Transcript Ribera2

Long-term observations in coastal
areas: a key tool for understanding the
functioning of a system and detecting
changes
Maurizio Ribera d’Alcalà
Stazione Zologica Anton Dohrn, Napoli, Italy
The Universe of Ptolemaeus
Message 1
Careful observations over time allows
for detecting patterns on which it is
possible to build semiempirical
reconstruction of the time course of a
process
The Universe of Kepler
Message 2 and 3
1. A reanalysis of the ‘same’
observations may stimulate an
improved reconstruction of the time
course of a process
2. In turn, the reconstruction may
become the basis for detecting a
mechanism beyond the pattern
Hidden and visible Sapropels
Cramp & Sullivan, 1999
Message 4
It is not necessary to observe a
process while it is occurring, as long as
it has left traces in time. Still, we have
to gather observations on the traces.
Isotopic signal in deep sea cores
Lourens, 2004
Laskar, 1993
Message 5
Also records may play the role of
suggesting mechanisms. Though, they
are not always testable
The basis of modern biological oceanography
Sverdrup, 1959
Message 6
A hypothesis on a mechanism behind a
process, again stimulated by
observations, may be tested with a
time series of other observations
Napoli time series Marechiara
°
40.90
Naples
N
200
MC
100
100
Ischia
200
°
40.70
500
Ad
r
Ty
rrh
en
ian
°
40.50
°
13.80
iat
ic
Se
a
Se
a
200
500
Capri
°
14.00
°
14.20
Sampled fortnightly in
1984-1991 and weekly
from 1995 to present
100
200
500
°
14.40
Monthly averages
Salinity 0-10 m
Temperature 0-10 m
Mixed Layer Depth (m)
Monthly averages
µmol L-1
TIN 0-10 m
SiO4 0-10 m
PO4 0-10 m
The seasonal cycle of chl.a
Chla - 0 m
mg m-3
Int Chla 0-70 m
mg m-2
Phytoplankton - 0m
Abundance
Biomass
μg C l-1
Cells ml
-
50000
1
5000
500
100
Percentage
Phytoplankton Biomass - 0m
100
80
60
40
20
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Diatoms
Dinoflagellates
Coccolithophores
Other flagellates
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
DIATOMS
DINOFLAGELLATES
COCCOLITHOPHORES
OTHER FLAGELLATES
Message 7
Phytoplankton may be differently
distributed in the water column in
different seasons, therefore surface
chl.a concentration may mislead the
interpretation of the seasonal cycle
Species phenology
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
J
F M
A M J
J
A S O N DJ
J
F M
A M J
J
A S O N JD F M
F M
A M J
J
A S O N DJ
F M
A M J
J
A S O N D
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
A M J
J
A S O N D J
F M
A M J
J
A S O N D
Message 8
Phytoplankton species display a
phenology
Trophic conditions
TIN
START OF THE BLOOMS
16
10
14
12
8
mol/dm3
DECLINE OF THE BLOOMS
10
6
8
4
6
4
2
2
0
0
Max
Min
75th %
25th %
Median
Zingone, Sarno, Nardella & Licandro, in preparation
Temperature
START OF THE BLOOMS
DECLINE OF THE BLOOMS
30
28
28
26
26
24
24
°C
22
20
22
20
18
18
16
16
14
14
12
12
Max
Min
75th %
25th %
Median
Zingone, Sarno, Nardella & Licandro, in preparation
Synchronous patterns in the occurrence of the species
FV
PO
FS
cells ∙ ml-1
NA
FP
PT
PL
30
1800
1800
20
1200
1200
10
600
600
0
0
0
400
800
800
200
100
400
0
0
0
400
4000
4000
200
2000
2000
0
200
0
4000
0
4000
100
2000
2000
0
1000
0
0
100
100
500
50
50
0
0
0
40
2
2
20
1
1
0
40
0
0
2
2
20
1
0
MA
2002
MA
‘02
MA
2003
MA
‘03
MA
2004
MA
‘04
MA
2005
MA
‘05
NA
FS
FP
P
PL
cells ∙ ml-1
PO
1
0
J
2001
J
‘01
AM J
2002
AM J
2003
AM J
2004
AM
2005
0
‘01
4500
600
400
3000
400
200
1500
200
0
400
0
0
20000
1600
200
10000
800
0
400
0
20000
1000
200
10000
500
0
0
0
1000
10000
600
500
5000
300
0
400
0
2000
100
200
1000
50
0
0
0
60
200
4
30
100
2
0
0
0
60
50
4
30
25
JJ
2001
JJ
‘01
M JJ
2002
MJJ
‘02
M JJ
2002
MJJ
‘03
M JJ
2002
MJJ
‘04
M
2005
M
‘05
0
MJ J
2002
MJJ
‘02
MJ J
2003
MJJ
‘03
MJ J
2004
M
2005
‘04
‘05
MJJ
M
Cerataulina pelagica
600
0
JJ
2001
JJ
AMJ AMJ AMJ AMJ
‘02
‘03
‘04
‘05
Leptrocylindrus danicus
Pseudoscurfieldia marina
FV
Chaetoceros throndsenii
Prorocentrum triestinum
Bacteriastrum furcatum
0
0
2
JA
2001
JA
‘01
JA
2002
JA
2003
JA
2004
‘02
‘03
‘04
JA
JA
JA
0
JA
2001
JA
‘01
JA
2002
JA
‘02
JA
2003
JA
‘03
JA
2004
JA
‘04
Siano et al., in prep.
