Catena 200 (2021) 105159

Contents lists available at ScienceDirect

Catena
journal homepage: [Link]/locate/catena

Is soil organic carbon underestimated in the largest mangrove forest

ecosystems? Evidence from the Bangladesh Sundarbans
Md. Saidur Rahman a, b, *, Daniel N.M. Donoghue a, Louise J. Bracken a
a
Department of Geography, Durham University, South Road, Durham DH1 3LE, United Kingdom
b
Forestry and Wood Technology Discipline, Khulna University, Khulna 9208, Bangladesh

A R T I C L E I N F O A B S T R A C T

Keywords: Globally, mangroves sequester a large amount of carbon into the sediments, although spatial heterogeneity exists
Soil organic carbon owing to a wide variety of local, regional, and global controls. Rapid environmental and climate change,
Salinity zones including increasing sea-level rise, global warming, reduced upstream discharge and anthropogenic activities,
Soil depth
are predicted to increase salinity in the mangroves, especially in the Bangladesh Sundarbans, thereby disrupting
Mangrove forest
The Sundarbans
this blue carbon reservoir. Nevertheless, it remains unclear how salinity affects the belowground soil carbon
despite the recognised effect on above ground productivity. To address this gap, research was undertaken in the
Bangladesh Sundarbans to compare total soil organic carbon (SOC) across three salinity zones and to explore any
potential predictive relationships with other physical, chemical properties and vegetation characteristics. Total
SOC was significantly higher in the oligohaline zone (74.8 ± 14.9 Mg ha− 1), followed by the mesohaline (59.3 ±
15.8 Mg ha− 1), and polyhaline zone (48.3 ± 10.3 Mg ha− 1) (ANOVA, F2, 500 = 118.9, p < 0.001). At all sites, the
topmost 10 cm of soil contained higher SOC density than the bottom depths (ANOVA, F3, 500 = 30.1, p < 0.001).
On average, Bruguiera sp. stand holds the maximum SOC measured, followed by two pioneer species Sonneratia
apetala and Avicennia sp. Multiple regression results indicated that soil salinity, organic C:N and tree diameter
were the best predictor for the variability of the SOC in the Sundarbans (R2 = 0.62). Despite lower carbon in the
soil, the study highlights that the conservation priorities and low deforestation rate have led to less CO2 emis­
sions than most sediment carbon-rich mangroves in the world. The study also emphasised the importance of
spatial conservation planning to safeguard the soil carbon-rich zones in the Bangladesh Sundarbans from
anthropogenic tourism and development activities to support climate change adaptation and mitigation
strategies.

1. Introduction sequestration, United Nations Environmental Programme (UNEP)

designated this ecosystem as “Blue Carbon” along with other coastal
Mangroves are recognised as one of the most carbon-dense forest vegetated ecosystems such as seagrass meadows and saltmarshes (Nel­
types in the world due to their efficient carbon sequestration capacity lemann et al., 2009; Lovelock and Duarte, 2019; Macreadie et al., 2019).
into both above and below ground carbon pools (Donato et al., 2011; This growing worldwide importance of mangroves has led to a sub­
Alongi, 2012; Sanderman et al., 2018). Recent assessments of soil carbon stantial reduction of mangrove loss leading to reductions in CO2 emis­
suggest that mangrove ecosystems contain, on an average, between 856 sions in the last three decades (Friess et al., 2019). At the same time,
and 1023 Mg of carbon per hectare, with the majority (~85%) of this mangroves have gained substantial traction in being managed, protected
carbon stored in the soil (Donato et al., 2011; Pendleton et al., 2012; and restored as part of national and global climate change mitigation
Sanderman et al., 2018; Kauffman et al., 2020). This large amount of soil policies and actions including Nationally Determined Contributions
carbon is of global importance due to its potential to store sequestered (NDC) towards the Paris Agreement and climate action goal (goal 13)
CO2 emissions for the long term and help mitigate adverse effects of under United Nations Sustainable Development Goals (SDG) (Taillardat
climate change (McLeod et al., 2011; Duarte et al., 2013; Abdullah et al., et al., 2018; Friess et al., 2020). However, variability and uncertainty in
2016). To recognise the importance of mangrove forests for carbon SOC estimation is a key barrier to the inclusion of mangroves (and other

* Corresponding author at: Department of Geography, Durham University, South Road, Durham DH1 3LE, United Kingdom.
E-mail address: msrahman@[Link] (Md.S. Rahman).

[Link]
Received 1 July 2020; Received in revised form 3 November 2020; Accepted 7 January 2021
Available online 21 January 2021
0341-8162/© 2021 Elsevier B.V. All rights reserved.
Md.S. Rahman et al. Catena 200 (2021) 105159

blue carbon) in national and international policy tools and frameworks. (Rogers et al., 2019). These physical and biological factors and
Despite covering only 0.1% of the world’s total landmass, mangroves geomorphic evolutionary processes promote and develop unique coastal
sequester more carbon per unit area than any other natural ecosystems environmental settings, which ultimately drive macroscale variation in
(Atwood et al., 2017; Lovelock and Duarte, 2019). With autochthonous SOC (Rovai et al., 2018). The site-specific variability in SOC is largely
inputs from the productive above-ground, mangrove soils store large attributed to differences in species composition (Ren et al., 2008), stand
quantities of carbon as a result of the low decomposition rate resulting age (Lovelock et al., 2010; Donato et al., 2011), sources of allochthonous
from anoxic conditions (Donato et al., 2011; Alongi, 2012). Mangroves particles (Bouillon and Boschker, 2006; Yang et al., 2014), soil physical
are also highly efficient traps for allochthonous inputs through their and chemical properties (Freeman et al., 2004; Kristensen et al., 2008;
dense network of above ground roots. The rising elevation of mangroves Banerjee et al., 2018), elevation and tidal regimes (Liu and Lee, 2006;
in response to sea-level rise allows large accommodation spaces to Spivak et al., 2019), plant-litter biochemistry (Kristensen et al., 2008;
sequester more carbon in the soil, which barely reaches saturation Brodersen et al., 2019) and plant–microbe interactions (Fontaine et al.,
(Krauss et al., 2014; Rogers et al., 2019). Therefore, mangroves act as an 2007; Alongi, 2014). Several soils and environmental characteristics
efficient carbon store despite continuous threats from deforestation, such as pH, salinity, organic matter, precipitation and tidal inundation
land-use change, sea-level rise, and climate change. influence the mangrove productivity and can also directly or indirectly
Blue carbon research across the globe has highlighted considerable influence SOC (Yando et al., 2016). Therefore, careful consideration of
spatial heterogeneity in soil organic carbon (SOC) at multiple scales relevant factors is vital for reliable estimation of SOC at a particular
(Atwood et al., 2017; Sanderman et al., 2018). At a regional and global spatial scale.
scale, SOC variability has been linked to net primary productivity The Sundarbans is the largest contiguous mangrove forest in the
(Alongi, 2012; Twilley et al., 2017), latitude/climate (Rovai et al., 2018; world and is situated in the lower delta plain of the Ganges-
Twilley et al., 2018; Kauffman et al., 2020), coastal geomorphology Brahmaputra-Meghna (GBM) delta, and stretches across political
(Rovai et al., 2018; Twilley et al., 2018) and Holocene sea-level trends boundaries between Bangladesh and India (Giri et al., 2011; Sarker

Table 1
Comparison of Soil Organic Carbon (SOC) density and stock among studies in the Sundarbans and globally.
Study area Study Sample Depth (cm) Methods Mean Soil organic Mean Soil organic Mean top m Soil
size carbon percentage carbon density (gm/ Organic Carbon
(%) (range) cm3) (range) Storage (Mg/ha)
(range)

Sundarbans Bangladesh Bomer et al. 56 100 cm Coring, CHN 0.9 (0.6–1.5) 0.010 (0.008–0.011) –
(2020a) analyser
Khan and Amin 35 15 cm (0–15) Coring, wet 0.6 (0.4–1.0) – –
(2019) oxidation
Sanderman – 100 cm Literature and – – 127 (74–463)
et al. (2018) Model based
Atwood et al. – 100 cm Literature and – – 118
(2017) Model based
Prasad et al. 400 100 cm (1 cm Coring, CN 1.25 (0.8–2.4) – –
(2017) interval) analyser
Hossain and 96 5 cm (0–5) Coring, wet 1.2 (0.6–2.0) – –
Bhuiyan (2016) oxidation
Rahman et al. 150 100 cm (0–30, Coring, wet – 0.011 (0.007–0.014) 112 (90–134)
(2015) 30–100) oxidation
Donato et al., 4 100 cm (0–30, Coring, wet 1.7 (1.6–1.7) 0.016 (0.015– 0.016) –
(2011) 30–100) oxidation
Allison et al. 4 600 cm (0–600) Coring, CHN 0.5–1.1 – –
(2003) analyser
India Dutta et al. 48 40 cm (0–10, 10–20, Coring, wet 1.25 (0.8–1.6) – –
(2019) 20–30, 30–40) oxidation
Prasad et al. 300 100 cm (1 cm Coring, CN 0.8–5.2 – –
(2017) interval) analyser
Dutta et al. 15 25 cm (0–5, 5–10, Coring, TOC 1.8 (1.2–2.1) 0.017 (0.013–0.019) –
(2013) 10–15,15–20, analyser
20–25)
Banerjee et al. 140 40 cm (0–10, 10–20, Coring, wet 1.0 (0.5–1.4) 0.011 (0.007–0.015) –
(2012) 20–30, 30–40) oxidation
Mitra et al. 120 40 cm (0–10, 10–20, Coring, wet 0.7 (0.4–1.1) 0.009 (0.006–0.012) –
(2012) 20–30, 30–40) cm oxidation
Ray et al. 16 30 cm (0–30) Coring, wet 0.6 (0.5–0.7) – –
(2011) oxidation
Global studies Kauffman et al. 190 sample plot data from 5 continents in different – – 334 (33–789)
(2020) soil depth
Sanderman Model based estimation of carbon from literature – – 361 (94–628)
et al. (2018) values of 1812 samples
Rovai et al. Model based estimation of carbon from literature – 0.033 (0.001–0.153) –
(2018) values of 932 samples
Atwood et al. Literature based estimation from 1230 sampling – – 283 (15–1527)
(2017) locations
IPCC (2014) Literature based estimation – – 428
Jardine and Model based estimation of carbon from literature 5.7 (0.1–43.3) 0.032 (0.014–0.115) 369 (272–703)
Siikamäki values of 932 samples
(2014)
Donato et al. Field based data from Indo-pacific area of 25 11.9 (1.7–21.5) 0.043 (0.016–0.076) –
(2011) samples

