The Future of Ocean Carbon Modelling in the Southern Ocean

1
The Future of Ocean Carbon Modelling in the
Southern Ocean
Richard Matear, CSIRO Marine and Atmospheric
Research
1
2
The Future of Ocean Carbon Modelling in the
Southern Ocean

• Outline
• Review the role of the oceans in the global
  carbon cycle
• Present the observed storage of carbon in the
  ocean
• Discuss the simulated CO2 uptake by the oceans -
  high variability in the Southern Ocean
• Present the projected response of the Ocean and
  Terrestrial Carbon Cycles to global warming
• Limitation of the present ocean carbon models and
  new questions to tackle
• Discuss the future developments of ocean carbon
  models

2
3
The Future of Ocean Carbon Modelling in the
Southern Ocean
The role of the oceans in the Global Carbon Cycle
and the uptake of anthropogenic CO2
3
4
Global Carbon Reservoirs

• Ocean are largest active reservoir of carbon
• Carbon stored in the ocean is 40 times greater
  than the anthropogenic carbon that has
  accumulated in the atmosphere in the last 200
  years.
• On millennium time-scales it is the ocean that
  determines the atmospheric level of CO2

5
The Oceans have the chemical capacity to
eventually take up 80-85 of the anthropogenic
carbon
Ultimate Fate of Anthropogenic CO2
6
Oceanic uptake is slow because of sluggish ocean
circulation
Sarmiento et al. (1992)
7
The observed oceanic storage of anthropogenic CO2
7
8
The southern mid- to high-latitude oceans are the
largest zonally-integrated storage region of
anthropogenic CO2.
Sabine et al., 2004
9
Anthropogenic CO2 is carried into the ocean
interior by water masses formed in the SO.
McNeil et al. 2000, Sabine et al. 2004
10
New estimates of Anthropogenic CO2 accumulation
between 1980-99 based on CFC-12 observations.
Pacific Basin
Atlantic Basin
McNeil et al, Science, 2003
11
Global Oceanic Uptake of Anthropogenic CO2 from
observations and models.
McNeil et al, Science, 2003
12
The changes in carbon storage (importance of
ocean observations)
Changes in the carbon storage
Sabine et al., 2004
12
13
The Simulated uptake of anthropogenic CO2 by the
oceans with and without climate change
13
14
OCMIP-2 simulations - ocean CO2 uptake w/o
climate effects
The Southern Ocean accounts for 50 of the
inter-model variance
Less agreementin the future
Agreementin the past
Orr et al. 2005
15
OCMIP-2 simulations - Global Oceanic Uptake of
Anthropogenic CO2 w/o climate effects
PgC/yr
Range between 2.15 to 2.82
Averaged uptake 2.0 0.4 and 2.5 0.5 PgC/yr,
for 1980s and 1990s, respectively
16
OCMIP-2 Simulated 1995 cumulative CO2 fluxes and
inventory
Large model differences in the Southern Ocean
(which accounts for gt40 of total uptake)
Orr et al. 2005
17
Assessing the oceanic anthropogenic CO2 uptake
Consistent models 1.9 0.2 and 2.2 0.2 PgC/yr
Matsummoto et al. 2004
18
Impact of Global Warming on the Global the Carbon
Cycle
19
Climate models suggest SO overturning will slow
down as a result of global warming.
Warming and freshening increases the high
latitude stratification, shutting down
AABW formation. Is this result realistic? Can we
observe the change in stratification? What are
the impacts?
Hirst and Matear (2001)
20
Climate Change Impact on Oceanic CO2 Uptake
No surface warming
Total Response
Warming Response
Matear and Hirst, 1999
21
(No Transcript)
22
460 PgC difference due Terrestrial Biosphere
23
Carbon Cycle response to global warming
The Ocean response to global warming is small (lt
100 PgC) compared to the terrestrial response
(460 Pg) But the marine biological
representation of the ocean carbon model is
extremely simple How big could the oceanic
uptake of carbon be?
23
24
(No Transcript)
25
Potential of Biology to increase Oceanic uptake
of Carbon
Increase export Production in Fe-limited
regions To exhaust all macro- nutrients 160
Pg C increase
Matear 1999
26
Global warming impacts on ocean carbon cycle
There is potential for the ocean carbon cycle to
feedback on climate but there are other important
questions independent of the oceanic uptake of
CO2 How will the marine ecosystems response to
direct impact of rising CO2 in the oceans? Will
other radiative important gases be affect by the
global warming? Both questions are very relevant
to the Southern Ocean
26
27
Chemistry Changes with Oceanic uptake of CO2
CO2
Increase from air-sea exchange
CO2 H2O ? CO32- 2H ? HCO3- H
Decrease
Important for calcification!
Ca2 CO32- ----gt CaCO3
Increasing atmospheric CO2
28
Pteropod kept in water undersaturated in
aragonite for 48 hours
Growing edge
Prismatic Layer
Apeture region
Control

