This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy Deep-Sea Research I 57 (2010) 1304–1313
Contents lists available at ScienceDirect
Deep-Sea Research I journal homepage: www.elsevier.com/locate/dsri
Marine snow originating from appendicularian houses: Age-dependent settling characteristics Fabien Lombard n, Thomas Kiørboe Technical University of Denmark, National Institute for Aquatic Resources, Oceanography Section, Kavalerg˚ arden 6, DK-2920 Charlottenlund, Denmark
a r t i c l e in f o
a b s t r a c t
Article history: Received 2 December 2009 Received in revised form 2 June 2010 Accepted 7 June 2010 Available online 20 June 2010
The evolution of size, sinking velocity, and dry weight of aging discarded appendicularian houses, a component of marine snow, were examined in laboratory experiments. The sizes of discarded houses decrease over time, with a rapid deflation during the first hour, followed by a slower rate of compression leading to a total of 60% and 87% decrease in diameter after 1 h and 5 d, respectively. The initial rapid deflation of the houses is accompanied by a massive loss of its particle content and a 10– 63% loss in weight. The initial weight loss is left as a trail of elevated particle and solute concentration in the wake of the sinking house. Subsequently the house weight decreases at a much lower rate that is consistent with bacterial degradation. The combined effect of weight losses and deflation–compression process is an increase in the sinking speed of the houses, by a factor of 1.7–6 after 1.5–3 d. These processes can provide a new insight on the sinking dynamic and flux of appendicularian produced marine snow from in situ observations. We applied our laboratory derived rates to field data from the East Atlantic Ocean and estimate that large (2000–4000 mm) houses account for about 1/3 of the 300–500 mm particles in the upper 100 m and loose 30% of their mass before leaving the upper 200 m. The observed deflation–compression process may have several consequences on the dynamics of appendicularian-derived marine snow particles. First, it may explain field observations that marine snow sinking velocities increase with depth. Second, an initial rapid loss of weight and particles will decrease the potential vertical flux of particulate carbon due to appendicularians. And finally, the trail of particles and solutes may guide zooplankton to the sinking house, and further increase its degradation due to grazing by detrivorous organisms. & 2010 Elsevier Ltd. All rights reserved.
Keywords: Marine snow Appendicularians Sinking speed Particle size Particles weight
1. Introduction Large marine snow particles ( 4500 mm) are abundant in pelagic ecosystems and account for a major fraction of the downward flux of particulate organic carbon (Alldredge and Silver, 1988; Kiørboe, 2001; Turner, 2002). The sedimentation speed of marine snow depends on particle characteristics, including the sizes and shapes of particles, their origin and composition, their apparent density and porosity, and on the inclusion of lithogenic or calcite material that act as a ballast (Francois et al., 2002; Klaas and Archer, 2002; De La Rocha and Passow, 2007). Studies on marine snow have mainly concentrated on three aspects: field observations with in situ imaging devices or sediment traps to quantify particle abundances and fluxes (Honjo et al., 1984; Guidi et al., 2008; Honjo et al., 2008), modelling studies to examine the effects of different processes, such as coagulation, degradation or disagregation, on marine
n
Corresponding author. Tel.: +45 29 71 12 79; fax: +45 33 96 34 34. E-mail address: fl
[email protected] (F. Lombard).
0967-0637/$ - see front matter & 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr.2010.06.008
snow dynamics and flux (Stemmann et al., 2004b; Burd and Jackson, 2009), and direct observations of marine snow, either in situ or in the laboratory on individual particles, artificially produced or brought in from the field (Alldredge and Gotschalk, 1988; Alldredge, 2000; Ploug et al., 2008). Imaging devices lead to snapshot observations of the water column, sediment traps to integrated fluxes over time for a limited range of depths, whereas studies using in situ collected marine snow mostly concentrate on particle properties near the surface. Such studies provide only limited insight in the modifications of individual particle during aging and sedimentation. Such changes, if significant, may have major implications to the dynamics and flux of particles in the ocean. Marine snow sedimentation velocities appear to be smaller near the surface than below the euphotic zone (Syvitski et al., 1995; Guidi et al., 2008) and particles collected in situ, but observed 4 h latter in laboratory conditions have higher settling speeds than when directly observed in situ (Alldredge and Gotschalk, 1988). Such observations suggest that particle properties have changed with age during the settling process. It has been hypothesized that particles may compact during sedimentation,
Author's personal copy F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
decreasing their porosity and increasing their density and sinking velocity (Logan and Kilps, 1995; Guidi et al., 2008). Similar increase of sedimentation speed have been observed from sediment trap sampling (Berelson, 2002; Fischer and Karakas, 2009), but it remains unclear if this change originates from a change in the size spectra of particles or from changes in the physical properties of the particles. It is well documented that marine snow are continuously degraded by microorganisms (Ploug and Grossart, 2000) and zooplankton grazing (Kiørboe, 2000), but how such degradation modifies the sinking characteristics is unknown and not trivial. Examining temporal changes in the characteristics of marine snow particles is nearly impossible in the field and complicated in laboratory, due to the difficulty of producing those in a realistic way in the laboratory (Logan and Kilps, 1995) or because field-collected particles may be modified by their collection and handling (Alldredge and Gotschalk, 1988). Marine snow originates from a large variety processes and primary particles. Appendicularians are a major producer of marine snow through the secretion and discarding of houses (Alldredge and Silver, 1988) and may contribute from 12% up to 83% of the total particulate organic carbon export in various sites from polar-euphotic to tropical-oligotrophic systems (Alldredge, 2005). Houses are external mucous filter used by appendicularians to collect food particles (Flood and Deibel, 1998) and are discarded and renewed up to 27 times a day (Sato et al., 2001, 2003). The discarded houses include concentrated non-ingested particles such as bacteria, phytoplankton, and lithogenic material. Many direct observations of marine snow, both in the field and in the laboratory, have been made on discarded appendicularians houses (Alldredge and Gotschalk, 1988; Alldredge and Gotschalk, 1990; Alldredge, 2000). Appendicularians, especially Oikopleura dioica, can easily be cultivated (Fenaux and Gorsky, 1985; Bouquet et al., 2009; Lombard et al., 2009a), which allow fast production of a large number of discarded houses with physical properties that must be very similar to those produced in the ocean. In this study, we used appendicularian houses, produced from a laboratory culture, to examine how the size, dry weight, density, and sinking speed of discarded appendicularian houses change as they age. We show that despite significant loss of weight and size, the sinking speed of discarded houses increase significantly as they age.