2
Md.S. Rahman et al. Catena 200 (2021) 105159

et al., 2016). It is either mostly excluded from the global estimates of 2. Material and methods
mangrove SOC (Table 1), or is underrepresented due to a limited num­
ber of samples or perceived poor data quality (Donato et al., 2011; 2.1. Study area
Jardine and Siikamäki, 2014; Atwood et al., 2017; Sanderman et al.,
2018; Twilley et al., 2018; Kauffman et al., 2020). A range of studies into The Sundarbans is the largest single block of mangrove forest in the
SOC content in mangrove soils of the Sundarbans have been carried out world and a Ramsar and UNESCO World Heritage site (Fig. 1) (Giri et al.,
(Table 1), but these all have limitations. The first comprehensive carbon 2011; Sarker et al., 2016). The Bangladesh Sundarbans is situated be­
inventory throughout the Sundarbans was completed by the Bangladesh tween 21◦ 30′ N and 22◦ 30′ N and 89◦ 00′ E and 89◦ 55′ E. The climate of
Forest Department (BFD) in 2009–10; however, the wider vertical the Sundarbans is warm, humid, and tropical, where annual precipita­
sample depth might have an effect on the SOC estimation within the top tion varies from 1474 to 2265 mm and mean annual minimum and
meter (Rahman et al., 2015). Allison et al. (2003) and Donato et al. maximum temperature is between 29 ◦ C and 31 ◦ C (Chowdhury et al.,
(2011) investigated soil organic carbon at greater depth (>1 m) in the 2016; Sarker et al., 2016). Based on the salinity variation, the Sundar­
Bangladesh Sundarbans, however, the number of samples (2 and 6 bans naturally divides into three distinct zones based on the soil salinity;
respectively) was not sufficient to address the variability inside the i) Oligohaline (LSZ) (<2 dS/m, ii) Mesohaline (2–4 dS/m) and iii) Pol­
forests. Studies by Khan and Amin (2019) and Hossain and Bhuiyan yhaline (>4 dS/m) (Siddiqi, 2001; Chanda et al., 2016b). Several studies
(2016) measured SOC from different parts of the Sundarbans, however, have identified a relationship between tree species abundance along the
the sampling was only performed within the top 15 cm. These all the east–west salinity gradient (Iftekhar and Saenger, 2008; Aziz and Paul,
previous studies of SOC in the Sundarbans have limitations resulting 2015; Sarker et al., 2016; Sarker et al., 2019a). Although Excoecaria
from low spatial sampling intensity and limited analysis of soil depth agallocha is abundant in all three salinity zones, Heritiera fomes (char­
range. Moreover, some global studies like Rovai et al. (2018) argued that acteristic species in Bangladesh Sundarbans) is dominant in both the
past climate-based estimation overestimated SOC by up to 86% for oligohaline and mesohaline zones, whereas Ceriops decandra is abundant
deltaic settings like the Sundarbans. Therefore, accurate investigation in the polyhaline zone (Sarker et al., 2019a). Some pioneer species, such
on the spatial variation of soil organic carbon and the identification of as Avicenna spp. and Sonneratia apetala are also present in the mudflats
major controls for such variation in the Bangladesh Sundarbans is ur­ all over the Sundarbans. A short description of all 23 tree species from 10
gently needed. families found in this study is presented in Table A.1.
Increasing salinity in the inundated mangroves stimulate a wide
range of biogeochemical reactions- including enhancing sulphate con­ 2.2. Geology and soils of the Sundarbans
centrations, cation exchange, ionic and osmotic stress, acidity, and
turbidity and at the same time reducing soil redox potential and oxygen The Sundarbans mangrove forest lies in the south-western part of the
levels (Setia et al., 2013; Luo et al., 2019). These soil biogeochemical Bengal Basin, one of the most extensive sediment reservoirs in the world
changes in turn alter sediment characteristics and modify plant and composed of unconsolidated Quaternary deposits (Rudra, 2014). The
microbe communities, which ultimately affect both the soil organic rapid sedimentation followed by the tectonic collision of the Indian plate
carbon pool and quality. Increased soil salinity affects organic matter with the Tibetan plate and the Burmese plate in the Miocene triggered
solubility by altering flocculation of different soil particles (Wong et al., the formation of the Bengal Basin (Alam, 1989). Since the Holocene, the
2009; Wong et al., 2010; Rath and Rousk, 2015). The Investigations of dynamic Ganges-Brahmaputra river system has been discharging sedi­
tidal wetlands across the world reveals a significant negative relation­ ments from the Sub-Himalaya and is still delivering >1 Gt of sediment to
ship between the soil organic carbon pool with salinity (Nyman et al., the delta plain of India and Bangladesh (Islam et al., 1999; Syvitski and
1990; Craft, 2007; Więski et al., 2010; Morrissey et al., 2014; Hu et al., Milliman, 2007). The Sundarbans is of relatively recent origin (3000-
2016). High soil salinity decreases decomposition rates by lowering year B.P.) and this mangrove has developed as a result of both fluvial
microbial activities in the soil and lowers autochthonous carbon input and tidal forces depositing sediments to the GBM river mouth (Goodbred
by reducing plant productivity leading to lower organic carbon in the and Kuehl, 2000; Allison and Kepple, 2001; Rogers et al., 2013). Pre­
soil (Baldwin et al., 2006; Marton et al., 2012; Setia et al., 2013; Liu viously, the Ganges was the main source of sediments in the Sundarbans,
et al., 2017; Zhao et al., 2017). High salinity in general acts as an in­ however, recent changes have resulted from the merging of the Ganges
hibitor of carbon mineralisation, however the opposite is also evident in and Brahmaputra which have now migrated to the eastward, far away
some studies suggesting that a small increase in salinity promotes min­ from the Sundarbans (Rudra, 2014; Islam, 2016). Together with the
eralisation process in the oligohaline zone, while in the mesohaline and eastward migration of the primary GBM delta, the construction of the
polyhaline zones, elevated salinity reduces the mineralisation rate (Luo Farakka barrage in the main Ganges River and earthen embankments
et al., 2019). Therefore, the impact of salinity on the soil organic carbon surrounding the Sundarbans have reduced freshwater flow, resulting in
pool and quality is not uniform in all wetland settings in the world, reduced fluvial sedimentation in the Bangladesh Sundarbans. This
rather it depends on the local geomorphology and hydrological geomorphological change, in turn, has led to increased remobilisation of
characteristics. sediments by tidal forces (Rogers et al., 2013; Hale et al., 2019; Bomer
The aim of the present study is to estimate soil organic carbon (SOC) et al., 2020b). The changed pattern of freshwater flow has resulted in a
in the Bangladesh part of the Sundarbans mangrove forest and to better salinity gradient increasing from the east to the west of the Sundarbans.
understand the relationship of SOC with three salinity zones (oligoha­ The soil is mainly fine-grained, grey coloured, slightly calcareous,
line, mesohaline and polyhaline) and major forest types. The study and mostly composed of silts to clayey silts (Allison et al., 2003; Bomer
hypothesises that higher salinity zones (polyhaline) would yield a lower et al., 2020a). The subsurface sediment extends up to 6 m in depth in the
organic soil carbon stock as a reflection of lower productive vegetation landward direction and up to 4 m in depth in the seaward direction
and altered soil physical and biological processes compared with the (Allison et al., 2003). The median grain size ranges between 16 and 32
lower salinity zone (oligohaline). The relationships between physical, µm reflecting the medium silt range. The average dry bulk density (0.81
chemical properties and vegetation characteristics with SOC are also g cm− 3) is higher in the Sundarbans in comparison to other mangroves in
investigated to develop dependable predictive models for this forest. The the world (Bomer et al., 2020a). The soil physical and chemical prop­
novelty of this study lies in the extensive stratified random sampling erties are varied from the eastern to the western part of Bangladesh
from all over Bangladesh Sundarbans combined with vertical investi­ Sundarbans, the eastern part is softer, more fertile and receives more
gation of soil depth up to 1 m. fresh sediments than the western part (Siddiqi, 2001). Soils are mostly
neutral to alkaline (pH 6.5–8.0), whereas the polyhaline zone is more
alkaline than the oligohaline zone. Soils of the western and southern

3
Md.S. Rahman et al. Catena 200 (2021) 105159

Fig. 1. Sundarbans mangrove forest, Bangladesh. Legend colour represents three major salinity zones (Chanda et al., 2016b). ESRI Basemap Sources: Esri, HERE,
Garmin, FAO, NOAA, USGS, © OpenStreetMap contributors, and the GIS User Community.

polyhaline zone are comparatively richer in P, K, Na, Mg, Cl- and Fe, but sub-samples were taken from the middle of each core section with fixed
lower in soil NH4- and Na than the eastern oligohaline zone (Siddiqi, 2.5 cm length, sealed in plastic and subsequently placed in an icebox to
2001; Sarker et al., 2016). This pronounced differences in soil nutrients reduce oxidation. The sub-samples were kept below 4 ◦ C in zip-sealed
and salinity trigger the diversity of vegetation composition in different plastic bags until laboratory processing.
parts of the Sundarbans. For vegetation data, the Diameter at Breast Height (DBH) and height
were measured for all trees within the 11.3 m radius plot. DBH was
2.3. Sediment and tree data collection measured at 1.3 m and height was measured with a Vertex-III hyp­
someter. For small trees with a DBH < 14.5 cm, a smaller circular plot
In the Bangladesh Sundarbans, permanent sample plots (PSP) were (radius 5 m) was nested within the 11.3 m plot. The elevation of each
established in 1986 by the ODA (Overseas Development Administration) plot was calculated by subtracting the mean tree height of the plot from
for monitoring growth, regeneration, and long-term ecological changes the Digital Surface Model (DSM) taken from the TanDEM-X 90 m sat­
(Chaffey et al., 1985). A total of 120 PSPs (20 m × 100 m) were ellite data (Hawker et al., 2019). For major forest types, single-species
established to assess growth rate, regeneration, stocking, and crop dominance was determined when the relative composition is >75%,
composition based on salinity, forest type and accessibility (Iftekhar and and the remaining forest types are termed as mixed type.
Saenger, 2008; Sarker et al., 2019b) (Fig. 1). In this study, sediment
samples were collected from 55 plots, of which 50 plots are from PSPs 2.4. Laboratory analysis
selected at random, and the remaining five plots are from outside PSPs to
represent areas outside PSP. Sampling was undertaken in two phases: In 2.4.1. Soil physical and chemical properties
the first phase, three sediment cores of 50 cm depth were taken from 18 For each core sub-samples, samples were freeze-dried and re-
PSPs. After laboratory analysis of the samples from the first 18 PSPs, it weighted to determine the bulk density. Bulk density was calculated
was decided to extend the sediment sampling depth to 1 m and take two by dividing the dry mass of the soil by the volume of the soil. Soil pH and
core samples from each plot because of little within-plot variation soil salinity (as soil conductivity) were measured from a portion of the
among the initial 54 core samples. In the second phase, an additional 37 homogenised dry soil for each core. Dried soils were diluted with
PSPs were sampled with two cores sampled at each plot. Altogether, 126 distilled water (1:5 ratio), and subsequently, soil pH was measured using
sediment cores from 55 plots were sampled across the whole of the a Jenway 3510 Standard Digital pH Meter and soil salinity by a hand­
Bangladesh Sundarbans (Fig. 1). held Jenway 470 Conductivity Meter (Hardie and Doyle, 2012).
The location of the cores within a PSP was decided by establishing a
random circular plot with a radius of 11.3 m (an area of 400 m2). Within 2.4.2. Total soil organic carbon (SOC)
each plot, a small circular plot was laid with 5 m radius and sediment To determine the soil organic carbon (SOC) and nitrogen content of
cores were taken from east, west and south side (east and west for two the soil, any large stones or twigs were removed from the sample and
cores) from the centre, which is perpendicular to each other. The cores subsequently homogenised and ground with a ball mill. A few milli­
were taken using an open-faced auger (6 cm diameter), which was grams (~40 mg) of the sediment was then passed through an ele­
further subdivided into four depths (0–10, 10–30, 30–50 and 50–100 mental analyser (Thermo Scientific Flash 2000-NC Soil Analyzer) to
cm), following the method of Kauffman and Donato (2012). Sediment derive the total carbon and nitrogen as a percentage. Inorganic carbon

