Variability of primary production and air sea CO 2 flux in the Southern Ocean

Transcription

1 Variability of primary production and air sea CO 2 flux in the Southern Ocean Shanlin Wang 1,2 and Keith Moore 2 1,2 National Center for Atmospheric Research 2 University of California, Irvine

2 Background flux ofacontemof natural CO2, so that the Southern is concentrated in the±region south ofthe 45!S (Figu long-term trend negative trend of!0.007 porary Ocean CO2 exhibits Although agreeme!2 of!0.007 ± negative trend porary CO2 exhibits a long-term Although the agreement between the different stud (Figure 1) Pg C a encouraging, all ocean Pg C a!2 (Figure 1). encouraging, all ocean modelsatmospheric were forcedwinds with and the trend of anthropogenic [10] Thus, relative to the uptake trend of anthropogenic [10] Thus, relative to the uptake atmospheric and fluxesncep-1. of heat and freshwate from our simulations suggest thatwinds the Southern CO2, results A recent set o suggest thatsink the Southern CO2, results from our simulations NCEP-1. recent set of model experiments performe weakened at the Ocean (<35!S) of contemporary CO2Ahas IPSL model (K.!2at has weakened Ocean (<35!S) sink of contemporary CO the IPSL model (K. Rodgers, personal communic 2 between 1979 and We a rate of Pg C a 2008) suggests that th and We reported a rate of Pg C a!2 between 2008) suggests that the use of ECMWF forcing compare1979 this trend with those in the ocean modsmaller change in theyi compare this trend with those reported in the ocean mod- smaller change in the Southern Ocean CO2 sink, pr 5826 JOURNAL OF CLIMATE VOLUME 21 (Lovenduski et al., 2008) 2. (a) Linear trendsco in2the flux linearly of natural CO2 andwith (b) trends Figure 2. (a) Linear trends infigure the air-sea flux of natural andair-sea (b) trends congruent the flux and (d) trend exp speed index. (c) in the anthropogenic CO flux and (d) trend expected from the atmospheric wind speed index. (c) Trend inwind the anthropogenic COTrend 2 2!2!2!2!2 a ). Trends are , and only those trends perturbation in anthropogenic perturbation CO2 (mol min (mol m a ). Trends are from anthropogenic COfrom 2 with significance #95% are with shown. Positive values trends towardvalues oceanindicate outgassing. significance #95% indicate are shown. Positive trendsthe towa corresponding figure for the entire period of model output ( ) can be found in Figure S4. can corresponding figure for the entire period of model output ( ) 3 of 9 (Le Quéré et al., 2007) 3 of 9 (Matear and Lenton, 2008) There has been a debate about whether there is a decreasing trend in the Southern Ocean carbon sink.

3 The Ocean Carbon Cycle Ø The carbon pumps: solubility pump and biological pump. Ø Variability and contribution of biological pump should be examined. CO 2 (gas) + H 2 O H 2 CO 3 HCO H + CO H + Ocean Circulation (hjp:// pump- biological- pump- physical- pump/)

4 Model Description Ecosystem Module 5 phytoplankton groups One zooplankton group DOM, sinking particulates Multiple potentially growth limiting nutrients Biogeochemistry Module Full carbonate chemistry, air- sea CO 2 flux and O 2 flux Ocean circulation component in Community Climate System Model 3.0 (CCSM 3) Coarse resolution Forced with NCEP/ NCAR COREv2 reanalysis data

5 Model Description The BEC model Atmospheric deposition Growth N fixation Calcification Inorganic NO 3, NH 4, PO 4, Si (OH) 3, Fe, O 2, DIC & alkalinity Remineralization Dissolution Phytoplankton Small phytoplankton (coccolithophores) Diatoms Diazotrophs Phaeocystis Excretion Mortality Aggregation Detritus DOM(C, N, P, Fe) Particulates (dust, POM, silica, CaCO 3 ) Sinking Grazing Mortality Feeding Chlorophyll (sp, diat. diaz, phaeo) Zooplankton Sedimentary fluxes

6 Differences from old BEC model: 1. Improved iron cycle representation 2. Incorporated an additional phytoplankton functional group, Phaeocystis, which has been reported to be an important phytoplankton species in the SO 3. Simulated mixed layer depths in the SO are more realistic, allowing for light limitation of growth (Moore and Braucher, 2008)

7 Phytoplankton Group Distinctions Ø Iron Uptake Efficiency: Small Phytoplankton > Diatoms >> Phaeocystis Ø Light Utilization Efficiency: Phaeocystis >> Small Phytoplankton > Diatoms Ø Grazing Pressure: Small Phytoplankton > Diatoms = Phaeocystis Ø Growth: Diatoms & Small Phytoplankton & Phaeocystis Phytoplankton community shifts in response to forcings.

