Assumptions of sampling

1
(No Transcript)
2
Assumptions of sampling

• Sample quantitatively represent the numbers and
  kinds of animals in the parcel of water sampled
• The water parcel sampled represents the area
  around the sampling location out to some assumed
  distance away.

3
Early history of net sampling

• Prior to Hensen 1895 sampling qualitative
• Hensens premise
• Plankton are evenly distributed in ocean waters
• Therefore, small samples can represent large
  oceanic areas provided
• The volume of water filtered by the net could be
  exactly determined
• Organisms would not escape through the net mesh
• Constructed nets and materials, measured net
  filtration efficiencies, employed nets at sea,
  and analyzed samples
• Identified many sources of error, quantified
  them, and developed means to reduce them

4
Ring nets

• Can be opening/closing, but usually not
• Often used vertically
• Lowered cod end first so no filtering on downcast
• Integrated cast of water column

5
Bongo nets

• Non-opening/closing (usually)
• Towed obliquely at 1-3 kts
• Larger volume sampled
• No obstruction to mouth opening
• (Reduces avoidance)

6
Computer-controlled nets

• e.g. MOCNESS
• Multiple Opening/Closing Net Environmental
  Sensing System
• Also, MultiNet (German), BIONESS (Canadian), and
  several others (non-commercial)
• Vertically stratified data

7
From Weibe and Benefield
8
Flowmeters

• Critical for quantitative assessment

9
Nets some definitions

• Mesh size/aperature, a distance (mm) along mesh
  holes
• Filament diameter, f (mm)
• Porosity fraction of mesh fabric that is hole
  a2/(af)2
• Mesh area amount of mesh fabric (m2)
• Filtering area mesh area porosity
• Filtering ration, R filtering area/net mouth
  area
• Filtering efficiency, - (water filtered/water
  encountered) 100

f
a
10
Problems with net sampling

• Escapement (extrusion)
• Avoidance
• Clogging

11
Escapement

• Organisms pass through mesh
• Selection of mesh size
• Smallest cross-sectional diameter of organisms of
  interest
• Not including limbs or hairs
• Mesh diagonal length is just less
• v2a2
• Saville (1958) tested theoretical calculations
• Used a coarse net surrounded by fine net
• Calanus fits rule of thumb well,
• Oithona retained slightly better than expected

www.sintef.no
www.sams.ac.uk/research/departments/ecology
12
Avoidance

• Jumping out of way of net
• Affected by
• Net diameter
• Warning of approach
• Extent of bridle
• Noise
• Forward pressure wave extent
• Towing speed
• Visibility

13
Estimating avoidance
R radius of net U velocity of net x0
distance of plankter from front of net r
distance of plankter from perpendicular through
center of net u velocity of plankter
U
if r 0, the net will catch the plankter if
u
R
Or, if r gt 0
x0
R - r
x0
lt
U
u
14
Estimating avoidance
R radius of net U velocity of net x0
distance of plankter from front of net r
distance of plankter from perpendicular through
center of net u velocity of plankter
U
if r 0, the net will catch the plankter if
u
R
Or, if r gt 0
x0
R - r
x0
lt
U
u
For copepods, u can be gt500 body lengths per
second, gt1m/s
So, we want to maximize U and R and minimize x0
15
Clogging

• Mesh holes obscured by sticky or ½-extruded
  particles/organisms
• Lower clogging when
• Greater water clarity
• Greater net aperature size (a)
• Higher R (mesh amount)
• Section of cylinder
• Smith et al 1968

16
Pump sampling

• Advantages
• Volume filtered measured reliably
• Depth of sampling controlled well
• Other parameters can be measured from same water
• Spatial/temporal variability can be assessed
• Sample small animals well
• Disadvantages
• Small volume sampled
• Animals must be dense
• Often used within 200 m of surface
• Bulky

17
Pumps on deck
18
Submersible pumps
Megapump (C. Miller)
9 50-µm filters on carousel 2 m3 per
minute Power from ship via cable Telemeters
depth, filter-flow, T, S net advance Pump
on-off net advance by command from computer on
deck
19
Subsampling

• Can have 10,000 to 1 million of one animal in jar
• Stir, then
• Snatch animals
• Ladle
• Stemple pipette
• Brinton method (dilute and slosh)
• or Split
• Motoda splitter
• Folsom splitter

20
Splitters
Folsom splitter
Motoda splitter
21
How much to subsample?
f - fraction of sample that is the subsample c -
individuals in the subsample N - number of
individuals in the whole sample
Nc/f
A perfect subsample will contain exactly cfN
organisms (never happens)

• Confidence limits
• Based on normal distribution if cgt10
  (otherwise, binomial)

22
Number in whole sample
23
(No Transcript)
24
(No Transcript)
25
Problems detecting patches with nets
26
(No Transcript)
27
(No Transcript)
28
References

• Miller, C.B. and Judkins, D.C., 1981. Design of
  pumping systems for sampling zooplankton, with
  descriptions of two high-capacity samplers for
  coastal studies. Biological Oceanography 1 (1),
  29-56.
• Sameoto, D.P., Wiebe, P., Runge, L., Postel, L.,
  Dunn, J., Miller, C.B. and Coombs, S., 2000.
  Collecting zooplankton. In R.P. Harris et al.
  (Eds), ICES Zooplankton Methodology Manual.
  Academic Press, San Diego.
• Saville, A., 1958. Mesh selection in plankton
  nets. Journal du Conseil - Conseil International
  pour L'exploration de la Mer 23, 192-201.
• Smith, P., Counts, R. and Clutter, R., 1968.
  Changes in filtering efficiency of plankton nets
  due to clogging under tow. Journal du Conseil -
  Conseil International pour L'exploration de la
  Mer 23, 232-248.
• Wiebe, P. and Benfield, M.C., 2003. From the
  Hensen net toward four-dimensional biological
  oceanography. Progress in Oceanography 56, 7-136.