Productivity in a Changing Southern Ocean Kevin R. Arrigo Stanford University 1
Productivity in a Changing Southern Ocean A Paleo-perspective Satellite view of the Southern Ocean Role of ice and iron Controls on production The Ross Sea The Amundsen Sea Future changes? 2
Glacials: Antarctic Zone Reduced overturning Low productivity High ice cover Little air-sea CO 2 exchange A Paleo-Perspective Subantarctic Front Antarctic Convergence 3 1 Subantarctic Zone Greater dust flux High productivity Strong biological pump Large net CO 2 sink 1..1 Antarctic Zone Subantarctic Zone Chlorophyll a (mg m -3 ).1
A Paleo-Perspective Interglacials: Antarctic Zone Greater overturning Higher productivity Lower ice cover Weak biological pump Large net CO 2 source Subantarctic Zone Reduced dust flux Lower productivity Weak biological pump Small net CO 2 sink Subantarctic Front Antarctic Convergence 3 1 1..1 Antarctic Zone Subantarctic Zone Chlorophyll a (mg m -3 ).1
A Paleo-Perspective Interglacials: Antarctic Zone Greater overturning Higher productivity Lower ice cover Weak biological pump Large net CO 2 source Subantarctic Front Antarctic Convergence 3 1 Subantarctic Zone Reduced dust flux Lower productivity Weak biological pump Small net CO 2 sink TODAY: Southern Ocean is neutral or a net CO 2 source Antarctic Zone Subantarctic Zone Chlorophyll a (mg m -3 ) 1..1.1
.5 Pg C yr -1 CO 2 sink -.6 Pg C yr -1 +.1 Pg C yr -1 NPP g C m -2 yr -1 15 125 1 Antarctic Convergence NPP Climatology 1998-213 75 5 25 Takahashi et al. (29) 6
.5 Pg C yr -1 CO 2 sink -.6 Pg C yr -1 +.1 Pg C yr -1 NPP g C m -2 yr -1 15 125 1 Antarctic Convergence 75 5 Doesn t account for coastal hot spots Very efficient biological pumps Takahashi et al. (29) 25 7
Ito et al. (21) Neither do most models (even high resolution ones)
-.6 Pg C yr -1 +.1 Pg C yr -1 NPP g C m -2 yr -1 15 125 1 Antarctic Convergence 75 5 Ross Sea: -.13 Pg C yr -1 CO 2 sink 25 Arrigo et al. (28) 9
21 2 19 18 17 16 15 14 Interannual variability in NPP Southern Ocean R² =.7 1 Annual NPP (Tg C) 1997-1998 1998-1999 1999-2 2-21 21-22 22-23 23-24 24-25 25-26 26-27 27-28 28-29 29-21 21-211 211-212 212-213 Mean = 1825 Tg C yr -1 Low interannual variability (CV = 4%) No significant change over time
21 2 19 18 17 16 15 14 R² =.7 65 6 55 R² =.71 5 45 4 1997-1998 1998-1999 1999-2 2-21 21-22 22-23 23-24 24-25 25-26 26-27 27-28 28-29 29-21 21-211 211-212 212-213 1998 2 22 24 26 28 21 212 Interannual variability in NPP Southern Ocean Arctic Ocean Mean = 1825 Tg C yr -1 Mean = 524 Tg C yr -1 Low interannual variability (CV = 4%) High interannual variability (CV = 11%) No significant change over time Significant 3% increase since 1998 11 Annual NPP (Tg C)
Interannual variability in NPP 65 Arctic Ocean R² =.74 Annual NPP (Tg C) 6 55 5 45 4 5.E+6 6.E+6 7.E+6 8.E+6 Open Water Area (km 2 ) 65 6 55 Arctic Ocean R² =.72 5 45 4 75 1 125 15 Growing Season Length (days) 12
Interannual variability in NPP Southern Ocean 21 Annual NPP (Tg C) 2 R² =.15 19 18 17 16 15 3.1E+7 3.2E+7 3.3E+7 3.4E+7 3.5E+7 Open Water Area (km 2 ) 21 R² =.14 2 19 Southern Ocean 18 17 16 15-2 2 4 SAM Index 13
Interannual variability in NPP Southern Ocean 21 Annual NPP (Tg C) 2 R² =.15 19 18 17 16 15 3.1E+7 3.2E+7 3.3E+7 3.4E+7 3.5E+7 Open Water Area (km 2 ) 21 2 19 Southern Ocean R² =.6 18 17 Difficult to predict trajectory of change 16 15-2 -1 1 2 3 ENSO Index 14
Interannual variability in NPP The exceptions are the continental shelves Ross Sea Continental Shelf Annual NPP (Tg C) 35 3 25 2 15 1 5 R² =.81 1 2 3 Open Water Area (km 2 ) The amount of ice cover is very important (Just like in the Arctic Ocean - which has a large shelf area) 15
Interannual variability in NPP Shelves are also the most variable Coefficient of Variation Pelagic Shelf Weddell Sea 12% 5% South Indian Ocean 6% 36% Southwest Pacific Ocean 6% 4% Ross Sea 7% 34% Bellingshausen-Amundsen Sea 5% 34% 16
