Transcript Climate change projection of phytoplankton concentrations and

1
Climate Change Projections of the
Tasman Sea from an Ocean Eddyresolving Model – the importance of
eddies
Richard Matear, Matt Chamberlain, Chaojiao Sun, Ming Feng
CSIRO Marine and Atmospheric Research
Sun et al 2012, Chamberlain et al 2012, Matear et al., 2013 in press JGR
Global Warming Trend
• Maximum warming in the Western Boundary Current regions
due shifts and intensification of the WBC
Wu et al., NCC 2012.
3
Outline
1. Why use Ocean Eddy-resolving Model?
• To resolve important processes like Boundary
Currents and Eddies
2. How do we project climate change with an Ocean
Eddy-resolving model
• Use climate anomalies from a global climate
model projection to drive high-resolution model.
3. Consequence of resolving boundary currents and
eddies
• Sea surface temperature
• Phytoplankton
4
Oceans around Australia – climate and highresolution models
• Climate model captures large-scale ocean circulation, but misses
the boundary currents (e.g. Leeuwin and East Australia Currents)
and eddies
Sun et al. 2012.
5
Method of projecting Climate Change in the
Ocean Eddy-Resolving Model
Models used:
Chamberlain et al. 2012
• Global Climate Model (GCM) – CSIRO Mk3.5
• 1°x 2° horizontal resolution ocean model
• Used output from the SRES A1b, “integrated world, balanced energy sources”
emission scenario, IPCC’s AR4
• calculated climate change anomalies from the GCM and used them to force the
ocean eddy resolving model (minimise the effect of model bias in eddy-resolution
projection). – Anomalies include change in Ocean State (T,S, N, Phytoplankton,
Zooplankton) and changes in forcing (Heat, Freshwater and winds)
• Simulations presented for the decade of 2060s
• Ocean Eddy-resolving Model (OEM) – BlueLINK’s Ocean Forecasting
Australia Model (OFAM1.0).
• Global domain with 10-km
resolution around Australia.
• Present-day state simulated with observed
forcings
• Future state adds anomalies to present
day
Change in Boundary Currents – e.g. EAC
6
GCM
OEM
Multi-year averages
• GCM missing fine structure shown in the OEM.
• GCM has increased in the EAC extension flow along the coast of
Australia
• OEM the increased flow in the EAC extension is due to eddies
Sun et al. 2012.
7
Change in upper ocean Temperature: OEM –
observation comparison
• Model
Observed
multiyear averages
Matear et al. 2013, JGR
8
Change in upper ocean Temperature: 2060s –
1990s
• OEM
GCM
multiyear averages
Matear et al. 2013, JGR
9
Observed SST trends (°C / century
Observed warming in
Tasman in last 30 years
closely resembles the
projected warming
Correlation with projected
change
GCM – 0.65
OEM – 0.74
Matear et al., 2013 JGR
10
Mixed Layer Depth Maximum (m): model
verus observations
• OEM
Observed
multiyear averages
Matear et al. 2013, JGR
11
Change in Mixed Layer Depth (m): Maximum
and January
• Seasonal Maximum
Jan.
Matear et al. 2013, JGR
12
Change in Mixed Layer Depth (m): 2060s –
1990s from OEM
• STZ 40S 145-170E
SAZ 50S 145-170E
Matear et al. 2013, JGR
13
Eddy Kinetic Energy: 1990s and change
• OEM
Observed
Matear et al. 2013, JGR
14
Change in Annual Mean Phytoplankton (mmol
N/m3)
• OEM
GCM
PP increases by 10% in the Tasman Sea (30Matear et al. 2013, JGR
50S and 145 – 170E
15
Change in Phytoplankton (mmol N/m3)
• STZ 40S 145-170E
SAZ 50S 145-170E
Matear et al. 2013, JGR
16
Conclusions
Ocean Eddy-resolving Model alters the climate projection from
the coarse resolution GCM by
•Changing the upper ocean warming (less warming along
Tasmania)
•Changing the East Australian Current (EAC) response –
increased EAC and increased EAC extension (more eddies)
•Increasing the phytoplankton concentrations north of the SubTropical Front due to increased nutrient supply from eddypumping.
•Primary Production in the oligotrophic Tasman Sea increases by
10% with climate change rather than declining as projected by
the GCM
Thank you
Phytoplankton– model vs observed
chlorophyll
• OEM
Observed
Phytoplankton– model vs observed
chlorophyll
• OEM
Observed
Change in Pacific phytoplankton
concentrations
2060s
Model BGC shows
region of subsurface
phytoplankton in
western Pacific.
Shoaling of thermocline
with climate change
makes relative
subsurface response
greater than surface.
2060s-1990s
mmol(NO3)/m3
10 ICSHMO. Climate Change Science to Adaption
mmol(NO3)/m2
Oceans around Australia - schematic
10 ICSHMO. Climate Change Science to Adaption
Website imos.org.au
22
Method of projecting Climate Change in the
Ocean Eddy-Resolving Model
GCM
Mk3.5
OEM
OFAM
• Use change in ocean state (2060s – 1990s) of
temperature, salinity and biogeochemistry to define
OEM projection initial condition.
• Use change in surface fluxes (heat, freshwater,
wind stress) to modify OEM fluxes.
• Spinup experiment
• Use ‘observed’ fluxes.
• Diagnose correction fluxes by restoring to
observed surface temperature + salinity.
• Projection (2060s)
• Modified initial condition.
• Observed fluxes + correction + climate
anomalies (+ feedback)
10 ICSHMO. Climate Change Science to Adaption
Chamberlain et al. 2012
EAC current
GCM
OEM
Multi-year averages
• GCM missing fine structure shown in the OEM.
10 ICSHMO. Climate Change Science to Adaption
Sun et al. 2012.
Change in Boundary Currents – southward
transport of EAC
• Both GCM (dashed) and OEM (solid)
show increase in EAC transport, but
differ in the details.
• GCM – increased EAC only south of
32S – increase in the southward
extension of the EAC
• OEM - increased EAC north of 32°S
with most of the increased flowing east
between 32 - 34° S. Increased EAC
extension flow south of 36S
Black – 1990s
Red – 2060s
10 ICSHMO. Climate Change Science to Adaption
Sun et al. 2012.