2. Materials and methods 2.1. Appendicularian collection and culture Appendicularians, Oikopleura dioica, were collected in the Øresund (Denmark) during September 2008 with a 100-mm mesh size plankton net with a large (30-L), non-filtrating cod-end. Living appendicularians, actively filtering inside their houses, were separated from the plankton using wide mouth pipettes (7 mm diameter), isolated in 40-mm filtered seawater, and cultivated as described in Lombard et al. (2005). A mixture of Thalassosira pseudonana, Isochrysis galbala and Rhodomonas salina with final equal cell concentration was used as food. The culture temperature was 16 1C. One hour before experiments the animals to be used were fed with an algal mixture at a concentration of 15,000 cells ml 1 which is representative for meso- to eutrophic conditions. This ‘stains’ the house with particles. 2.2. Experimental setup Two types of experiments were designed to (i) observe the size change over time of discarded appendicularians houses and (ii) to
1305
investigate the evolution of settling speed. The first experiment consisted of repeated measurements under a dissecting microscope of appendicularian houses. Because observations under the dissecting microscope lead to a slight warming of the houses that biases settling speed, the temporal evolution of sedimentation speed of discarded houses was recorded in the separate, second experiment. Size was also monitored during sedimentation observations, but with less accuracy. To ensure constant salinity and temperature in all experiments, seawater for experiments was collected in the appendicularian culture beakers 12 h prior to the experiment and stored in a closed glass jar in the thermostated room (20 1C), where experiments were conducted. Both experiments made repeated observations on the same houses for up to 6 d after they had been discarded. Between observation, houses were stored in closed bottles (500 mL) on a rotating table with a rotation speed assuring minimal contacts of the house with the bottles walls. Houses were gently transferred with wide mouth pipettes that ensured minimal physical disturbance. For the first experiment (size observations), houses occupied by appendicularians were gently transferred to a 50 mm diameter Petri dish filled with seawater at the observation room temperature. Measurements of the house diameter and appendicularian trunk length (excluding gonads) were done and the animal was then provoked to discard the house by slightly vibrating the Petri dish. Repeated house diameter observations were done during the following time at: 2, 5, 10, 15, 20, 30 min, 1, 2, 4, 5 h, 1, 2, 4 and 5 d. The house was placed in the incubation bottle 5 min after being discarded and immediately after each repeated following observation. Size measurements were made in a Leica Mz6 dissection microscope equipped with a uEyes UI-1540-C camera. Images were analysed using UTHSCSA Image Tool software. We recorded size-change observations in 17 individual houses ranging in diameter between 350 and 6340 mm. The 2nd experiment, settling observations were conducted in a 500 mL transparent Plexiglas chamber (dimensions 5 5 20 cm). The sinking house was recorded using a b/w CCD camera (Watex, WAT-535EX, 25 Hz frame rate) equipped with a 35 mm lens. Houses were illuminated using a red (650 nm wavelength) laser sheet that was shined through the settling house perpendicular to viewing direction of the camera. This illumination allowed us also to see the small algal particles (5–10 mm) in the water. Both the camera and the laser beam were mounted on a motorized support, which allows moving the whole system in two dimensions without modifying the setting of the camera, which was kept constant during the entire experiment. Each experiment was initiated by transferring an appendicularian from the culture to a Petri dish, where its trunk length (excluding gonads) and house diameter were measured in the dissection microscope. The animal was left until it discarded its house and the freshly discarded house was then gently placed in the surface and center of the settling chamber. The house was followed by the video camera until the house reached the bottom (approximately 10–20 min depending on the house). Sedimentation of the house was repeated two times within the first hour and several measurements were done at different time intervals (e.g. for house #6: 0.5, 1, 2, 3, 10, 15, 20, 35, 47 and 60 min) and the house was then transferred to the 500 ml bottle. Repeated observations on the same houses were done at increasing time intervals for up till 3 d. All houses that lost their typical spherical shape (mostly when accidentally aggregated with fiber) or hit the bottom of the rotating bottles were removed from the experiment. Altogether, temporal development of settling velocities were examined in nine houses with initial diameters ranging 1200–7800 mm. Due to the difficulty of locating the houses in the observation chamber after each sedimentations, without creating physical stress, the
Author's personal copy 1306
F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
sinking experiment was not feasible on houses smaller than 1000 mm.
to water and the dry weight of the house was calculated by subtracting the weight of water occupying the entire house volume from the estimated wet weight of the house.
2.3. Image proceeding and calculations Settling speeds of the discarded houses were estimated from video-sequences recorded when the house was in the middle of the settling chamber and over periods that corresponded to the time taken to sink ca. 20 house diameters. Settling speeds were corrected for ambient fluid motions as described below. The selected video sequence was transferred to the computer. The center of the discarded house was calculated and 20–40 small particles ( 10 mm) located in the field of view, but not in the vicinity of the house as well as the house itself was tracked on the successive images. The vertical settling velocity of each particle was calculated from their change of position over time. The vertical velocities of the small particles, which we assume to be neutrally buoyant and to faithfully follow water motions, were used to estimate the background water movement at the position of the house, which was then removed from the settling speed of the discarded house. House settling speeds were 7–100 times larger than the ambient fluid velocities. House diameters were also measured on the video. All the following calculations assumed that appendicularian houses are spherical. Simultaneous measurements of the diameter of houses of different ages showed that depending on the position, the measured diameter can vary by 4% (SD). If considering houses as ellipsoids, their axial ratio was o1.2 and their dynamical shape factor ko1.01 (Davies 1979, Table 7), which means that houses are close to be spheres (k¼1) and that the error in estimating Reynolds number from settling speed should be less than 1%. The Reynolds numbers (Re) of the settling houses were calculated as Re ¼
