See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/274116107

Carbon sequestration in mangrove forests

Article  in  Carbon Management · April 2014

DOI: 10.4155/cmt.12.20

CITATIONS READS

470 34,341

1 author:

Daniel M. Alongi
Tropical Coastal & Mangrove Consultants
218 PUBLICATIONS   15,862 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Capturing carbon in mangroves, climate regulation View project

Guest Editor Special Issue "Climate Change Impacts on Mangrove Ecosystems" View project

All content following this page was uploaded by Daniel M. Alongi on 31 May 2015.

The user has requested enhancement of the downloaded file.

 Review

Carbon sequestration in mangrove forests

Carbon Management (2012) 3(3), 313–322

Daniel M Alongi*
Mangrove forests are highly productive, with carbon production rates equivalent to tropical humid forests.
Mangroves allocate proportionally more carbon belowground, and have higher below- to above-ground
carbon mass ratios than terrestrial trees. Most mangrove carbon is stored as large pools in soil and dead
roots. Mangroves are among the most carbon-rich biomes, containing an average of 937 tC ha-1, facilitating
the accumulation of fine particles, and fostering rapid rates of sediment accretion (~5  mm  year -1) and
carbon burial (174 gC m-2 year -1). Mangroves account for only approximately 1% (13.5 Gt year -1) of carbon
sequestration by the world’s forests, but as coastal habitats they account for 14% of carbon sequestration
by the global ocean. If mangrove carbon stocks are disturbed, resultant gas emissions may be very high.
Irrespective of uncertainties and the unique nature of implementing REDD+ and Blue Carbon projects,
mangroves are prime ecosystems for reforestation and restoration.

Mangrove forests are the only woody halophytes that live forests represent a continuum of types in relation to
in salt water along the world’s subtropical and tropical gradients in their physical settings. Variations can be
coastlines. Coincidentally, poverty and dense human expressed within a single estuary, where there are usually
populations flourish along these low-latitude coasts, upstream–downstream changes in geo­morphology, salin-
partly explaining the high (1–3%) annual de­forestation ity, waves, tides and river flow, with these factors affecting
rates of these tidal forests. Mangroves are true ecotones, water circulation by generating mixing and trapping of
having some components of both marine and terrestrial coastal water [2] . The development of mangrove forests
biomes, but have also developed a number of unique occurs where near-horizontal topography coincides with
structural and functional adaptations, such as viviparous sea level; a relatively stable period of sea level is, thus, a
embryos, physiological mechanisms to tolerate salt and prerequisite for the development of old-growth forests
aerial roots that enable the plants to respire in anoxic, [3] . The response of mangroves to environmental change
waterlogged soils [1] . Mangroves are architecturally sim- is, therefore, often indicative of past changes in coastal
ple compared with terrestrial forests, usually harboring conditions, especially in sea level. Comparing present
few tree species and lacking an understory of ferns and patterns in forest species with paleoecological informa-
scrubs. However, the standing biomass of some mangrove tion provides considerable insight, not only into how
forests in equatorial regions can be immense, rivaling the mangroves responded to past sea level changes, but how
height and weight of many tropical rainforests [1] . they may respond to climate change in the future.
Mangroves are ultimately limited by temperature but, Human disturbance obscures natural change and
at local and regional scales, variations in pre­cipitation, our ability to distinguish one from the another is lim-
tides, waves and river flow greatly determine their expanse ited, as most forests have a history of both natural and
and biomass. Attempts have often been made to classify human disturbances, and are often intertwined and
the sequential changes in forest structure and species dis- indistinguish­able. Mangroves are naturally disturbed by
tribution parallel to shore but, in reality, most mangrove tsunamis, floods, cyclones, lightning, pests and disease,

*Australian Institute of Marine Science, PMB 3, Townsville MC, Queensland 4810, Australia
E-mail: d.alongi@aims.gov.au

future science group 10.4155/CMT.12.20 © 2012 Future Science Ltd ISSN 1758-3004 313
 Review  Alongi