4
Md.S. Rahman et al. Catena 200 (2021) 105159

content was deducted from the total carbon to get organic carbon per­ 3. Results
centage, according to Howard et al. (2014). The inorganic carbon con­
tent was measured from some random samples across all salinity zones 3.1. Soil organic carbon, salinity zones and soil depth
using an Analytik Jena Multi EA (Elemental Analyser) 4000. Soil organic
carbon density (gm cm− 3) for each sample and total organic carbon The average soil organic carbon (SOC) density significantly varied
content (Mg ha− 1) of each depth and core were measured according to from 0.003 gm cm− 3 to 0.009 gm cm− 3 in different salinity zones and
Howard et al. (2014). soil depths (two-way ANOVA for Ln (SOC density), salinity zones, F2, 500
= 112.3, p < 0.001 and soil depths, F3, 500 = 30.1, p < 0.001) (Fig. 2,
2.5. Statistical analysis Table A.2). Both salinity zone and soil depth had a significant interaction
effect on the variability of SOC density in the Sundarbans (F6, 500 = 3.5,
All statistical analysis and graphics used R 3.6.1 for Windows (R Core p < 0.01) (Table A.1). Significantly higher SOC density was found in the
Team, 2019). Total organic carbon (Mg ha− 1), organic carbon density topmost depth followed by the subsequent three depths; however, SOC
(gm cm− 3) and bulk density (gm cm− 3) among three salinity zones and density in the intermediate depths (between 10–30 cm and 30–50 cm)
four depths were compared with two-way analysis of variance (ANOVA) are not significantly different (Fig. 2b), which indicates the unequal
test using the ‘car’ package (Fox and Weisberg, 2019). In order to variability of SOC with soil depth. The oligohaline zone comprises
compare soil organic carbon among vegetation types, total organic higher SOC density (gm cm− 3) followed by mesohaline and polyhaline
carbon (Mg ha− 1) was compared with one-way analysis of variance zone indicating higher soil organic carbon in the low salinity areas.
(ANOVA). The results of ANOVA are summarized in Supplementary The bulk density (BD) of the soil revealed an opposite trend as
Information. To derive the relationship among organic carbon density (g significantly higher bulk density was observed in the higher salinity
cm− 3), bulk density and total nitrogen content, data from all the core zones and in the 50–100 cm soil depth (two-way ANOVA for Ln (bulk
subsections (n = 512) were used. To examine the relationship among density (gm cm− 3)), salinity zones, F2, 500 = 22.2, p < 0.001 and soil
SOC and soil physical and chemical parameters (soil salinity, pH, bulk depth, F3, 500 = 46.2, p < 0.001) (Fig. A.1, Table A.3). Likewise, SOC
density, Total N, organic C:N, elevation, latitude and longitude) and density, the soil organic carbon storage (SOC) for the different depths
vegetation characteristics (species richness, tree density, mean DBH and was significantly different among the three salinity zones and the four
mean height), stepwise multiple linear regression analysis was under­ soil depths (two-way ANOVA for Ln (SOC), salinity zones, F2, 500 =
taken. SOC was considered as the dependant variable, whereas all the 118.9, p < 0.001 and soil depth, F3, 500 = 526.2, p < 0.001) (Fig. 3 &
selected parameters were independent variables. Correlation analysis Fig. 4, Table A.4). However, higher amounts of SOC were found in the
and principal component analysis (PCA) were carried out to decrease the 50–100 cm depth in comparison to the top three depths (Fig. 4). The top
number of explanatory variables and to reduce collinearity in the meter SOC ranges from 26.2 Mg ha− 1 to 107.9 Mg ha− 1 in the Sundar­
regression model. All the variables were standardised before PCA ac­ bans, where oligohaline zone comprises the highest SOC (74.8 Mg ha− 1),
cording to Legendre and Legendre (2012). Eigenvalues greater than one followed by the mesohaline (59.3 Mg ha− 1), and the polyhaline zone
were retained and variables with factor loadings >0.35 were treated as (48.3 Mg ha− 1) (Table 2).
potential explanatory variables for the regression model (Jackson,
1993). In all cases, the data were logarithmic (natural) transformed (if
needed) to meet the assumptions of normality and equal variances by 3.2. Soil organic carbon and forest types
using Shapiro Wilk and Levene’s tests, respectively and subsequently
back-transformed to present graphically. The graphical output of the One-way ANOVA revealed that SOC varied with major forest types in
linear model was generated using the ‘ggplot2’ package (Wickham, the Sundarbans (F7, 47 = 3.3, p < 0.01) (Table A.5). As shown in Fig. 5,
2016). the average SOC content in the Bruguiera sp. stand was the highest, with
an average of 105.3 Mg ha− 1, followed by Sonneratia sp. and Avicennia
sp., with an average of 68.7 Mg ha− 1 and 67.1 Mg ha− 1, respectively.
The Tukey HSD test showed that the other forest types had no significant

Fig. 2. (A) The distribution of soil organic carbon (SOC) density (gm cm− 3) in four soil depths presented as violin-box plot, where the black vertical line represents
the median and black dots are outliers. Here, the width of violin plot represents the proportion of the data located there as a measure of kernel probability density. (B)
Average SOC density (gm cm− 3) in three salinity zones and four soil depths.

5
Md.S. Rahman et al. Catena 200 (2021) 105159

Fig. 3. Spatial distribution of total soil organic carbon (SOC) (Mg ha− 1) and soil salinity (dS/cm) in the Sundarbans. Note that circle represents the amount of SOC
and gradual colour ramp reveals soil salinity indicating green to red as from low to high salinity. Three major salinity zones are represented according to Chanda
et al., (2016b). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

expected, oligohaline zones had relatively low average soil bulk density,
pH, and soil salinity, in comparison to higher salinity zones. Addition­
ally, significantly higher SOC and lower total N contributes higher
organic C:N in the oligohaline zone, although it is similar to the meso­
haline zone (p < 0.05). BD and SOC density showed a statistically sig­
nificant negative relationship (r = − 0.47, p < 0.001) (Fig. 6A).
However, the soil organic carbon (SOC) density and soil nitrogen density
is significantly positively correlated with soil nitrogen density across the
Sundarbans (r = 0.66, p < 0.001) (Fig. 6B). Analysis from the satellite
and tree height data reveals that the average elevation is higher in the
polyhaline zone. The average DBH and height of the trees were statis­
tically significantly higher in both the oligohaline and mesohaline zone,
whereas the average stem density was higher in both the mesohaline and
polyhaline zone (p < 0.05). Bivariate relationship between SOC and
other soil physical, chemical properties and vegetation characteristics
are presented in the supplementary Fig. A.2.

Fig. 4. Average soil organic carbon (Mg ha− 1) in different soil depths in three 3.4. Relationship of SOC with soil and vegetation properties
salinity zones.
SOC content was positively correlated with tree DBH, tree height,
effect on SOC content, which ranges from 50.2 Mg ha− 1 to 67.0 Mg ha− 1 organic C:N, latitude, and longitude, but negatively correlated with soil
for Ceriops and Heritiera forest types, respectively (Table A.6). salinity, bulk density, soil pH, tree density and elevation (p < 0.05)
(Fig. 7). As total nitrogen and species richness did not show any sig­
nificant correlation with SOC content, these two parameters were dis­
3.3. Soil physical, chemical properties and vegetation characteristics carded from the subsequent PCA analysis. The measured properties also
showed a significant positive and negative correlation among them­
The soil physical, chemical properties and vegetation characteristics selves, which indicates a source of multicollinearity, a phenomenon
varied considerably among the three salinity zones (Table 2). As which makes multiple regression unreliable. Therefore, principal

6
Md.S. Rahman et al. Catena 200 (2021) 105159

Table 2
Overview of measured soil parameters and vegetation characteristics. Values are presented as mean (±SD), where n ≥ 3. Lowercase letters indicate significant
variability among salinity zones, according to least-significant difference (LSD) test at α = 0.05.
Salinity Bulk density Soil pH Soil salinity Total Soil Total Nitrogen Organic Elevation Stem Height DBH (Diameter at
zone (gm cm− 3) (EC dS/cm) Organic Carbon (Mg ha− 1) C:N (m) Density (m) Breast Height)
(Mg ha− 1) (ha− 1) (cm)

Oligohaline 0.58 (0.07) b 7.06 1.49 (0.32) c 74.77 (14.93) a 2.66 (1.19) b
21.30 3.39 (1.78) 5,009 7.98 8.12 (2.41) a
c
(0.26) (7.23) a b
(2,485) b (2.03) a
Mesohaline 0.62 (0.04) 7.43 3.07 (0.56) b 59.30 (15.80) b 3.52 (1.08) a
17.30 3.67 (1.01) 6,876 7.88 8.60 (5.36) ab
ab b
(0.19) (6.87) a ab
(3,290) ab (2.63) a
Polyhaline 0.63 (0.05) a 7.80 5.56 (0.85) a 48.25 (10.32) c 3.81 (0.98) a
13.08 4.79 (1.52) 8,750 5.98 6.72 (4.12) b
a
(0.26) (3.00) b a
(4,798) a (1.66) b

Fig. 5. Integrated violin-box plot shows average soil organic carbon (SOC) in major forest types in the Sundarbans. The black vertical line of box plot represents the
median and the width of violin plot represents the proportion of the data located there as a measure of kernel probability density.

Fig. 6. (A) Relationship between bulk density and soil organic carbon density. (B) Relationship between soil nitrogen density and soil organic carbon density.