rl dv m
ð1Þ
where rl is the density (1.024 g cm 3) and m the viscosity (0.01085 g cm 1 s 1) of the seawater, both computed from observed temperature and salinity, and d the diameter of the house. In our study, Re was between 0.3 and 6, meaning that Stokes’ law does not properly describe the settling speed. Equating the gravitational and drag forces yields (Newton– Rittinger equation) 4 rs rl v2 ¼ dg ð2Þ 3CD rl where rs is the density of the particle (g cm 3), g the gravitational acceleration (981 cm s 2) and CD the drag coefficient. Appendicularian houses are complex structures, including numerous chambers surrounded by impermeable walls and the water can only enter in the house through two small apertures located on the same side (Fenaux, 1986; Flood and Deibel, 1998). Thus, even though appendicularian houses are very porous, they are not permeable, i.e., water does not flow through them as they sink. We therefore assume a drag coefficient as for a solid sphere. The drag coefficient for a sphere at Reo100 can be approximated by (Cheng, 2009): 24 CD ¼ ð1 þ0:27ReÞ0:43 Re
3. Results When the house was still occupied by the appendicularian, its diameter was directly proportional to the appendicularian trunk length (Fig. 1A). Immediately after being discarded, house sinking velocities ranged 10–64 m d 1 for 1.2–7.8 mm diameter houses, and settling velocity was proportional to house diameter (Fig. 1B). The dry weight of newly discarded houses (estimated from their settling velocity) increases as a power function of its diameter (Fig. 1C). The exponent of the relationship between diameter and weight of a particle is its fractal dimension (D; Logan and Kilps, 1995; Guidi et al., 2008). Then D of newly discarded houses is close to 2.4 (Fig. 1C, Table 1). Fig. 2 shows an example of the changes observed over time of a discarded house (house no. 6; Table 1). Immediately after being discarded, the house quickly deflates and shrinks from an initial size of 5.30 mm when still occupied by the appendicularian to 4.16 mm just after being discarded, 1.25 mm after 1 h, and 0.66–0.69 mm after 3 d of shrinking (Fig. 2A). The first fast decrease in size is partly due to the loss of internal pressure in the house, no longer maintained by the beating tail of the appendicularian. Together with the shrinking, the sinking speed of the house increases over time, from an initial sinking speed around 40 m d 1 just after being discarded, to 72 m d 1 1 h later, and up to 92–109 m d 1 after 3 d (Fig. 2B). This increase of sinking velocity implies an increase in the average density of the house (Fig. 2C). At the same time, the dry weight of the house decreases over time (Fig. 2D). When plotted on semi-log axis, the slopes of the linear regressions between weight and time are estimates of the specific weight loss rates. Two phases can be distinguished: a fast, initial weight loss occurring during the first hour with a loss rate of 5 d 1, and a slower weight loss rate, about 10 2 d 1, during the remaining of the observation period. Two such distinct phases were observed in 6 of the 9 houses, with initial loss rates of 0.9–7.3 d 1 (mean 4.41 d 1), followed by slower weight loss with rates between 0.01 and 0.18 d 1 (mean 0.06 d 1). In these six observations, the discarded houses lost between 10% and 63% of their initial weight during the first hour. The loss of weight resulted in a visible plume of elevated concentration of small particles in the wake of the sinking house (Fig. 3). The pattern of temporal changes observed in house #6 (Fig. 2) was found also in all the other houses examined (Figs. 4 and 5, Table 1). All the houses deflate in a similar way independent of their initial size (Fig. 4A and B) and lost on an average 60% and 87% of their diameter and 93% and 99.7% of their volume after 1 h and 5 d, respectively. Their sinking speed increased significantly over time (Fig. 5A, Table 1), by a factor of 1.7–6 (average 3) after 1.5–3 d. Despite the high variability between different houses, a general relationship relating the sinking speed V(t) (m d 1) normalized by the initial sinking speed (Vi) as a function of age (t, days) could be expressed in the form of VðtÞ=Vi ¼ 2:059t 0:1 ðR2 ¼ 0:26Þ
ð4Þ
ð3Þ
Assuming that discarded houses are spherical and using the observed settling speed (v), Eqs. (1)–(3) can be used to estimate the average density (rs) and the wet weight of the observed house. As marine snow particles have high porosity, we assume that the volume occupied by solid material is negligible compared
The loss of dry weight over time was similarly significant in all but two cases (Fig. 5B, Table 1). When combining the observations in age intervals, the relationship between the size of the house and its settling speed (Fig. 5C, Table 1) illustrates the combined changes in shrinking and increase of settling speed over time. For example, a 1000 mm sized particle originating from a discarded house will sink at
Author's personal copy F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
Using similar age intervals, the powers of the size-dry weight relations (the fractal dimension, D) decreased significantly as the houses aged (Fig. 5E), from an initial D around 2.4 to a value of 1.6 after 1 d. The last data point (age 1.5–3 d) was not considered because the reduced house diameter range may have lead to a biased estimation of the fractal dimension. In summary, the compaction and weight loss of discarded appendicularian houses over time lead to smaller, denser, and faster sinking marine snow particles of lower fractal dimension.
10000
House diameter (m)
8000
6000
4000
2000
4. Discussion and conclusions
0 0
500 1000 Appendicularian trunk length (m)
1500
80 Alldredge (2005) Present observations
Sinking velocity (m d -1)
70 60 50 40 30 20 10 0 0
2000
4000 6000 House diameter (m)
8000
10000
101
Dry weight (g)
1307
100
10-1 103
104 House diameter (m)
Fig. 1. House diameter (d, mm) as a function of the trunk length (TL, mm; excluding gonads) of the appendicularian (A), sinking velocity (v, m d 1) of newly discarded houses as a function of house diameter (B) and dry weight (Dw, mg) of newly discarded houses as a function of its diameter (C). All observations made just before (A) or immediately after the house had been discarded (to 1 min) (B, C). In (B), the data from Alldredge (2005) were added for comparison. The fitted regression lines are: A: d ¼7.22 TL 595; R2 ¼ 0.96; B: v¼ 0.0064 d1.036; R2 ¼ 0.73; C: Dw ¼4.7 10 9 d2.38; R2 ¼0.81.
around 15 m d 1 if it has just been produced, 50–60 m d 1 if the house is 30 min–6 h old and up to 80–160 m d 1 if the house was aged from 1 to 1.5–3 d.
