Geo-Marine Letters

, Volume 38, Issue 6, pp 481–496 | Cite as

Sediment transport by tropical cyclones recorded in a submarine canyon off Bangladesh

  • Hermann R. KudrassEmail author
  • Björn Machalett
  • Luisa Palamenghi
  • Inka Meyer
  • Wenyan Zhang


Frequent cyclones originating in the Bay of Bengal landfall on the delta coast of the Ganges and Brahmaputra rivers. The cyclones are well recorded in the sediments of a canyon that is deeply incised into the shelf offshore Bangladesh. The large mud supply by the two rivers forms temporary deposits on the innermost shelf, where they are mobilized by waves and currents during the passage of cyclones. The resulting, highly concentrated fine sand-silt-clay suspension is moved by wind-induced currents and eventually plunges into the shelf canyon. These gravity flows are deposited as graded beds on the broad canyon floor. In a 362-cm-long section of a dated sediment core covering the period from 2006 to 1985, nearly all 59 graded beds can be correlated with 42 cyclones observed in that period. The threefold decrease in the sedimentation rate of the last decade compared to the period from 1994 to 1954 is due to the decreased number and power of cyclones. Compared to the sediment transfer by cyclones, the input by local sediment slumps, tidal currents, and monsoonal floods is small. Thus, cyclones dominate the mobilization and distribution of sediment on the Bangladesh shelf. This sediment dispersal mechanism is probably also typical for other shelf areas crossed by tropical cyclones.


Cyclones have a massive impact on the coast because of high wind velocities, storm surges, and heavy rainfall. Before landfall, waves and currents associated with the cyclones mobilize and disperse sediment from the impacted shallow-water area landward to the coastal plain and seaward to the shelf, the continental slope and into canyons (cf. reviews by Bonnin et al. 2008; Puig et al. 2014; Shanmugan 2008; Ulses et al. 2008; Walsh and Nittrouer 2009; Zhang et al. 2016). In the northern Bay of Bengal, tropical cyclones frequently occur during the pre- and post-monsoon seasons. Most cyclones move north to northeastward from the central Bay of Bengal and impact the estuarine coast of Bangladesh along its entire width (Fig. 1). Only few tropical depressions trend westward.
Fig. 1

Tracks of tropical cyclones from 1985 to 1994 (a) and 1994–2006 (b) with less than 300 km distance from the core position SO 188 336 KL (marked by x). Dots on the track show the center of the cyclone for every 6 h (track data from the NOAA database IBTrACS, Knapp et al. 2010). Map drawn with ArcMap 10.4.1 using World Ocean Base Map and General Bathymetric Chart of the Oceans (GEBCO), NOAA National Centers for Environmental Information (NCEI), Natural Earth

The combination of strong onshore cyclonic winds and intense precipitation produces floods and storm surges as high as 12 m with severe consequences for the low-lying coastal region of the Ganges-Brahmaputra delta (e.g., Antony et al. 2014; Auerbach et al. 2015; Kim et al. 2008; Krien et al. 2017). Since 1954, the number and intensity of cyclones recorded by the International Best Track Archive for Climate Stewardship (IBTrACS, Knapp et al. 2010) have varied greatly. Cyclones in the period 1954 to 1977 are more frequent and intense compared to those between 1991 and 2006. Various attempts have been made to investigate the changing short- and mid-term frequency of these high-energy events with respect to climatic processes and anticipated future global climate variability (Emanuel 2011; Klotzbach and Landsea 2015; Webster et al. 2005). However, a historical 60-year-long cyclonic record from the Indian Ocean (Hoarau et al. 2012; Knapp et al. 2010; Mooley 1980) is not long enough to test multi-decadal trends and variations.

In this context, the aim of this investigation is to demonstrate that cyclones crossing the coast of Bangladesh dominate sediment dynamics and the transfer of sediment to the shelf canyon “Swatch of No Ground.” Therefore, the sedimentary record in the canyon is investigated because it should serve as a long-term archive for the frequency of cyclonic events.

Physical setting

The inner shelf of Bangladesh receives about 30% of the 1 billion tons of suspended material, which is mostly transported by the monsoonal floods in the rivers Ganges and Brahmaputra (Goodbred and Kuehl 2000; Kuehl et al. 1997; Lupker et al. 2013). The fluvial load of fine sand, silt, and clay predominantly accumulates at the mouth of the Ganges-Brahmaputra located along the eastern coastal islands (Allison 1998; Rahman et al. 2011; Sarwar and Woodroffe 2013; Wilson and Goodbred 2015). A portion of the fluvial load is transported southwestward by tidal currents (Barua et al. 1994) and temporarily settles out on the topset beds of the submarine delta in shallow water of less than 20 m (Fig. 2) (Kuehl et al. 1989). The topset beds, composed of clayey fine sand and silt, extend up to a water depth of about 30 m (Kuehl et al. 2005). The clinoform foreset beds aggrade in 30–60 m water depth at a rate of a few centimeters per year, receiving about 13–20% of the total fluvial supply of terrigenous material (Kuehl et al. 1997; Michels et al. 1998; Palamenghi et al. 2011; Suckow et al. 2001). The foreset beds consist of bioturbated mud with some intercalated layers of fine sand (Michels et al. 1998; Kuehl et al. 2005). The outer shelf extends from the toe of the muddy bottom-set beds at about 80 m water depth to the shelf edge at 150 m. It is covered by a discontinuous layer of palimpsest muddy sand, which contains biogenic debris and fluvial deposits of the last glacial period reworked during lowered sea level (Michels et al. 1998).
Fig. 2

Bathymetry of the northern Bay of Bengal and the position of the sediment core SO 188 336 KL within the shelf canyon “Swatch of No Ground.” Recent sedimentation rates are based on various authors (see text)

The “Swatch of No Ground” canyon (SONG) is the most prominent bathymetric feature on the Bengal shelf. It is deeply incised into the shelf and ends in a 20-km-wide amphitheater at the 20-m isobaths, where it intersects the prograding foreset beds (Fig. 2). The landward rim of the canyon is dissected by some small meandering channels (Kottke et al. 2003; Rogers and Goodbred 2010; and Palamenghi et al. unpublished). The canyon walls at the northeastern end of the canyon dip at about 10–20° toward the southwest. The slope and the 10-km-broad canyon floor are partly covered by acoustically finely stratified, parallel-bedded sequences (Facies “Ia” of Kottke et al. 2003). The sediments of this seismic facies consist of centimeter- to millimeter-thick laminated graded beds of fine sand, silt, and clay. The most prominent graded sand beds were assumed to have been deposited by intense tropical cyclones affecting the northern Bay of Bengal (Kudrass et al. 1998; Michels et al. 1998; Michels et al. 2003).