7
Md.S. Rahman et al. Catena 200 (2021) 105159

Fig. 7. Correlation matrix among SOC and other physicochemical, geophysical and vegetation properties. The number of each block shows the Spearman’s rank
correlation coefficients at p < 0.05, where red and violet colour represents respective positive and negative correlations. The white block indicates the correlation
coefficient is statistically insignificant. The soil properties: SOC = Soil organic carbon, SS = soil salinity, BD = Bulk density, pH = soil pH, TN = Total Nitrogen, C:N
= organic C- total Nitrogen ratio, the vegetation characteristics: SR = Species richness, TD = Tree density, DBH = mean Diameter at Breast Height, H = Mean height
and geophysical properties: E = Elevation, LAT = Latitude and LONG = Longitude. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)

component analysis was used to identify and group those properties that stepwise multiple linear regression (MLR). The regression results
influence SOC the most and to overcome the influence of showed that soil salinity alone could explain 50% SOC variability in the
multicollinearity. Sundarbans, however, the percentage increases to 57% and 62% when
Principal component analysis (PCA) was performed with ten vari­ soil C:N or soil C:N and tree DBH are added to the model (Table 3).
ables to assemble and isolate the smallest possible of subsets to explain Although all three regression models are highly significant (Table A.8),
the variation of the dataset (Fig. 8). The PCA result indicates that the the best subset of MLR model was selected based on the largest adjusted
first two principal components explained more than two-thirds of the R2 value and the smallest Mallow’s Cp, AIC (Akaike Information
total variation with an eigenvalue greater than 1. The most important Criteria) and RMSE (Root Mean Squared Error) and presented in Eq. (1).
component (PC1) explained 49.5% with the highest loadings (>0.35) for
Ln (SOC) = 3.439 − 0.077 SS + 0.274 Ln (C : N) + 0.017 DBH (1)
soil SS (soil salinity) and pH. On the other hand, the second component
showed higher leadings for tree H, DBH and soil C:N with 20% explained
4. Discussion
variation (Table A.7). As soil SS and pH are highly correlated with each
other (r = 0.76, p < 0.05) (Fig. 7), the variable with the highest loading,
The reported average soil organic carbon (SOC) density in this study
soil SS, was selected from the first component for the regression model.
is lower than previous estimates from the Sundarbans and far lower than
Similarly, tree H was discarded due to collinearity with tree DBH and
average estimates of SOC density from global mangroves (Table 1). SOC
therefore, tree DBH and soil C:N was selected from the second
density, the standardized carbon stock measurement with depth, is the
component.
most useful parameter to compare SOC between different forests
By using the PCA-derived subset of variables, the relationship be­
(Donato et al., 2011; Weiss et al., 2016). However, due to unreported
tween SOC and soil and vegetation properties was obtained by using
bulk density, it was not possible to convert from the reported organic

8
Md.S. Rahman et al. Catena 200 (2021) 105159

Fig. 8. Principal component analysis (PCA) biplot of soil physicochemical, geophysical and vegetation characteristics as vectors (n = 10) and mangroves areas are
coloured coded as three salinity zones (n = 55). The soil physicochemical properties included SS = soil salinity, pH = soil pH, BD = Bulk density, C:N = organic C-
total Nitrogen ratio, geophysical properties comprised LAT = Latitude, LONG = Longitude, E = Elevation, and the vegetation characteristics included TD = Tree
density, DBH = mean Diameter at Breast Height, H = Mean height. Here, perpendicular direction signifies uncorrelated relationship, while negative and positive
correlated vectors are presented in the opposite vectors and small angle vectors, respectively.

concentration is a function of soil depth and shows considerable vari­

Table 3
ability (Wuest, 2009; Kauffman and Donato, 2012; Jandl et al., 2014).
Summary statistics of regression model. Here, SS = Soil salinity, C:N = Soil
Moreover, using coring for sampling might have an influence on soil
organic carbon: Nitrogen and DBH = tree Diameter at breast height.
bulk density estimation leading to lower SOC stock estimation in deeper
Model R2 Adjusted R2 C(p) AIC RMSE soils (Rau et al., 2011; Gross and Harrison, 2018).
1. SS 0.508 0.498 18.217 − 10.457 0.212 In comparison to global studies, the estimated top 1 m SOC stock is
2. SS and C:N 0.590 0.574 8.650 − 18.513 0.196 lower in the Sundarbans than the reported average from sites distributed
3. SS, C:N and DBH 0.637 0.616 4.00 − 23.255 0.186
all over the world (Table 1). Based on model-based georeferenced
database of mangrove SOC, the global SOC map showed that the Sun­
carbon (%) to SOC density for most of the local and world studies. darbans contains the lowest SOC stocks per ha in the world (Sanderman
Despite a greater range of soil organic carbon (SOC) percentage in this et al., 2018). Compared to direct estimates from 190 sites across the
study (0.3–4.4%), the average value (1.2%) is in line with most previous world by Kauffman et al. (2020), the Sundarbans contains higher SOC
studies, although higher than estimates published by Ray et al. (2011), than only two other mangrove forests, the Porto Céu mangrove in Brazil
Banerjee et al. (2012), and Allison et al. (2003). These differences are (48 Mg ha− 1) and the Bu Tinah Janoub in the United Arab Emirates (33
likely to be attributed to variable sampling strategies along with variable Mg ha− 1), located in lower and higher latitudes respectively than the
soil depth or different methods used for carbon estimation. Likewise Sundarbans. However, global comparison in soil carbon among tropical,
SOC density, the average top 1 m SOC storage in the Bangladesh Sun­ subtropical and temperate mangroves showed a contrasting relationship
darbans (50.9 ± 15.2 Mg ha− 1) is almost half of the previous estimate by with latitudes (Atwood et al., 2017; Twilley et al., 2018; Kauffman et al.,
Rahman et al. (2015), Sanderman et al. (2018) and Atwood et al. (2017). 2020; Ouyang and Lee, 2020). Both Kauffman et al. (2020) and Ouyang
Estimates of soil organic carbon could fluctuate based on the differences and Lee (2020) found significantly lower soil carbon in mangroves
in sampling design, choice of analytical method and soil depth (Howard >20◦ N, although the former study had fewer samples largely limited to
et al., 2014; Nayak et al., 2019). In the case of mangroves, Passos et al. the middle east hyper arid mangroves. On the other hand, Atwood et al.
(2016) found overestimation of organic carbon measured with the (2017) and Twilley et al. (2018) documented the poor relationship be­
oxidation method in comparison to the elemental analyser. Anaerobic tween latitude and SOC stocks. This poor relationship might be attrib­
microbial decomposition yields reduced soil compounds (i.e., Fe2+, S2− , uted due to the poor representation of samples in the studies from the
Mn2+, and Cl− ) in mangroves, which might interfere with organic car­ subtropical mangroves like the Sundarbans.
bon determination with chemical oxidation method (Nelson and Som­ The low soil carbon in the Sundarbans is largely due to high mineral
mers, 1996; Bisutti et al., 2004; De Vos et al., 2007; Nóbrega et al., sediment deposition (Sanderman et al., 2018; Twilley et al., 2018), low
2015). Apart from using different methods, the SOC variation originates burial rate (Ray et al., 2011), rapid turnover rate (Ray et al., 2018),
from the consideration of soil depth in sample design as the SOC historical logging, stand age (Marchand, 2017), plant litter quality

9
Md.S. Rahman et al. Catena 200 (2021) 105159

(Rovai et al., 2018) and biological processes. Being both a tide and river- composition of mangrove forests in different tidal regimes yield variable
dominated ecosystem, the carbon allocation in the above and below SOC stock (Lacerda et al., 1995; Gleason and Ewel, 2002). Moreover, the
ground is very complex, largely dependent on the local and regional long and short-term resilience and resistance of microbial communities
geomorphic and geophysical drivers (Twilley et al., 2018). In riverine are largely dependent on the structure and zonation of mangrove com­
deltas, trees invest much of the carbon to the above ground to keep pace munities reflecting environmental gradients (Capdeville et al., 2019). In
with sedimentation and sea-level rise, which is evident in the oligoha­ this study, the species with higher SOC stock such as Bruguiera sp.,
line zone with greater forest productivity (Twilley et al., 2018; Sarker Sonneratia sp. and Avicennia sp. are frequently inundated due to prox­
et al., 2019a; Sarker et al., 2019b). Moreover, research has highlighted imity to the river and low land than other species in the Sundarbans
that mangroves subjected to frequent cyclones leading to temporary (Siddiqi, 2001; Sarker et al., 2016). These high inundation regimes, in
losses of above ground carbon are usually followed by rapid below turn, lead to increased microbial activity and a higher level of dissolved
ground carbon gains during recovery process according to the organic carbon in the sediment (Wang et al., 2013; Chambers et al.,
‘Ecosystem Development’ theory (Odum, 1969; Danielson et al., 2017; 2014; Chambers et al., 2016). Regular tides also bring sediments along
Kominoski et al., 2018). These rapid carbon gains in the above ground with high allochthonous input whereas the raised less-inundated areas
and the disturbance from the catastrophic cyclones could be the source foster autochthonous SOC and less microbial activity (Lovelock et al.,
of higher autochthonous input to the below ground. Nonetheless, higher 2015; Woodroffe et al., 2016). Rao et al. (1994) found almost double C:N
tidal amplitude in the Sundarbans leads to higher carbon export total­ ratio in fresh leaves of Bruguiera sp. compared with other mangrove
ling 7.3 Tg C yr− 1 to the adjacent Bay of Bengal, which is higher than any species, suggesting higher input of autochthonous carbon. Being the
other mangroves in the world (Ray et al., 2018). This rapid carbon pioneer species in the succession of the Sundarbans, both Sonneratia sp.
turnover results in reduced burial of organic matter (0.18%) in the soil and Avicennia sp. are resilient to disturbances leading to higher SOC than
(Ray et al., 2011). Moreover, the pronounced tidal cycles in the Sun­ climax and seral species (Table A.1) and accumulate a large quantity of
darbans affects carbon burial process by altering soil water chemistry organic litter in the tidal channel close to the river or seafront (Sarker
(Chatterjee et al., 2013; Spivak et al., 2019). Besides high carbon turn­ et al., 2016; Bomer et al., 2020a). The variability of SOC stocks among
over rate, the Sundarbans is believed to have become tidally active in the forest types followed a similar pattern to the global studies by Atwood
recent past due to reduced freshwater flow from the Ganges- et al. (2017), except for Sonneratia sp. which was found to hold less SOC
Brahmaputra-Meghna river (Rogers et al., 2013; Hale et al., 2019). stock than Heritiera and Ceriops. On the other hand, Kauffman et al.
However, despite the historical reduction of sedimentation, the Sun­ (2020) found significantly lower below ground carbon stocks in Avi­
darbans is itself still keeping pace with sea-level rise with the highest cennia sp., especially in the arid mangroves of Middle-East Asia, which is
average surface elevation and vertical accretion rate (0.74 and 2.71 cm solely occupied by this species. Therefore, the impact of above ground
yr− 1) compared to the worldwide average (Bomer et al., 2020a; Bomer vegetation of below ground is largely site-specific, and it depends on a
et al., 2020b). This high sedimentation rate is the outcome of the wide range of factors.
massive flux of clastic sediments which attenuates the amount of organic The unexplained variation of the best multiple regression models (R2
carbon per unit area. = 0.64) highlights the necessity of including other soil and environ­
The century-long historical exploitation in the Sundarbans before the mental parameters such as soil cations and anions, clay characteristics
felling moratorium in 1989 has largely decreased the populations of and texture, precipitation, temperature, and river discharge. This study
threatened tree species (Siddiqi, 2001; Sarker et al., 2011). This in turn did not address these properties and suggests future studies incorporate
is likely to have lessened the continuous autochthonous input of organic a wider range of parameters to gain a better understating of organic
matter in the forest and reduced the overall stand age. Studies also carbon dynamics in the Sundarbans. In particular, for better ecosystem
showed that historical harvesting had altered the species composition in management, future research should include information relating to
the Sundarbans, with decreasing abundances of Heritiera fomes, Ceriops contextualising soil (e.g. soil texture, grain size and minerology),
decandra and Xylocarpus mekongensis and increasing for Excoecaria biogeochemical (e.g. important properties of soil and pore-water
agallocha (Sarker et al., 2016). The SOC stock also depends on the age of chemistry such as sulphate, oxygen, nitrate, ferric oxides in case of
the stands as evident in the Chrono sequence study on SOC stocks in mangroves) and ecological (e.g. vegetation and plat-microbe interac­
French Guiana which revealed that the SOC varied from 4 to 107 Mg tion) properties (Luo et al., 2019; Spivak et al., 2019). However, soil
ha− 1 from young stand to senescent stage (Marchand, 2017). In addi­ salinity is considered as the outcome of the combined impact of these
tion, studies have suggested that lower organic carbon in the soil is climatic and environmental variables in the Sundarbans resulting pro­
mostly associated with higher C:N of the plant litter which has resulted nounced differences of SOC stock among the three salinity zones (Sarker
from lowering decomposition speed and decreasing carbon-use effi­ et al., 2016; Sarker et al., 2019b; Rahman et al., 2020). Several previous
ciency of the decomposer (Bouillon et al., 2003; Zhou et al., 2019). studies have confirmed that salinity determines the strong zonation of
Compared to mangrove associates, the senescent leaves of true man­ tree species and diversity in the Sundarbans, which in turn leads to
groves contain considerably higher C:N (~33) in the Indian part of comparatively higher diversity and taller tree species in the oligohaline
Sundarbans (Chanda et al., 2016a). Kamruzzaman et al. (2019) observed followed by mesohaline and polyhaline zone (Aziz and Paul, 2015;
a decreasing trend of C:N of the leaf litter in both forest floor and buried Sarker et al., 2016; Sarker et al., 2019a; Sarker et al., 2019b; Rahman
condition starting from 40, but barely reached below 30 after 196 days et al., 2020). Comparatively higher productive trees (e.g. higher DBH
of decomposition study, suggesting N limitation in the oligohaline zone and higher height) promotes organic matter accumulation through
of the Bangladesh Sundarbans. The low organic carbon can also be producing higher litter mass and increases SOC stock by forming stable
attributed by the abundance of leaf-consuming organisms ingesting aggregates from roots and pneumatophores (Lange et al., 2015). The
organic litter detritus both at surface and subsurface in burrows. The three salinity zones also comprise differential soil physical, chemical
Sundarbans encompasses a wide range of gastropod species (e.g., Cer­ properties and vegetation characteristics that usually affects SOC stor­
ithedia cingulata, Cymia lacera) that predominantly consume mangrove age by influencing microbial decomposition, soil water chemistry,
detritus (Nayak et al., 2014). plant–microbe interaction, and plant litter quality. While comparing
Variation in SOC stocks among different forest types is often medi­ nutrient concentration in the leaf litter of Sonneratia apetala, one of the
ated by the primary productivity, resources allocation in different parts major pioneer species in the Sundarbans, Nasrin et al. (2019) found
(e.g. above and below ground) and microorganism activity is driven by a lowest concentrations of N, P and K and the highest concentrations of Na
number of biological (e.g. bioturbation and species composition) and in the polyhaline zone, reflecting higher C:N in the leaf litter. However,
physical (e.g. soil texture, salinity, inundation and nutrients) factors the low SOC in the polyhaline zone is also coincided with the low C:N
(McLeod et al., 2011). Therefore, differing stand structure and indicting inwelling of marine and terrestrial suspended particulate