Key terms and become more susceptible when primary production (NPP) based on different methods
Mangroves: Trees and associated human stressors such as pollutants range from 0.5 to 112.1 t dry weight (DW) ha-1 year -1
plants, microbes and animals that live at are introduced. However, mangroves but most methods either significantly overestimate (the
the interface between land and sea. often exhibit considerable resilience light attenuation method) or underestimate (litterfall)
These tidal ecosystems have both
to disturbance, undergoing perpet- the true rates of production.
semi-terrestrial and marine
components. ual change in ecosystem develop- The most reasonable means at present to assess NPP of
Coastal: Land, water and aquatic
ment commensurate with the evo- forests is to measure aboveground biomass accumulation
habitats that reside where the lution of the environmental settings plus litterfall, and there are quite a number of such mea-
continents meet the ocean. These they inhabit, and are, thus, mosaics surements for both mangroves and tropical terrestrial
habitats are usually only a few of successional stages arrested or forests. For mangroves, the mean rate of above­ground
kilometers in width but are highly
dynamic and interactive with respect to
interrupted over time and space by NPP is 11.1 t DW ha-1 year -1 with a median value of
energy and material flow between land natural ecological responses in rela- 8.1 t DW ha-1 year -1; for tropical terrestrial forests, the
and sea. tion to disturbances both large and mean rate of aboveground NPP is 11.9 t DW ha-1 year -1
Carbon sequestration: Term used to small [4] . with a median value of 11.4 t DW ha-1 year -1; for both
describe the acquisition and storage of Mangrove forests are a valuable mangroves and terrestrial forests, NPP declines with
carbon. Refers most often in relation to ecological and economic resource, increasing latitude [1] . Considering the differences within
the ability of ecosystems to reduce the
impact of increasing CO2 concentrations providing essential services such as and between both forest groups in biomass, height, age
in the atmosphere. food and fuel resources; nursery and species, the rates are very close and clearly imply
grounds for fish, mammals and that rates of NPP are equivalent between mangroves and
other semi-terrestrial and aquatic fauna; depocenters other forests.
for sediment, carbon and other elements; and, in some Like other forests, mangroves vary in size and age and,
instances, offering some protection from coastal erosion therefore, vary in rates of production and in the bal-
due to tsunamis and intense tropical storms [1] . Despite ance between carbon production and respiration. The
their uses to humans, approximately 50% of the world’s few studies that have measured mangrove tree growth
mangrove forests have disappeared over the past 50 years over time or in stands of known age have observed stand
[5] , ironically reflecting their importance as a valuable dynamics similar to other forests, identifying stages of
economic resource. Major causes for this destruction early rapid growth during colonization and early estab-
have been urban development, aquaculture, mining, and lishment, followed by a slow decline in growth rate into
overexploitation of timber, fish, crustaceans and shell- maturity and senescence [1,10,11] . The stable-state matu-
fish. The average monetary value of mangroves has been rity phase can be prolonged in some mangrove stands
estimated as second only to the value of estuaries and and may represent an alternate succession state in which
seagrass meadows, and greater than the economic value the clock for the climax stage is reset by successive dis-
of coral reefs, continental shelves and the open sea [6] . turbances [10] . The relationship between mangrove forest
Of greater eventual value is the role of mangroves in age and photosynthetic production [11] suggests prolonga-
storing carbon to help ameliorate the impact of climate tion or arrested progression when forests are disturbed;
change. There is a growing consensus that it will be Rhizophora apiculata forests in southeast Asia show log-
impossible to achieve significant cuts in GHG emis- phase photosynthetic rates until approximately 20 years,
sions without passive and active means to capture and after which photosynthesis levels off but does not signifi-
store CO2 [7] . The role of carbon storage in mangroves cantly decline for nearly a century [1] . These data imply
has often been overlooked and either underestimated that mangroves might indeed constitute a carbon sink
or overestimated [1] , and it is the purpose of this review for up to a century if left relatively undisturbed.
to critically assess the role of mangroves in carbon Other primary producers inhabit mangrove forests
sequestration and its global significance. and their rates of NPP can be significant, especially in
comparatively open canopies and on tidal banks where
Carbon production sufficient light penetrates to the forest floor [1] . Various
Mangroves are usually highly productive forests and, as autotrophic and mixotrophic microbes and microalgae,
a significant fraction of their soil carbon is plant-derived as well as macroalgae, live on the soil surface and as
[8] , it is crucial to assess rates of net primary productivity epiphytes on tree parts, especially aerial roots and decom-
of mangroves and associated plants, especially benthic posing wood. The quantitative contribution of these
microalgae. Measurement of primary production in smaller autotrophs is dwarfed by tree production, but
mangrove forests is limited by methodological short- belies their importance as food and refugia for consum-
comings, but the best estimates suggest that mangrove ers. However, some evidence suggests that they can play
carbon production is more rapid than other estuarine an important role in soil carbon and nitrogen cycling,
and marine primary producers [9] . Rates of mangrove net especially when found as intact mats [12] .

314 Carbon Management (2012) 3(3) future science group

 Carbon sequestration in mangrove forests  Review

Carbon allocation & ecosystem storage A recent assessment of carbon stored in various forest
Critical to our ability to estimate the role of mangroves domains found that in comparison with boreal, temper-
in coastal and global carbon cycling is an accurate under- ate and tropical upland forests, mangroves throughout
standing of where carbon fixed by the trees is allocated. the Indo-Pacific are among the most carbon­-rich forests
Like other woody plants, mangroves construct new foli- in the tropics containing, on average, 1023 tC ha-1, most
age, reproductive organs, stem, branches and root tissues of which is stored in soils >30 cm deep [21] . Adding pub-
and maintain existing tissue, as well as creating storage lished and unpublished data by authors from southern
reserves and providing chemical defense. Approximately China, Vietnam, Indonesia, arid Western Australia,
half of all CO2 assimilated by mangroves is returned to Queensland, Thailand and Malaysia (Table 1) to the data
the atmosphere via above- and below-ground respiration set of Donato et al. [21] to diversify the geographical,
[1,11] . This is only a crude estimate owing to the lack subtropical and arid-zone forest domains, we obtain a
of empirical data and the difficulty of measuring root revised mean whole-ecosystem carbon storage estimate
processes and respiration of woody parts. The propor- of 937 tC ha-1 (Figure 1) , which still indicates that man-
tional allocation of fixed carbon within trees varies with groves are among the world’s most carbon-rich forests.
many factors, such as light intensity, species composi- It is possible, of course, this statement may not hold true
tion, nutrient and water availability, salinity, tides, waves, globally, especially when data is obtained from Central
temperature and climate [11] . and South America and Africa, and from more forests in
The greatest unknown with regard to carbon alloca- the arid tropics and subtropics where fringing mangroves
tion is root production, which is difficult to measure, and mangroves growing on hard and/or substrates of
especially in waterlogged soils. The few studies that have limited depth are common. Nevertheless, throughout the
measured root growth in situ estimated rates ranging from equatorial regions (e.g., the wet tropics of southeast Asia)
18 to 1145 g DW m-2 year -1 with most estimates between it is true that mature mangrove stands attain highest car-
300 and 380 g DW m-2 year -1 [1] . These estimates are at bon mass compared with other carbon-rich ecosystems,
the lower end of the range of values measured in tropical such as tropical rainforests.
terrestrial forests [13] . However, most measurements were What does inarguably appear to be a global pattern
made in mangrove fringe stands, so it is likely that the among mangrove forests is that their belowground pools
growth and production of mangrove roots is similar to of root and soil carbon are large, having a higher below-
their terrestrial counterparts. A recent ana­lysis of carbon to above-ground carbon mass ratio than any other woody
allocation suggests that mangroves allocate proportion- vegetation [22] .
ally more carbon belowground than terrestrial trees [14] . With the bulk of belowground carbon stored in dead
Carbon inventories from a number of mangrove eco- roots and soil rather than in live roots, mangroves have
systems show that both above- and below-ground biomass a tendency to accumulate carbon relatively quickly.
increases, and that the ratio of below- to above-ground Belowground roots may only represent approximately
biomass decreases with increasing stand age (Table 1) . 10–15% of total tree biomass, but the allocation of fixed
These data show that belowground carbon biomass is, carbon to replace sloughed root hairs and fine roots is
on average (mean = 1.3), equivalent to carbon allocated considerably greater [23,24] . Moreover, carbon concen-
aboveground; other studies have indicated that more car- trations in dead roots are greater than in live roots,
bon biomass is allocated belowground [15–18] supporting suggesting that dead roots store proportionally more
the notion that mangroves store a disproportionate frac- carbon [19,20] .
tion of fixed carbon underground. Further, the amount Vertical profiles of live versus dead root matter in a
of soil carbon increases with forest age (see Figure 5.1 in number of mangroves show that most living roots are
Alongi [1]). shallow, within the upper 0–40 cm of soil [1] . Most fine
Complete inventories of ecosystem components show roots are dead, probably the net result of rapid root turn-
that carbon fixed within the forest, as well as carbon over coupled with slow rates of root decomposition [23] .
imported from adjacent terrestrial and marine waters, Rates of belowground decomposition of fine and coarse
are stored as large pools of soil carbon [19,20] . Analysis of mangrove roots are indeed slow, with most rates ranging
carbon in Rhizophora stylosa and Avicennia marina in arid from 0.07 to 0.17% root mass lost per day; only roots of
coastal areas of Western Australia [19] and in R. apiculata A. marina decompose more quickly at rates varying from
forests in southern Thailand [20] showed that although 0.09 to 0.34% root mass lost per day [1] . Roots decom-
most carbon was unassociated with roots, the majority pose at equivalent rates regardless of intertidal elevation,
(75–95%) of tree carbon belowground was vested in but coarse roots decompose less quickly than fine roots.
dead, rather than live, roots. The Thai study also showed These slow decay rates explain the formation of peat in
that the soil and dead root carbon pools increased in size many mangrove forests as inputs must exceed decay rates
with increasing stand age [20] . in order for peat to accumulate [23–25] .