Controls on NPP Pelagic Southern Ocean NPP largely limited by iron availability Evidenced by numerous ocean fertilization experiments What about continental shelves? Boyd et al. (212) 17
Controls on NPP Shelves are more light-limited than pelagic Southern Ocean More sensitive to variations in ice cover Shelves have higher Fe concentrations more Fe sources Circumpolar Sediments (CDW) Smith et al. (212) 18
Controls on NPP The Ross Sea and Amundsen Sea Sea ice cover changes Phytoplankton responses to Fe Amundsen Sea Ross Sea 19
Changes in Sea Ice Cover Bellingshausen Sea Amundsen Sea Ross Sea Length of sea ice season (since 1979) Stammerjohn et al. (212) 2
Changes in Sea Ice Cover Influenced by Southern Annular Mode (SAM) + SAM Positive phase: Cold southerly winds blow out of Ross Sea H H More sea ice Warm northerly winds blow into Amundsen Sea Less sea ice H L Ross Sea + SAMs are becoming more common 21
Ross Sea Continental Shelf 22
Ross Gyre CDW 23
1997-1998 1998-1999 1999-2 2-21 21-22 22-23 23-24 24-25 25-26 26-27 27-28 28-29 29-21 21-211 211-212 212-213 2 18 16 14 12 1 8 6 4 2 Ross Sea Pelagic Shelf MIZ-Pelagic MIZ-Shelf 5 45 4 35 3 25 2 15 1 5 Pela She MIZ- MIZ- 24 Annual NPP (g C m -2 yr -1 ) 1997-1998 1998-1999 1999-2 2-21 21-22 22-23 23-24 24-25 25-26 26-27 27-28 28-29 29-21 21-211 211-212 212-213 Annual NPP (Tg C)
Ross Sea Shelf and pelagic respond differently to SAM Retreat (DJF) Advance (MAM) +SAM L L Earlier Later 25
Ross Sea Stronger southerly winds blow ice off shelf when SAM is more positive Winds are cold so ice doesn t melt and it piles up in pelagic Open Water Area (1 5 km 2 ) 2. 1.5 1..5 72 7 68 Shelf R 2 =.36 p=.17 66 64 Pelagic R 2 =.31 p=.3-2 -1 1 2 3 SAM (Oct-Mar) 26
Ross Sea Primary Production (Tg C) 5 1 15 2 25 NPP (Tg C yr -1 ) Primary Production (Tg C) 5 1 15 2 25 NPP (Tg C yr -1 ) R 2 =.32 p=.26-2 -1 1 2 3 SAM (Oct-Mar) R 2 =.85 p<.1 SAM (Oct-Mar) NPP on the shelf significantly correlated with SAM Higher NPP when SAM is more positive Due to: Strong relationship between open water area and NPP.5 1. 1.5 2. Open Water Area (1 5 km 2 ) Open water area (1 5 km 2 ) Less ice = More NPP 27
Ross Sea Response to Fe? Iron limits phytoplankton growth in Ross Sea Especially at high temperature Rose et al. (29) 28
Ross Sea Response to Fe? Diatoms come to dominate community Phaeocystis antarctica declines, especially at high temperature Are only diatoms Felimited? 19 Hex/Chl a Fuco/Chl a Rose et al. (29) T control High Fe High Temp High Both 29
Ross Sea Response to Fe? Diatoms respond strongly to Fe addition at both high and low light Phaeocystis did not respond to Fe additions Apparently, only diatoms are Felimited in Ross Sea There is no impact of CO 2 concentration Feng et al. (21) 3
Amundsen Sea Continental Shelf 31
Amundsen Sea 32
Amundsen and Pine Island polynyas Amundsen Sea ANTARCTICA 25 1 Pine Island Polynya 1 Amundsen Polynya.1 Pine Island Glacier (PIG) just calved in July 213 72 km2 Chlorophyll a (mg m-3).1 33
Amundsen Sea Ice edge Amundsen Polynya Photo: Dave Munroe Highest chlorophyll a concentrations in Southern Ocean Intense blooms dominated by colonial Phaeocystis antarctica 34
Amundsen Sea Annual NPP (Tg C yr -1 ) Annual NPP (g C m -2 yr -1 ) Amundsen Polynya (gray) Pine Island Polynya (black) Light ice year Heavy ice year Arrigo et al (212) 35
Amundsen Sea Amount of annual NPP controlled by sea ice cover Amundsen Polynya Pine Island Polynya Annual NPP (Tg C yr -1 ) Annual NPP (g C m -2 yr -1 ) Arrigo et al (212) Open water area (km 2 ) Open water area (km 2 ) 36
Amundsen Sea Arrigo et al (212) Light ice year - Negative SAM Cold winds blow offshore clear out sea ice from coast 37