10
Md.S. Rahman et al. Catena 200 (2021) 105159

materials (Bouillon et al., 2003). The strong positive correlation (r = gratitude to the Bangladesh Forest Department for giving permission for
0.66, p < 0.001) between carbon and nitrogen density indicates that the fieldwork in the Sundarbans and Forestry and Wood Technology Disci­
source of carbon and nitrogen is likely to be same and can vary spatially pline, Khulna University for allowing study leave to the first author. We
(Matsui et al., 2015). also thank everyone involved the fieldwork in the Sundarbans and we
Although the Sundarbans is considered to be of recent origin, the recognised that without their help, this work could not have been
large accommodation space exists due to accretion and erosion with completed. We are also grateful to the Institute of Hazard, Risk and
historical relative sea-level variability (Goodbred and Kuehl, 2000; Resilience for a small research grant and the Department of Geography,
Tyagi and Sen, 2019). Therefore, the Sundarbans might have a 3 m Durham University for allowing laboratory analysis and to Ustinov
organic layer in the seaward direction and much more in the landward College, Durham University for a travel grant. Lastly, we give thanks to
(Allison et al., 2003). By considering this vertical depth and the area Ms. Robyn Shilland for her brilliant comments in the manuscript.
covered by mangrove forest, the Sundarbans are likely to contain
considerable volumes of soil organic carbon. Previous research has Appendix A. Supplementary material
demonstrated that mangroves holding higher carbon storage also have a
higher rate of deforestation with 50% mangrove loss attributed to Supplementary data to this article can be found online at [Link]
Indonesia, which holds about 25% of soil carbon in the world’s man­ org/10.1016/[Link].2021.105159.
groves; the figure increases to 75% when Malaysia and Myanmar are
considered (Atwood et al., 2017). Therefore, mangroves from these References
countries are considered as a significant source of emissions due to high
deforestation and forest conversion (Hamilton and Friess, 2018). On the Abdullah, A.N., Stacey, N., Garnett, S.T., Myers, B., 2016. Economic dependence on
mangrove forest resources for livelihoods in the Sundarbans, Bangladesh. For. Policy
other hand, in Bangladesh, despite the lower SOC stock in the Sundar­ Econ. 64, 15–24. [Link]
bans mangrove forest demonstrated by this paper, recent positive trends Alam, M., 1989. Geology and depositional history of Cenozoic sediments of the Bengal
in forest cover demonstrate the value of blue carbon conservation and an Basin of Bangladesh. Palaeogeogr., Palaeoclimatol Palaeoecol. 69, 125–139. https://
[Link]/10.1016/0031-0182(89)90159-4.
improved understanding of carbon storage will be of benefit to the in­ Allison, M., Kepple, E., 2001. Modern sediment supply to the lower delta plain of the
clusion of mangroves in national and international climate strategies Ganges-Brahmaputra River in Bangladesh. Geo-Mar. Lett. 21, 66–74. [Link]
and policies. org/10.1007/s003670100069.
Allison, M.A., Khan, S.R., Goodbred, S.L., Kuehl, S.A., 2003. Stratigraphic evolution of
the late Holocene Ganges-Brahmaputra lower delta plain. Sediment. Geol. 155,
5. Conclusion 317–342. [Link]
Alongi, D.M., 2012. Carbon sequestration in mangrove forests. Carbon Manage. 3,
313–322. [Link]
The top meter of soil organic carbon (SOC) per area in the
Alongi, D.M., 2014. Carbon cycling and storage in mangrove forests. Ann. Rev. Mar. Sci.
Bangladesh Sundarbans is lower than has previously been reported. 6, 195–219. [Link]
However, the total SOC will likely to be greater if total vertical depth is Atwood, T.B., Connolly, R.M., Almahasheer, H., Carnell, P.E., Duarte, C.M., Lewis, C.J.E.,
considered. The soil organic carbon stock (SOC) in the Sundarbans is Irigoien, X., Kelleway, J.J., Lavery, P.S., Macreadie, P.I., Serrano, O., Sanders, C.J.,
Santos, I., Steven, A.D.L., Lovelock, C.E., 2017. Global patterns in mangrove soil
largely influenced by soil salinity, probably by amending the forest carbon stocks and losses. Nat. Climate Change. 7, 523–528. [Link]
productivity and microbial activity. The results highlighted that 10.1038/nclimate3326.
increasing salinity as result of predicted sea-level rise will likely have Aziz, A., Paul, A.R., 2015. Bangladesh Sundarbans: present status of the environment and
biota. Diversity. 7, 242–269. [Link]
pronounced effects on future soil carbon accumulation rates by altering Baldwin, D.S., Rees, G.N., Mitchell, A.M., Watson, G., Williams, J., 2006. The short-term
the soil environment and vegetation characteristics. The study un­ effects of salinization on anaerobic nutrient cycling and microbial community
derlines the importance of spatial conservation planning measures and structure in sediment from a freshwater wetland. Wetlands. 26, 455–464. https://
[Link]/10.1672/0277-5212(2006)26[455:TSEOSO][Link];2.
initiatives to conserve and maximize carbon accumulation and to Banerjee, K., Bal, G., Mitra, A., 2018. How soil texture affects the organic carbon load in
contribute to global climate change adaptation and mitigation strate­ the mangrove ecosystem? A case study from Bhitarkanika, Odisha. In: Singh, V.P.,
gies. Results suggest that high sediment carbon zone in the eastern part Yadav, S., Yadava, R.N. (Eds.), Environmental Pollution. Springer Singapore,
Singapore, pp. 329–341.
of the Sundarbans is highly vulnerable to tourism and economic devel­ Banerjee, K., Roy Chowdhury, M., Sengupta, K., Sett, S., Mitra, A., 2012. Influence of
opment activities. In terms of climate change mitigation and adaptation, anthropogenic and natural factors on the mangrove soil of Indian Sundarbans
the conservation of the existing carbon stock should receive much higher wetland. Arch. Environ. Sci. 6, 80–91.
Bisutti, I., Hilke, I., Raessler, M., 2004. Determination of total organic carbon – an
priority rather than the debates of high-low carbon stock. The
overview of current methods. TrAC Trends Anal. Chem. 23, 716–726. [Link]
Bangladesh Sundarbans can act as an important blue carbon hotspot due org/10.1016/[Link].2004.09.003.
to the high sedimentation and carbon sequestration rate and conserva­ Bomer, E.J., Wilson, C.A., Elsey-Quirk, T., 2020a. Process controls of the live root zone
tion priority by the government. However, disturbances such as sea- and carbon sequestration capacity of the Sundarbans mangrove forest, Bangladesh.
Sci. 2, 35. [Link]
level rise, global warming, eutrophication, and landscape development Bomer, E.J., Wilson, C.A., Hale, R.P., Hossain, A.N.M., Rahman, F.M.A., 2020b. Surface
might hinder this conservation activities in the future. elevation and sedimentation dynamics in the Ganges-Brahmaputra tidal delta plain,
Bangladesh: evidence for mangrove adaptation to human-induced tidal
amplification. Catena. 187, 104312 [Link]
Funding Bouillon, S., Boschker, H.T.S., 2006. Bacterial carbon sources in coastal sediments: a
cross-system analysis based on stable isotope data of biomarkers. Biogeosciences. 3,
The study was funded by Commonwealth Scholarship Commission in 175–185. [Link]
Bouillon, S., Dahdouh-Guebas, F., Rao, A.V.V.S., Koedam, N., Dehairs, F., 2003. Sources
the UK (Grant number: BDCS-2017-55). of organic carbon in mangrove sediments: variability and possible ecological
implications. Hydrobiologia. 495, 33–39. [Link]
Declaration of Competing Interest 1025411506526.
Brodersen, K.E., Trevathan-Tackett, S.M., Nielsen, D.A., Connolly, R.M., Lovelock, C.E.,
Atwood, T.B., Macreadie, P.I., 2019. Oxygen consumption and sulfate reduction in
The authors declare that they have no known competing financial vegetated coastal habitats: effects of physical disturbance. Front. Mar. Sci. 6, 00014.
interests or personal relationships that could have appeared to influence [Link]
Capdeville, C., Pommier, T., Gervaix, J., Fromard, F., Rols, J.-L., Leflaive, J., 2019.
the work reported in this paper. Mangrove facies drives resistance and resilience of sediment microbes exposed to
anthropic disturbance. Front. Microbiol. 9 [Link]
Acknowledgements fmicb.2018.03337.
Chaffey, D.R., Miller, F.R., Sandom, J.H., 1985. A forest inventory of the Sundarbans,
Bangladesh. Project Report No.140, Overseas Development Authority (ODA),
We are grateful to Commonwealth Scholarship Commission in the London.
UK for the PhD scholarship to the first author. We want to extend our