• Evidence of dissolution of the aragonite shell
• Limacina helicina

Orr et al., 2005
29
Stability of CaCO3 in the oceans
Feely et al., 2004
30
Project response to elevated CO2
CO2.
Atmospheric CO2 Scenarios
Zonal pH changes
Zonal carbonate ion changes
Orr et al., 2005
31
Change in the Stability of Aragonite (CaCO3)
Orr et al., 2005
32
Excess CO3-2 for aragonite from the OCMIP
median for the IS92a scenario at 2100
Under- saturated
??1, (2100, S650)
??1, (2100, IS92a)
Orr et al., 2005
33
Sensitivity of Aragonite Stability to future
Atmospheric CO2 scenarios
600
Orr et al., 2005
34
Direct Impact of Elevated CO2 on Phytoplankton
(Coccolithophores)

• pCO2300ppm
• pCO2780ppm

Elevated CO2 levels in the upper ocean reduces
the calcification rate of phytoplankton
Riebesell et al, 2000.
35
Links between carbon cycling, biological
processes and cycling other elements
Interaction beyond just CO2 DMS increases by
100 in the Southern Ocean with Global Warming
(Gabric et al 2004)
36
Foodweb based carbon cycle models

• Question
• How will ocean biology respond to global warming
  and environmental change?
• Approach
• Three potential ways of pursuing this
• Mechanistic model of marine biology - difficult
  to validate and highly uncertain (e.g. Bopp et
  al. 2001)
• Relating climate change response patterns to
  observed modes of interannual variability(i.e.
  ENSO) - assumes pattern of global warming
  corresponds to the observed patterns of
  interannual variability (e.g. Boyd and Doney,
  2002)
• Develop empirical models which relate
  observations of chlorophyll and primary
  production to variables from the Climate model
  simulations (e.g. Platt and Sathyendranath, 1988)

36
37
Surface Phosphate (mmol/kg)

• Levitus
• Transient Run 1880-89
• Transient Run at 2570-80

38
Changes in Export production at 2100 with Global
Warming (NPZD foodweb model)
Bopp et al 2001
39
Optical Impact of phytoplankton on Ocean Dynamics
(comparison for with and without phytoplankton
absorption)
1.0
Potential for biology to Impact ocean dynamics
-1.0
200
-20
Manifredi et al (2005)
40
Plankton Functional Types
41
Biological model complexity
42
Statistical Approach

• Approach - divide the ocean into biomes based on
  physical characteristics which are further
  divided into biogeographical provinces within
  each a separate statistical chlorophyll model is
  developed
• Chlorophyll concentration estimate from the
  multiple linear regression of the log of SeaWiFS
  Chlorophyll with observed variables. Seven
  different variables are explored
• Sea surface properties SST, SSS, SSs,
• Nitrient supply upwelling, stratification, mixed
  layer depth
• sea ice cover
• Empirical based primary production (PP) algorithm
  is applied to estimate PP as a function of
  temperature, light supply, euphotic depth, and
  chlorophyll concentration
• Climate model output used to predict both the
  biogeographical province boundaries and the
  chlorophyll concentration in each province
• Chlorophyll distributions are used to calculate
  PP

43
Biome Definitions

• Divide the ocean into major biomes based on
  physical characteristics
• maximum winter mixed layer depth, MMLD
  (observations based on Levitus et al., 1998)
• The sign of Ekman divergence (observations based
  on Hellerman and Rosenstein 1983 model based on
  vertical velocity at 50 m
• Simple definition neglects the role of light
  supply and the supply of macro- and micro-
  nutrients

