Sediment trap time series from the North Pacific Ocean - PICES

Sediment trap time series from
the North Pacific
Ocean
David Timothy
data analyses/
presentation
CS Wong
Program organiser
Frank Whitney
lab and sampling logistics
Janet Barwell-Clarke
John Page
Linda White
lab and data analyses
Tim Soutar
Ron Bellegay
Darren Tuele
trap and mooring
preparation
Alex Bychkov
Oleg Ibadulaev
Logistical support for
western Pacific traps
Officers and Crew
of the JP Tully

TIME SERIES FROM:
‣Ocean Ocean Station PAPA
‣Alaska Alaska Gyre and North Pacific Western Gyre
‣Line Line P with emphasis at La Perouse Bank

Marine sediments are composed of…
biogenic silica
calcium carbonate
organic matter
and rocks!

OSP: 2002-2003
overlapping deployments
3700 m - 3800 m
total dry weight (mg m -2 d -1 )
400 3700 m
3800 m
300
200
100
0
July A S O N D J F M A M June
2002 2003
200
total dry weight (mg m -2 d -1 )
150
100
50
0
Sept Oct Nov Dec
2002

Sediment-trap moorings deployed by IOS
60
50
latitude
40
30
20
10+ yrs
~3 yrs
~1 yr
220 140
200 160
OSP: 1982 to present (3 depths)
La Perouse Bank: 1986 to present
Alaska Gyre: 1990 to 1992
180
longitude
160
140
120
3 stations in the Western Pacific:
August, 1991 to June, 1992
3 stations along line P:
sporadically from 1995 to 1997

TIME SERIES FROM:
‣Ocean Station PAPA
‣Alaska Alaska Gyre and North Pacific Western Gyre
‣Line Line P with emphasis at La Perouse Bank

200 m
1 year of fluxes from OSP
1000 m
3800 m

total dry weight (mg m -2 d -1 )
1982 84 86 88 90 92 94 96 98 00 2002
600
400
200
600
400
200
600
400
200
0
"21 years" of
sediment-trap flux
at OSP
1000
3800 m
200 m
1000 m
1982 84 86 88 90 92 94 96 98 00 2002

characterising seasonality and “filling gaps”
to generate a single time-series

POC (mg m -2 d -1 ) CaCO 3
(mg m -2 d -1 ) BSi (mg m -2 d -1 ) TDW (mg m -2 d -1 )
600
400
200
0
200
100
0
200
100
0
80
60
40
20
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
deployments per day
200 m fluxes at OSP
sorted by Julian day
10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
9
8
7
6
5
4
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
7.7 years of data

TDW (mg m -2 d -1 )
600
400
200
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1000 m fluxes at OSP
sorted by Julian day
POC (mg m -2 d -1 ) CaCO 3
(mg m -2 d -1 ) BSi (mg m -2 d -1 )
0
300
200
100
0
300
200
100
0
30
20
10
deployments per day
15 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
14
13
12
11
10
9
8
7
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
11 years of data
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
TDW (mg m -2 d -1 )
400
300
200
100
3800 m fluxes at OSP
sorted by Julian day
CaCO 3
(mg m -2 d -1 ) BSi (mg m -2 d -1 )
0
200
100
0
200
100
deployments per day
16 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
14
12
10
8
POC (mg m -2 d -1 )
0
20
15
10
5
31
6
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
13 years of data
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

TIME-LAG IN FLUX BETWEEN
SHALLOW AND DEEP TRAPS
For each constituent, maximum
correlation between 1000 m and
3800 m occurred at a 12-16 day
shift, which is close to the average
deployment period.
TDW (mg m -2 d -1 )
250
200
150
100
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Material took ~8 to 24 days to sink
2800 m, for sinking rates of about
100 to 400 m d -1 .
The shift for CaCO 3
(12 days) was
less than the shift for TDW, BSi
and POC (16 days). Is CaCO 3
sinking faster than BSi, and is
POC associated with BSi moreso
than with CaCO 3
Correlations (lower fig) were determined
by shifting the entire time series, not the
seasonally averaged fluxes as implied by
the upper figure.
50
0
1000 m
3800 m
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
corr coefficient (r)
0.82
0.80
0.78
0.76
0.74
0.72
0.70
0 5 10 15 20
shift (d)