Message 9
Species growth and accumulation in
phytoplankton is not linearly linked
with proximate abiotic factors. Indeed
plankton display the cabability to
extert control on their life cycle
Looking for trends
µg L-1
Chla 0m
PSU
Salinity 0m
Looking for trends
Message 10
Time series, even of basic
hydrographic parameters highlight
mid-term trends, if any
cells ml-1
Looking for trends
Phytoplankton abundance
Looking for trends
Diatom abundance
Mean diatom size
100
90
19
17
15
13
11
50
9
40
7
30
20
5
10
3
0
1
ESD (µm)
-1
60
cells 10 ml
70
3
80
Total carbon
< 10 µm
10 - 15 µm
> 15 µm
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1991
1990
1989
1988
1987
1986
1985
1984
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Message 11
Size may be an indicator of possible
change in the structure of plankton
community
Factors/Issues
Climate Change
Atmospheric Deposition
Basin Scale Oscillations Alien Species
Land Based Inputs
Hydrological Cycle (and flushing)
Seasonality
Coherence
Episodic vs. Chronic
Salinity and Precipitation
Seasonal shift in wind direction
Interannual variability in wind direction
Nutrient load
tonn y-1
Indirectly estimated loads
6000
5000
4000
3000
2000
1000
0
N load
P load
1971
1981
1991
1996
Large scale dynamics
Rio et al., 2006
Message 12
There is a wealth of easily accessible,
easily produceable data that can help
understanding what modulates the
system functioning
Salinity and Chlorophyll a shifts
2 sample Kolmogorov-Smirnov test
Null Hyp.: Parameter 1984-1988 true distrib. funct. is not > true distrib. funct. 2003-2006
Chl.a (0 m) 1984-1988 is stochastically > Chl.a (0 m) 2003-2006 p=0.003
Chl.a (0-70 m) 1984-1988 is stochastically > Chl.a (0-70 m) 2003-2006 p=0.03
S (0-10 m) 1984-1988 is stochastically > S (0-10 m) 2003-2006 p=0.06
Phase variability (decade of the month)
T min
Start of
Stratification
Max time for Tsurf=III°
End low
S
1984
March II
April II
November II
July 4
1985
March II
April II
November II
July 4
1986
March II
April II
November II
July 4
1987
March I
April I
November II
July 4
1988
March I
April I
November I
July II
1989
March I
April I
October I
July I
1995
April II
November I
July I
1996
March I
August II
1997
March I
May I
October II
July I
1998
March II
April II
October II
July I
1999
February II
April III
October III
July I
2001
March I
May I
November III
August I
2002
January I
April I
October III
July I
2003
March II
April II
October II
giugno I
2004
February I
April II
November II
July III
2000
Idealized shapes
Thalassiosira rotula
Calciopappus caudatus
Protoperidinium diabolus
Phaeocystis sp.