11
Md.S. Rahman et al. Catena 200 (2021) 105159

Chambers, L.G., Davis, S.E., Troxler, T., Boyer, J.N., Downey-Wall, A., Scinto, L.J., 2014. Hardie, M., Doyle, R., 2012. Measuring soil salinity. In: Shabala, S., Cuin, T.A. (Eds.),
Biogeochemical effects of simulated sea level rise on carbon loss in an Everglades Plant Salt Tolerance: Methods and Protocols. Humana Press, Totowa, NJ,
mangrove peat soil. Hydrobiologia. 726, 195–211. [Link] pp. 415–425. [Link]
013-1764-6. Hawker, L., Neal, J., Bates, P., 2019. Accuracy assessment of the TanDEM-X 90 Digital
Chambers, L.G., Guevara, R., Boyer, J.N., Troxler, T.G., Davis, S.E., 2016. Effects of Elevation Model for selected floodplain sites. Remote Sens. Environ. 232, 111319
salinity and inundation on microbial community structure and function in a [Link]
mangrove peat soil. Wetlands. 36, 361–371. [Link] Hossain, G.M., Bhuiyan, M.A.H., 2016. Spatial and temporal variations of organic matter
0745-8. contents and potential sediment nutrient index in the Sundarbans mangrove forest,
Chanda, A., Akhand, A., Manna, S., Das, S., Mukhopadhyay, A., Das, I., Hazra, S., Bangladesh. KSCE J. Civil Eng. 20, 163–174. [Link]
Choudhury, S.B., Rao, K.H., Dadhwal, V.K., 2016a. Mangrove associates versus true 0333-0.
mangroves: a comparative analysis of leaf litter decomposition in Sundarban. Howard, J., Hoyt, S., Isensee, K., Telszewski, M., Pidgeon, E. (Eds.), 2014. Coastal blue
Wetlands Ecol. Manage. 24, 293–315. [Link] carbon: methods for assessing carbon stocks and emissions factors in mangroves,
Chanda, A., Mukhopadhyay, A., Ghosh, T., Akhand, A., Mondal, P., Ghosh, S., tidal salt marshes, and seagrasses. Conservation International, Intergovernmental
Mukherjee, S., Wolf, J., Lázár, A.N., Rahman, M.M., Salehin, M., Chowdhury, S.M., Oceanographic Commission of UNESCO, International Union for Conservation of
Hazra, S., 2016b. Blue carbon stock of the Bangladesh Sundarban mangroves: what Nature, Arlington, Virginia, USA, pp. 184.
could be the scenario after a century? Wetlands. 36, 1033–1045. [Link] Hu, Y., Wang, L., Fu, X., Yan, J., Wu, J., Tsang, Y., Le, Y., Sun, Y., 2016. Salinity and
10.1007/s13157-016-0819-7. nutrient contents of tidal water affects soil respiration and carbon sequestration of
Chatterjee, M., Shankar, D., Sen, G., Sanyal, P., Sundar, D., Michael, G., Chatterjee, A., high and low tidal flats of Jiuduansha wetlands in different ways. Sci. Total Environ.
Amol, P., Mukherjee, D., Suprit, K., 2013. Tidal variations in the Sundarbans 565, 637–648. [Link]
estuarine system, India. J. Earth Syst. Sci. 122, 899–933. Iftekhar, M., Saenger, P., 2008. Vegetation dynamics in the Bangladesh Sundarbans
Chowdhury, M.Q., De Ridder, M., Beeckman, H., 2016. Climatic signals in tree rings of mangroves: a review of forest inventories. Wetlands Ecol. Manage. 16, 291–312.
Heritiera fomes Buch.-Ham. in the Sundarbans, Bangladesh. Plos One. 11, e0149788 [Link]
[Link] IPCC, 2014. 2013 Supplement to the 2006 IPCC guidelines for national greenhouse gas
Craft, C., 2007. Freshwater input structures soil properties, vertical accretion, and inventories: Wetlands. IPCC, Geneva, p. 354.
nutrient accumulation of Georgia and U.S tidal marshes. Limnol. Oceanogr. 52, Islam, M.R., Begum, S.F., Yamaguchi, Y., Ogawa, K., 1999. The Ganges and Brahmaputra
1220–1230. [Link] rivers in Bangladesh: basin denudation and sedimentation. Hydrol. Process. 13,
Danielson, T.M., Rivera-Monroy, V.H., Castañeda-Moya, E., Briceño, H., Travieso, R., 2907–2923. [Link]
Marx, B.D., Gaiser, E., Farfán, L.M., 2017. Assessment of Everglades mangrove forest hyp906>[Link];2-e.
resilience: implications for above-ground net primary productivity and carbon Islam, S.N., 2016. Deltaic floodplains development and wetland ecosystems management
dynamics. For. Ecol. Manage. 404, 115–125. [Link] in the Ganges–Brahmaputra–Meghna Rivers Delta in Bangladesh. Sustainable Water
foreco.2017.08.009. Resour. Manage. 2, 237–256. [Link]
De Vos, B., Lettens, S., Muys, B., Deckers, J.A., 2007. Walkley-Black analysis of forest soil Jackson, D.A., 1993. Stopping rules in principal components analysis: a comparison of
organic carbon: recovery, limitations and uncertainty. Soil Use Manage. 23, heuristical and statistical approaches. Ecology. 74, 2204–2214. [Link]
221–229. [Link] 10.2307/1939574.
Donato, D.C., Kauffman, J.B., Murdiyarso, D., Kurnianto, S., Stidham, M., Kanninen, M., Jandl, R., Rodeghiero, M., Martinez, C., Cotrufo, M.F., Bampa, F., van Wesemael, B.,
2011. Mangroves among the most carbon-rich forests in the tropics. Nat. Geosci. 4, Harrison, R.B., Guerrini, I.A., Richter, D.d., Rustad, L., Lorenz, K., Chabbi, A.,
293–297. [Link] Miglietta, F., 2014. Current status, uncertainty and future needs in soil organic
Duarte, C.M., Losada, I.J., Hendriks, I.E., Mazarrasa, I., Marbà, N., 2013. The role of carbon monitoring. Sci. Total Environ. 468–469, 376–383. [Link]
coastal plant communities for climate change mitigation and adaptation. Nat. [Link].2013.08.026.
Climate Change. 3, 961–968. [Link] Jardine, S.L., Siikamäki, J.V., 2014. A global predictive model of carbon in mangrove
Dutta, J., Banerjee, K., Agarwal, S., Mitra, A., 2019. Soil Organic Carbon (soc): a proxy to soils. Environ. Res. Lett. 9, 104013 [Link]
assess the degree of anthropogenic and natural stress. J. Interrupted Stud. 2, 90–102. 104013.
Dutta, M.K., Chowdhury, C., Jana, T.K., Mukhopadhyay, S.K., 2013. Dynamics and Kamruzzaman, M., Basak, K., Paul, S.K., Ahmed, S., Osawa, A., 2019. Litterfall
exchange fluxes of methane in the estuarine mangrove environment of the production, decomposition and nutrient accumulation in Sundarbans mangrove
Sundarbans, NE coast of India. Atmos. Environ. 77, 631–639. [Link] forests, Bangladesh. Forest Sci. Technol. 15, 24–32. [Link]
10.1016/[Link].2013.05.050. 21580103.2018.1557566.
Fontaine, S., Barot, S., Barré, P., Bdioui, N., Mary, B., Rumpel, C., 2007. Stability of Kauffman, J.B., Adame, M.F., Arifanti, V.B., Schile-Beers, L.M., Bernardino, A.F.,
organic carbon in deep soil layers controlled by fresh carbon supply. Nature 450, Bhomia, R.K., Donato, D.C., Feller, I.C., Ferreira, T.O., Jesus Garcia, M.d.C.,
277–280. [Link] MacKenzie, R.A., Megonigal, J.P., Murdiyarso, D., Simpson, L., Hernández Trejo, H.,
Fox, J., Weisberg, S., 2019. An R companion to applied regression, third ed. Sage, 2020. Total ecosystem carbon stocks of mangroves across broad global
Thousand Oaks CA. environmental and physical gradients. Ecol. Monogr. 90 [Link]
Freeman, C., Ostle, N.J., Fenner, N., Kang, H., 2004. A regulatory role for phenol oxidase ecm.1405.
during decomposition in peatlands. Soil Biol. Biochem. 36, 1663–1667. [Link] Kauffman, J.B., Donato, D., 2012. Protocols for the measurement, monitoring and
org/10.1016/[Link].2004.07.012. reporting of structure, biomass and carbon stocks in mangrove forests. Working
Friess, D.A., Rogers, K., Lovelock, C.E., Krauss, K.W., Hamilton, S.E., Lee, S.Y., Lucas, R., Paper 86. Center for International Forestry Research (CIFOR), Bogor, Indonesia, pp.
Primavera, J., Rajkaran, A., Shi, S., 2019. The state of the world’s mangrove forests: 50.
past, present, and future. Annu. Rev. Env. Resour. 44, 89–115. [Link] Khan, M., Amin, M., 2019. Macro nutrient status of Sundarbans forest soils in southern
10.1146/annurev-environ-101718-033302. region of Bangladesh. Bangladesh J. Sci. Ind. Res. 54, 67–72. [Link]
Friess, D.A., Yando, E.S., Abuchahla, G.M.O., Adams, J.B., Cannicci, S., Canty, S.W.J., 10.3329/bjsir.v54i1.40732.
Cavanaugh, K.C., Connolly, R.M., Cormier, N., Dahdouh-Guebas, F., Diele, K., Kominoski, J.S., Gaiser, E.E., Baer, S.G., 2018. Advancing theories of ecosystem
Feller, I.C., Fratini, S., Jennerjahn, T.C., Lee, S.Y., Ogurcak, D.E., Ouyang, X., development through long-term ecological research. Bioscience. 68, 554–562.
Rogers, K., Rowntree, J.K., Sharma, S., Sloey, T.M., Wee, A.K.S., 2020. Mangroves [Link]
give cause for conservation optimism, for now. Curr. Biol. 30, R153–R154. https:// Krauss, K.W., McKee, K.L., Lovelock, C.E., Cahoon, D.R., Saintilan, N., Reef, R., Chen, L.,
[Link]/10.1016/[Link].2019.12.054. 2014. How mangrove forests adjust to rising sea level. New Phytol. 202, 19–34.
Giri, C., Ochieng, E., Tieszen, L.L., Zhu, Z., Singh, A., Loveland, T., Masek, J., Duke, N., [Link]
2011. Status and distribution of mangrove forests of the world using earth Kristensen, E., Bouillon, S., Dittmar, T., Marchand, C., 2008. Organic carbon dynamics in
observation satellite data. Global Ecol. Biogeogr. 20, 154–159. [Link] mangrove ecosystems: a review. Aquat. Bot. 89, 201–219. [Link]
10.1111/j.1466-8238.2010.00584.x. [Link].2007.12.005.
Gleason, S.M., Ewel, K.C., 2002. Organic matter dynamics on the forest floor of a Lacerda, L.D., Ittekkot, V., Patchineelam, S.R., 1995. Biogeochemistry of mangrove soil
micronesian mangrove forest: an investigation of species composition shifts. organic matter: a comparison between Rhizophora and Avicennia soils in south-
Biotropica. 34, 190–198. [Link] eastern Brazil. Estuar. Coast. Shelf Sci. 40, 713–720. [Link]
Goodbred, S.L., Kuehl, S.A., 2000. The significance of large sediment supply, active ecss.1995.0048.
tectonism, and eustasy on margin sequence development: late Quaternary Lange, M., Eisenhauer, N., Sierra, C.A., Bessler, H., Engels, C., Griffiths, R.I., Mellado-
stratigraphy and evolution of the Ganges-Brahmaputra delta. Sediment. Geol. 133, Vázquez, P.G., Malik, A.A., Roy, J., Scheu, S., Steinbeiss, S., Thomson, B.C.,
227–248. [Link] Trumbore, S.E., Gleixner, G., 2015. Plant diversity increases soil microbial activity
Gross, C.D., Harrison, R.B., 2018. Quantifying and comparing soil carbon stocks: and soil carbon storage. Nat. Commun. 6, 6707. [Link]
underestimation with the core sampling method. Soil Sci. Soc. Am. J. 82, 949–959. ncomms7707.
[Link] Legendre, P., Legendre, L.F., 2012. Numerical ecology, third ed. Elsevier, Great Britain,
Hale, R., Bain, R., Goodbred Jr, S., Best, J., 2019. Observations and scaling of tidal mass p. 1006.
transport across the lower Ganges-Brahmaputra delta plain: implications for delta Liu, X., Ruecker, A., Song, B., Xing, J., Conner, W.H., Chow, A.T., 2017. Effects of salinity
management and sustainability. Earth Surface Dyn. 7, 231–245. [Link] and wet–dry treatments on C and N dynamics in coastal-forested wetland soils:
10.5194/esurf-7-231-2019. Implications of sea level rise. Soil Biol. Biochem. 112, 56–67. [Link]
Hamilton, S.E., Friess, D.A., 2018. Global carbon stocks and potential emissions due to 10.1016/[Link].2017.04.002.
mangrove deforestation from 2000 to 2012. Nat. Climate Change. 8, 240–244. Liu, Z., Lee, C., 2006. Drying effects on sorption capacity of coastal sediment: the
[Link] importance of architecture and polarity of organic matter. Geochim. Cosmochim.
Acta. 70, 3313–3324. [Link]