Our study is the first to quantify changes in size and sedimentation velocity over time of marine snow produced by appendicularians (discarded houses). Similar process was observed, but not quantified, on bathypelagic giant appendicularians (Hamner and Robison, 1992), and different studies have observed sedimentation of epipelagic appendicularians houses (Gorsky et al., 1984; Hansen et al., 1996; Alldredge, 2005). The sedimentation velocity of freshly discarded houses observed in situ is comparable to those that we observed (Fig. 1, Alldredge and Gotschalk, 1988; Alldredge, 2005). For 400–800 mm trunk length sized, O. dioica at 16–19 1C, Gorsky et al. (1984) observed a 50–120 m d 1 settling velocity. Similarly, houses produced by 680 mm trunk length O. dioica had sedimentation velocities between 69 and 138 m d 1 when observed under various conditions, including houses aged for 21 h on a rotating table (Hansen et al., 1996). These latter values are slightly higher than our observation for similarly sized appendicularians (20–50 m d 1). The differences observed may be due to both methodological differences and in the age of the examined houses. First, both Gorsky et al. (1984) and Hansen et al. (1996) ignored possible convection during the measurements, which may lead to either a decrease or increase of the estimated sedimentation velocity. In our case, the variability due to the convection was equal to 720 m d 1 and would have resulted in an average 7 m d 1 overestimation of the sinking velocity, if it had not been corrected for. Only studies that correct for convection effect, using a spot of neutrally buoyant dye, to serve as a reference point, to measure the sedimentation velocity, came up with measurement similar to ours (Alldredge and Gotschalk, 1988; Alldredge, 2005). Secondly, the field collected houses observed in Hansen et al. (1996) were of unknown age. Houses produced by similarly sized appendicularians (600–800 mm) in our experiments had sedimentation velocities between 50 and 100 m d 1 when aged between 30 min and 3 h and between 80 and 115 m d 1 after 1 d (Fig. 5C), which correspond to the results of Hansen et al. (1996). Different factors can influence the sedimentation velocity of appendicularian houses. When food concentration increases, the ingestion efficiency decreases and the quantity of particles ˜ a and Kiefer, trapped in the discarded houses increases (Acun 2000; Sato et al., 2005; Lombard et al., 2009b). Thus, the density of houses produced under high particles concentration conditions may be higher leading to higher sedimentation velocities. The nature of the filtered particles may also have an impact on sedimentation velocities and the inclusion of ballast materials such as coccolithophorid shells, opal originating from diatoms or lithogenic material may increase the houses’ sedimentation velocities (Francois et al., 2002; Klaas and Archer, 2002; De La Rocha and Passow, 2007; Ploug et al., 2008). However, our observations suggest that the age of the discarded houses is a major determinant of the sinking velocity and accounts for almost an order of magnitude variation, in sinking velocity of houses, of a given size (Fig. 5C).
Author's personal copy 1308
F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
Table 1 Parameters for the relationships between age and house sinking speed (Fig. 5A) or mass (Fig. 5B), and between house diameter and settling speed (Fig. 5C) or dry weight (Fig. 5D). All relationships are expressed in the form Y¼ aXb. Parameter b have been tested as significantly different between the different regressions (covariance analysis on log–log transformed data; F8,92 ¼3.68 p ¼0.0008 for Fig. 5A, F8,92 ¼2.65 p ¼0.012 for Fig. 5B, F7,107 ¼2.41 p ¼0.024 for Fig. 5C) only for house diameter vs. mass relationships b is not significantly different between the different age intervals (F7,107 ¼1.98; p ¼ 0.063), whereas a is significantly different (F7,114 ¼ 29.92; po 0.0001). House no.
1 2 3 4 5 6 7 8 9
Initial house size (mm)
Time vs. settling rate (Fig. 5A)
3009 3655 2257 5181 2924 5302 3200 1200 7852
Time since discarding
Time vs. dry weight (Fig. 5B) 2
a
b
r
a
b
r2
56.23 119.12 50.58 86.7 81.28 114.82 63.1 16.22 108.89
0.28 0.09 0.16 0.08 0.08 0.12 0.19 0.09 0.09
0.73 0.29 0.52 0.9 0.58 0.9 0.84 0.3 0.74
0.51 2.17 0.52 1.4 0.8 0.97 0.32 0.06 4.46
0.087 0.178 0.008 0.121 0.104 0.093 0.009 0.049 0.075
0.32 0.73 0 0.77 0.77 0.73 0.01 0.11 0.49
House diameter vs. settling rate (Fig. 5C) a
0 0–5 min 5–15 min 15–30 min 30 min–3 h 3–6 h 0.9–1.5 d 1.5–3 d
0.007 0.007 0.03 0.026 1.574 0.687 3.279 0.055
b
r
a
b
r2
1.02 1.08 0.89 0.97 0.47 0.61 0.46 1.16
0.72 0.82 0.57 0.56 0.43 0.46 0.36 0.6
4.91 10 09 4.12 10 08 1.75 10 07 5.81 10 07 4.53 10 06 6.47 10 06 2.17 10 05 3.33 10 07
2.36 2.18 2.02 1.93 1.68 1.66 1.57 2.28 Mean 1.85
0.82 0.93 0.95 0.87 0.88 0.84 0.85 0.83
140 Sinking velocity (m d-1)
4000 3000 2000 1000
120 100 80 60 40
0
20
1.031
1.2
1.03
1 log (Dry weight (g))
House diameter (m)
5000
1.029 Density
House diameter vs. dry weight (Fig. 5D) 2
1.028 1.027 1.026 1.025
0.8 0.6 0.4 0.2 0
1.024
-0.2 0
1
2 Time (days)
3
4
0
1
2 Time (days)
3
4
Fig. 2. Temporal changes in diameter (A), sinking velocity (B), density (C) and dry weight (D) of an example appendicularian house (house #6, Table 1). The fitted lines in A–C correspond to the best power adjustment (Table 1). In (D) semi-log regression has been fitted to the apparent 2-phases of the weight loss process.
The dry weight of discarded appendicularian houses varies with of age and size of the houses (Fig. 5), but also between studies. Thus, dry weights of field collected appendicularian houses reported by Alldredge and Gotschalk (1988) are up to an
order of magnitude higher than those reported here. We do not know whether this difference is real (due to different origin of houses) or caused by methodological differences, but it does not change the qualitative observation of a weight decrease as well as
Author's personal copy F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
Fig. 3. Plume of particles trailing after a sinking appendicularian house. Such trails can be observed during the first hour after the house has been discarded.
