Methodology used for quantitative plankton observations from “Mir” submersibles.
A.L. Vereshchaka, project no 911102(IMR)
1. Introduction .
Use of the standard gear for the plankton sampling (nets, trawls, and big bottles) leads to the underestimation of the proper plankton concentrations. Many actively swimming plankton animals demonstrate avoiding behaviour; gelatinous animals (medusas, ctenophores), especially those fragile and dwelling at greater depths, are usually destroyed during the sampling. Thus, estimated values of some animals may appear to be several times less than in the reality, while other groups, although numerous in the oceanic depths, may not be sampled at all. In addition, standard methods require many hours for descending, sampling, and retrieval of the gear. Given the movability of the water layers (respective to each other and to the research vessel), the sample sets usually derive from different sites distanced horizontally from each other by several kilometres.
Direct visual observations give us opportunity to study animal distribution with any desired vertical precision. Most of groups (except nectonic fishes and squids) do not demonstrate avoidance behaviour and may be counted directly from the view port. Below a brief description of methodology developed during the dives in 1994-2002 may be found.
Plankton observations made twice above the same site provide very similar pictures of the distribution of dominant groups that gives a good evidence for representativity of the methods in use.
2. General characteristics and equipment
Russian Academy of Sciences has 2 twin manned submersibles “Mir-1”and “Mir-2”, both with working depth 6000 m and based at R/V ”Akademik Mstislav Keldysh”. If compared with other 6000-m submersibles acting in 1980s-1990s (table 1), Mirs have much greater energy supply and maximal speed (shaded in table 1). Alvin, not included in table 1, also had lesser working depth and energy supply than “Mir”s. The energy supply of “Mir-1” and “Mir-2” allows illumination of the water column during descent and ascent of submersibles (2-4 hours each depending on depth of the ocean) and 5-6 hours of the bottom observations. During ascent or descent, visual observation of the plankton communities are possible.

Table 1. Technical parameters of the manned submersibles acting in 1980s-1990s at depths
0-6000 m
Parameters
Nautile
(France)
18.5
Submersibles
Mir-1 and Mir-2
(Russia)
18.6
Seacliff
(USA)
29.0
Shinkai-6500
(Japan)
25.2 Dry weight
Length
Width
8.0
2.7
7.8
3.8
8.6
3.6
8.2
3.6
Height
Energy supply (kW*h)
Life supply (person*h)
Maximal speed (knots)
Buoyancy (kg)
Equipage (No of persons)
3.45
50
390
2.5
200
3
2.1
3.0
100
246
5.0
290
3
2.1
3.4
60
300
2.0
250
3
2.1
3.45
55
300
2.0
220
3
2.1 Diameter of the principal sphere (m)
Material of the principal sphere
Ti - alloy Ni - steel Ti - alloy Ti - alloy
Type of accumulators Pb - acid Ni-Cd Ag-Zn Ag-Zn
Both submersibles provide a possibility of visual observation of all visible animals (3 mm and larger) in the water column (pelagic communities), near-bottom layer (benthopelagic communities), on the sea floor (benthic communities). Gelatinous plankton is well observed and its abundances may be adequately estimated. Animal may be counted, their sized estimated (precision 1 cm).
Abundances may be calculated and later converted to biomasses (both wet and carbon weigh) with use of available algorithms.
Video and still photoes are possible in the near-bottom layer and on the sea-floor. Each submersible is equipped with 2 manipulators that can sample sessile and slowly moving macrobenthic animals on any substrate as well as meiobenthos in the soft sediments (with use of special grabs and sieves).
One of submersibles is equipped with a slurp-gun that can sample selected planktonic and benthic animals near or on the bottom. Sampling of mesoplankton in the water column is also possible by means of unselective pumping of water.