Amundsen Sea Positive SAM year Heavy sea ice Much lower productivity Arrigo et al (212) 38
Amundsen Sea Response to Fe? Amundsen Sea ANTARCTICA 25 1 DynaLiFe 13 Jan 18 Feb 29 Pine Island Polynya 1 ASPIRE 14 Dec 21 5 Jan 211 Amundsen Polynya.1 Arrigo et al (212) Chlorophyll a (mg m -3 ).1 39
Satellite Primary Production Open Water Area (x 1 km 2 ) 8 6 4 2 8 6 4 2 Pine Island Polynya Amundsen Polynya ASPIRE DynaLiFe NPP OW 3 2.5 2 1.5 1.5 3 2.5 2 1.5 1.5 NPP (g C m -2 d -1 ) Arrigo et al (212) Oct Nov Dec Jan Feb Mar 4
Amundsen Sea Amundsen Polynya ASPIRE pco 2 (ppm) Chl a (µg L -1 ) Getz Dotson Getz Dotson 3 1 1..1 5 3 1 Upwelled MCDW outflow in front of Dotson Ice Shelf Low phytoplankton biomass High biomass in central polynya (>2 µg Chl a L -1 ) 41
Fe addition bioassay experiments POC (µg L -1 ) 2 15 1 5 Amundsen Sea Amundsen Polynya 14 Dec * 2 1 15 1 5 -Fe +Fe -Fe +Fe 26 Dec 3 * -Fe +Fe -Fe +Fe Surface Sub surface Alderkamp et al. (in prep.) POC (µg L -1 ) 2 15 1 5 2 1 Jan 4 5 Jan 5 15 1 5 -Fe +Fe -Fe +Fe Amundsen Polynya is Fe-limited in some locations 42
DynaLiFe Amundsen Sea Pine Island Polynya Mills et al. (212) 43
Amundsen Sea Pine Island Polynya pco 2 (µatm) Chl a (µg L -1 ) Tortell et al. (212) 44
Amundsen Sea Pine Island Polynya CDW upwells onto shelf CDW upwells to the surface in front of the PIG Depth (m) Depth (m) Depth (m) 1 2 3 1 2 3 1 2 3 CDW CDW CDW Temperature ( C) 1. Salinity (g kg -1 ) 25 2 15 1 5 Distance from PIG (km) Gerringa et al. (212) 45. -1. -2. 34.8 34.4 34. 33.6 Chl a (mg m -3 ) 15 1 5
Amundsen Sea Pine Island Polynya DFe (nm).8 Depth (m) 1 2.6.4.2 3 TDFe (nm) 4 Depth (m) 1 2 3 2 1 3 Gerringa et al. (212) 25 2 1 5 Distance from PIG (km) 46
Amundsen Sea Pine Island Polynya Dissolved iron Total dissolvable iron Distance from Pine Island Glacier (km) Gerringa et al. (212) Concentrations decrease with distance from Pine Island Glacier Both dilution and phytoplankton uptake 47
Amundsen Sea Pine Island Polynya Fe addition bioassay experiments Chl a (µg L -1 ) 2 15 1 5 1 m water -Fe +Fe -Fe +Fe -Fe +Fe 5 days 2 2 Jan 31 Jan 15 1 5 Chl a (µg L -1 ) 2 15 1 5 2 15 1 5 Mills et al. (212) 15 Jan -Fe +Fe 7 Feb -Fe +Fe -Fe +Fe Phytoplankton in Pine Island Polynya not Fe-limited NO 3 drawn down to zero -Fe +Fe 48
Amundsen Sea Pine Island Polynya Sea ice.7 3 31 66 Phyto need 2-1 12 53 88 Vertical diffusion.9 Upwelling.4 Fe fluxes in µmol m -2 d -1 Gerringa et al. (212) 49
Amundsen Sea Fe limitation Amundsen Polynya Fe limitation offshore Pine Island Polynya No Fe limitation 5
Conclusions Southern Ocean a source of CO 2 to the atmosphere during interglacials Not clear if this applies to anthropogenic warming conditions (e.g., today) Little interannual variation in NPP in pelagic Southern Ocean Largest changes in NPP and biggest CO 2 sinks are on shelves Positive SAMs are becoming more common Increases ice cover in pelagic Ross Sea Decreases ice cover in pelagic Amundsen Sea Very small net change in annual NPP Decreases ice cover on Ross Sea shelf Large increase in annual NPP 51
Conclusions Shelves likely to experience largest NPP changes due to anthropogenic warming More positive SAM will reduce ice cover on Ross Sea shelf And other shelves as well NPP likely to rise More glacial melt will increase Fe input into shelf waters Already large sinks for CO 2 May increase NPP in Fe-limited shelf regions Phytoplankton community composition could change Diatoms dominate pelagic likely to continue Warmer temperatures and additional Fe could favor diatoms on shelves Would reduce efficiency of biological pump But Pine Island Bay experiences high Fe and Phaeocystis dominates 52
THANK YOU! Gert van Dijken Kate Lowry Casey Smith Anne-Carlijn Alderkamp Matt Mills Tish Yager (ASPIRE) Hein de Baar Loes Gerringa Philippe Tortell Ocean Biology and Biogeochemistry Cryosphere Science Program 53