future science group www.future-science.com 315

 Review  Alongi

Table 1. Whole-ecosystem inventories of above- and below-ground carbon biomass and soil carbon for natural and replanted
mangrove forests.
Location Dominant species Age Total AGB BGB and Roots/AGB Roots Soil Soil depth
(years) (tC ha-1) (tC ha-1) soil (tC ha-1) (tC ha-1) (tC ha-1) (tC ha-1) (cm)
Peninsular Malaysia Rhizophora apiculata 80 2205 312 1893 NA NA NA 3800
R. apiculata 18 1117 193 924 NA NA NA 4000
R. apiculata 5 479 87 392 NA NA NA 2800
Southern Vietnam R. apiculata 6 1179 54 1125 NA NA NA 3400
R. apiculata 20 979 72 907 NA NA NA 2750
R. apiculata 35 1904 153 1752 NA NA NA 3600
Southern Kandelia candel NA 619 64 555 2.0 130 425 1850
China K. candel NA 391 43 348 2.2 94 254 1900
K. candel NA 332 7 325 1.1 8 317 1175
Indonesia Avicennia marina NA 437 24 413 NA NA NA 80
Rhizophora stylosa NA 703 19 684 NA NA NA 62
Sonneratia caseolaris NA 654 28 626 NA NA NA 1450
Southern Thailand R. apiculata 25 808 138 670 1.0 142 528 1900
R. apiculata 5 579 20 559 2.9 57 502 800
Ceriops decandra 3 600 29 571 4.4 127 444 1000
Western Australia R. stylosa NA 863 115 621 1.1 127 621 1500
A. marina NA 662 55 515 1.7 92 515 775
Queensland, Australia R. stylosa NA 2139 297 1842 1.1 312 1530 3500
AGB: Aboveground biomass; BGB: Belowground biomass; NA: Not available.
Data from [48,50–54,101].

Why do mangrove forests have such large amounts of very tight cycling between roots and microbes several
carbon vested belowground compared with terrestrial meters deep into the soil, possibly to curtail losses and
forests? The presence of a large pool of dead roots can to minimize energetic costs.
serve as a nutrient conserving mechanism, and even large
dead roots may serve this purpose. For instance, old root Mechanisms facilitating sediment accumulation
channels have been found in mangroves in central Belize Lying at the interface between land and sea, it is hardly
with a proliferation of living roots among the decaying surprising that mangroves accumulate sediment and
roots, taking paths of least resistance and recovering associated particulate elements, such as inorganic and
nutrients released from decomposing roots [25] . A large organic carbon. What is surprising is that their presence
pool of belowground live and dead root biomass mixed actively facilitates the accumulation of materials [26] .
with rich soils may reflect their numerous physiological Carbon is accumulated in mangroves by direct inputs
and morphological adaptations to life in a harsh, saline of mangrove carbon to the soil pool and by increasing
waterlogged environment. Salt negatively affects water rates of mass sediment accumulation. Carbon produced
use and under such conditions it may be advantageous by mangroves does have other flow pathways, such as
for mangrove trees to invest more fixed carbon in grow- consumption by living organisms, especially microbes.
ing very expensive root systems that turnover rapidly in Carbon consumed is remineralized and either emitted
order to maximize water gain. Large reservoirs beneath back to the atmosphere as CO2 or exported by dissolved
the forest floor may also help to stabilize the trees and inorganic carbon. Dissolved and particulate organic
the entire ecosystem from the continual push and pull carbon is also exported by tides where it can be either
of the tides, wave action, coastal winds and tropical deposited or eaten or mineralized offshore.
storms. It makes evolutionary sense for mangroves to The amount of carbon stored in mangrove soils varies
invest in a large belowground pool of carbon biomass widely, from <0.1% by soil dry weight to >40% with a
as an effective counterbalance to litter and carbon dis- grand median of 2.2% [8] . A highly variable proportion
solved in interstitial water that is of this carbon is mangrove-derived as organic matter is
Key term lost via the tides. Whereas tropical brought in by the tides from adjacent seagrass mead-
Flocculation: Physical, chemical and humid forests recycle nutrients by ows, coral reefs, macroalgae, rivers and from land-based
microbial processes by which particles rapid soil decomposition of litter in sources, and other marine environments [8] . The frac-
are cemented together; the term ‘floc’ a relatively thin humus layer, man- tion of mangrove-derived carbon in forest soils depends
refers to the cemented tuft-like mass.
groves reclaim elements by way of on a number of factors, including location of the forest