Based on five (up to 18 m long) piston cores recovered from this stratified facies, the annual sedimentation rates, determined by the 137Cs concentrations since 1954, decrease seaward from a proximal rate of 40 cm/year to a distal rate of 14 cm/year (Michels et al. 2003). High sedimentation rates with highly reactive organic carbon contents (Galy et al. 2007) and restricted circulation below the low-salinity surface waters result in canyon water enriched in methane and depleted in dissolved oxygen (Berner et al. 2003). The terrestrial organic matter is relatively old compared to the young depositional age indicative of millennial residence time in the delta (French et al. 2018). Consequently, the texture of the bottom sediments is not disturbed by burrowing. Based on density and magnetic susceptibility logs, and a few supportive grain-size analyses, individual sets of laminae can be combined into graded bed units indicative of cyclone activity (Kudrass et al. 1998; Michels et al. 2003).

Materials and methods

Setting time marks

In February 1994, an 11.3-m-long piston core (SO 93 96 KL) was recovered at 270 m water depth on the northern slope of the canyon (Fig. 2; 21.3507° N; 89.5777° E). The recovered sediments consist of graded beds deposited at an average sedimentation rate of 40 cm/year (Kudrass et al. 1998). In August 2006, a second piston core with a length of 18.3 m (SO 188 336 KL) was recovered at the same position. Both cores were logged by a multi-sensor core logger at 1 cm steps, and 137Cs concentrations were determined in steps of 10 or 50 cm. The magnetic susceptibility values correlate positively with the sand content of the sediment (Michels et al. 1998). Starting from a core depth of 174 cm in the second core, the log fits that of the first core, and the 174 cm horizon in core SO 188 336 KL is thus a reliable time mark for February 1994 when core SO 93 96 KL was collected (Fig. 3). Another time mark is the significant increase in the concentration of 137Cs between 17.60 and 17.00 m in core SO 188 336 KL, which marks the radioactive emissions of atomic bombs in 1954 ad (Suckow et al. 2001). 137Cs concentrations were determined using a high-purity germanium gamma spectrometer.
Fig. 3

Correlation of core SO 188 336 KL with core SO 93 96 KL using magnetic susceptibility logs (in SI units of the multi-sensor core logger). The water-rich surface sediments in core 96 KL cause the low-amplitude variations in the magnetic logs

Grain-size analysis and sediment texture

Samples of uniform dimensions (0.5 × 0.5 × 0.5 cm) were taken from sediment slices cut off for radiography from core SO 188 336 KL. On the samples from the upper, 362 cm continuous high-resolution grain-size analyses (n = 724) were carried out. Particle-size distributions were measured by means of a Beckman-Coulter LS 13320 PIDS laser diffraction particle-size analyzer following the method described in Machalett et al. (2008) and Machalett (2017), resulting in 116 particle-size classes per sample. An auto-prep station was used to measure each sample (2–3 aliquots per sample, 5 measurements per aliquot) under identical conditions using the defined standard operating method BM-v13 (Machalett 2011; Serno et al. 2014). Grain-size distributions were subdivided according to the Wentworth classification scheme. In addition, the texture of the 0.5-cm-thick core slices from core SO 188 336 KL was investigated by X-radiography. Furthermore, a high-resolution XRF scan was carried out at 0.2 mm step intervals on a 12-cm-long core section.

Establishing a proxy for cyclonic sediment transfer into the canyon

The amount of sediment mobilized by tropical cyclones depends on many variables (Dail et al. 2007; Keen and Slingerland 1993; Miles et al. 2015; Palanques et al. 2006; Warner et al. 2008), which may change substantially during the passage of an individual storm and from one storm to another. The most important parameters are the following:
  • cyclonic wind velocity, which determines the wind-induced wave height, as well as the velocity and direction of wind-induced currents and, finally, also the amount of mobilized and laterally transported sediment;

  • track and diameter of the cyclone, which determines the size of the impacted shelf area and the transport path of the mobilized sediment (Fig. 4);

  • forward velocity of the cyclone, which defines the duration of cyclonic impact;

  • local bathymetry, which defines the source and sink areas for the mobilized sediment;

  • availability and grain size of sediment within the reach of the cyclonic impact.

Fig. 4

Two hypothetical tracks of cyclones with the ring of cyclonic winds (50–100 km) demonstrate the different directions of sediment transport with respect to the canyon

In the Bay of Bengal, cyclones occur at an average frequency of 1.5 cyclones per year mainly in the months before and after the summer monsoon (Alam et al. 2003; Girishkumar and Ravichandran 2012). Most cyclones move northward until making landfall on the Ganges-Brahmaputra Delta (Fig. 1). As the cyclones greatly differ with respect to wind velocity, track location and forward velocity, a model was used to estimate the relative transport capacity of each cyclone from the inner shelf into the SONG.

The model for this transport is based on the cyclone track location, track direction, wind velocity, and forward velocity. The corresponding data from NOAA ( which are reported at 6-h intervals contain the geographic center position of the cyclone and the estimated maximum wind velocity (Knapp et al. 2010). However, for an estimation of sediment transfer into the canyon, additional parameters, especially the radius of the cyclone and the distribution of wind velocity within the gyre, have to be assumed. Recent satellite observations of cyclones in the Bay of Bengal have shown the great variability of these two parameters, which additionally change along the track of a cyclone (Krien et al. 2017). As this information is not available for former cyclones, a standard size of 300 km radius for each cyclone is assumed. The maximum cyclone diameter was controlled from the daily wind velocity data of six onshore meteorological stations (Satkhira, Khulna, Barisal, Bhola, Patuakhali, Chittagong) supplied by the Bangladesh Meteorological Department for the period 1966–1999. In cases where the stations near the path of the approaching cyclone did not record a high daily cumulative wind velocity (> 200 km per day), the radius of the cyclone was correspondingly reduced. This had to be done for less than 10% of the cyclones.

The maximum wind velocity (based on the NOAA data) is then set for the radial segment of 50–100 km from the center point (Carrasco et al. 2014). The distribution of wind velocity within the outer gyre is assumed to decrease linearly in steps of 50 km to zero velocity at the outer limit of the 300-km radius (Table 1). Kim et al. (2008), McPhaden et al. (2009), and Chang et al. (2010) reported similar decreases in wind velocity within the cyclone gyres investigated by them. The wind-induced current is assumed to be 3% of the wind velocity (Chang et al. 2010; Mairs et al. 1992; Open University 1989), and the direction of the cyclone-induced surface current is assumed to propagate in the direction of the wind vector, as was observed in the Gulf of Mexico (Chang et al. 2010; Shay et al. 1989), the Great Barrier Reef (Carter et al. 2009), the Gulf of Maine (Miles et al. 2015; Warner et al. 2008), and as modeled for the Bangladesh cyclone of April 1991 (Kim et al. 2008). In this way, the transport distance during a 6-h-long period is estimated for each of the six 50-km-wide cyclonic rings (Table 1). In the next step, the geometry of the radial transport path was considered. The position of a ring segment with respect to the canyon head (the dotted line in Fig. 5a, b) and the angle of the incoming cyclonic current are used to estimate the impacted area that moved across the 30-km-long border of the canyon. The transport was generally calculated for every 6 h during the passage of the cyclone, but in cases where the cyclone center passed closer than 100 km from the canyon head, the interval was reduced to 3 h in order to take the rapid change in current directions into account.
Table 1

