their ability to deal with rising sea levels (Snedaker, 1993).
R O
ere are both large and small opportunities for mangrove restoration. Both have
a role in mangrove forest management, but it is important to understand that very
small restoration projects, such as the typical shoreline plantings of a few hundred
or thousand mangroves are generally neither as cost-effective on a per hectare basis
nor as ecologically valuable as larger hydrologic restoration projects, such as restor-
ing normal seasonal freshwater flows to mangroves in Florida Bay (Ley et al., 1994;
Ley and McIvor, 2002) or Biscayne Bay (Braun et al., 2004). Large scale hydrologic
restoration, such as that described by Brockmeyer et al. (1997), has produced real
ecological benefits for US$250 ha–1. Small local citizen-based restorations still have
educational value and can play a supplementary role, especially in regard to rees-
tablishing species and community diversity along the urban interface (Layman and
Arringon, 2005). But where money is limited and/or fish habitat optimization is the
goal, we should listen to the words of Spurgeon (1998): “[I]f coastal habitat reha-
bilitation/creation is to be widely implemented, greater attempts should be made to:
find ways of reducing the overall costs of such initiatives; devise means of increas-
ing the rate at which environmental benefits accrue; and to identify mechanisms for
appropriating the environmental benefits.” ere are many opportunities to restore
mangroves worldwide, including at least several hundred thousand hectares of aban-
doned shrimp aquaculture ponds, primarily in Southeast Asia, but also in Ecuador
and Brazil (Stevenson et al., 1999).
C S
Aerial photographs of the same mangrove restoration site over a period of 4 yrs
represent a portion of a 500 ha hydrologic restoration project at West Lake (WL) in
Hollywood, Florida (Fig. 1; described in Lewis (1990)). Success resulted from using a
reference site, and targeting final constructed grades to be the same as the adjacent
undisturbed forest. is resulted in a final sloped grade from +27 cm to +42 cm
mean sea level (MSL). Extensive constructed tidal creeks and shallow mudflats were
also added to the original plans which had been designed and originally permit-
ted without them. No planting of mangroves took place or was necessary. All three
of the Florida species of mangroves (red mangrove Rhizophora mangle Linnaeus,
black mangrove Avicennia germinans (Linnaeus) Linnaeus, and white mangrove La-
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
832
Figure 1. Oblique aerial photographs of a portion of the hydrologic mangrove restoration project
at West Lake Park, Hollywood, Florida over three time periods: (A) Preconstruction, July 1987,
(B) completion of construction, August 1989, and (C) 2 yrs post-construction, May 1991. No
planting of mangroves occurred. All vegetation derived from volunteer mangrove propagules.
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 833
guncularia racemosa (Linnaeus) Gaertn.f.) volunteered on their own. High marshes
dominated by B. maritima were also hydrologically restored. is type of major ex-
cavation to restore mangrove habitat is similar to the successful efforts reported by
Barimo and Serafy (2003) for Biscayne Bay. Another form of hydrologic restoration
is to reconnect impounded mangroves to normal tidal influence (Brockmeyer et al.,
1997; Turner and Lewis, 1997; Layman and Arringon, 2005).
Comparative sampling of fish populations within both restored and natural refer-
ence areas at WL, and adjacent natural and restored sites at John U. Lloyd Park (JUL;
Roberts, 1994), resulted in the collection of 9937 fishes representing 34 species in the
JUL reference sites, 6835 fishes representing 28 species at the JUL mangrove restora-
tion sites, 9141 fishes representing 36 species at the WL reference sites, and 11, 374
fishes representing 31 species from the WL restoration site. e predominant species
were bay anchovy Anchoa mitchilli Valenciennes, 1848; juvenile mullet Mugil spp.;
yellowfin mojarra Gerres cinereus (Walbaum, 1792); silver jenny Eucinostomus gula
(Quoy and Gaimand, 1824); striped mojarra Eugerres plumie (Cuvier, 1830); eastern
gambusia; sailfin molly; marsh killifish Fundulus confluentus Goode and Bean; 1879,
and goldspotted killifish. Statistical tests of the differences in fish assemblages, and
total mean abundance in the natural and 3–5 yr old restored mangroves sites were
not significant. Lewis (1992) noted in reviewing similar reports on fish use of tidal
marsh restoration sites in Florida, North Carolina, and California that “…rapid (3–5
yrs) establishment of comparable fish communities in created and restored coastal
wetlands, when compared with natural wetlands, is a generally documented obser-
vation…”
C
Ellison (2000) asks the question “mangrove restoration: do we know enough?” His
answer is that “[R]estoration of mangal does not appear to be especially difficult…”
and comments that in contrast to the complexity of restoring inland wetlands, “…it is
more straightforward to restore tidal fluctuations and flushing to impounded coastal
systems where mangroves could subsequently flourish…” us, ecological restora-
tion of mangrove forests is feasible, has been done on a large scale in various parts
of the world, and can be done cost-effectively. Fish utilization of these restored eco-
systems has, however, not been a routine measure of successful restoration, and we
recommend more use of this criterion in the future.