4.
Animal counting: what to do during the dives . Animals are counted during vertical sets of observations with use of a counting frame. In the oligotrophic areas like the Mid-Atlantic Ridge south of 40 N, a counting frame is bordered by 2 extended manipulators and a rope between the
“palms” (Fig. 1); in this case the counting area is equal to 3 m 2 . In the meso- or eutrophic areas, much smaller counting frame may be made from wire and then fixed in one of manipulators; in this case the counting area may range from 0.1 to 0.6 m 2 . Each frame has marks on boarders that allows rough (with precision ½-1 cm) estimation of the observed animal size. The counting volume is illuminated from two points, from left and from right, in order to meet the following requirements:
(1) illumination is at a minimum necessary for certain recording of transparent (ctenophores, etc.) and small (5-mm copepods) animals; (2) illumination of the volume beyond the counting frame is minimised to dusk in order not to induce avoidance behaviour of animals.
Vertical observations are usually performed between 10 a.m. and 21 p.m., plankton is thoroughly counted during descents of submersible, vertical speed ranges from 10 (meso- or eutrophic areas) to 30 m/min (oligotrophic areas). During the dives, any information of observed animals (first of all, taxon and approximate length) is tape-recorded. Information about depth is also synchronously tape-recorded (any desired precision, usually each 5-10 m in the meso- or eutrophic areas, each 20-50 m in the meso- or eutrophic areas).
5. Treatment of results: what to do after the dives . After decoding the tape on board the ship, the observer knows how many animals of each group have been observed within selected depth range. For example, if 6 cyclothones (fishes) have been observed between 400 and 420 m, actual abundances of this animals in the layer 400-420 m may be calculated as 6 ind/examined volume = 6 ind/(3 m 2 * 20 m) = 0.1 ind/m 3 . Treatment of results may be done in any of frequently used programs. We recommend a specially designed program PLANKTY (contact the author) that allows fast treatment of plankton data. Each dive is usually resulted in a set of pictures illustrating vertical distribution of animals (species, higher taxa, and ecological groups) with any desired precision. Figs 2-6 show how the abundances of the main dominant plankton groups change with depth at one of studies sites, fig. 7 represents distribution of the ecological group "gelatinous animals".
These data have been obtained above the Mid-Atlantic Ridge on 11 July 2002 in the site
37

17' N; 32

16 W, st. 4384. Distribution of examined groups appears to be associated with the salinity/temperature/density and turbidity gradients (data of the Neil-Brown soundings made several hours before the dive). This is especially clear for the gelatinous animals' maxima in the depth range 400-800 m.

6. Comparison of visual and net data. We can compare vertical distribution of animals obtained with use of standard gears (plankton nets) and visual observations. Fig. 8 (right side) and Fig. 9 illustrate vertical distribution of the planktonic fish of the genus Cyclothone . We can see that the general tendencies are similar but visual observations provide more detailed information. For example, the gap in the depth range 600-650 m is visible in Fig 9 but vanishes in Fig. 8, where the general picture is more rough.
7. Further approaches.
Described methods may be used for more specialised tasks, for example, for the studies of the animal distribution in the near-bottom layer by means of oblique sets. During these sets, submersible moves with a horizontal speed of 2 knots and synchronously descends or ascends with a vertical speed 10 m/min. The wire frame is oriented horizontally and positioned aside the submersible in order to avoid influence of the hydrodynamic «pillow» just ahead the horizontally moving vehicle (horizontal speed was about 1.0-1.5 m/c). Animals are observed through the lateral illuminator of the submersible.
During each biological dive, about 4000 m 3 of the near-bottom water may be observed in detail.
Another possibility to advance described method leads to the estimation of actual biomasses. When rough sizes are recorded, special regressions and programs (please contact the author) may be used to calculate wet and carbon weights of counted animals. The procedures are analogous to those described for the estimation of animal abundances.
7. References to papers based upon described methodology
1.
Vereshchaka, A. L., Vinogradov G. M., 1999. Visual observations of the vertical distribution of plankton throughout the water column above Broken Spur vent field,
Mid-Atlantic Ridge.

Deep-Sea Research I, 46: 1615

1632.
2.
Vinogradov M. E. and Shushkina E. A., 1992. Peculiarities of the meso- and macroplankton vertical distribution in the central tropical areas of the North Pacific.
Okeanologiya, 30 (1): 115-127 (in Russian)
3.
Vinogradov M. E. and Shushkina E. A., 1994. Study on the vertical distribution of the
North Pacific zooplankton, based on quantitative estimations from deep-water manned submersibles (DMS) "Mir". In: (Eds. M. E. Vinogradov and A. L. Vereshchaka)
Plankton and Benthos of the Northern Pacific: Study with Deep Manned Submersibles
“Mir”,
Trudy Instituta okeanologii, 131 : 41-63 (in Russian)

4.
Vinogradov M. E. and Tchindonova Yu. G., 1994 Notes on the vertical senility of pelagic fauna, based on the direct observations from DMS "Mir".
In: (Eds. M. E.
Vinogradov and A. L. Vereshchaka) Plankton and Benthos of the Northern Pacific:
Study with Deep Manned Submersibles “Mir”,
Trudy Instituta okeanologii, 131 : 64-75.
(in Russian)
5.
Vinogradov, M. E., A. L. Vereshchaka & E. A. Shushkina. 1996. Vertical structure of the zooplankton communities of the oligotrophic areas of the North Atlantic and the impact of the hydrothermal ecosystems.