Estimation of transport distance in the cyclonic gyre

Ring segment (km)

Maximum wind velocity (km/h)

Induced water current at 3% of wind (km/h)

Transport distance in 6 h (km)


Vmax × 0.562

Vmax × 0.562 × 0.03

Vmax × 0.562 × 0.03 × 6


Vmax × 0.908

Vmax × 0.908 × 0.03

Vmax × 0.908 × 0.03 × 6


Vmax × 0.732

Vmax × 0.732 × 0.03

Vmax × 0.732 × 0.03 × 6


Vmax × 0.503

Vmax × 0.503 × 0.03

Vmax × 0.503 × 0.03 × 6


Vmax × 0.368

Vmax × 0.368 × 0.03

Vmax × 0.368 × 0.03 × 6


Vmax × 0.184

Vmax × 0.184 × 0.03

Vmax × 0.184 × 0.03 × 6

Fig. 5

a Bathymetry of the canyon head with the position of sediment core SO 188 336 KL. The 30-km-long border line along the 30 m isobath (dotted) was used to estimate the transport of wind-induced currents into the canyon (arrow). b Estimation of the turbid water mass transported across the 30-km-wide border line (left red dotted line) of the canyon head by a 50-km-wide cyclonic ring. The water mass depends on the position of the ring with respect to the border and its incident angle with the border

The relative sediment transport capacity of each cyclone segment into the canyon (Qs) is finally assessed using the equation: Qs = AU2.33, where A represents an ensemble effect of the impacted area and regionally constant factors like grain size and bathymetry, and U is the wind velocity of the segment. For a more detailed account, see the Supplement at the end of the paper.

Two important parameters are not considered in the above ranking procedure: tidal currents and availability of sediment. The semidiurnal tides with residual currents setting in southwesterly directions (Barua et al. 1994) certainly modify the sediment transport direction temporarily on the inner shelf, but they probably have only a transient effect because the durations of the cyclonic impacts usually last longer than 24 h. The availability of clayey silt on the inner shelf may significantly increase during the high summer monsoon runoff and would thus increase the amount of transported sediment during post-monsoonal cyclones. However, the turbid water observed on satellite images during the non-monsoonal periods (Barua 1990) indicates a permanent supply of fine sandy mud, which accumulates on the innermost shelf as an ephemeral 2–3-m-thick veneer (Kuehl et al. 1997). In addition, cyclones may erode the fine sandy muds of the coastal islands (Allison 1998; Hanebuth et al. 2013; Rahman et al. 2011; Sarwar and Woodroffe 2013). Different cyclonic tracks will also affect different source areas. It is thus assumed that there is no limitation for the availability of fine-grained sediment.


Grain size and texture

The entire core SO 188 336 KL is composed of silty sand and clayey silt, forming laminated graded beds. The lamina thickness varies from a few centimeters in the sandy bottom portion to predominantly millimeter-sized laminae in the silty clay sequences in the top portion of a graded bed. The rapid increase of the sand fraction (volume-%) defines the bottom of a graded bed. The maximum sand content frequently occurs a few centimeters above its base (e.g., Figs. 6 and 7). This inversely graded portion at the bottom is composed of several centimeter-thick laminae that are internally normally graded (Fig. 6). The mean grain size of each subsequent lamina increases to form an inversely graded basal sequence. Another set of normally graded centimeter-sized laminae with generally decreasing grain-size modes forms the middle part of the graded bed. The upper part is devoid of sand or coarse silt. It consists of many millimeter-thick laminae variably composed of fine to medium silt and clay. The overall fining upward trend continues, and the finest grain-size mode is used to define the upper limit of the graded bed (Figs. 7, 9, and 10).
Fig. 6

X-radiograph of section 180–196 cm of core SO 188 336 KL (darker color indicates higher density or coarser grain size) combined with Zr-counts of the in situ-XRF scanner in 0.2 mm steps (thin line) and 5 points-moving average (thick line). Higher Zr-counts indicate coarser sediment (Flood et al. 2018). Graded sections are marked with arrows (normal grading is shown by thick/thin lines, inverse grading by dotted lines)

Fig. 7

a Grain-size distributions between core depths 266.5 and 252.5 cm showing the abrupt transition of two graded beds and the stepwise grain-size decrease in the overlying section. b Detailed grain-size distributions between core depths 265.5 and 262 cm. From 264.5 to 263 cm, the increasing contributions of coarse silt and fine sand indicate the increasing sediment transport during the onset of a cyclone

In most graded beds, the sand content is low, mostly less than 15%, while silt is the dominant fraction. Even the coarsest sample (at 154 cm) contains only 31% sand, the remainder being composed of silt (64%) and clay (5%). The thickness of graded beds ranges from 1.5 (by definition) to 17 cm, with the average sand content varying from 0.7–11.7%. The average sand content of individual graded beds generally increases with their thickness, albeit with a high variability (Fig. 8).
Fig. 8

Average sand content (volume-%) of graded beds versus thickness (cm)

Cross-bedding was not observed in the sandy silt of the bottom section. As the fine-grained top laminae of the graded beds are preserved, erosion at the concordant contacts between the graded beds seems to be small. The boundaries of most graded beds are well defined (Figs. 9 and 10), although some have a more complex structure, for instance with two sand maxima (194–208 cm) or fluctuating very low sand content in the top part. In any case, the finest grain-size mode is used to define their top. Using this definition, 27 graded beds have been identified in the core section 0–174 cm and 21 graded beds in the lower part of the core 174–362 cm (Fig. 10).
Fig. 9

Grain-size variations of five graded beds and their assignment to historical cyclones

Fig. 10

Assignment of graded beds to the record of tropical cyclones (2006–1985) and their transport capacity

Regional variations of the cyclonic transport capacity

The ranking of the cyclonic relative transport capacity into the canyon varies from 0.05 to 642 (see Supplement for the compilation of the data). The most important variable is the track of the cyclone with respect to the location of the canyon (Figs. 1 and 4). According to the model, most cyclones start with low-volume transport, reach a maximum at their closest position to the canyon, and then fade out with landfall. Cyclones passing east of the canyon mobilize and transport more sediment from the shallow inner shelf into the canyon than cyclones passing west of the canyon. During the passages of cyclones west of the canyon, the sediment is mainly transported into the delta. The estimated transport directions of some cyclones passing eastward of the canyon head at distances < 100 km change from westward to northward transport (without input into the canyon) and eventually end with eastward transport into the canyon.