316 Carbon Management (2012) 3(3) future science group

 Carbon sequestration in mangrove forests  Review

in relation to the open coast, dis-

1200
tance to adjacent aquatic habitats,

Ecosystem carbon storage (tonnes ha-1)

Aboveground carbon storage
tidal amplitude, forest position in
the tidal seascape and productivity 1000 Belowground carbon storage
of primary producers [27] .
Unconsolidated sediments accu- 800
mulate in relation to the movement
of the turbidity maximum zone, 600
where incoming bottom flow meets
outward river flow. Tidal mixing 400
and pumping within the moving
zone facilitate particle flocculation
200
and settlement. Flocculation of
particles begins at salinities <1, and
0
small flocs and free particles move Boreal Temperate Tropical Mangroves
downstream where they aggregate
with local particles [28] . As flocs get Figure 1. Differences in whole-ecosystem carbon stocks among boreal, temperate and
larger, they move toward the river tropical terrestrial forests, and subtropical and tropical mangrove forests.
bed where they are entrained back Reprinted with permission from Macmillan Publishers Ltd: Nature Geoscience [21] © (2011).
upstream by baroclinic circulation Mangrove data taken from supplementary data [102] in [21], and [48,50–54].
and even further upstream at flood
tide due to tidal pumping [28] . As these flocs move into carbon losses equal carbon either buried or unaccounted
the forest on flood tides, turbulence generated by flow for) or reflect mostly modern rates of accumulation
around the trees helps to maintain flocs in suspension (measurement of short-term sediment accumulation
[2] . Settling occurs quickly, facilitated by the sticking of using sediment traps or changes in depth of the soil
microbial mucus in the soil surface and by pelletization profile). Analysis of radioactive elements produced by
by invertebrate excreta. Large quantities of nonfloccu- fallout (excess 210Pb and 137Cs) from atomic bomb test-
lated particles are re-exported from the forest on ebb ing in the atmosphere coupled with estimates of soil
tide, but most stick to mucus at the water surface. carbon concentrations provide longer term estimates
Mangroves thus actively capture silt, clay and organic of accumulation and a chronology of sedimentation of
particles, and are not just passive importers of fine par- up to a century. Such methods also have their pitfalls,
ticles [2,28] ; mangrove vegetation has a profound impact including reliance on expensive analytical equipment,
on sedimentation. Large trees with complex root sys- difficulty in interpreting radiotracer profiles in biotur-
tems, such as Rhizophora species, facilitate the deposi- bated and disturbed soils and in soils where there are
tion of particles to a much larger extent than trees that vertical changes in grain size, and problems with error
are smaller and of much simpler architecture, such as induced by compaction of sediments as a result of the
Ceriops species. Until slack water, turbulent wakes cre- coring process [29] .
ated by tree trunks, prop roots and pneumatophores The rate of soil accretion in mangrove forests averages
maintain particles in suspension, but most flocs settle 5 mm year -1, with 94 measurements out of a total of
within 30 min just before slack high tide [28] . Despite the 139 ranging from 0.1 to 10.0 mm year -1 (Figure 2) . The
pull of the ebb tide, most flocs are retained within the median value is 2.7 mm year -1 with a few measurements
forest as water motion and turbulence necessary for their showing net erosion (minimum value = -11.0 mm year -1)
resuspension is inhibited by the high vegetation density. or massive accretion (46.3 mm year-1) in highly-impacted
estuaries, such as those in southern China [30] .
Rates of soil accretion & carbon sequestration Frequency of tidal inundation is the primary factor
Mangroves accumulate carbon in tree biomass, but controlling the rate of accretion [31–33] . Less frequent
much of this carbon is eventually lost in the short- and inundation by tides means less input of sediment par-
medium-term by way of clear-cutting and human use, ticles; forests located in the high intertidal area experi-
decomposition and export to adjacent ecosystems. Over ence less soil accretion than forests closer to mean sea
the long term, carbon is stored primarily belowground as level, such as fringing stands at the sea–forest interface.
soil carbon and, eventually, under the right conditions, In fact, mangrove carbon often accumulates on adjacent
as peat. There are a number of methods to measure margins and intertidal mudflats [34] . Often overlooked,
soil accretion [29] , but some are either highly inaccurate because empirical data are rare, are contributions to ver-
(a mass balance approach where carbon inputs minus tical accretion from the growth of belowground roots

future science group www.future-science.com 317

 Review  Alongi

accretion rates at a number of these

100
Mean = 4.996 mm year -1 endangered forests is higher than
SD of mean = 7.402 mm year -1 the eustatic sea level rise.
Confidence interval of mean = 1.241 mm year -1 Available data on burial rates of
80 Median = 2.7 mm year -1 carbon in mangrove ecosystems
Minimum = -11.0 mm year -1 were first compiled by Twilley et al.
Number of measurements