44
Biome Definitions

• Equatorial Biomes
• 5S to 5N, sub-divided into 2 biogeographical
  provinces according to whether the vertical
  velocity is up or down
• Permanently Stratified Subtropical Biome
• Vertical velocity down and MMLD less than 150m -
  low chlorophyll concentrations
• Seasonal mixed Subtropical Biome
• Downward vertical velocity and MMLD greater than
  150m
• Low Latitude upwelling biome
• 35S to 30N but excluding equatorial biome where
  vertical velocity is up
• Sub-Polar Biome
• South of 35S and North of 30N where vertical
  velocity is up
• Marginal Sea-Ice Biome
• At least partially covered by sea-ice
• Separate into basins gives a total of 33
  biogeographical provinces

45
Equatorial Downwelling
Equatorial Upwelling
Subtropical Permanently Stratified
Subtropical Seasonally Stratified
Low latitude upwelling
Sub-Polar
Marginal Sea-ice
46
Zonally integrated response of PP using empirical
chlorophyll and BF algorithm
Temperature dominates the response
47
Development of foodweb based ocean carbon model
will require
Observational approaches to monitor the
system How will we do this? Biogeochemical
models can be used to test monitoring approaches
and to identify observational strategies
47
48
How is new production measured?
Chlorophyll a
Sediment traps
Surface nutrients
Thorium234
49
New production estimates deviate most in the
Southern hemisphere
Global OBGC estimates
Satellite based estimate
Source Bopp et al. 2001, Global Biogeochemical
Cycles

• Schlitzer et. al, 2002 Model estimates higher
  than satellite-based estimates by a factor of 2
  and 5.

50
Calculating SNObio
SNObio Biological Seasonal Net Outgassing of O2
O2
CO2 H2O ? CH2O O2
Base of seasonal mixed layer
summer
spring
autumn
winter
51
Simulated SNObio
Units PgC/year
5
Atmospheric method
SNObio
Oceanic method (MOM3)
Perfect retrieval, g 0.5
0
NP
0
5
Roy et al 2005, prep
52
Observed changes in Dissolved Oxygen in the
Southern Ocean

• Observed changes between 1968 and 1995
• Dashed Lines are the predicted changes between
  the Control and Greenhouse Experiments for the
  1990s

59S
55S
59S
s1500
-10
-5
-15
-25
0
5
-20
Oxygen Change mmol/kg
Matear et al., 2000 G-cube
53
Temperature Change

• Change at 2000-2010
• Oxygen change

54
Observed Averaged Southern Ocean Oxygen
Concentrations Changes in the 1990s
K. Keller et al, draft
55
Southern Ocean Carbon Cycle

• SO is a high nitrate concentrations in the
  surface water - potential to stimulate the
  biological pump and increase uptake
• SO is an Important uptake region of CO2 which is
  connected to ocean dynamics and the uptake is
  sensitive to climate change
• SO is the first region where aragonite becomes
  unstable in the surface ocean under rising CO2
  levels
• SO phytoplankton have the potential to impact
  ocean dynamics
• Changes in SO sea-ice may affect phytoplankton
  production and ecosystem structure
• In the SO, DMS production is an important
  contributor to CCN - CCN doubles at 3x CO2
  atmosphere
• SO circulations changes are evident in BGC fields
  like dissolved oxygen

56
The Future of Ocean Carbon Modelling in the
Southern Ocean

• Present carbon models are insufficient to deal
  with important global warming impacts questions
  such as
• Ecosystem Impact of elevated CO2
• Potential changes in the structure and production
  of marine foodwebs
• The future carbon models will incorporate a
  foodweb based approaches (multiple phytoplankton
  functional groups and multiple elemental cycles)
• Such models will be more complex with many
  parameters
• Increased model complexity will require the
  development of metrics to assess the model and
  targeted observations to monitor the ocean carbon
  system
• Models can help design observing strategies

56
57
(No Transcript)
58
Simulated Oceanic uptake of Anthropogenic CO2 for
the 1980s (no climate change)
Global Oceanic Uptake For the 1980s
Princeton Hadley IPSL MPI CSIRO Average 1.8
Uptake (Gt C /yr) 2.2 2.1 1.5 1.6 1.6
Zonal Uptake (Gt C/yr/deg)

• Largest variability occurs in the Southern Ocean

Orr et al (2001)
59
3-Component Foodweb Model
Input
1 phytoplankton and 1 zooplankton functional
group with varying PNC uptake
Export
60
Phosphate Concentration (mmol/kg)
Surface concentration
61
The SOIREE algal bloom
March, 1999 80 km across 61 South 140 East Boyd
et al., Nature 2000. Technically sweet! .