Relation of cyclonic events and graded beds

The correlation between historical and sedimentological records is based on the two joint time markers (August 2006, February 1994) and on the assumption that the thickness of graded beds increases with cyclonic transport capacity. In a first step, the graded beds near the two time marks of 2006 and 1994 were assigned to the respective cyclones of the meteorological record. Then the four prominent graded beds at 25 cm, 150 cm, 260 cm, and 355 cm were correlated with the cyclones having high transport capacities (May 2004, November 1995, April 1991, and May 1985). In the next step, the smaller graded beds were assigned to respective weaker cyclones (Fig. 10).

The assignment of graded beds and cyclones to the estimation of the transport capacity is complicated by uncertainties in both time series. However, the number of cyclones in the two well-defined periods generally corresponds to the number of graded beds. From 2006 to 1994, 21 cyclones correspond to 27 graded beds, whereas in the period from 1994 to 1985, the 21 known cyclones correspond to 22 graded beds. Several, usually thin, graded beds that could not be assigned to a cyclone were probably produced by waves radiating from remotely passing cyclones, local strong winds (not all tropical depressions are documented by the data of the National Oceanic and Atmospheric Administration (NOAA) supported by the semidiurnal tides, or extreme monsoonal floods (for a more rigorous test of the model see “Long-term sedimentation rates and relative transport capacity” section).

According to the assignment process, the properties of individual graded beds (total thickness, sand thickness, and average sand content) are positively but not highly correlated with the transport capacity (Fig. 11), all three parameters showing high scatter. The sand thickness in the graded beds, which had not been used for the correlation, increases by three orders of magnitude and shows the best correlation with transport capacity. Especially this last relation indicates that high-energy cyclones transport coarser material into the canyon compared to weaker cyclones. In toto, these relations confirm the concept that cyclones are the main process delivering sediment to the canyon. In summary, about 85% of the core section can be attributed to the impact of cyclones.
Fig. 11

Average sand content (volume-%), thickness of the sand portion (0.05 mm), and total thickness (cm) of graded beds versus the relative transport capacity of their assigned cyclones

Long-term sedimentation rates and relative transport capacity

The estimation of sedimentation rates of the graded beds in core SO 188 336 KL is based on three time markers: the ages of the sediment surface (August 2006) and that of the previous surface (February 1994) (Fig. 3). The third time mark is the rapid increase in 137Cs concentration at 17.3 m core depth, which indicates the onset of atmospheric fallout of radioactive material from atom bomb tests in 1954 (Fig. 12). The validity of the transport model was tested by the comparison of the sedimentation rates of the two dated core sections with the cumulative values of the relative transport capacity (Table 2). The pronounced threefold reduction of sedimentation rates from those in the period 1954–1994 to those from 1994 to 2006 is almost identical with the coeval changes of the estimated cyclonic transport capacity (Table 2). The sedimentation rate at the western foreset beds of the submarine delta, i.e., close to the SONG, also decreases by a factor of three (Palamenghi et al. 2011), which also demonstrates a relatively lower westward cyclonic transport toward the canyon.
Fig. 12

137Cs concentration versus depth in core SO 188 336 KL

Table 2

Sedimentation rates and modeled transport capabilities


Segment of core 336 KL

Sedimentation rate core 336 KL

Modeled relative transport capacity

Sedimentation rate of western foreset beds (Palamenghi et al. 2011)


0–174 cm

14.1 cm/year

14 volume/year

1.8 cm/year


174–1730 cm

38.9 cm/year

48 volume/year








Sediments from the shallow shelf can be mobilized and transported into canyons by different processes such as cascading dense water, tides, floods, storms, and cyclones. To what extend these processes transport sediment into the SONG is discussed below.

Dense shelf water cascading into the SONG

Cooling of surface water during winter can result in an intense flow of dense water from the shelf into canyons as observed in the Golf of Lions (Puig et al. 2008). The flows are channelized into canyons, and the resulting currents can erode and transport even coarse sand-sized particles. A similar process can be excluded for the shelf of Bangladesh because the high flux of freshwater from the Ganges-Brahmaputra produces a permanent low-salinity cover, which prevents a cooling and overturning of the water stratification on the shelf. The permanent anoxic condition in the SONG, as documented by the lack of bioturbation (Michels et al. 2003), indicates that stratification persisted throughout the last few centuries.

Sediment transport into the SONG by tides

The semidiurnal tidal currents continually produce clouds of turbid water in the tidal channels and along the coast of the Ganges-Brahmaputra delta. During the calm winter period, the turbid water generally extends up to the 10-m isobath, whereas during the high-discharge period of the summer monsoon, in combination with more wind-generated waves and turbulence, the plumes can extend up to the 80-m isobath. The weak residual tidal currents are generally directed to the southwest (Barua et al. 1994). Consequently, most suspended sediment probably settles out on the shallow inner shelf to form a meter-thick ephemeral mud layer (Kuehl et al. 1997). The graded beds in the SONG cannot have been deposited from such suspensions because the differential settling of the particles through the water column would have produced distinctly separate layers of sand, silt, and clay.

Sediment transport into the SONG by river floods

The Ganges and Brahmaputra transport most of the sediment during floods of the high-discharge summer monsoon season. A significant part of this is deposited onshore (Goodbred and Kuehl 2000; Kuehl et al. 1997; Lupker et al. 2013) and up to 30% in the submarine delta (Kuehl et al. 1997; Michels et al. 1998; Palamenghi et al. 2011; Suckow et al. 2001). Severe floods, which occasionally drown up to 70% of the country (Mirza 2002), disperse much of their sediment load in the large flood plains of the delta (Auerbach et al. 2015). Some of the sediment reaches the coast near the combined mouth of the two rivers and is mainly deposited along the estuarine islands lining barrier the northeastern Bengal coast (Allison 1998; Rahman et al. 2011; Sarwar and Woodroffe 2013; Wilson and Goodbred 2015). The sediment reaching the mouth of the rivers cannot be directly transferred to the SONG because it is located 130 km away. The numerous tidal channels ending north of the SONG, if anything, would be more effective transport pathways. However, most of the sediment seems to be effectively trapped in the flooded swamps of the Sundarban mangrove forests (Rogers et al. 2013).

Sediment transport into the SONG by storms and cyclones

In contrast to tides and river floods, storms and especially cyclones are much more effective in mobilizing and transporting sediment into the canyon. Especially waves generated by intense storms and cyclones produce highly turbulent, suspension-rich bottom water during their passage over the inner shelf (Bonnin et al. 2008; Dalyander and Butman 2015; Keen and Slingerland 1993; Miles et al. 2015; Palanques et al. 2006; Puig et al. 2004; Puig et al. 2014; Ross et al. 2009; Ulses et al. 2008; Zhang et al. 2016). According to numerous in situ monitoring studies, most suspended sediment is concentrated in the bottom water (Wright and Friedrichs 2006; Wright et al. 2001). In the case of the northern Bay of Bengal, the up to 150-km-wide, shallow (less than 30 m water depth) inner shelf forms an almost unlimited source for cyclone-induced resuspension of the fine sandy mud (Michels et al. 1998). Especially the ephemeral muds delivered by the monsoonal floods and tidal currents (Kuehl et al. 1989) constitute a reservoir from which suspension-rich bottom waters can be generated. In addition, erosion of the coastal mud islands by cyclonic waves mobilizes additional fine-grained sediments (Allison 1998; Hanebuth et al. 2015; Sarwar and Woodroffe 2013).