Maximum = 46.3 mm year -1 [39] , later updated by Jennerjahn

and Ittekot [40] and Duarte et  al.
60
[9] , based on data in Chmura et al.
[41] . Despite the different databases
and methods used, all derived a
40 similar estimate of a global car-
bon burial rate of approximately
23  TgC  year -1, which is equiva-
lent to a rate of 167 gC m-2 year -1
20 assuming a total mangrove area of
137,760 km2 [5] . Bouillon et al. [26,27]
and Alongi [1] derived carbon
0 burial rates of 18.4  TgC  year -1
-20 -10 0 10 20 30 40 50 (= 134 g C m -2
 year-1) and 29 TgC year-1
Sediment accretion rates (mm year -1) (= 211 g C m -2
 year -1), respectively.
Adding more recent data derived
from radiochemical methods, we
Figure 2. Sediment accretion rates measured in various mangrove forests
can revise the mean global burial
worldwide (n = 139).
rate for soil carbon to 24 TgC year -1,
Data from [1,24,28–31,33,35,38,48,51,55].
equivalent to 174 g C m-2 year -1 with
and surface growth of microbial mats and turf algae values ranging from 10 to 920 g C m-2 year -1; the median
and accumulation of litter. In some cases, such as in the burial rate was 16 TgC year -1 (= 115 g C m-2 year -1).
Caribbean, contributions from these biological sources Like the sediment accretion data, the standard devia-
can be greater than accretion of mineral particles [33] . tion exceeds the mean reflecting the high level of
On other islands such those in the Federates States of variability (and uncertainty) in carbon burial rates
Micronesia, natural subsidence plays a key role in overall among forests worldwide. Nevertheless, most individ-
rates of net elevation [35] , although not the actual rates ual estimates (47 of a total of 66 measurements) are
of soil accretion; nevertheless, such changes are impor- <200 g C m-2 year -1, with a minority of forests accumu-
tant in determining the susceptibility of mangroves lating soil carbon faster (Figure 3) , mostly in catchments
to changes in sea level [35] . Rates of soil accretion can heavily impacted by human activities, such as those in
vary over long timescales. In lagoon mangroves on the southern China [30] and in southeast Asia [42] .
Yucatan Peninsula in Mexico, for instance, natural vari-
ations in accumulation rates and sources of soil carbon Significance of mangroves to terrestrial &
were detected over the past 160 years [36] . These changes marine carbon sequestration
corresponded to fluctuations in climatic variability in How do these new estimates of carbon sequestration
the region. compare with other forested and coastal ecosystems?
Key terms
Mangrove sedimentation in rela- Globally, are mangroves a significant sink for carbon?
tion to sea level rise was assessed Does their loss represent a significant return of CO2
REDD+: Acronym for Reducing
Emissions from Deforestation and
by Alongi, who found that most to the atmosphere?
Forest Degradation. The + refers to the mangrove forests were currently The data presented here confirm the notion that
additional steps of conservation and the keeping pace with local rises in sea mangroves are among the most carbon-rich eco­systems
sustainable management of forests and level [37] . However, there are a num- in the tropics. But at a global level, mangroves occupy
enhancement of forest carbon stocks.
ber of regions where sedimentation only approximately 137, 760 km 2, and a simple scal-
Blue carbon: Term coined to refer to rates are lower than the rates of ing up of the mean carbon burial rate equates to a
steps designed to enhance the
acquisition and storage of carbon in regional relative sea level rise, such global carbon sequestration rate of 13.53 Gt year -1. The
aquatic ecosystems, especially in coastal as on some Pacific Islands [36] and same exercise for boreal, temperate and tropical terres-
habitats such as seagrass beds and at a number of mangrove stands trial forests extrapolates to global sequestration rates
mangrove forests.
in the Caribbean [33,38] , although of 451.1, 327.6, and 422.4 Gt year -1, respectively  [43] .
 

318 Carbon Management (2012) 3(3) future science group

 Carbon sequestration in mangrove forests  Review

Thus, mangroves account for

approximately 3% of carbon seques- 30
tered by the world’s tropical forests, Mean = 174.1 mm year -1
although they account for <1% of SD of mean = 184.1 mm year -1
total area of tropical forests. 25 Confidence interval of mean = 45.3 mm year -1
These data do, however, suggest Median = 114.5 mm year -1
the potential for significant GHG Minimum = 10.0 mm year -1

Number of measurements
emissions if the high per-hectare 20
Maximum = 920.0 mm year -1
carbon stocks of mangroves are
disturbed. Losses of mangroves by
clearing, conversion to industrial 15
estates/aquaculture and changes in
drainage patterns lead to dramatic
changes in soil chemistry and usu-
10
ally result in rapid emission rates of
GHGs, especially CO2. For exam-
ple, deforesting mangroves that
5
grow on peat soils results in CO2
emissions comparable to rates esti-
mated from collapse of terrestrial
0
peat soils [44] . Lovelock et al. mea- 0 200 400 600 800 1000
sured CO2 emissions from cleared
Annual rates of carbon burial (g m-2 year -1)
mangrove peat soils in Belize on
the order of 2900  tC  km-2  year -1 

[44] ; this value compares well

with CO 2 emissions measured Figure 3. Annual rates of carbon burial estimated in various mangrove forests
from hurricane-damaged and worldwide (n = 66).
aquaculture-impacted mangroves Data from [1,9,15–18,21,25,28,30–32,34,38,41,42,48,51,55–59].
(1500–1750  tC  km-2  year -1), rain-
 

forests drained for agriculture (3200 tC km-2 year -1) much greater than that from all other habitats, except
 

and thawed Arctic tundra (150–430 tC km-2 year -1). for salt marshes. Therefore, considering the data in
 

Donato et al. [21] calculated a plausible range of CO2 Figure 1 and in Table 2, mangrove forests have the high-
emissions of 112–392 tC released per hectare of man- est area rates of carbon sequestration compared with
grove forest and soils cleared, which gives a global emis- any other ecosystem, terrestrial or marine, contributing
sions range of 0.02–0.12 PgC year -1, assuming current disproportionately as a carbon sink.
 

deforestation rates (1–2% per year) and global area.