12
Md.S. Rahman et al. Catena 200 (2021) 105159

Lovelock, C.E., Cahoon, D.R., Friess, D.A., Guntenspergen, G.R., Krauss, K.W., Reef, R., zones. Wetlands Ecol. Manage. 23, 269–283. [Link]
Rogers, K., Saunders, M.L., Sidik, F., Swales, A., Saintilan, N., Thuyen, L.X., Triet, T., 9379-x.
2015. The vulnerability of Indo-Pacific mangrove forests to sea-level rise. Nature. Rahman, M.S., Sass-Klaassen, U., Zuidema, P.A., Chowdhury, M.Q., Beeckman, H., 2020.
526, 559–563. [Link] Salinity drives growth dynamics of the mangrove tree Sonneratia apetala Buch.
Lovelock, C.E., Duarte, C.M., 2019. Dimensions of Blue Carbon and emerging -Ham. in the Sundarbans, Bangladesh. Dendrochronologia. 62, 125711 [Link]
perspectives. Biol. Lett. 15, 20180781. [Link] org/10.1016/[Link].2020.125711.
Lovelock, C.E., Sorrell, B.K., Hancock, N., Hua, Q., Swales, A., 2010. Mangrove forest and Rao, R.G., Woitchik, A.F., Goeyens, L., van Riet, A., Kazungu, J., Dehairs, F., 1994.
soil development on a rapidly accreting shore in New Zealand. Ecosystems. 13, Carbon, nitrogen contents and stable carbon isotope abundance in mangrove leaves
437–451. [Link] from an east African coastal lagoon (Kenya). Aquat. Bot. 47, 175–183. [Link]
Luo, M., Huang, J.-F., Zhu, W.-F., Tong, C., 2019. Impacts of increasing salinity and org/10.1016/0304-3770(94)90012-4.
inundation on rates and pathways of organic carbon mineralization in tidal Rath, K.M., Rousk, J., 2015. Salt effects on the soil microbial decomposer community and
wetlands: a review. Hydrobiologia. 827, 31–49. [Link] their role in organic carbon cycling: a review. Soil Biol. Biochem. 81, 108–123.
017-3416-8. [Link]
Macreadie, P.I., Anton, A., Raven, J.A., Beaumont, N., Connolly, R.M., Friess, D.A., Rau, B.M., Melvin, A.M., Johnson, D.W., Goodale, C.L., Blank, R.R., Fredriksen, G.,
Kelleway, J.J., Kennedy, H., Kuwae, T., Lavery, P.S., Lovelock, C.E., Smale, D.A., Miller, W.W., Murphy, J.D., Todd, D.E.J., Walker, R.F., 2011. Revisiting soil carbon
Apostolaki, E.T., Atwood, T.B., Baldock, J., Bianchi, T.S., Chmura, G.L., Eyre, B.D., and nitrogen sampling: quantitative pits versus rotary cores. Soil Science. 176,
Fourqurean, J.W., Hall-Spencer, J.M., Huxham, M., Hendriks, I.E., Krause- 273–279. [Link]
Jensen, D., Laffoley, D., Luisetti, T., Marbà, N., Masque, P., McGlathery, K.J., Ray, R., Baum, A., Rixen, T., Gleixner, G., Jana, T.K., 2018. Exportation of dissolved
Megonigal, J.P., Murdiyarso, D., Russell, B.D., Santos, R., Serrano, O., Silliman, B.R., (inorganic and organic) and particulate carbon from mangroves and its implication
Watanabe, K., Duarte, C.M., 2019. The future of Blue Carbon science. Nat. Commun. to the carbon budget in the Indian Sundarbans. Sci. Total Environ. 621, 535–547.
10, 3998. [Link] [Link]
Marchand, C., 2017. Soil carbon stocks and burial rates along a mangrove forest Ray, R., Ganguly, D., Chowdhury, C., Dey, M., Das, S., Dutta, M.K., Mandal, S.K.,
chronosequence (French Guiana). For. Ecol. Manage. 384, 92–99. [Link] Majumder, N., De, T.K., Mukhopadhyay, S.K., Jana, T.K., 2011. Carbon sequestration
10.1016/[Link].2016.10.030. and annual increase of carbon stock in a mangrove forest. Atmos. Environ. 45,
Marton, J.M., Herbert, E.R., Craft, C.B., 2012. Effects of salinity on denitrification and 5016–5024. [Link]
greenhouse gas production from laboratory-incubated tidal forest soils. Wetlands. Ren, H., Jian, S., Lu, H., Zhang, Q., Shen, W., Han, W., Yin, Z., Guo, Q., 2008. Restoration
32, 347–357. [Link] of mangrove plantations and colonisation by native species in Leizhou bay, South
Matsui, N., Meepol, W., Chukwamdee, J., 2015. Soil organic carbon in mangrove China. Ecol. Res. 23, 401–407. [Link]
ecosystems with different vegetation and sedimentological conditions. J. Marine Sci. Rogers, K., Kelleway, J.J., Saintilan, N., Megonigal, J.P., Adams, J.B., Holmquist, J.R.,
Eng. 3, 1404–1424. [Link] Lu, M., Schile-Beers, L., Zawadzki, A., Mazumder, D., Woodroffe, C.D., 2019.
McLeod, E., Chmura, G.L., Bouillon, S., Salm, R., Björk, M., Duarte, C.M., Lovelock, C.E., Wetland carbon storage controlled by millennial-scale variation in relative sea-level
Schlesinger, W.H., Silliman, B.R., 2011. A blueprint for blue carbon: toward an rise. Nature. 567, 91–95. [Link]
improved understanding of the role of vegetated coastal habitats in sequestering Rogers, K.G., Goodbred, S.L., Mondal, D.R., 2013. Monsoon sedimentation on the
CO2. Front. Ecol. Environ. 9, 552–560. [Link] ‘abandoned’ tide-influenced Ganges-Brahmaputra delta plain. Estuar. Coast. Shelf
Mitra, A., Banerjee, K., Sett, S., 2012. Spatial variation in organic carbon density of Sci. 131, 297–309. [Link]
mangrove soil in Indian Sundarbans. Natl. Acad. Sci. Lett. 35, 147–154. [Link] Rovai, A.S., Twilley, R.R., Castañeda-Moya, E., Riul, P., Cifuentes-Jara, M., Manrow-
org/10.1007/s40009-012-0046-6. Villalobos, M., Horta, P.A., Simonassi, J.C., Fonseca, A.L., Pagliosa, P.R., 2018.
Morrissey, E.M., Gillespie, J.L., Morina, J.C., Franklin, R.B., 2014. Salinity affects Global controls on carbon storage in mangrove soils. Nat. Climate Change. 8,
microbial activity and soil organic matter content in tidal wetlands. Global Change 534–538. [Link]
Biol. 20, 1351–1362. [Link] Rudra, K., 2014. Changing river courses in the western part of the Ganga-Brahmaputra
Nasrin, S., Hossain, M., Rahman, M.M., 2019. Adaptive responses to salinity: nutrient delta. Geomorphology. 227, 87–100. [Link]
resorption efficiency of Sonneratia apetala (Buch.-Ham.) along the salinity gradient geomorph.2014.05.013.
in the Sundarbans of Bangladesh. Wetlands Ecol. Manage. 27, 343–351. [Link] Sanderman, J., Hengl, T., Fiske, G., Solvik, K., Adame, M.F., Benson, L., Bukoski, J.J.,
org/10.1007/s11273-019-09663-6. Carnell, P., Cifuentes-Jara, M., Donato, D., Duncan, C., Eid, E.M., Ermgassen, P.Z.,
Nayak, A.K., Rahman, M.M., Naidu, R., Dhal, B., Swain, C.K., Nayak, A.D., Tripathi, R., Lewis, C.J.E., Macreadie, P.I., Glass, L., Gress, S., Jardine, S.L., Jones, T.G.,
Shahid, M., Islam, M.R., Pathak, H., 2019. Current and emerging methodologies for Nsombo, E.N., Rahman, M.M., Sanders, C.J., Spalding, M., Landis, E., 2018. A global
estimating carbon sequestration in agricultural soils: a review. Sci. Total Environ. map of mangrove forest soil carbon at 30 m spatial resolution. Environ. Res. Lett. 13
665, 890–912. [Link] [Link]
Nayak, B., Zaman, S., Gadi, S.D., Raha, A.K., Mitra, A., 2014. Dominant gastropods of Sarker, S.K., Deb, J.C., Halim, M.A., 2011. A diagnosis of existing logging bans in
Indian Sundarbans: a major sink of carbon. Int. J. Adv. Pharm. Biol. Chem. 3, Bangladesh. Int. For. Rev. 13, 461–475. [Link]
282–289. 146554811798811344.
Nellemann, C., Corcoran, E., Duarte, C., Valdés, L., De Young, C., Fonseca, L., Grimsditch, Sarker, S.K., Matthiopoulos, J., Mitchell, S.N., Ahmed, Z.U., Mamun, M.B.A., Reeve, R.,
G. (Eds.), 2009. Blue carbon. A rapid response assessment. United Nations 2019a. 1980s–2010s: The world’s largest mangrove ecosystem is becoming