TDW (mg m -2 d -1 )
250
200
150
100
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
total dry weight flux at OSP
with 16 day shift between
1000 m and 3800 m
50
1000 m
3800 m
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1000 m TDW
3800 m TDW
2
1
0
1000 m flux / 3800 m flux
31 day smooth
flux decreases
with depth
flux increases
with depth
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
seasonally averaged
change in flux
with depth

BSi (mg m -2 d -1 )
120
100
80
60
40
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
biogenic silica flux at OSP
with 16 day shift between
1000 m and 3800 m
20
1000 m
3800 m
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1000 m BSi
3800 m BSi
2
1
flux decreases
with depth
seasonally averaged
change in flux
with depth
0
1000 m flux / 3800 m flux
31 day smooth
flux increases
with depth
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

CaCO 3
(mg m -2 d -1 )
120
100
80
60
40
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
CaCO 3 flux at OSP
with 12 day shift between
1000 m and 3800 m
20
1000 m
3800 m
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
1000 m CaCO3
3800 m CaCO 3
2
1
0
1000 m flux / 3800 m flux
31 day smooth
flux decreases
with depth
flux increases
with depth
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
seasonally averaged
change in flux
with depth

POC (mg m -2 d -1 )
12
10
8
6
4
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
POC flux at OSP
with 16 day shift between
1000 m and 3800 m
Shoaling of seasonal thermocline
in Aug high primary production
(D Crawford, pers. comm.)
2
0
1000 m
3800 m
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
similar double peak but different
remineralisation dynamics than
presented yesterday at HOTS
1000 m POC
3800 m POC
3
2
1
flux decreases
with depth
seasonally averaged
change in flux
with depth
0
1000 m flux / 3800 m flux
31 day smooth
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

600
400
200
0
average 1982 1983 1984 1985 1986
1000 m fluxes
at OSP with merged
3800 m fluxes
TDW
mg m -2 d -1
600
400
600
400
200
0
1987 1988 1989 1990 1991
1000 m (measured)
3800 m (translated to 1000 m)
deployments per day
19 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
18
17
16
15
14
200
13
0
1992 1993 1994 1995 1996 1997
12
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
600
16+ years
400
200
0
1998 1999 2000 2001 2002 2003

1982 1990 2000
Southern Oscillation
Index
2
0
-2
Pacific Decadal
Oscillation
2
0
-2
BSi flux
anomaly
CaCO 3
flux
anomaly
1
0
-1
1
0
-1
POC flux
anomaly
1
0
-1
1982 1990 2000

1982 1990 2000
Southern
Oscillation
Index
2
0
-2
La Niña
El Niño
2
BSi/CaCO 3
flux
1
anomaly 0
1982 1990 2000
CONCERNS and ONGOING WORK
1. Large anomalies in general. Need to average fluxes over longer periods
How long
2. Large anomalies in winter. Longer averaging
3. 1988 is a period when 3800 m fluxes were translated to 1000 m.
Need to assure 1988 anomaly exists relative to the 3800 m time-series.

200 m fluxes at OSP
300
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
160
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
TDW (mg m -2 d -1 )
250
200
150
100
TDW
CaCO 3
(mg m -2 d -1 )
140
120
100
80
60
CaCO 3
1000 m
3800 m
200 m
50
1000 m
3800 m
200 m
40
20
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
120
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
40
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
BSi (mg m -2 d -1 )
100
80
60
40
20
BSi
1000 m
3800 m
200 m
POC (mg m -2 d -1 )
30
20
10
1000 m
3800 m
200 m
POC
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

OSP conclusions
• The seasonal cycle of flux at OSP occurs with spring and
summer peaks in the fluxes of BSi and POC, but only a
spring peak in the flux of CaCO 3 .
• Particles sink at rates of 100-400 m d -1 below 1000 m, with
the possibility CaCO 3 sinks faster than BSi and POC.
• POC follows BSi more closely than it follows CaCO 3 .
• The time series from OSP is long enough for correlations
with SOI and PDO. Preliminary analysis shows a greater
predominance of diatoms during La Niña a and
coccolithophorids during El Niño, consistent with Wong
and Crawford (2002).