Lewis (2005) recommends than an ecological engineering approach be applied to
mangrove ecosystem restoration projects. e simple application of the five steps
outlined by Stevenson et al. (1999) would at least insure an analytical thought pro-
cess and minimize resources wasted on futile mangrove “gardening” efforts. To opti-
mize fish use of restored mangrove ecosystems, we recommend planning restoration
projects around the three key strategies outlined earlier: establishing plant cover,
a network of tidal channels, and a landscape mosaic, all of which mimic natural
undisturbed systems and reflect the zonation and diversity of the local mangrove
ecosystem, including freshwater inputs to the landward edge of a normal estuarine
gradient.
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
834
L C
Barimo, J. F. and J. E. Serafy. 2003. Fishes of a restored mangrove habitat on Key Biscayne,
Florida. Fla. Sci. 66: 12–22.
Benaka, L. R. 1999. Fish habitat: Essential Fish Habitat and Rehabilitation. American Fisheries
Society, Bethesda, Maryland. 459 p.
Blaber, S. J. M. 1997. Fish and fisheries of tropical estuaries. Fish and Fisheries Series 22. Chap-
man & Hall, New York. 367 p.
Braun, G., R. Lewis, and J. Watson. 2004. Freshwater flow and ecological relationships in Bis-
cayne Bay. Tasks 1-6. South Florida Water Management District Contract C-15967-Wo0-
06. West Palm Beach, Florida. Available from: http://www.sfwmd.gov/org/wsd/mfl/biscay-
nebay/project_doc.htm. Accessed 15 January 2007.
Brockmeyer, R. E. Jr., J. R. Rey, R. W. Virnstein, R. G. Gilmore, and L. Ernest. 1997. Rehabilita-
tion of impounded estuarine wetlands by hydrologic reconnection to the Indian River La-
goon, Florida (USA). Wetl. Ecol. Manage. 4: 93–109.
Callaway, J. C. 2001. Chapter three. Hydrology and substrate. Pages 89–117 in J. B. Zedler, ed.
Handbook for restoring tidal wetlands. CRC Press, Boca Raton. 439 p.
Cintrón-Molero, G. 1992. Restoring mangrove systems. Pages 223–277 in G. W. ayer, ed.
Restoring the nation’s marine environment, Maryland Seagrant Program, College Park,
Maryland. 716 p.
Crewz, D. W. and R. R. Lewis. 1991. An evaluation of historical attempts to establish emergent
vegetation in marine wetlands in Florida. Florida Sea Grant Technical Publication No. 60.
Florida Sea Grant College, Gainesville, Florida. 76 p + appends.
Daniel, P. A. and A. I. Robertson. 1990. Epibenthos of mangrove waterways and open embay-
ments: community structure and the relationship between exported mangrove detritus and
epifaunal standing stocks. Estuar. Coast. Shelf Sci. 31: 599–619.
Davis, J. H. 1940. e ecology and geologic role of mangroves in Florida. Carnegie Inst. Wash.
Pap. Tortugas Lab. No. 32. Publ. 517: 305–412.
De Leon, T. O. D. and A. T. White. 1999. Mangrove rehabilitation in the Philippines. Pages
37–42 in W. Streever, ed. An international perspective on wetland rehabilitation. Kluwer
Academic Publishers, Dordrecht. 338 p.