Particle diameter (m)
104
103
102 10-4
10-2
100
100
//
80 Percent of diameter lost
102
//
60
40
20
0 0.0
// 0.2 0.7 4.0 Time after discarding (day)
7.4
Fig. 4. Change in appendicularian houses diameter (A) and percent of diameter lost (B) as a function of age. The regression line in B relates house diameter (d, in mm) to time (t, in day) by the following relationship: d(t) ¼ d(t0)0.188t 0.235, R2 ¼0.86.
the separation in two distinct phases. We assumed in our calculations that appendicularian houses are impermeable to flow. However, if the houses indeed are permeable, this will lead to higher weight estimates. Assuming a hypothetical permeability
1309
of 1/100, representing the area ratio between inlet filters surface to total house surface, the drag force of the house can be calculated according to Michaelides (2006). This leads to a slight increase (2–15%) of the estimated weight of the houses, so this cannot explain the discrepancy. Our results demonstrate that the density, size, and sedimentation velocity of marine snow particles produced by appendicularians change over time (Figs. 2, 4 and 5). This process can be separated in two phases, an initial rapid deflation process with large weight loss that lasts for 1 h and a subsequent phase with much slower compression and weight loss rates. While the initial rapid deflation is probably specific to discarded houses, the subsequent decline in size and sinking characteristics may be more representative for other kinds of marine snow. This deflation-compaction-weight-loss process has several consequences. First of all, the deflation–compression causes a change in settling characteristics. The discarded houses evolve in a couple of days from large and slowly sinking particles to small and fast sinking ones. This is contrary to the usual admitted concept that large particles settle faster than small ones. This remains true for discarded appendicularian houses within age classes (Fig. 5C), but if the age is not taken in account, any relation between size and sinking velocity disappear in our data, because small, old houses may sink faster than larger recently discarded ones. The deflation–compression process described here may explain the differences in the size-sinking speed relationships reported in the literature for marine snow collected at different depths (Fig. 5F). The newly discarded houses have settling velocities corresponding to sinking speeds of marine snow particles observed in surface water (Alldredge and Gotschalk, 1988; Alldredge and Gotschalk, 1989); after a few hours their settling speeds correspond to deeper estimates obtained between 70 and 100–1000 m (Syvitski et al., 1995; Guidi et al., 2008) and after 1 d even exceed the highest observed in the field. Our observations, however, remain within the theoretical limits defined by the density of primary particles and the fractal dimension of the aggregate (Stemmann et al., 2004b). This does not mean that all types of the marine snow behave in a similar way. Especially, the rapid deflation during the first hour is specific to appendicularian-derived marine snow. However, appendicularian houses are important constituents of marine snow (Alldredge and Silver, 1988) and their change during sedimentation may significantly modify the mean sinking speed of marine snow observed at different depths in the ocean. The compactness and fractal dimension of diatom flocks depend on the physical characteristic of the environment, in which they were formed (Logan and Kilps, 1995) and may similarly change during sinking and aggregation. Our results, in conjunction with the observed increase of particles sinking speed with depth (Berelson, 2002; Fischer and Karakas, 2009), suggest that compaction of marine snow particles is significant in the ocean and highlights a need for studies of the temporal modifications of other types of marine snow particles as they sink and degrade. The temporal changes in the size-dependent characteristics of aggregates are conveniently expressed in their fractal dimension, which is consequently a very useful parameter in modelling the dynamics (formation, sinking, and degradation) of marine snow particles. In lack of knowledge about how particles change as they age, modellers have assumed a time-independent fractal dimension (Stemmann et al., 2004a, 2004b). Such an assumption appears not to be warranted, at least not for marine snow formed from discarded appendicularian houses, where the fractal dimension decreases significantly during the first 24 h after their formation (Fig. 5E). The range of values that we observed for D in particles formed from appendicularian houses, 1.6–2.4, is
Author's personal copy 1310
F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
101
Dry weight (g)
150
100
0 10-4
200 Sinking velocity (m d-1)
100
10-1
50
10-2 100 Time (days)
100
10-2 100 Time (days)
102
103 House diameter (m)
104
101
t = 0 min 0-5min 5-15min 15-30min 30min-3h 3-6h 0.9-1.5d 1.5-3d
150
10-2 10-4
102
Dry weight (g)
Sinking velocity (m d-1)
200
50
100
10-1
0 102
103 House diameter (m)
104
102
104
t=0 0.9-1.5d 1.5-3d
Settling speed (m d-1)
Fractal dimension
2.6 2.4 2.2 2 1.8 1.6
6
1 5
102
4 3 2
100
1.4 1.2
10-2 10-2
100 Time (day)
100
101
102 103 Diameter (m)
104
105
Fig. 5. Temporal changes in sinking velocity (A) and dry weight (B) of discarded houses, sinking velocity (C) and dry weight (D) as a function of the diameter of the house, and the fractal dimension of houses as a function of age (E). In panels (A–B), each house is represented with a different colour and the parameters of the regression are shown in Table 1. In panels (C–D), each different age intervals is represented with a different colour. The regression in panel (E), shown with 95% confidence limits, is D ¼1.48 (7 0.03) T 0.061 ( 7 0.004), where T is age (days). In panel (F), the observed size-sinking speed relations for houses aged 0 (black), 0.9–1.5 d (red) and 1.5–3 d (blue) have been compared with reported settling velocities for field-collected aggregates and with settling velocities obtained from a coagulation model (Stemmann et al. 2004 with different parameter values; 1: Dr ¼ 0.05, D ¼ 2.33; 2: Dr ¼0.01, D ¼1.79). The field data stem from line 3: Alldredge and Gotschalk (1988); 4: Alldredge and Gotschalk (1989); 5: Syvitski et al. (1995), 6: Guidi et al. (2008).