Environment Programme (UNEP), GRID-Arendal, pp. 78. homogeneous. Biol. Conserv. 236, 79–91. [Link]
Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon, and organic matter. In: biocon.2019.05.011.
Sparks, D. et al. (Eds.), Methods of soil analysis: Part 3 Chemical methods, pp. Sarker, S.K., Reeve, R., Paul, N.K., Matthiopoulos, J., 2019b. Modelling spatial
961–1010. [Link] biodiversity in the world’s largest mangrove ecosystem—The Bangladesh
Nóbrega, G.N., Ferreira, T.O., Artur, A.G., de Mendonça, E.S., Leão, R.A. de O., Sundarbans: a baseline for conservation. Divers. Distrib. 25, 729–742. [Link]
Teixeira, A.S., Otero, X.L., 2015. Evaluation of methods for quantifying organic org/10.1111/ddi.12887.
carbon in mangrove soils from semi-arid region. J. Soils Sed. 15, 282–291. https:// Sarker, S.K., Reeve, R., Thompson, J., Paul, N.K., Matthiopoulos, J., 2016. Are we failing
[Link]/10.1007/s11368-014-1019-9. to protect threatened mangroves in the Sundarbans world heritage ecosystem? Sci.
Nyman, J.A., Delaune, R.D., Patrick, W.H., 1990. Wetland soil formation in the rapidly Rep. 6 [Link]
subsiding Mississippi River Deltaic Plain: mineral and organic matter relationships. Setia, R., Gottschalk, P., Smith, P., Marschner, P., Baldock, J., Setia, D., Smith, J., 2013.
Estuar. Coast. Shelf Sci. 31, 57–69. [Link] Soil salinity decreases global soil organic carbon stocks. Sci. Total Environ. 465,
P. 267–272. [Link]
Odum, E.P., 1969. The strategy of ecosystem development. Science. 164, 262–270. Siddiqi, N.A., 2001. Mangrove forestry in Bangladesh. Institute of Forestry &
[Link] Environmental Sciences, University of Chittagong, Chittagong, Bangladesh, pp. 301.
Ouyang, X., Lee, S.Y., 2020. Improved estimates on global carbon stock and carbon pools Spivak, A.C., Sanderman, J., Bowen, J.L., Canuel, E.A., Hopkinson, C.S., 2019. Global-
in tidal wetlands. Nat. Commun. 11, 317. [Link] change controls on soil-carbon accumulation and loss in coastal vegetated
14120-2. ecosystems. Nat. Geosci. 12, 685–692. [Link]
Passos, T.R.G., Artur, A.G., Nóbrega, G.N., Otero, X.L., Ferreira, T.O., 2016. Comparison Syvitski, J.P.M., Milliman, J.D., 2007. Geology, geography, and humans battle for
of the quantitative determination of soil organic carbon in coastal wetlands dominance over the delivery of fluvial sediment to the coastal ocean. J. Geol. 115,
containing reduced forms of Fe and S. Geo-Mar. Lett. 36, 223–233. [Link] 1–19. [Link]
10.1007/s00367-016-0437-7. Taillardat, P., Friess, D.A., Lupascu, M., 2018. Mangrove blue carbon strategies for
Pendleton, L., Donato, D.C., Murray, B.C., Crooks, S., Jenkins, W.A., Sifleet, S., Craft, C., climate change mitigation are most effective at the national scale. Biol. Lett. 14,
Fourqurean, J.W., Kauffman, J.B., Marbà, N., Megonigal, P., Pidgeon, E., Herr, D., 20180251. [Link]
Gordon, D., Baldera, A., 2012. Estimating global “blue carbon” emissions from Twilley, R.R., Castañeda-Moya, E., Rivera-Monroy, V.H., Rovai, A., 2017. Productivity
conversion and degradation of vegetated coastal ecosystems. PLOS ONE. 7, e43542 and carbon dynamics in mangrove wetlands. In: Rivera-Monroy, V.H., Lee, S.Y.,
[Link] Kristensen, E., Twilley, R.R. (Eds.), Mangrove Ecosystems: A Global Biogeographic
Prasad, M.B.K., Kumar, A., Ramanathan, A.L., Datta, D.K., 2017. Sources and dynamics Perspective: Structure, Function, and Services. Springer International Publishing,
of sedimentary organic matter in Sundarban mangrove estuary from Indo-Gangetic Cham, pp. 113–162. [Link]
delta. Ecol. Process. 6, 8. [Link] Twilley, R.R., Rovai, A.S., Riul, P., 2018. Coastal morphology explains global blue carbon
R Core Team, 2019. R: A Language and Environment for Statistical Computing. R distributions. Front. Ecol. Environ. 16, 503–508. [Link]
Foundation for Statistical Computing, Vienna, Austria. Tyagi, N.K., Sen, H.S., 2019. Development of Sundarbans through estuary management
Rahman, M.M., Khan, M.N.I., Hoque, A.K.F., Ahmed, I., 2015. Carbon stock in the for augmenting freshwater supply, improved drainage and reduced bank erosion. In:
Sundarbans mangrove forest: spatial variations in vegetation types and salinity Sen, H.S. (Ed.), The Sundarbans: A Disaster-Prone Eco-Region: Increasing Livelihood

13
Md.S. Rahman et al. Catena 200 (2021) 105159

Security. Springer International Publishing, Cham, pp. 375–402. [Link] Woodroffe, C.D., Rogers, K., McKee, K.L., Lovelock, C.E., Mendelssohn, I.A., Saintilan, N.,
10.1007/978-3-030-00680-8_13. 2016. Mangrove sedimentation and response to relative sea-level rise. Ann. Rev.
Wang, G., Guan, D., Peart, M.R., Chen, Y., Peng, Y., 2013. Ecosystem carbon stocks of Mar. Sci. 8, 243–266. [Link]
mangrove forest in Yingluo Bay, Guangdong Province of South China. For. Ecol. Wuest, S.B., 2009. Correction of bulk density and sampling method biases using soil mass
Manage. 310, 539–546. [Link] per unit area. Soil Sci. Soc. Am. J. 73, 312–316. [Link]
Weiss, C., Weiss, J., Boy, J., Iskandar, I., Mikutta, R., Guggenberger, G., 2016. Soil sssaj2008.0063.
organic carbon stocks in estuarine and marine mangrove ecosystems are driven by Yando, E.S., Osland, M.J., Willis, J.M., Day, R.H., Krauss, K.W., Hester, M.W., 2016. Salt
nutrient colimitation of P and N. Ecol. Evol. 6, 5043–5056. [Link] marsh-mangrove ecotones: using structural gradients to investigate the effects of
ece3.2258. woody plant encroachment on plant–soil interactions and ecosystem carbon pools.
Wickham, H., 2016. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag, New J. Ecol. 104, 1020–1031. [Link]
York. Yang, J., Gao, J., Liu, B., Zhang, W., 2014. Sediment deposits and organic carbon
Więski, K., Guo, H., Craft, C.B., Pennings, S.C., 2010. Ecosystem functions of tidal fresh, sequestration along mangrove coasts of the Leizhou Peninsula, southern China.
brackish, and salt marshes on the Georgia coast. Estuaries Coasts. 33, 161–169. Estuar. Coast. Shelf Sci. 136, 3–10. [Link]
[Link] Zhao, Q., Bai, J., Lu, Q., Zhang, G., 2017. Effects of salinity on dynamics of soil carbon in
Wong, V.N.L., Dalal, R.C., Greene, R.S.B., 2009. Carbon dynamics of sodic and saline degraded coastal wetlands: Implications on wetland restoration. Phys. Chem. Earth,
soils following gypsum and organic material additions: a laboratory incubation. Parts A/B/C. 97, 12–18. [Link]
Appl. Soil Ecol. 41, 29–40. [Link] Zhou, G., Xu, S., Ciais, P., Manzoni, S., Fang, J., Yu, G., Tang, X., Zhou, P., Wang, W.,
Wong, V.N.L., Greene, R.S.B., Dalal, R.C., Murphy, B.W., 2010. Soil carbon dynamics in Yan, J., Wang, G., Ma, K., Li, S., Du, S., Han, S., Ma, Y., Zhang, D., Liu, J., Liu, S.,
saline and sodic soils: a review. Soil Use Manage. 26, 2–11. [Link] Chu, G., Zhang, Q., Li, Y., Huang, W., Ren, H., Lu, X., Chen, X., 2019. Climate and
10.1111/j.1475-2743.2009.00251.x. litter C/N ratio constrain soil organic carbon accumulation. Natl. Sci. Rev. 6,
746–757. [Link]

14