Wind-induced, tide-modified currents supported by wave action move the suspension-rich water masses on the shallow shelf in various directions depending on the position of a cyclonic gyre (Fig. 4) (Chang et al. 2001; Kim et al. 2008; Miles et al. 2015; Shay et al. 1989). Once these sediment-laden water masses which initially move along the gentle slope (0.4°) of the submarine delta reach the rim of the canyon, they descend down the steeper slopes (10–20°) of the SONG. In addition, some gravity flows may be channelized into the canyon via the small, up to 20-m-deep meandering gullies incised into the canyon head. However, the small size of their cross-sections (200–300 m) and their high sinuosity (6 meanders along a 9-km channel length covering a straight distance of 5.5 km) does not support a high-volume transport of sediment through these channels (Shahjahan 1970). Conversely, the intense cyclone Sidr, which passed directly over the SONG in 2007, destabilized the foreset sediments along the rim of the upper canyon and caused massive downslope displacements (Rogers and Goodbred 2010). These mass flows probably also generate suspension-rich density flows interfering with the cyclonic transport.

Deposition of graded beds

The stratified seismic facies of the canyon floor und its flanks consists of numerous graded bed sequences (Kottke et al. 2003; Michels et al. 2003). Graded beds are commonly related to turbidity currents and their occurrence in the SONG, which leads to the world’s largest deep-sea turbidite fan, seems to favor such an explanation. However, the graded beds in the SONG do not fit the classic turbidite model in which the coarsest, well-sorted, and cross-bedded sediment at the erosional bottom is followed by laminated silt sequences that fine upward into clay (Stow and Piper 1984). Instead, the graded beds in the canyon, especially the thicker ones, frequently start with inverse grading. In addition, grain-size sorting in the basal layers is extremely poor, even the bottom laminae consisting of fine sandy, clayey silt (Figs. 5 and 6).

Based on these characteristic signatures, it is assumed that the graded beds are deposited by gravity-driven, suspension-rich turbulent flows moving downslope in close contact with the bottom. The flows must have a large horizontal and vertical extent to cover the rough topography of the canyon floor. They must also be in contact with the bottom because settling from suspension would produce a much better separation of grain sizes. On the other hand, there are no signs of erosion at the bottom contacts of the graded beds. The “soft” gradual start of cyclonic input events may produce a thin sediment layer that protects the underlying fine-grained top sequences, and which may act as a fluid bed of reduced bottom friction and shear stress for the main body of the gravity flow (Parsons et al. 2009). The increasing sand content and/or the inverse grading at the base probably reflect the increasing impact of the cyclonic gyre during its approach on the inner shelf. High sand content layers reflect maximum sediment transport periods from the inner shelf to the canyon. The pronounced and generally fining upward trend after the sand maximum may indicate the waning period of the cyclonic supply and the final slow settling of the suspended fine silt and clay within the canyon. The lamination in the graded beds can be caused by internal eddy plume dynamics (Mullenbach and Nittrouer 2006) or tides. As the impact of most cyclones usually extends over a period of about 30 h, the grain-size variations within individual graded beds (Figs. 6 and 9) are probably caused by the semidiurnal tide, which modifies the transport into the canyon and thereby influences the depositional conditions of the graded beds. In addition, rhythmic deposition from the eddies of the bypassing sediment plumes might further modify the lamination.

Monitoring the variance of suspended sediment of wind-driven gravity flows in other canyons has shown similar patterns. Thus, the highest suspended sediment concentrations are reached after a few hours, whereas the subsequent decline in suspended sediment concentration usually takes much longer and could last for a day or more (Allison et al. 2005 for the Louisiana shelf; Palanques et al. 2008, 2012 for the canyons of the Golf of Lions; Puig et al. 2004 for the Eel Canyon; Symons et al. 2017; Xu et al. 2002 for the Monterey Canyon). Modeling indicates that most of the suspended sediment is concentrated at the bottom (Ulses et al. 2008), although the turbid layer can extend some tens of meters above the sea floor as observed on the Iberian Shelf (Chang et al. 2001) and in the Monterey Canyon (Symons et al. 2017).

Correlation of graded beds with cyclones

The two time series derived from the historical record of cyclones and the sedimentary sequence of graded beds have two common time markers (August 2006, February 1994). An additional time marker is set by the intense cyclone of May 1985, which is correlated to the thick graded bed at the core depth of 357 cm. The cyclone intensity during the 3 years prior to 1985 was low as suggested by the thin graded beds in the deeper core segment. About 85% of the core section can be attributed to the input of cyclones. Overall, six thin graded beds in the period 2006–1994 and some graded beds in the period 1994–1985 could not be assigned to the cyclone record. These beds were probably produced by local strong winds (not all tropical depressions are documented by IBTrACS), waves radiating from remotely passing cyclones, or extreme monsoonal floods.

In general, the properties of individual graded beds (total thickness, sand thickness, and average sand content) are positively correlated with the transport capacity. However, the individual properties of each graded bed show high scatter when plotted against the transport capacity (Fig. 11). Different processes of mobilization, transport, dispersal, and deposition, which are only partly considered in the transport model, may cause this variability. On the inner shelf, the amount of transported material depends on the variable extent of the low-salinity surface layer of the pre- or post-monsoonal periods, the variable thickness of the ephemeral sediment mud layer, tidal influences, etc. In the canyon, the different transport paths, the downslope velocity, and shifting depositional centers influence the final deposition and composition of a graded bed. The sand thickness in the graded beds, which increases by three orders of magnitude, shows the best correlation with transport capacity. Especially this last relation indicates that high-energetic cyclones transport coarser material into the canyon compared to weaker cyclones. The total thickness or the average sand content represents the finer grain-size fractions and seem to be influenced by depositional processes. In toto, these relations support the concept that cyclones are the main process of delivering sediment into the canyon.


From the results of this investigation, the following main conclusions can be drawn:
  • The laminated sediment of the SONG shelf canyon mainly consists of graded beds that are well characterized by high-resolution grain-size analyses.

  • The sequence of graded beds can be correlated with the historical record of cyclones by using the relative transport capacity of each cyclone. This approach is validated by identical relative changes in sedimentation rates and the estimated transport capacity for the two investigated time slices (2006–1994, 1994–1954).

  • The number of cyclones and the sedimentation rates decreased by a factor of about three over the last decade.

  • Sedimentation in the canyon off the Ganges-Brahmaputra delta is dominated by cyclonic sediment mobilization and lateral transport. Thus, the graded sequences in the canyon contain a reliable record of cyclones affecting the northern Bay of Bengal. This record is mainly produced by the optimal setup involving a huge sediment supply, an efficient sediment trap and ideal preservation in the anaerobic shelf canyon.