consistent with observations of marine snow from both natural environment and from the laboratory (1.59–2.67, Logan and Wilkinson, 1990; Jackson, 1995; Li et al., 1998) and with the range of D-values explored in the model of Stemmann et al. (2004a). Other types of marine snow aggregates may similarly change properties as they age and are degraded. Thus, observations have shown that D of marine diatom aggregate decreases during the time course of a mesocosm phytoplankton bloom (Li and Logan, 1995) and estimates of D are also smaller in on old diatoms flocs as compared to diatom aggregates at the onset of aggregation
(Li et al., 1998). However, the compaction and fractal dimension of an aggregate is a function not only of degradation processes, but also of the fluid mechanical environment, in which they were formed and in which they sink (Logan and Kilps, 1995). Marine snow particles experience different hydrodynamical regimes as they sink, from turbulent environments during their initial formation in the surface (while not yet sinking), to a flow field environment generated by their sinking through the water column. This complicates the interpretation of field observed variation in the fractal dimension of marine snow aggregates and
Author's personal copy F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
emphasises the need for experimental examinations of temporal changes in the properties also of marine snow other than those formed from appendicularian houses. A second consequence of the deflation process is that the house releases most of the enclosed water into the ambient, with all its constituents of particles and solutes. This leads to elevated concentrations of particles and solutes in the trail of the sinking house (Fig. 3), which may guide chemosensory bacteria (Jackson, 1989; Kiørboe and Jackson, 2001) and detrivorous zooplankters (Kiørboe and Thygesen, 2001) to the discarded house, and thus accelerate its degradation. During the first hour, the house thus typically looses 93% of its water content (Fig. 4). The concentration of dissolved organic carbon (DOC) inside marine snow, including discarded appendicularian houses, is more than one order of magnitude higher than in the surrounding water (Alldredge, 2000). Because both the concentration of compounds and the length of the detectable chemical trail are directly related to the leakage rate from the particle (Kiørboe and Thygesen, 2001), the deflation process will intensify the chemical signal over that from a non-deflating particle, which leaks DOC only due to microbial dissolution of an organic particle. Assuming an amino acid leakage rate (L) from a non-deflating particle, as in Kiørboe 1 and Thygesen (2001) (L¼1 10 12 r1.5 , where rp is the p mol s radius of the particle), a background concentration of amino acids of 10 10 mol cm 3, and an internal concentration of amino acids 30 times higher in houses than in the environment (Alldredge, 2000), the enhancement of the chemical signal due to deflation can be estimated using the observed deflation rate and the ‘moving point source’ model of Jackson and Kiørboe (2004). Taking house #6 as an example, the length of the detectable trail will be elevated after 1 min by a factor of 6 over that of a nondeflating particle; by 40% after 15 min, and by o5% after 1 h. In larger houses, the enhancement is larger (Fig. 6), and for large house, the effect is dramatic. In the extreme case of bathypelagic appendicularians that may have houses with diameter 430 cm (Robison et al., 2005), the trail length may thus be elevated by a factor of 1800 when discarded, and even after 3 d there is still an 11% enhancement. This increase leads to a similar elevation of colonization of particularly zooplankton, which may explain the high disappearance rate (70%) of appendicularian houses observed between 10 and 30 m water depths in the ocean (Vargas et al., 2002) as well as the important number of
12 1000 m 5000 m 10000 m
Trail length increase
10 8 6 4 2 0 10-3
10-2
10-1 Time
100
101
Fig. 6. Enhancement of the length of the chemical trail left by a deflating, appendicularian-derived marine snow aggregate over that of a similarly sized, non-deflating aggregate. The enhancement changes as a function of time and increases with the size of the discarded house.
1311
zooplankton organisms found in and on bathypelagic appendicularian houses (Alldredge, 1979; up to 500 metazoans house 1; Steinberg et al., 1994). The concentration of dissolved organic material in the plume trailing behind the sinking particle is enhanced by the same factor as the plume length. These solute plumes are utilized by free-living bacteria (Kiørboe and Jackson, 2001; Kiørboe et al., 2001). Thus, appendicularians houses, due to the deflation process, may degrade faster than other marine snow particles due to zooplankton grazing and microbial activity, and better support both microbial and zooplankton communities in the upper water column. A last consequence of the rapid initial deflation process to be considered is the leakage of small particles that may be significant enough to make the trail visible (Fig. 3). The leaking particles are initially attached to the filters of the appendicularian house, but the deflation process is so intense that the water flux back-flushes the particles accumulated on the concentrating filter. After 1 h the space between the two layers of the filter is too reduced to allow further release of particles. In our experiments, this particle leakage has been responsible for a 10–60% total loss of house dry weight within the first hour. The subsequent rate of weight loss, between 0.01 and 0.2 d 1, is comparable to the weight loss rate observed in other marine snow particles, which is attributable to bacterial dissolution of the aggregate (Ploug and Grossart, 2000). The particle loss may have several consequences: (1) it may add to the effect of the chemical trail in guiding zooplankton and microbes to the house; (2) it is a source of small-scale patchiness of small particles in the ocean; and (3) it detracts from the effect of sinking appendicularian house on the vertical particle flux in the ocean, because a significant fraction of the particle load is left in the upper layer. These are all potentially significant effects that may have major influence on particle dynamics and material fluxes in the ocean, particularly in regions, where appendicularians abound. Future estimation of appendicularians contribution to the POC export has to consider these processes. To finally illustrate the potential significance of the agerelated changes of discarded appendicularian houses for vertical flux process in the ocean, we compared predicted depth-, size distribution of appendicularian houses with the simultaneously observed depth- and size-distribution of particles at a station in the NE Altantic Ocean (Fig. 7). We used data collected in May at station S4, where the maximal appendicularian abundance (135 ind m 3) was recorded at 75 m depth. Our predictions (Fig. 7A) shows that within a few hours, the large (2000–4000 mm) houses produced are quickly compacted and accumulate in the smaller size class of particles (o500 mm) and that the steady-state distribution is dominated by particleso500 mm occurring at depths 4100 m. Such a decrease of marine snow size can be seen in the marine snow profiles (Fig. 7B), where large (500–4000 mm) particles are more abundant in the 50–100 m interval and decrease afterward, but is mostly masked by the massive occurrence of smaller particles. Our calculation further leads to the estimate that up to 20–40% of 300–500 mm particles in the upper 100 m depths may be of appendicularian origin, stressing the importance of those animals as important particles producers in the ocean. Using the mean observed bacterial degradation rate of house (0.06 d 1) and additional leakage rate during the first hour (4.35 d 1), 14% of discarded house material is directly lost in particles trail during the first hour of sinking, whereas another 16% is consumed by bacteria before the discarded house leave the upper 200 m. These estimates, however, neglect other potential loss processes such as consumption, aggregation or fragmentation of particles by zooplankton (Goldthwait et al., 2004; Koski et al., 2007), possible pressure effects and effects of changing microbial community
Author's personal copy 1312
F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
Fig. 7. A: estimated steady state concentration and size- and depth distribution of discarded appendicularian houses (particles m 3) at station S4 of the POMME 2 cruise in the NE Atlantic Ocean (May 2002; 42.521N, 17.971W). We used the observed size- and depth distribution of appendicularians as well as a physiological model of house production to estimate the input rate of discarded houses (Lombard et al., 2010) and the age-related changes in size and settling characteristics of discarded houses reported here to estimate the steady-state distribution of houses on the assumption of no other house degradation processes B: observed concentration and depth distribution of marine snow particles at the same site and date (Guidi et al., 2007). Note the different colour codes in panels (A) and (B). The circles in panel (A) describe the fate of an imaginary cohort of discarded houses, assuming that each appendicularian discards one house at time t ¼0.
during sedimentation (Tamburini et al., 2009). All these processes can have a significant impact in modifying the vertical flux of material and needs to be quantified.