TIME SERIES FROM:
‣Ocean Ocean Station PAPA
‣Alaska Gyre and North Pacific
Western Gyre
‣Line Line P with emphasis at La Perouse Bank

Alaska Gyre: 1990 to 1992
60
50
latitude
40
30
20
10+ yrs
~3 yrs
~1 yr
140 220
160 200
180
longitude
160
140
120

POC (mg m -2 d -1 ) CaCO 3
(mg m -2 d -1 ) BSi (mg m -2 d -1 ) TDW (mg m -2 d -1 )
300 Alaska Gyre
200
100
0
150
100
50
0
150
100
50
0
12
9
6
3
0
1990 1991 1992
mg m -2 d -1
200
180
160
140
120
100
80
60
40
20
0
55 0 00’ N
145 0 00’ W
trap depth = 3700 m
bottom depth = 3900 m
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
TDW
BSi
CaCO 3
POC x 10
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

3 stations in the Western Pacific
August, 1991 to June, 1992
60
50
WP-D
latitude
40
30
20
10+ yrs
~3 yrs
~1 yr
WP-A
WP-B
220 140
200 160
180
160
140
120
longitude
stations WP aka GEMS

North Pacific Western Gyre sediment-trap fluxes
WP D
WP A
WP B
51 0 25’ N
165 0 13’ E
4300 m
bottom depth = 5200 m
44 0 58’ N
165 0 00’ E
5300 m
bottom depth = 5860 m
45 0 05’ N
176 0 53’ W
5200 m
bottom depth = 5755 m
1500
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
250
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
250
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
BSi (mg m -2 d -1 ) TDW (mg m -2 d -1 )
1200
900
600
300
0
1200
900
600
300
1992 1991
200
150
100
50
0
200
150
100
50
1992 1991
200
150
100
50
0
200
150
100
50
1992 1991
CaCO 3
(mg m -2 d -1 )
0
120
90
60
30
0
40
30
20
10
0
40
30
20
10
POC (mg m -2 d -1 )
0
15
12
9
6
3
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
4
3
2
1
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
4
3
2
1
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

sediment composition
at deep traps
%BSi
100 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
75
50
25
W Pacific
4300 m – 5300 m
OSP
3800 m
Alaska Gyre
3700 m
%CaCO 3
0
100
75
50
25
0
6
%POC
4
2
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

TIME SERIES FROM:
‣Ocean Ocean Station PAPA
‣Alaska Alaska Gyre and North Pacific Western Gyre
‣Line P with emphasis at
La Perouse Bank

54
53
OSP
P20
P16
P12
P4
British
Columbia
52
51
50
49
Latitude (N)
48
146
144
142
140
138
136
134
132
130
128
126
124
122
Longitude (W)

La Perouse Bank
48 0 35’ N
126 0 24’ W
prior to 1991: 500 m
after 1991: 700 m
bottom depth = 1300 m
2000
1500
1000
500
0
2000
1986 1987 1988 1989 1990 1991
deployments per day
14 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
13
12
11
10
9
8
7
6
1500
1000
500
0
2000
1500
TDW
mg m -2 d -1
1992 1993 1994 1995 1996 1997
5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
11 yrs of data
1000
500
0
1998 1999 2000 2001 2002 2003

Resuspended fluxes at La Perouse Bank

no signal of primary
production in the bulk flux
at La Perouse Bank, and
up to 75% of the
sediment was lithogenic.
lithogenic sediment =
total - biogenic
2000
1500
1000
500
0
1000
500
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
TDW
mg m -2 d -1
lithogenic
mg m -2 d -1
biogenic sediment =
BSi + CaCO 3
+ 1.85 (POC)
0
1000
500
biogenic
mg m -2 d -1
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Resuspended sediment at La Perouse Bank
The composition of resuspended
sediment was modelled after the
samples where fluxes were
more than 65% lithogenic and the
total flux was > 900 mg m -2 d -1
100
80
the mean composition of
these samples was:
73% lithogenic
13% biogenic silica
8.3% CaCO 3
3.5% POC
%lithogenic
60
40
20
These were all in the winter
when the “downward” flux
would have been minimal.
0
0 500 1000 1500 2000
TDW (mg m -2 d -1 )