  • Transferring the results from the study area to other shelves suggests that cyclone activity is the main mechanism for sediment transfer from the shallow shelf to deeper water depocenters.



BM acknowledges the support by the German Science Foundation (DFG), the German Federal Environmental Foundation (DBU), and the German Academic Exchange Service (DAAD). IM was supported by CARIMA/BMBF. WZ is supported by the research program “Marine, Coastal and Polar Systems” (PACES II) of the Hermann von Helmholtz-Gemeinschaft Deutscher Forschungszentren e.V. Grain-size analyses were performed by BM at LIAG/Hannover. U. Röhl at MARUM/Bremen did the RFA scan. 137Cs concentrations were determined by the Bundesamt für Seeschifffahrt und Hydrographie/Hamburg. A former shorter version of the manuscript benefited from the comments of P. Puig, Institut de Ciencies del Mar, Barcelona/Spain, and J. P. Walsh, East Carolina, Greenville/USA. The present manuscript greatly profited from the suggestions of B. Flemming, Senckenberg am Meer, Wilhelmshaven/Germany, and two anonymous reviewers.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

367_2018_550_MOESM1_ESM.docx (47 kb)
ESM 1 (DOCX 47.4 kb)
367_2018_550_MOESM2_ESM.xlsx (105 kb)
ESM 2 (XLSX 105 kb)