Ellison, A. M. 2000. Mangrove restoration: do we know enough? Rest. Ecol. 8: 219–229.
Erftemeijer, P. L. A. and R. R. Lewis. 2000. Planting mangroves on intertidal mudflats: habitat
restoration or habitat conversion? Pages 156–165 in Proc. ECOTONE VIII Seminar En-
hancing Coastal Ecosystems Restoration for the 21st Century, Ranong, ailand, 23–28
May 1999, Royal Forest Department of ailand, Bangkok, ailand. 201 p.
Food and Agricultural Organization (FAO). 2003. New global mangrove estimate. Avail-
able from: http://www.fao.org/documents/show_cdr.asp?url_file=/docrep/007/j1533e/
j1533e52.htm. Accessed 1 October 2006.
Fore, P. L. and T. W. Schmidt. 1973. Biology of juvenile and adult snook, Centropomus undeci-
malis, in the Ten ousand Islands, Florida. Chapter 16 in M. R. Carter, ed. Ecosystems
Analyses of the Big Cypress Swamp and Estuaries. U.S. Environmental Protection Agency,
Surveillance and Analysis Division, Athens, Georgia. 356 p.
Gilmore, R. G. 1987a. Fish, macrocrustacean and avian population dynamics and cohabitation
in tidally influenced impounded subtropical wetlands. Pages 373–394 in W. R. Whitman
and W. H. Meredith, eds. Proc. Symp. on Waterfowl and Wetlands Management in the
Coastal Zone of the Atlantic Flyway. Delaware Depart. Nat. Res. and Environ. Control,
Dover, Delaware. 567 p.
___________. 1987b. Subtropical-tropical seagrass communities of the Southeastern United
States: Fishes and fish communities. Pages 117–137 in M. J. Durako, R. C. Phillips, and R.
R. Lewis, eds. Proc. Symp. on Subtropical-tropical seagrasses of the Southeastern United
States. Marine Research Publication 42, Fla. Dept. Nat. Res. 209 p.
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 835
___________ and S. C. Snedaker. 1993. Chapter 5: Mangrove forests. Pages 165–198 in W. H.
Martin, S. G. Boyce, and A. C. Echternacht, eds. Biodiversity of the southeastern United
States: lowland terrestrial communities. John Wiley & Sons, Inc., Publishers, New York.
502 p.
____________, D. W. Cooke, and C. J. Donohoe. 1982. A comparison of the fish populations
and habitats in open and closed salt marsh impoundments in east-central Florida. North-
east Gulf Sci. 5: 25–27.
____________, C. J. Donohoe, and D. W. Cooke. 1983. Observations on the distri-
bution and biology of east-central Florida populations of the common snook,
Centropomus undecimalis (Bloch). Fla. Sci. 46: 313–336.
Hamilton, L. S. and S. C. Snedaker. 1984. Handbook for mangrove area management. East-west
Center, Honolulu, Hawaii. 123 p.
Harrington, R. W., Jr. and E. S. Harrington. 1961. Food selection among fishes invading a high
subtropical salt marsh: from onset of flooding through the progress of a mosquito brood.
Ecology 41: 646–666.
_________________ and _____________. 1982. Effects on fishes and their forage organisms of
impounding a Florida salt marsh to prevent breeding by salt marsh mosquitoes. Bull. Mar.
Sci. 32: 523–531.
Heald, E. J. 1971. e production of organic detritus in a south Florida estuary. Seagrant Tech-
nical Bulletin No. 6., University of Miami Sea Grant Program, Miami, Florida. 110 p.
Layman, C. A. and D. A. Arringon. 2005. Community-based collaboration restores tidal flow to
an island estuary (Bahamas). Ecol. Rest. 23: 58–59.
____________, ____________, R. B. Langerhans, and D. R. Silliman. 2004. Degree of fragmenta-
tion affects assemblage structure in Andros Island (Bahamas) estuaries. Caribb. J. Sci. 40:
232–244.
Lewis, R. R. 1982a. Mangrove forests. Pages 153–172 in R. R. Lewis, ed. Creation and restora-
tion of coastal plant communities. CRC Press, Boca Raton. 219 p.