Okeanologija 36(1): 76

79. [In Russian, completely translated into English]
6.
Vinogradov, M. E., A. L. Vereshchaka, E. A. Shushkina & G. N. Arnautov. 1999.
Structure of zooplanktonic communities in the frontal zone of Gulf Stream and Labrador
Current.

Okeanologija 39(4): 555

566. [In Russian, completely translated into English in Oceanology]

Fig. 1. Plankton counting during the «Mir» dives, by G.M. Vinogradov.

0
0
5
0.04
Ind / m 3
0.08
10
0.12
T, °C 15
0.16
500
1000
S, o / oo
(solid line);
T, °C (dotted line)
(st. 4375
13
)
Lucky Strike, st. 4384.
Appendicularium "houses".
1500
Nephels
(st. 4375
10
)
520 600
Bottom
35.2
35.6
36
S, o / oo
Fig. 2. Distribution of the larvacean "houses" above the Mid-Atlantic Ridge, site 37

17' N;
32

16 W (above Lucky Strike vent field).

0
0
5
0.04
Ind / m 3
0.08
10 T, °C 15
0.12
500
1000
S, o / oo
(solid line);
T, °C (dotted line)
(st. 4375
13
)
Lucky Strike, st. 4384.
Chaetognatha.
1500
Nephels
(st. 4375
10
)
520 600
Bottom
35.2
35.6
36
S, o / oo
Fig. 3. Distribution of the chaetognaths above the Mid-Atlantic Ridge, site 37

17' N; 32

16
W (above Lucky Strike vent field).

0
0
5
0.02
Ind / m 3
0.04
10 T, °C 15
0.06
500
1000
S, o / oo
(solid line);
T, °C (dotted line)
(st. 4375
13
)
Lucky Strike, st. 4384.
Radiolarians and their colonies > 5 mm.
1500
Nephels
(st. 4375
10
)
520 600
Bottom
35.2
35.6
36
S, o / oo
Fig. 4. Distribution of the radiolarians above the Mid-Atlantic Ridge, site 37

17' N; 32

16 W
(above Lucky Strike vent field).

0
0
5
0.02
Ind / m 3
0.04
10 T, °C 15
S, o / oo
(solid line);
T, °C (dotted line)
(st. 4375
13
)
0.06
500
1000
Lucky Strike, st. 4384.
Siphonophora.
1500
Nephels
(st. 4375
10
)
520 600
Bottom
35.2
35.6
36
S, o / oo
Fig. 5. Distribution of the syphonophoras above the Mid-Atlantic Ridge, site 37

17' N; 32

16
W (above Lucky Strike vent field).

0
0
5
0.04
0.08
Ind / m 3
0.12
0.16
10 T, °C 15
S, o / oo
(solid line);
T, °C (dotted line)
(st. 4375
13
)
0.2
500
1000
Lucky Strike, st. 4384.
Cyclothones.
1500
Nephels
(st. 4375
10
)
520 600
Bottom
35.2
35.6
36
S, o / oo
Fig. 6. Distribution of the cyclothones above the Mid-Atlantic Ridge, site 37

17' N; 32

16 W
(above Lucky Strike vent field).

0
0 0.05
5
0.1
Ind / m 3
0.15
0.2
10
0.25
T, °C
15
0.3
500
S, o / oo
(line);
T, °C (dash line)
1000
Lucky Strike: all gelatinous animals, including appendicularium “houses”
1500
Turbidity
520 600
35.2
35.6
36
S, o / oo
Fig. 7. Distribution of the gelatinous animals above the Mid-Atlantic Ridge, site 37

17' N;
32

16 W (above Lucky Strike vent field).

,
5 0 0
1 0 0 0
0
0 . 9
R a in b o w
0 . 6
5 0 0
,
1 0 0 0
0 . 3 0 . 3 0 . 6
L u c k y S t r ik e
0 . 9
1 5 0 0
2 0 0 0
Fig. 8. Distribution of the Cyclothone biomass (mg/m 3 ) relative to depth (m) above the sites Rainbow (36 N, left) and Lucky Strike (37 N, right) according to the net samples (BR plankton net with opening 1 m 2 and mesh size 500 mkm).
0 0 . 0 4 0 . 0 8 0 . 1 2 0 . 1 6
0
5 1 0 T ,
° C
1 5
0 . 2
1 5 0 0
5 2 0 6 0 0
3 5 . 2 3 5 . 6
S , o / o o
Fig. 9. Distribution of the Cyclothone abundances (ind/m 3 ) relative to depth (m) above the site Lucky Strike (37 N) according to the visual observations from «Mir».
3 6