This range is equivalent to at least 2–10% of global Future perspective
deforestation emissions (~1.2 PgC year -1 [45]) and up to Mangroves are currently being advanced as an essential
 

50% of emissions from the world’s tropical peatlands component of climate change strategies such as REDD+
(0.24 PgC year -1 [46]). These values are only indicative, and blue carbon. McLeod et al. [47] and Alongi [48] have
 

as large uncertainties remain, including the accuracy recently identified specific actions and issues that need
of forest areas, temporal and spatial variations in fluxes to be addressed in blue carbon projects:
and standing stocks, local and regional differences in
ƒƒ Careful site selection, preferably at the seaward edge,
the modes of disturbance, and variations in the depth
based on drivers thought to affect carbon sequestra-
to which soil is dredged.
tion rates, such as frequency of tidal inundation, pri-
If the contribution of mangroves to global forest car-
mary productivity and rates of exchange with adjacent
bon sequestration is very small, their contribution to
ecosystems, as not all mangroves accumulate carbon;
carbon burial in the global coastal ocean is considerably
greater. Compared with other coastal ecosystems, man- ƒƒ Measure and map the spatial and temporal variations
groves contribute an average of 14% to carbon seques- in carbon stocks and burial rates, relating these fac-
tration in the world’s oceans, although accounting for tors to environmental and ecological drivers, possibly
only 0.5% of total coastal ocean area (Table 2) . determining a set of indicators that can be used to
Even considering the large uncertainties in these esti- quickly estimate changes in carbon stocks and
mates, the average burial rate of carbon in mangroves is fluxes;

future science group www.future-science.com 319

 Review  Alongi

Table 2. Global contribution of mangroves and other coastal habitats to carbon sequestration in the global
coastal ocean.
Habitat Area (1012 m2) Sequestration rate Global carbon sequestration
(gC m-2 year -1) (Tg year -1)
Mangroves 0.14 (0.5%) 174 24 (14%)
Salt marshes 0.22 (0.8%) 150 33 (20%)
Seagrasses 0.3 (1.1%) 54 16 (10%)
Estuaries 1.1 (4.0%) 45 50 (30%)
Shelves 26 (93.6%) 17 44 (26%)†
Total 167
†
Assumes that depositional areas cover 10% of total shelf area [9].
Data from [41,60–62].

ƒƒ Remote sensing and aerial photography may be useful Future climate scenarios for the ocean are subject to
to facilitate changes in restoration/rehabilitation large uncertainties, but regional changes in ocean cir-
strategies, and in identifying changes in land use; culation, temperature, salinity and pH patterns, and in
sea level, must be considered as likely to have a strong
ƒƒ Standardization of methods used to measure biomass
impact on the ability of mangroves to sequester carbon
and soil carbon stocks and rates of carbon burial; [49] . Large uncertainties exist in our knowledge of carbon

ƒƒ The execution of any scheme must consider modeled sequestration in mangroves, and such limitations must be
predictions of future climate changes, such as regional factored into the blueprints of any payment for ecosystem
predicted rises in sea level; services, blue carbon or REDD+ schemes. Only then
will management of mangrove ecosystems be sustainable.
ƒƒ Planting of mixed species to maximize biodiversity,
food web connectivity and net ecosystem production; Financial & competing interests disclosure
The author has no relevant affiliations or financial involvement with
ƒƒ Priority must be given to REDD+ schemes that give
any organization or entity with a financial interest in or financial
priority to old-growth forests as mangrove carbon
conflict with the subject matter or materials discussed in the manu-
stocks increase with stand age;
script. This includes employment, consultancies, honoraria, stock
ƒƒ Studies should be conducted concurrently to assess ownership or options, expert t­estimony, grants or patents received or
the conditions that determine whether or not climate pending, or royalties. No writing assistance was utilized in the
change impacts such as changes in sea. production of this manuscript.

Executive summary
Carbon production
ƒƒ Mangrove net primary production averages 11.1 t dry weight ha-1 year-1, roughly equivalent to tropical terrestrial forests.
ƒƒ Mangroves may constitute a carbon sink for up to a century.
Carbon allocation & storage
ƒƒ Belowground biomass is equivalent to aboveground biomass in mangroves.
ƒƒ Most carbon in mangroves is stored as large pools of soil carbon and belowground roots.
ƒƒ Storage of carbon in mangroves averages 937 tC ha-1.
Mechanisms facilitating sediment accretion
ƒƒ Mangroves actively facilitate accumulation of carbon and other elements associated to fine particles.
Rates of soil accretion & carbon sequestration
ƒƒ Rates of soil accretion in mangroves average 5 mm year -1.
ƒƒ Frequency if tidal inundation is the main factor controlling accretion.
ƒƒ Global carbon burial rates for mangroves approximate 24 TgC year -1.
Significance of mangroves to terrestrial & marine carbon sequestration
ƒƒ Mangroves account for 3% of carbon sequestered by the world’s tropical forests, but 14% of carbon sequestered in the world’s ocean.
ƒƒ If disturbed, mangroves may emit 0.02–0.12 PgC year -1, equal to 2–10% of global deforestation emissions.
Future perspective
ƒƒ Mangroves are prime candidates for REDD+ and blue carbon projects, but a number of issues and specific actions must be carefully
addressed prior to commencement of such projects.

320 Carbon Management (2012) 3(3) future science group

 Carbon sequestration in mangrove forests  Review

References 14 Lovelock CE. Soil respiration and 28 Furukawa K, Wolanski E. Sedimentation in