  1. Alam M, Alam MM, Curray JR, Chowdhury MLR, Gani MR (2003) An overview of the sedimentary geology of the Bengal Basin in relation to the regional tectonic framework and basin-fill history. Sediment Geol 155:179–208CrossRefGoogle Scholar
  2. Allison MA (1998) Historical changes in the Ganges-Brahmaputra Delta front. J Coast Res 14:1269–1275Google Scholar
  3. Allison MA, Sheremet A, Goñi MA, Stone GW (2005) Storm layer deposition on the Mississippi–Atchafalaya subaqueous delta generated by hurricane Lili in 2002. Cont Shelf Res 25:2213–2232CrossRefGoogle Scholar
  4. Antony C, Testut L, Unnikrishnan AS (2014) Observing storm surges in the Bay of Bengal from satellite altimetry. Estuar Coast Shelf Sci 151:131–140CrossRefGoogle Scholar
  5. Auerbach LW, Goodbred Jr SL, Mondal DR, Wilson CA, Ahmed KR, Roy K, Steckler MS, Small C, Gilligan JM, Ackerly BA (2015) Flood risk of natural and embanked landscapes on the Ganges-Brahmaputra tidal delta plain. Nat Clim Chang 5:153–157CrossRefGoogle Scholar
  6. Barua DK (1990) Suspended sediment movement in the estuary of the Ganges-Brahmaputra-Meghna river system. Mar Geol 91:243–253CrossRefGoogle Scholar
  7. Barua DK, Kuehl SA, Miller RL, Moore WS (1994) Suspended sediment distribution and residual transport in the coastal ocean off the Ganges-Brahmaputra river mouth. Mar Geol 120:41–61CrossRefGoogle Scholar
  8. Berner U, Poggenburg J, Faber E, Quadfasel D, Frische A (2003) Methane in ocean waters of the Bay of Bengal: its sources and exchange with the atmosphere. Deep-Sea Res Part II: Topical Studies in Oceanography 50:925–950CrossRefGoogle Scholar
  9. Bonnin J, Heussner S, Calafat A, Fabres J, Palanques A, Durrieu de Madron X, Canals M, Puig P, Avril J, Delsaut N (2008) Comparison of horizontal and downward particle fluxes across canyons of the Gulf of Lions (NW Mediterranean): meteorological and hydrodynamical forcing. Cont Shelf Res 28:1957–1970CrossRefGoogle Scholar
  10. Carrasco CA, Landsea CW, Lin Y-L (2014) The influence of tropical cyclone size on its intensification. Weather Forecast 29:582–590CrossRefGoogle Scholar
  11. Carter RM, Larcombe P, Dye JE, Gagan MK, Johnson DP (2009) Long-shelf sediment transport and storm-bed formation by cyclone Winifred, central Great Barrier Reef, Australia. Mar Geol 267:101–113CrossRefGoogle Scholar
  12. Chang GC, Dickey TD, Williams AJ (2001) Sediment resuspension over a continental shelf during hurricanes Edouard and Hortense. J Geophys Res Oceans 106:9517–9531CrossRefGoogle Scholar
  13. Chang Y-C, Tseng R-S, Centurioni LR (2010) Typhoon-induced strong surface flows in the Taiwan strait and pacific. J Oceanogr 66:175–182CrossRefGoogle Scholar
  14. Dail MB, Reide Corbett D, Walsh JP (2007) Assessing the importance of tropical cyclones on continental margin sedimentation in the Mississippi delta region. Cont Shelf Res 27:1857–1874CrossRefGoogle Scholar
  15. Dalyander PS, Butman B (2015) Characteristics of storms driving wave-induced seafloor mobility on the U.S. East Coast continental shelf. Cont Shelf Res 104:1–14CrossRefGoogle Scholar
  16. Emanuel K (2011) Global warming effects on U.S. hurricane damage. Weather, Climate, and Society 3:261–268CrossRefGoogle Scholar
  17. Flood RP, BarrID WGJ, Roberson S, Russell MI, Meneely J, Orford JD (2018) Provenance and depositional variability of the thin mud facies in the lower Ganges-Brahmaputra delta, West Bengal Sundarbans, India. Mar Geol 395:198–218CrossRefGoogle Scholar
  18. French KL, Hein CJ, Haghipour N, Wacker L, Kudrass HR, Eglington TI, Galy V (2018) Millennial soil retention of terrestrial organic matter deposited in the Bengal Fan. Nature. Scientific Reports 8: 11997| DOI:
  19. Galy V, France-Lanord C, Beyssac O, Faure P, Kudrass H, Palhol F (2007) Efficient organic carbon burial in the Bengal fan sustained by the Himalayan erosional system. Nature 450:407–410CrossRefGoogle Scholar
  20. Girishkumar MS, Ravichandran M (2012) The influences of ENSO on tropical cyclone activity in the Bay of Bengal during October–December. J Geophys Res Oceans 117:C02033. CrossRefGoogle Scholar
  21. Goodbred JSL, Kuehl SA (2000) Enormous Ganges-Brahmaputra sediment discharge during strengthened early Holocene monsoon. Geology 28:1083–1086CrossRefGoogle Scholar
  22. Hanebuth TJJ, Kudrass HR, Linstädter J, Islam B, Zander AM (2013) Rapid coastal subsidence in the central Ganges-Brahmaputra Delta (Bangladesh) since the 17th century deduced from submerged salt-producing kilns. Geology 41:987–990CrossRefGoogle Scholar
  23. Hanebuth TJJ, Lantzsch H, Nizou J (2015) Mud depocenters on continental shelves—appearance, initiation times, and growth dynamics. Geo-Mar Lett 35:487–503. CrossRefGoogle Scholar
  24. Hoarau K, Bernard J, Chalonge L (2012) Intense tropical cyclone activities in the northern Indian Ocean. Int J Climatol 32:1935–1945CrossRefGoogle Scholar
  25. Keen TR, Slingerland RL (1993) Four storm-event beds and the tropical cyclones that produced them; a numerical hindcast. J Sediment Res 63:218–232Google Scholar
  26. Kim KO, Yamashita T, Choi BH (2008) Coupled process-based cyclone surge simulation for the Bay of Bengal. Ocean Model 25:132–143CrossRefGoogle Scholar
  27. Klotzbach PJ, Landsea CW (2015) Extremely intense hurricanes: revisiting Webster et al. (2005) after 10 years. J Clim 28:7621–7629CrossRefGoogle Scholar
  28. Knapp KR, Kruk MC, Levinson DH, Diamond HJ, Neumann CJ (2010) The international best track archive for climate stewardship (IBTrACS). Bull Am Meteorol Soc 91:363–376CrossRefGoogle Scholar
  29. Kottke B, Schwenk T, Breitzke M, Wiedicke M, Kudrass HR, Spiess V (2003) Acoustic facies and depositional processes in the upper submarine canyon Swatch of No Ground (Bay of Bengal). Deep-Sea Res Part II: Topical Studies in Oceanogr 50:979–1001CrossRefGoogle Scholar
  30. Krien Y, Testut L, Islam AKMS, Bertin X, Durand F, Mayet C, Tazkia AR, Becker M, Calmant S, Papa F, Ballu V, Shum CK, Khan ZH (2017) Towards improved storm surge models in the northern Bay of Bengal. Cont Shelf Res 135:58–73CrossRefGoogle Scholar
  31. Kudrass HR, Michels KH, Wiedicke M, Suckow A (1998) Cyclones and tides as feeders of a submarine canyon off Bangladesh. Geology 26:715–718CrossRefGoogle Scholar
  32. Kuehl SA, Allison M-A, Goodbred SL, Kudrass H (2005) The Ganges-Brahmaputra Delta In: Giosan L, Bhattacharya JP (Eds) River deltas–concepts, models, and examples. SEPM Spec Publ 83, pp 413–434Google Scholar
  33. Kuehl SA, Hariu TM, Moore WS (1989) Shelf sedimentation off the Ganges-Brahmaputra river system: evidence for sediment bypassing to the Bengal fan. Geology 17:1132–1135CrossRefGoogle Scholar
  34. Kuehl SA, Levy BM, Moore WS, Allison MA (1997) Subaqueous delta of the Ganges-Brahmaputra river system. Mar Geol 144:81–96CrossRefGoogle Scholar
  35. Lupker M, France-Lanord C, Galy V, Lavé J, Kudrass H (2013) Increasing chemical weathering in the Himalayan system since the Last Glacial Maximum. Earth Planet Sci Lett 365:243–252CrossRefGoogle Scholar
  36. Machalett B (2011) Past atmospheric circulation patterns and Aeolian dust dynamics recorded in Eurasian loess: utilizing high-resolution particle size analysis and amino acid geochronology. Mensch & Buch Verlag, Berlin 120 ppGoogle Scholar
  37. Machalett B (2017) WEBINAR: particle size characterization by laser diffraction analysis in geoscience and soil science - background, analyses, application, and interpretation. Beckman Coulter Life Sciences, Particle Size Research Letters 1Google Scholar
  38. Machalett B, Oches EA, Frechen M, Zöller L, Hambach U, Mavlyanova NG, Markovic SB, Endlicher W (2008) Aeolian dust dynamics in central Asia during the Pleistocene: driven by the long-term migration, seasonality, and permanency of the Asiatic polar front. Geochem Geophys Geosyst 9:Q08Q09. CrossRefGoogle Scholar
  39. Mairs HL, Koch SP, Gordon RB, Cuellar R (1992) The storm current response of Gulf of Mexico hurricanes. In: OnePetro (Eds) Offshore technology conference, 4–7 May 1992, Houston TX, OTC 6833, pp 235–241Google Scholar