__________. 1982b. Low marshes, peninsular Florida. Pages 147–152 in R. R. Lewis, ed. Cre-
ation and restoration of coastal plant communities. CRC Press, Boca Raton. 219 p.
__________. 1990. Creation and restoration of coastal plain wetlands in Florida. Pages 73–101
in J. A. Kusler and M. E. Kentula, eds. Wetland creation and restoration: the status of the
science. Island Press, Washington, D. C. 591 p.
__________. 1992. Coastal habitat restoration as a fishery management tool. Pages 169–173
in R. H. Stroud, ed. Stemming the tide of coastal fish habitat loss, Proc. Symp. on Conser-
vation of Coastal Fish Habitat, Baltimore, Md., March 7–9, 1991. National Coalition for
Marine Conservation, Inc., Savannah, Georgia, 321 p.
__________. 2000. Ecologically based goal setting in mangrove forest and tidal marsh restora-
tion in Florida. Ecol. Eng. 15: 191–198.
__________. 2005. Ecological engineering for successful management and restoration of man-
grove forests. Ecol. Eng. 24: 403–418.
__________ and W. Streever. 2000. Restoration of mangrove habitat. Tech Note ERDC TN-
WRP-VN-RS-3. U.S. Army, Corps of Engineers, Waterways Experiment Station, Vicksburg,
Mississippi. 7 p.
__________, A. B. Hodgson, and G. S. Mauseth. 2005. Project facilitates the natural reseeding
of mangrove forests (Florida). Ecol. Rest. 23: 276–277.
__________, R. G. Gilmore, Jr., D. W. Crewz, and W. E. Odum. 1985. Mangrove habitat and
fishery resources of Florida. Pages 281–336 in W. Seaman, ed. Florida aquatic habitat and
fishery resources, Florida Chapter, American Fisheries Society, Eustis, Florida. 543 p.
Ley, J. A. and C. C. McIvor. 2002. Linkages between estuarine and reef fish assemblages: en-
hancement by the presence of well-developed mangrove shorelines. Pages 539–562 in J. W.
Porter and K. G. Porter, eds. e Everglades, Florida Bay, and coral reefs of the Florida Keys
– an ecosystem sourcebook. CRC Press, Boca Raton. 1000 p.
BULLETIN OF MARINE SCIENC E, VOL. 80, NO. 3, 2007
836
________, C. L. Montague, and C. C. McIvor. 1994. Food habits of mangrove fishes: a compari-
son along estuarine gradients in northeastern Florida Bay. Bull. Mar. Sci. 54: 881–899.
Lorenz, J. J., J. C. Ogden, R. D. Bjork, and G. V. N. Powell. 2002. Nesting patterns of Roseate
spoonbills in Florida Bay 1935–1999: implications of landscape scale anthropogenic im-
pacts. Pages 563–606 in J. W Porter and K. G. Porter, eds. e Everglades, Florida Bay, and
coral reefs of the Florida Keys: an ecosystem source book. CRC Press, Boca Raton. 1000 p.
Luczkovich, J. J., S. F. Norton, and R. G. Gilmore. 1995. e influence of oral anatomy on prey
selection during the ontogeny of two percoid fishes, Lagodon rhomboides and Centropomus
undecimalis. Environ. Biol. Fish. 44: 79–95.
Manson, F. J., N. R. Loneragan, G. A. Skilleter, and S. R. Phinn. 2005. An evaluation of the
evidence for linkages between mangroves and fisheries: a synthesis of the literature and
identification of research directions. Oceanogr. Mar. Biol., Annu. Rev. 43: 485–515.
Mumby, P. J., A. J. Edwards, J. E. Arias-Gonzalez, K. C. Lindeman, P. G. Blackwell, A. Gall, M. I.
Gorczynska, A. R. Harborne, C. L. Pescod, H. Renken, C. C. C. Wabnitz, and G. Llewellyn.
2004. Mangroves enhance the biomass of coral reef fish communities in the Caribbean.
Nature 27: 533–536.
Odum, W. E. 1971. Pathways of energy flow in a south Florida estuary. Seagrant Technical Bul-
letin No. 7., University of Miami Sea Grant Program (Living Resources), Miami, Florida.
162 p.