belowground carbon allocation in mangrove mangrove forests. Mangroves Salt Marshes 1,
1 Alongi DM. The Energetics of Mangrove
forests. Ecosystems 11, 342–354 (2008). 3–10 (1996).
Forests. Springer, Amsterdam, The
Netherlands (2009). 15 Fujimoto K, Imaya A, Tabuchi R et al. 29 Bird MI, Fifield LK, Chua S et al. Calculating
Belowground carbon storage of Micronesian sediment compaction for radiocarbon dating
2 Mazda Y, Wolanski E. Hydrodynamics and
mangrove forests. Ecol. Res. 14, 409–413 of intertidal sediments. Radiocarbon 46,
modeling of water flow in mangrove areas. In:
(1999). 421–435 (2004).
Coastal Wetlands: An Integrated Ecosystem
Approach. Perillo GME, Wolanski E, Cahoon 16 Ren H, Chen H, Li Z et al. Biomass 30 Alongi DM, Pfitzner J, Trott LA et al. Rapid
DR, Brinson MM (Eds). Elsevier, accumulation and carbon storage of four sediment accumulation and microbial
Amsterdam, The Netherlands, 231–262 different aged Sonneratia apetala plantations mineralization in forests of the mangrove
(2009). in southern China. Plant Soil 327, 279–291 Kandelia candel in the Jiulongjiang Estuary,
(2010). China. Estuar. Coast. Shelf Sci. 63, 605–618
3 Ellison JC. Geomorphology and
17 Nguyen HT, Yoneda R, Ninomiya I et al. The (2005).
sedimentology of mangroves. In: Coastal
Wetlands: An Integrated Ecosystem Approach. effects of stand age and inundation on carbon 31 Mahmood H, Misri K, Sidik BJ et al.
Perillo GME, Wolanski E, Cahoon DR, accumulation in mangrove plantation soil in Sediment accretion in a protected mangrove
Brinson MM (Eds). Elsevier, Amsterdam, Namdinh, Northern Vietnam. Tropics 14, forest of Kuala Selangor, Malaysia. Pakistan
The Netherlands, 565–591 (2009). 21–37 (2004). J. Biol. Sci. 8, 149–151 (2005).

4 Berger U, Adams M, Grimm V et al. 18 Ray R, Ganguly D, Chowdhury C et al. 32 Ceron-Breton JG, Ceron-Breton RM,
Modelling secondary succession of subtropical Carbon sequestration and annual increase of Rangel-Marron M et al. Determination of
mangroves: causes and consequences of carbon stock in a mangrove forest. Atmos. carbon sequestration rate in soil of a
growth reduction in pioneer species. Persp. Environ. 45, 5016–5024 (2011). mangrove forest in Campeche, Mexico.
Plant Ecol. Evol. Syst. 7, 243–252 (2006). 19 Alongi DM, Clough BF, Dixon P et al. WSEAS Trans. Environ. Develop. 7, 55–64
Nutrient partitioning and storage in arid-zone (2011).
5 Giri C, Ochieng E, Tiszen LL et al. Status
and distribution of mangrove forests of the forests of the mangroves Rhizophora stylosa 33 McKee KL. Biophysical controls on accretion
world using earth observation satellite data. and Avicennia marina. Trees 17, 51–60 (2003). and elevation change in Caribbean mangrove
Global Ecol. Biogeogr. 20, 154–159 (2010). 20 Alongi DM, Wattayakorn G, Tirendi F et al. ecosystems. Estuar. Coast. Shelf Sci. 91,
Nutrient capital in different aged forests of the 475–483 (2011).
6 Costanza RR, d’Arge R, deGroot R et al. The
value of the world’s ecosystem services and mangrove Rhizophora apiculata. Bot. Mar. 47, 34 Sanders CJ, Smoak JM, Naidu AS et al.
natural capital. Ecol. Econ. 25, 3–15 (1998). 116–124 (2004). Organic carbon burial in a mangrove forest,
21 Donato DC, Kauffman JB, Murdiyarso D margin and intertidal mud flat. Estuar. Coast.
7 Mills RM. Capturing Carbon: The New
et al. Mangroves among the most carbon-rich Shelf Sci. 90, 168–172 (2010).
Weapons in the War against Climate Change.
Columbia University Press, NY, USA, 465 forests in the tropics. Nat. Geosci. 4, 293–297 35 Krauss KW, Cahoon DR, Allen JA et al.
(2011). (2011). Surface elevation change and susceptibility of
22 Saenger P. Mangrove Ecology, Silviculture and different mangrove zones to sea-level rise on
8 Kristensen E, Bouillon S, Dittmar T et al.
Conservation. Kluwer, Dordrecht, The Pacific high islands of Micronesia. Ecosystems
Organic carbon dynamics in mangrove
Netherlands, 342 (2002). 13, 129–143 (2010).
ecosystems: a review. Aq. Bot. 89, 201–219
(2008). 23 McKee KL, Faulkner PL. Restoration of 36 Gonneea ME, Paytan A, Herrera-Silva JA.
biogeochemical function in mangrove forests. Tracing organic matter sources and carbon
9 Duarte CM, Middelburg JJ, Caraco N. Major
Restor. Ecol. 8, 247–259 (2000). burial in mangrove sediments over the past
role of marine vegetation on the oceanic
160 years. Estuar. Coast. Shelf Sci. 61,
carbon cycle. Biogeoscience 2, 1–8 (2005). 24 Cahoon DR, Hensel P, Rybczyz J et al. Mass
211–227 (2004).
10 Fromard F, Puig C, Mougin E et al. tree mortality leads to mangrove peat collapse
at Bay Islands, Honduras after Hurricane 37 Alongi DM. Mangrove forests: resilience,
Structure, above-ground biomass and
Mitch. J. Ecol. 91, 1093–1105 (2003). protection from tsunamis, and responses to
dynamics of mangrove ecosystems: new data
global climate change. Estuar. Coast. Shelf Sci.
from French Guiana. Oecologia 115, 39–53 25 McKee KL. Root proliferation in decaying
76, 1–13 (2008).
(1998). roots and old root channels: a nutrient
conservation mechanism in oligotrophic 38 Sanders CJ, Smoak JM, Naidu AS et al.
11 Clough BF, Ong JE, Gong WK. Estimating
mangrove forests? J. Ecol. 89, 876–887 Mangrove forest sedimentation and its
leaf area index and photosynthetic production
(2001). reference to sea level rise, Cananeia, Brazil.
in canopies of the mangrove Rhizophora
Environ. Earth Sci. 60, 1291–1301 (2010).
apiculata. Mar. Ecol. Prog. Ser. 159, 285–292 26 Bouillon S, Rivera-Monroy VH, Twilley RR
(1997). et al. Mangroves. In: The Management of 39 Twilley RR, Chen RH, Hargis T. Carbon
Natural Coastal Carbon Sinks. Laffoley D, sinks in mangroves and their implications to
12 Joye SB, Lee RY. Benthic microbial mats:
Grimsditch G (Eds). International Union for carbon budget of tropical coastal ecosystems.
important sources of fixed nitrogen and
Conservation of Nature, Gland, Switzerland, Water Air Soil Pollut. 64, 265–288 (1992).
carbon to the Twin Cays, Belize ecosystem.
Atoll Res. Bull. 53, 1–24 (2004). 13–20 (2009). 40 Jennerjahn TC, Ittekot V. Relevance of
27 Bouillon, S, Borges AV, Castaneda-Moya E mangroves for the production and deposition
13 Perry DA, Oren R, Hart SC. Forest Ecosystems
et al. Mangrove production and carbon sinks: of organic matter along tropical continental
(2nd Edition). Johns Hopkins University
a revision of global budget estimates. Global margins. Naturwissenschaften 89, 23–30
Press, Baltimore, MD, USA, 596 (2008).
Biogeochem. Cycles 22, GB2013 (2008). (2002).