Acknowledgments We greatly thank L. Guidi, G. Gorsky, and G.A. Jackson for their constructive discussions and the Marie Curie Intra-European Fellowship No. 221696 to FL, the Danish Science Research Council as well as the Niels Bohr Foundation for funding.
References ˜ a, J.L., Kiefer, M., 2000. Functional response of the appendicularian Oikopleura Acun dioica. Limnol. Oceanogr. 45, 608–618. Alldredge, A.L., 2000. Interstitial dissolved organic carbon (DOC) concentrations within sinking marine aggregates and their potential contribution to carbon flux. Limnol. Oceanogr. 45, 1245–1253. Alldredge, A.L., 2005. The contribution of discarded appendicularian houses to the flux of particulate organic carbon from oceanic surface waters. In: Gorsky, G., Youngbluth, M.J., Deibel, D. (Eds.), Response of Marine Ecosystems to Global Change: Ecological impact of Appendicularians. GB Scientific Publisher, Paris, pp. 309–326. Alldredge, A.L., Gotschalk, C., 1988. In situ settling behavior of marine snow. Limnol. Oceanogr. 33, 339–351. Alldredge, A.L., Gotschalk, C.C., 1989. Direct observations of the mass flocculation of diatom blooms—characteristics, settling velocities and formation of diatom aggregates. Deep-Sea Res. Part A 36, 159–171. Alldredge, A.L., Gotschalk, C.C., 1990. The relative contribution of marine snow of different origins to biological processes in coastal waters. Cont. Shelf Res. 10, 41–58. Alldredge, A.L., Silver, M.W., 1988. Characteristics, dynamics and significance of marine snow. Prog. Oceanogr. 20, 41–82. Berelson, W.M., 2002. Particle settling rates increase with depth in the ocean. Deep-Sea Res. Part II 49, 237–251. Bouquet, J.M., Spriet, E., Troedsson, C., Ottera, H., Chourrout, D., Thompson, E.M., 2009. Culture optimization for the emergent zooplanktonic model organism Oikopleura dioica. J. Plankton Res. 31, 359–370. Burd, A.B., Jackson, G.A., 2009. Particle aggregation. Annu. Rev. Mar. Sci. 1, 65–90. Cheng, N.-S., 2009. Comparison of formulas for drag coefficient and settling velocity of spherical particles. Powder Technol. 189, 395–398. Davies, C.N., 1979. Particle–fluid interaction. J. Aerosol Sci. 10, 477–513. De La Rocha, C.L., Passow, U., 2007. Factors influencing the sinking of POC and the efficiency of the biological carbon pump. Deep-Sea Res. Part I 54, 639–658. Fenaux, R., 1986. The house of Oikopleura dioica (Tunicata, Appendicularia): Structure and functions. Zoomorphology 106, 224–231. Fenaux, R., Gorsky, G., 1985. Nouvelle technique d’e´levage des appendiculaires. Rapp. P–V Re´un.—Comm. Int. Explor. Sci. Mer Me´diterr. 29, 291–292.
Fischer, G., Karakas, G., 2009. Sinking rates and ballast composition of particles in the Atlantic Ocean: implications for the organic carbon fluxes to the deep ocean. Biogeosciences 6, 85–102. Flood, P.R., Deibel, D., 1998. The appendicularian house. In: Bone, Q. (Ed.), The Biology of Pelagic Tunicates. Oxford University Press, Oxford, pp. 105–124. Francois, R., Honjo, S., Krishfield, R., Manganini, S., 2002. Factors controlling the flux of organic carbon to the bathypelagic zone of the ocean. Global Biogeochem. Cy. 16, 1087. doi:10.1029/2001GB001722. Goldthwait, S., Yen, J., Brown, J., Alldredge, A., 2004. Quantification of marine snow fragmentation by swimming euphausiids. Limnol. Oceanogr. 49, 940–952. Gorsky, G., Fisher, N.S., Fowler, S.W., 1984. Biogenic debris from the pelagic tunicate, Oikopleura dioica, and its role in the vertical transport of a transuranium element. Estuarine Coast. Shelf Sci. 18, 13–23. Guidi, L., Stemmann, L., Legendre, L., Picheral, M., Prieur, L., Gorsky, G., 2007. Vertical distribution of aggregates (4110 mu m) and mesoscale activity in the northeastern Atlantic: effects on the deep vertical export of surface carbon. Limnol. Oceanogr. 52, 7–18. Guidi, L., Jackson, G.A., Stemmann, L., Miquel, J.C., Picheral, M., Gorsky, G., 2008. Relationship between particle size distribution and flux in the mesopelagic zone. Deep-Sea Res. Part I 55, 1364–1374. Hamner, W.M., Robison, B.H., 1992. In situ observations of giant appendicularians in Monterey-Bay. Deep-Sea Res. Part A 39, 1299–1313. Hansen, J.L.S., Kiørboe, T., Alldredge, A.L., 1996. Marine snow derived from abandoned larvacean houses: sinking rates, particle content and mechanisms of aggregate formation. Mar. Ecol. Prog. Ser. 141, 205–215. Honjo, S., Doherty, K.W., Agrawal, Y.C., Asper, V.L., 1984. Direct optical assessment of large amorphous aggregates (marine snow) in the deep ocean. Deep-Sea Res. Part A 31, 67–76. Honjo, S., Manganini, S.J., Krishfield, R.A., Francois, R., 2008. Particulate organic carbon fluxes to the ocean interior and factors controlling the biological pump: a synthesis of global sediment trap programs since 1983. Prog. Oceanogr. 76, 217–285. Jackson, G.A., 1989. Simulation of bacterial attraction and adhesion to falling particles in an aquatic environment. Limnol. Oceanogr. 34, 514–530. Jackson, G.A., 1995. Comparing observed changes in particle-size spectra with those predicted using coagulation theory. Deep-Sea Res. Part II 42, 159–184. Jackson, G.A., Kiørboe, T., 2004. Zooplankton use of chemodetection to find and eat particles. Mar. Ecol. Prog. Ser. 269, 153–162. Kiørboe, T., 2000. Colonization of marine snow aggregates by invertebrate zooplankton: abundance, scaling, and possible role. Limnol. Oceanogr. 45, 479–484. Kiørboe, T., 2001. Formation and fate of marine snow: small-scale processes with large-scale implications. Sci. Mar. 65, 57–71. Kiørboe, T., Jackson, G.A., 2001. Marine snow, organic solute plumes, and optimal chemosensory behavior of bacteria. Limnol. Oceanogr. 46, 1309–1318. Kiørboe, T., Ploug, H., Thygesen, U.H., 2001. Fluid motion and solute distribution around sinking aggregates. I. Small-scale fluxes and heterogeneity of nutrients in the pelagic environment. Mar. Ecol. Prog. Ser. 211, 1–13. Kiørboe, T., Thygesen, U.H., 2001. Fluid motion and solute distribution around sinking aggregates. II. Implications for remote detection by colonizing zooplankters. Mar. Ecol. Prog. Ser. 211, 15–25. Klaas, C., Archer, D.E., 2002. Association of sinking organic matter with various types of mineral ballast in the deep sea: implications for the rain ratio. Global Biogeochem. Cy. 16, 1116. doi:10.1029/2001GB001765.