___________ and E. J. Heald. 1975. e detritus based food web of an estuarine mangrove com-
munity. Page 265–286 in I. E. Cronin, ed. Estuarine research. Academic Press, Inc., New
York. 468 p.
Peterson, M. S. and R. Grant Gilmore, Jr. 1991. Eco-physiology of juvenile snook Centropomus
undecimalis (Bloch): Life-history implications. Bull. Mar. Sci. 48: 46–57.
Rey, J. R., J. Shaffer, D. Tremain, R. A. Crossman, and T. Kain. 1990. Effects of re-establishing
tidal connections in two impounded subtropical marshes on fishes and physical conditions.
Wetlands 10: 27–45.
Roberts, K. 1994. e distribution of fish populations in the natural and mitigated mangrove
forests in Southwest Florida. M.Sc. thesis, Nova Southeastern University, Ft. Lauderdale,
Florida. 174 p.
Robertson, A. I and N. C. Duke. 1990. Recruitment, growth and residence time of fishes in a
tropical Australian mangrove system. Estuar. Coast. Shelf Sci. 31: 723–743.
Royce, W. F. 1987. Fishery development. Academic Press, New York. 248 p.
Saenger, P. 2002. Mangrove ecology, silvaculture and conservation. Kluwer Academic Publish-
ers, Dordrecht. 360 p.
Sheridan, P. and C. Hays. 2003. Are mangroves nursery habitat for transient fishes and deca-
pods? Wetlands 23: 449–458.
Snedaker, S. C. 1993. Impact on mangroves. Pages 282–305 in G. A. Maul, ed. Climatic change
in the intra-American seas: implications of future climate change on the ecosystems and
socio-economic structure of the marine and coastal regimes of the Caribbean Sea, Gulf of
Mexico, Bahamas and N.E. Coast of S. America. United Nations Environment Program.
Edward Arnold, London. 389 p.
Stevenson, N. J., R. R. Lewis, and P. R. Burbridge. 1999. Disused shrimp ponds and mangrove
rehabilitation. Pages 277–297 in W. J. Streever, ed. An international perspective on wetland
rehabilitation. Kluwer Academic Publishers, Dordrecht. 338 p.
Stroud, R. 1983. Marine recreational fisheries at the crossroads. Marine recreational fisheries
8: Proc. Eighth Annual Marine Recreational Fisheries Symp. Sport Fishing Institute, Wash-
ington, D.C. 236 p.
Spurgeon, J. 1998. e socio-economic costs and benefits of coastal habitat rehabilitation and
creation. Mar. Pollut. Bull. 37: 373–382.
Sullivan, G. 2001. Chapter four. Establishing vegetation in restored and created coastal wet-
lands. Pages 119–155 in J. B. Zedler, ed. Handbook for restoring tidal wetlands. CRC Press,
Boca Raton. 439 p.
LEWIS AND GILMORE: MANGROVE RE STORATION AND FISH HABITAT 837
Turner, R. E. and R.R. Lewis. 1997. Hydrologic restoration of coastal wetlands. Wetl. Ecol.
Manage. 4: 65–72.
___________ and B. Streever. 2002. Approaches to coastal wetland restoration: northern Gulf
of Mexico. SPB Academic Press bv. e Hague. 147 p.
Vivian-Smith, G. 2001. Box 2.1. Reference data for use in restoring Tijuana Estuary. Pages 59–63
in J. B. Zedler, ed. Handbook for restoring tidal wetlands. CRC Press, Boca Raton. 439 p.
Wolanski, E. and K. Boto. 1990. Mangrove swamp oceanography and links with coastal waters.
Estuar. Coast. Shelf Sci. 31: 503–504.
Whitman, R. L., Jr. and R. G. Gilmore, Jr. 1993. Comparative evaluation of fisheries community
structure and habitat relationships in natural and created saltmarsh ecosystems. Southwest
Florida Water Management District, Brooksville, FL. 46 p.
A: (R.R.L.) Lewis Environmental Services, Inc., P.O. Box 5430, Salt Springs, Florida
32134. (R.G.G.) Estuarine, Coastal and Ocean Science, Inc., 5920 First Street SW, Vero Beach,
Florida 32968. C A: (R.R.L.) E-mail: <lesrrl3@aol.com>.