future science group www.future-science.com 321

 Review  Alongi

41 Chmura GL, Anisfield SC, Cahoon DC et al. World’s Climate. Henderson-Sellers A, ecosystems estimated by 210Pb chronology.
Global carbon sequestration in tidal, saline McGuffie K (Eds). Elsevier, MA, USA, Radioprotection 40, S527–S532 (2005).
wetland soils. Global Biogeochem. Cycles 17, 141–165 (2012). 57 Xiaonan D, Xiaoke W, Lu F, Zhiyun O.
1111 (2003). 50 Alongi DM, Trott LA, Rachmansyah et al. Primary evaluation of carbon sequestration
42 Fujimoto K. Below-ground carbon Growth and development of mangrove forests potential of wetlands in China. Acta Ecol.
sequestration of mangrove forests in the overlying smothered coral reefs, Sulawesi and Sinica 28, 463–469 (2008).
Asia–Pacific region. In: Mangrove Sumatra, Indonesia. Mar. Ecol. Prog. Ser. 370, 58 Kauffman JB, Heider C, Cole TG et al.
Management and Conservation: Present and 97–109 (2008). Ecosystem carbon stocks of Micronesian
Future. Vannucci M (Ed.). United Nations 51 Alongi DM, Dixon P. Mangrove primary mangrove forests. Wetlands 31, 343–352
University Press, Tokyo, Japan, 138–146 production and above- and below-ground (2011).
(2004). biomass in Sawi Bay, southern Thailand. 59 Mitra A, Sengupta K, Banerjee K. Standing
43 IPCC. Good Practice Guidance for Land Use, Phuket Mar. Biol. Cent. Sp. Publ. 22, 31–38 biomass and carbon storage of above-ground
Land-Use Change, and Forestry. Penman J, (2000). structures in dominant mangrove trees in the
Gytarsky M, Hiraishi T et al. (Eds). IPCC, 52 Alongi DM, Sasekumar A, Chong VC et al. Sunderbans. Forest Ecol. Manag. 261,
Kamiyamaguchi, Japan, 632 (2003). Sediment accumulation and organic material 1325–1335 (2011).
44 Lovelock CE, Ruess RW, Feller IC. CO2 flux in a managed mangrove ecosystem: 60 Matsui N. Estimated stocks of organic carbon
efflux from cleared mangrove peat. PLoS ONE estimates of land–ocean–atmosphere in mangrove roots and sediments in
6, e21279 (2011). exchange in peninsular Malaysia. Mar. Geol. Hinchinbrook Channel, Australia. Mangroves
45 van der Werf GR, Morton DC, DeFries RS 208, 383–402 (2004). Salt Marshes 2, 199–204 (1998).
et al. CO2 emissions from forest loss. Nat. 53 Alongi DM, Tirendi F, Trott LA et al. 61 Kennedy H, Beggins J, Duarte CM et al.
Geosci. 2, 737–738 (2009). Benthic decomposition rates and pathways in Seagrass sediments as a global carbon sink:
46 Page SE, Rieley JO, Banks CJ. Global and plantations of the mangrove Rhizophora isotopic constraints. Global Biogeochem. Cycles
regional importance of the tropical peatland apiculata in the Mekong delta, Vietnam. Mar. 24, GB4026, (2010).
carbon pool. Global Change Biol. 17, 798–818 Ecol. Prog. Ser. 194, 87–101 (2000).
62 Cai W-J. Estuarine and coastal ocean carbon
(2011). 54 Clough BF. Mangrove forest productivity and paradox: CO2 sinks or sites of terrestrial
47 McLeod E, Chmura GL, Bouillon S et al. biomass accumulation in Hinchinbrook carbon incineration? Annu. Rev. Mar. Sci. 3,
A blueprint for blue carbon: toward an Channel, Australia. Mangroves Salt Marshes 2, 123–145 (2011).
improved understanding of the role of 191–198 (1998).
vegetated coastal habitats in sequestering 55 Duarte CM, Kennedy H, Marba N et al. ƒƒ Websites
CO2. Front. Ecol. Environ. 9, 552–560 (2011). Assessing the capacity of seagrass meadows for
101 Data archives of the Australian Institute of
48 Alongi DM. Carbon payments for mangrove carbon burial: current limitations and future
Marine Science, 1979–1999.
conservation: ecosystem constraints and strategies. Ocean Coast. Manag. doi:10.1016/
http://data.aims.gov.au/metadataviewer/
uncertainties of sequestration potential. j.ocecoaman.09.001 (2011) (In press).
faces/search.xhtml
Environ. Sci. Policy 14, 462–470 (2011). 56 Tateda Y, Nhan DD, Wattayakorn G et al.
102 Nature Geoscience. Supplementary data.
49 Sen Gupta A, McNeil B. Variability and Preliminary evaluation of organic carbon
www.nature.com/naturegeoscience
change in the ocean. In: The Future of the sedimentation rates in Asian mangrove coastal

322 Carbon Management (2012) 3(3) future science group

View publication stats