The downward flux (total – resuspended) at La Perouse Bank
BSi (mg m -2 d -1 ) "downward" TDW (mg m -2 d -1 )
1000 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
800
600
400
200
0
600
400
200
"downward" TDW (mg m -2 d -1 )
BSi (mg m -2 d -1 )
300 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
200
100
0
150
100
50
0
0
CaCO 3
(mg m -2 d -1 )
300
200
100
CaCO 3
(mg m -2 d -1 )
75
50
25
0
0
POC (mg m -2 d -1 )
80
60
40
20
POC (mg m -2 d -1 )
20
10
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

compilation from Line P
54
53
OSP
trap: 3450 m
bottom: 3890 m
P20
trap: 3100 m
bottom: 3550 m
P16
trap: 2700 m
bottom: 3300 m
P12
P4
British
Columbia
52
51
50
49
Latitude (N)
48
146
144
142
140
138
136
134
132
130
128
126
124
122
Longitude (W)

fluxes were recorded sporadically along line P
between 1995 and 1998
500
400
300
200
100
250
200
150
100
50
2000
150
100
50
0
station P 12
TDW
mg m -2 d -1
station P 16
TDW
mg m -2 d -1
station P 20
TDW
mg m -2 d -1
1995 1996 1997 1998
clogs!!!

ut sorted by Julian day…
500
400
300
200
100
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
P 12
TDW
mg m -2 d -1
300
200
100
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
P 12
fall peaks occurred
on different years
and were mostly
biogenic
0
400
300
P 16
0
200
P 16
200
100
100
0
0
400
300
P 20
200
P 20
200
100
100
0
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Summary of fluxes in the North Pacific:
mean fluxes measured at deepest traps at each station
distance from
Vancouver Island
maximum coccolithophorid
biomass along Line P
periodically occurs at P12
(M Lipsen, pers. comm.).
POC (mg m -2 d -1 ) CaCO BSi (mg m -2 d -1 ) TDW (mg m -2 d -1 3
(mg m -2 d -1 )
)
400
300
200
100
0
250
200
150
100
50
0
50
40
30
20
10
0
20
15
10
5
0
GD
GA
GB
AG P20
OSP P16
P12
P4
total flux at
La Perouse Bank
“downward”
flux

distance from
Vancouver Island
highest BSi
contributions occur
In the Western
Pacific
%BSi
%CaCO 3
BSi/CaCO 3
100
75
50
25
0
50
40
30
20
10
0
10
8
6
4
2
0
12
GD
GA
GB
AG P20 P12
OSP P16 P4
highest carbonate
contribution occurs
from P12 into the
Alaska Gyre
Foram
continuous plankton
Recorder might help
%POC
9
6
3
0

conclusions
• Fluxes in the Alaska Gyre at 55 o N were similar in
magnitude and composition to fluxes at OSP.
• Fluxes at 3 stations west of the dateline were composed
almost entirely of BSi. Total fluxes were very low near
the dateline but in the NW, ~3x higher than in the
eastern North Pacific.
• Fluxes along Line P decreased with distance from the
coast. The “downward” flux at La Perouse Bank shows
sping and summer peaks similar to those at OSP.
Highest carbonate fluxes throughout the North Pacific
were measured at P12 and P16.

light at OSP: 1960-1980
35000 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
30000
25000
MJ m -2 d -1
20000
15000
10000
5000
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

This slide will contain…
• Four figures (one for each constituent), exactly
like upper right of the previous four slides, but
with 200 m fluxes along with 1000 m and 3800
m. Point will be that the same signal exists at
200 m. A lot less attenuation than expected for
the 200 m – 1000 m depth interval. There are
some analytical issues we’re dealing with, so we
haven’t yet pushed the 200 m time series
through the calculations.

distance from
Vancouver Island
BSi/CaCO litho (mg m -2 d -1 3
)
100
10
1
0.1
10
8
6
4
2
0
2.0
GD
GA
GB
AG P20 P12
OSP P16 P4
CaCO 3
/POC
1.5
1.0
0.5
0.0
12
BSi/POC
9
6
3
0

La Perouse Bank: 1986 to present
60
50
latitude
40
30
20
10+ yrs
~3 yrs
~1 yr
220 140
200 160
180
longitude
160
140
120

1982 1990 2000
Southern Oscillation
Index
2
0
-2
Pacific Decadal
Oscillation
2
0
-2
BSi/CaCO 3
flux
anomaly
POC/CaCO 3
flux anomaly
BSi/POC flux
anomaly
1
0
-1
2
1
0
-1
1
0
BSi/CaCO 3
POC/CaCO 3
BSi/POC
-1
1982 1990 2000