  40. McPhaden MJ, Foltz GR, Lee T, Murty VSN, Ravichandran M, Vecchi GA, Vialard J, Wiggert JD, Yu L (2009) Ocean-atmosphere interactions during cyclone Nargis. Eos, Transact AGU 90:53–54CrossRefGoogle Scholar
  41. Michels KH, Kudrass HR, Hübscher C, Suckow A, Wiedicke M (1998) The submarine delta of the Ganges–Brahmaputra: cyclone-dominated sedimentation patterns. Mar Geol 149:133–154CrossRefGoogle Scholar
  42. Michels KH, Suckow A, Breitzke M, Kudrass HR, Kottke B (2003) Sediment transport in the shelf canyon “Swatch of No Ground” (Bay of Bengal). Deep-Sea Rese Part II: Topical Studies Oceanogr 50:1003–1022CrossRefGoogle Scholar
  43. Miles T, Seroka G, Kohut J, Schofield O, Glenn S (2015) Glider observations and modeling of sediment transport in hurricane Sandy. J Geophys Res Oceans 120:1771–1791CrossRefGoogle Scholar
  44. Mirza MMQ (2002) Global warming and changes in the probability of occurrence of floods in Bangladesh and implications. Glob Environ Chang 12:127–138CrossRefGoogle Scholar
  45. Mooley DA (1980) Severe cyclonic storms in the Bay of Bengal, 1877–1977. Mon Weather Rev 108:1647–1655CrossRefGoogle Scholar
  46. Mullenbach BL, Nittrouer CA (2006) Decadal record of sediment export to the deep sea via Eel Canyon. Cont Shelf Res 26:2157–2177CrossRefGoogle Scholar
  47. Open University (1989) Ocean circulation, 1st edn. Pergamon Press, OxfordGoogle Scholar
  48. Palamenghi L, Schwenk T, Spiess V, Kudrass HR (2011) Seismostratigraphic analysis with centennial to decadal time resolution of the sediment sink in the Ganges–Brahmaputra subaqueous delta. Cont Shelf Res 31:712–730CrossRefGoogle Scholar
  49. Palanques A, Durrieu de Madron X, Puig P, Fabres J, Guillén J, Calafat A, Canals M, Heussner S, Bonnin J (2006) Suspended sediment fluxes and transport processes in the Gulf of Lions submarine canyons. The role of storms and dense water cascading. Mar Geol 234:43–61CrossRefGoogle Scholar
  50. Palanques A, Guillén J, Puig P, Durrieu de Madron X (2008) Storm-driven shelf-to-canyon suspended sediment transport at the southwestern Gulf of Lions. Cont Shelf Res 28:1947–1956CrossRefGoogle Scholar
  51. Palanques A, Puig P, Durrieu de Madron X, Sanchez-Vidal A, Pasqual C, Martín J, Calafat A, Heussner S, Canals M (2012) Sediment transport to the deep canyons and open-slope of the western Gulf of Lions during the 2006 intense cascading and open-sea convection period. Prog Oceanogr 106:1–15CrossRefGoogle Scholar
  52. Parsons JD, Friedrichs CT, Traykovski PA, Mohrig D, Imran J, Syvitski JPM, Parker G, Puig P, Buttles JL, García MH (2009) The mechanics of marine sediment gravity flows. In: Nittrouer CA, Austin JA, Field ME, Kravitz JH, Syvitski JPM, Wiberg PL (eds) Continental margin sedimentation: from sediment transport to sequence stratigraphy. IAS Spec Publ, vol 37, pp 275–337Google Scholar
  53. Puig P, Ogston AS, Mullenbach BL, Nittrouer CA, Parsons JD, Sternberg RW (2004) Storm-induced sediment gravity flows at the head of the Eel submarine canyon, northern California margin. J Geophys Res Oceans 109:C03019. CrossRefGoogle Scholar
  54. Puig P, Palanques A, Martin J (2014) Contemporary sediment-transport processes in submarine canyons. Annu Rev Mar Sci 6:53–77CrossRefGoogle Scholar
  55. Puig P, Palanques A, Orange DL, Lastras G, Canals M (2008) Dense shelf water cascades and sedimentary furrow formation in the Cap de Creus Canyon, northwestern Mediterranean Sea. Cont Shelf Res 28:2017–2030CrossRefGoogle Scholar
  56. Rahman AF, Dragoni D, El-Masri B (2011) Response of the Sundarbans coastline to sea level rise and decreased sediment flow: a remote sensing assessment. Remote Sens Environ 115:3121–3128CrossRefGoogle Scholar
  57. Rogers KG, Goodbred JSL (2010) Mass failures associated with the passage of a large tropical cyclone over the Swatch of No Ground submarine canyon (Bay of Bengal). Geology 38:1051–1054CrossRefGoogle Scholar
  58. Rogers KG, Goodbred SL, Mondal DR (2013) Monsoon sedimentation on the ‘abandoned’ tide-influenced Ganges–Brahmaputra delta plain. Estuar Coast Shelf Sci 131:297–309CrossRefGoogle Scholar
  59. Ross CB, Gardner WD, Richardson MJ, Asper VL (2009) Currents and sediment transport in the Mississippi Canyon and effects of hurricane Georges. Cont Shelf Res 29:1384–1396CrossRefGoogle Scholar
  60. Sarwar MGM, Woodroffe CD (2013) Rates of shoreline change along the coast of Bangladesh. J Coast Conserv 17:515–526CrossRefGoogle Scholar
  61. Serno S, Winckler G, Anderson RF, Hayes CT, McGee D, Machalett B, Ren H, Straub SM, Gersonde R, Haug GH (2014) Eolian dust input to the subarctic North Pacific. Earth Planet Sci Lett 387:252–263CrossRefGoogle Scholar
  62. Shahjahan M (1970) Factors controlling the geometry of fluvial meanders. Int Ass Sci Hydrol Bull 15:13–24CrossRefGoogle Scholar
  63. Shanmugan G (2008) The constructive functions of tropical cyclones and tsunamis during deep-water sand deposition during sea level highstand. AAPG Bull 92:443–471CrossRefGoogle Scholar
  64. Shay LK, Elsberry RL, Black PG (1989) Vertical structure of the ocean current response to a hurricane. J Phys Oceanogr 19:649–669CrossRefGoogle Scholar
  65. Stow DAV, Piper DJW (1984) Deep-water fine-grained sediments: facies models. Geol Soc Lond Spec Publ 15:611–646CrossRefGoogle Scholar
  66. Suckow A, Morgenstern U, Kudrass H-R (2001) Absolute dating of recent sediments in the cyclone-influenced shelf area off Bangladesh: comparison of gamma spectrometric (137Cs, 210Pb, 228Ra), radiocarbon, and 32Si ages. Radiocarbon 43:917–927CrossRefGoogle Scholar
  67. Symons WO, Sumner EJ, Paull CK, Cartigny MJB, Xu JP, Maier KL, Lorenson TD, Talling PJ (2017) A new model for turbidity current behavior based on integration of flow monitoring and precision coring in a submarine canyon. Geology 45:367–370CrossRefGoogle Scholar
  68. Ulses C, Estournel C, Puig P, Durrieu de Madron X, Marsaleix P (2008) Dense shelf water cascading in the northwestern Mediterranean during the cold winter 2005: quantification of the export through the Gulf of Lion and the Catalan margin. Geophys Res Lett 35:L07610. CrossRefGoogle Scholar
  69. Walsh JP, Nittrouer CA (2009) Understanding fine-grained river-sediment dispersal on continental margins. Mar Geol 263:34–45CrossRefGoogle Scholar
  70. Warner JC, Sherwood CR, Signell RP, Harris CK, Arango HG (2008) Development of a three-dimensional, regional, coupled wave, current, and sediment-transport model. Comput Geosci 34:1284–1306CrossRefGoogle Scholar
  71. Webster PJ, Holland GJ, Curry JA, Chang H-R (2005) Changes in tropical cyclone number, duration, and intensity in a warming environment. Science 309:1844–1846CrossRefGoogle Scholar
  72. Wilson CA, Goodbred SLJ (2015) Construction and maintenance of the Ganges-Brahmaputra-Meghna Delta: linking process, morphology, and stratigraphy. Annu Rev Mar Sci 7:67–88CrossRefGoogle Scholar
  73. Wright LD, Friedrichs CT (2006) Gravity-driven sediment transport on continental shelves: a status report. Cont Shelf Res 26:2092–2107CrossRefGoogle Scholar
  74. Wright LD, Friedrichs CT, Kim SC, Scully ME (2001) Effects of ambient currents and waves on gravity-driven sediment transport on continental shelves. Mar Geol 175:25–45CrossRefGoogle Scholar
  75. Xu JP, Noble M, Eittreim SL, Rosenfeld LK, Schwing FB, Pilskaln CH (2002) Distribution and transport of suspended particulate matter in Monterey Canyon, California. Mar Geol 181:215–234CrossRefGoogle Scholar
  76. Zhang W, Cui Y, Santos AI, Hanebuth TJJ (2016) Storm-driven bottom sediment transport on a high-energy narrow shelf (NW Iberia) and development of mud depocenters. J Geophys Res Oceans 121:5751–5772CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.MARUMBremenGermany
  2. 2.Department of GeosciencesUniversity of MassachusettsAmherstUSA
  3. 3.Department of Natural and Applied SciencesBentley UniversityWalthamUSA
  4. 4.Institute of Geography, ClimatologyHumboldt-Universität zu BerlinBerlinGermany
  5. 5.Renard Centre of Marine GeologyUniversity of GentGhentBelgium
  6. 6.Institute of Coastal ResearchGeesthachtGermany

Personalised recommendations