Shape relative abundance
Interannual variability in composition
Message 13
Very simple analyses with easy to
determine indicators and public
domain data may help in
characterizing patterns and, possibly,
in unveiling underlying mechanisms
Phenology of Centropages typicus
Mazzocchi et al., 2007
Large scale forcing
Mazzocchi et al., 2007
The Zooscan
• Copepod automatic Identifier (TP 93%)
• Size as a community descriptor
Number
Copepod size
spectrum
Size (ESD)
Digitalisation of
>600 samples
with the
ZOOSCAN
Total copepod
abundance
Diversity of size
classes
Shannon index
(Parson, 1969, Ruiz, 1994)
1 Size class ~ 1 Species
H '    p log
p
i
2 i
Message 14
Not always new technologies are
expensive and some may give
significant information
HABs dynamics
A misleading view regarding HABs is that
they are always irregular, unpredictable
and conspicuous events involving the
accumulation of highly concentrated
populations. In fact, many of the highly
toxic species often constitute a regularly
occurring component of normal
phytoplankton populations and can exert
their impact at low cell concentrations
(100-1000 cells·l-1)
Zingone & Wyatt, 2006
HABs dynamics
Harmful species are not equally dangerous
throughout the year, rather they have
generally one/several
predictable/unpredictable periods of the
year when they may exert their harmful
effects
Zingone & Wyatt, 2006
The history is recorded in the sediments
Zingone & Wyatt, 2006
HAB forming species and HABs
Zingone & Wyatt, 2006
Message 15
The presence of harmful species at
given sites is a necessary but not
sufficient condition for the
development of harmful algal blooms,
so that the geographic distributions of
HABs do not necessarily strictly reflect
those of the causative species
Zingone & Wyatt, 2006
Responses to HABs
Intense monitoring activities of causative
organisms and/or toxins
Development of alert systems based on
automated observations coupled with
predictive models that can expand the
lead-time to harmful events, so as to allow
more cost effective mitigation operations
Progress in modelling is however seriously
hampered by the lack of knowledge on the
basic mechanisms underlying the
development of specific algal blooms
Zingone & Wyatt, 2006
An intriguing species
80
55
75
50
70
45
65
40
60
35
55
30
50
25
45
700
40
600
35
20
30
500
Percentage of cells
Size classes (µm)
P. multistriata cell size distribution 1996-2006
15
10
100
200
300
400
500
600
100
200
300
400
500
600
5
Cell/ml
400
300
200
100
0
0
Time
‘96 ‘97 ‘98 ‘99 ‘00 ‘01 ’02 ’03 ’04 ’05 ’06
D’Alelio et al., in prep.
Possible scenarios
a
100
b
c
50
0
0
0
300
300
300
360
360
360
660
660
660
720
720
720
1020
1020
1020
1080
1080
1080
1380
1380
1380
1440
1440
1440
1740
1740
1740
100
50
100
50
100
50
100
50
%
100
50
100
50
100
50
100
50
100
50
100
50
0
1800
30
40
50
60
70
80
1800
30
40
50
60
70
80
1800
30
40
50
60
70
80
Cell size (µm)
D’Alelio et al., in prep.
Model for hindcasting
80
Cell size (µm)
70
60
50
40
30
28-Oct-95 11-Mar-97
24-Jul-98
06-Dec-99 19-Apr-01 01-Sep-02 14-Jan-04 28-May-05 10-Oct-06
Time
D’Alelio et al., in prep.
Message 16
Very simple models integrate
observations and help in testing
hypothesis on mechanisms
A highly seasonal species
Temora stylifera
total population abundance at st. 'MC' (Gulf of Naples)
900
800
700
ind. m
-3
600
500
400
300
200
100
0
1984
’85
’86
’87
’88
’89 ’90
’97
’98 ‘99
Mazzocchi et al., 2006
A highly seasonal species
Mean seasonal cycle of T. stylifera at st. ‘MC’
400
40
350
juv.
ind. m
-3
300
30
250
200
20
150
100
10
50
0
0
J
F
M
A
M
J
J
A
S
O
N
D
Mazzocchi et al., 2006
Another hindcasting model
THE MODEL STRUCTURE
stages
1
eggs
2
NI
3
NII-NVI
4
CI
5
CII-CV
6
adults (f, m)
Mazzocchi et al., 2006
Simple rules
Physiological age of an individual in stage i at time t :
Average duration in stage i : D
For each individual, assumed known its stage i and its
X tj
j
physiological age, X tj the age at time t  t is given by:
X
j
t  t
 t

j
j
 X t  max0;     t  t 
 Dj

j
Mortality
• Physiological/food dependent mortality
• Predation mortality
Mazzocchi et al., 2006
Simple rules
For the adult stage (stage 6):
qt
X 
F
n
t
where:
q t = number of eggs produced by a female at time t
tnL
F   f ( s )ds
n
t
with
L = female average life span
f(t) = average reproductive profile of the female
Mazzocchi et al., 2006
Optimal diet
3500
RESULTS
Total population
3000
PRO
ind. m-3
2500
2000
1500
1000
500
0
60
30
150
120
90
180
days
60
400
stage 5
350
stage 6
50
300
ind. m-3
ind. m-3
40
250
200
150
30
20
100
10
50
0
0
30
30
60
90
120
days
150
180
60
90
120
150
180
days
Mazzocchi et al., 2006
The real world
Mazzocchi et al., 2006
Message 17
Analyzing life cycle of key species may
help understanding causative factors
that modulate the cycle
Synthesis
1. Marine environment is complex, dynamic and
builds its own history
2. What part of the history we want to read
depends on our priorities, but without events
there is no history
3. Many events can be detected, monitored,
characterized while they are occurring or
reconstructed by their traces with very little
effort
4. A huge amount of information has been
accumulated in the last decades and is freely
avaliable, given a web access
5. Helping in selecting priorities and sharing the
information is the needed contribution we
can provide