Author's personal copy F. Lombard, T. Kiørboe / Deep-Sea Research I 57 (2010) 1304–1313
Koski, M., Moller, E.F., Maar, M., Visser, A.W., 2007. The fate of discarded appendicularian houses: degradation by the copepod, microsetella norvegica, and other agents. J. Plankton Res. 29, 641–654. Li, X.Y., Logan, B.E., 1995. Size distributions and fractal properties of particles during a simulated phytoplankton bloom in a mesocosm. Deep-Sea Res. Part II 42, 125–138. Li, X.Y., Passow, U., Logan, B.E., 1998. Fractal dimensions of small (15–200 mm) particles in Eastern Pacific coastal waters. Deep-Sea Res. Part I 45, 115–131. Logan, B.E., Kilps, J.R., 1995. Fractal dimensions of aggregates formed in different fluid mechanical environments. Water Res. 29, 443–453. Logan, B.E., Wilkinson, D.B., 1990. Fractal geometry of marine snow and other biological aggregates. Limnol. Oceanogr. 35, 130–136. Lombard, F., Legendre, L., Picheral, M., Sciandra, A., Gorsky, G., 2010. Prediction of ecological niches and carbon export by appendicularians using a new multispecies ecophysiological model. Mar. Ecol. Prog. Ser. 398, 109–125. Lombard, F., Renaud, F., Sainsbury, C., Sciandra, A., Gorsky, G., 2009a. Appendicularian ecophysiology. I. Food concentration dependent clearance rate, assimilation efficiency, growth and reproduction of Oikopleura dioica. J. Mar. Sys., 10.1016/j.jmarsys.2009.1001.1004. Lombard, F., Sciandra, A., Gorsky, G., 2009b. Appendicularian ecophysiology. II. Modeling nutrition, metabolism, growth and reproduction of the appendicularian Oikopleura dioica. J. Mar. Sys., 10.1016/j.jmarsys.2009.1001.1005. Lombard, F., Sciandra, A., Gorsky, G., 2005. Influence of body mass, food concentration, temperature and filtering activity on the oxygen uptake of the appendicularian Oikopleura dioica. Mar. Ecol. Prog. Ser. 301, 149–158. Michaelides, E.E., 2006. Particles, Bubbles and Drops—Their Motion, Heat and Mass Transfer. World Scientific Publishing Co. Pte. Ltd., Singapore. Ploug, H., Grossart, H.P., 2000. Bacterial growth and grazing on diatom aggregates: respiratory carbon turnover as a function of aggregate size and sinking velocity. Limnol. Oceanogr. 45, 1467–1475. Ploug, H., Iversen, M.H., Fischer, G., 2008. Ballast, sinking velocity, and apparent diffusivity within marine snow and zooplankton fecal pellets: implications for substrate turnover by attached bacteria. Limnol. Oceanogr. 53, 1878–1886.
1313
Robison, B.H., Reisenbichler, K.R., Sherlock, R.E., 2005. Giant larvacean houses: rapid carbon transport to the deep sea floor. Science 308, 1609–1611. Sato, R., Tanaka, Y., Ishimaru, T., 2001. House production by Oikopleura dioica (Tunicata, Appendicularia) under laboratory conditions. J. Plankton Res. 23, 415–423. Sato, R., Tanaka, Y., Ishimaru, T., 2003. Species-specific house productivity of appendicularians. Mar. Ecol. Prog. Ser. 259, 163–172. Sato, R., Tanaka, Y., Ishimaru, T., 2005. Clearance and ingestion rates of three appendicularian species, Oikopleura longicauda, O. rufescens and O. fusiformis. In: Gorsky, G., Youngbluth, M.J., Deibel, D. (Eds.), Response of Marine Ecosystems to Global Change: Ecological Impact of Appendicularians. GB Scientific Publisher, Paris, pp. 189–206. Steinberg, D.K., Silver, M.W., Pilskaln, C.H., Coale, S.L., Paduan, J.B., 1994. Midwater Zooplankton Communities on Pelagic Detritus (Giant Larvacean Houses) in Monterey Bay, California. Limnol. Oceanogr. 39, 1606–1620. Stemmann, L., Jackson, G.A., Gorsky, G., 2004a. A vertical model of particle size distributions and fluxes in the midwater column that includes biological and physical processes—Part II: application to a three year survey in the NW Mediterranean Sea. Deep-Sea Res. Part I 51, 885–908. Stemmann, L., Jackson, G.A., Ianson, D., 2004b. A vertical model of particle size distributions and fluxes in the midwater column that includes biological and physical processes—Part I: model formulation. Deep-Sea Res. Part I 51, 865–884. Syvitski, J.P.M., Asprey, K.W., Leblanc, K.W.G., 1995. In-situ characteristics of particles settling within a deep-water estuary. Deep-Sea Res. Part II 42, 223–256. Tamburini, C., Goutx, M., Guigue, C., Garel, M., Lefevre, D., Charriere, B., Sempere, R., Pepa, S., Peterson, M.L., Wakeham, S., Lee, C., 2009. Effects of hydrostatic pressure on microbial alteration of sinking fecal pellets. Deep-Sea Res. Part II 56, 1533–1546. Vargas, C.A., Toennesson, K., Sell, A., Maar, M., Moeller, E.F., Zervoudaki, T., Giannakourou, A., Christou, E., Satapoomin, S., Petersen, J.K., Nielsen, T.G., Tiselius, P., 2002. Importance of copepods versus appendicularians in vertical carbon fluxes in a Swedish fjord. Mar. Ecol. Prog. Ser. 241, 125–138.