8 Model validation Ø Correlations between modeled and observed major nutrient concentrations are higher than 0.8 with some significant biases. Ø Correlations between modeled and observed CO 2 flux and sinking POC are higher than 0.5.

9 Ø The BEC model reasonably reproduces the patterns of surface chlorophyll concentrations and phytoplankton biomass.

10 Climate impacts on the air-sea CO 2 flux in the SO Ø The Southern Ocean CO 2 sink has weakened relative to the expected carbon sink from rising atmospheric CO 2 and constant climate. Ø The weakening trend in simulated SO carbon sink (0.05 PgC/yr/dec) is similar to previous studies (Lovenduski et al., 2007;2008). Ø The weakening trend in SO carbon sink is a combined effect of an increasing anthropogenic carbon sink and a natural CO 2 outgassing.

11 Biological pump contributes significantly Air-sea CO 2 flux trends Sinking POC flux trends Correlation between sinking POC and CO 2 flux Ø Ø Ø Strong correlations (0.7) between the variance in sinking POC and sea-air CO 2 flux in both natural CO 2 simulation and contemporary CO 2 simulation. The biological pump strongly influenced trends, something not seen in previous studies Contributions of biological pump show a non-uniform spatial pattern.

12 Windspeed (m/s) Chlorophyll(mgChl/m3) PAR (W/m2) Trends and driving factors in different regions Mixed layer depth (m) POC FLUX (mol/m2) Iron (nm) A.)Cro Kerg._Trend_yr79 03 Sea Air CO2 (mol/m2) The increased CO 2 outgassing was mainly driven by stronger winds and upwelling in Mid-s.Pacific. Production (mol/m2) NO3(uM) Windspeed (m/s) Chlorophyll(mgChl/m3) PAR (W/m2) In the Crozet-Kerguelen region, there is high correlation between primary production and carbon uptake. Mixed layer depth (m) POC FLUX (mol/m2) Iron (nm) C.)mid s.pac_trend_yr79 03 Sea Air CO2 (mol/m2) Production (mol/m2) NO3(uM)

13 Trends and driving factors in different regions F.)Ross Sea_Trend_yr Chlorophyll(mgChl/m3) PAR (W/m2) Windspeed (m/s) Mixed layer depth (m) POC FLUX (mol/m2) Iron (nm) Sea Air CO2 (mol/m2) Ice Fraction Production (mol/m2) Variations in carbon uptake are largely driven by biological pump, though the trend in primary production is not significant. Ø Ø Ø Drivers of the variations are different in different regions Variability in primary production and sinking POC is tightly linked to variations in nutrient availability and the light regime. Productivity near Antarctica was also influenced by sea ice cover (e.g. the Ross Sea and the western Antarctic Peninsula).

14 Trends in primary production Ø Phytoplankton groups show different spatial patterns of trends. Ø Increase in primary production south of 60 S, contributed mainly by the larger phytoplankton groups. Ø Strong decreasing trend in primary production between 40 S-60 S mainly contributed by decreased production of small phytoplankton and diatoms.

15 Effects of variations in dust deposition Differences in dust, hist. const. Differences in sinking POC flux, hist. const. Differences in CO 2 flux, hist. const. Ø Variability in dust deposition had little influence on total air-sea CO 2 flux and primary production in the Southern Ocean. Ø Variations in dust deposition can alter regional patterns of biological production and the ocean carbon sink.

16 Summary Ø There is a weakening trend in the Southern Ocean carbon sink (0.05 PgC/ yr/dec), similar to previous studies; Ø Different from previous studies, significant variations of the Southern Ocean primary production were found and those variations contributed significantly to changes in CO 2 fluxes in my simulations. Ø There are still caveats in this work. Ø This study emphasizes the important role of biological pump in the SO carbon cycle. It is necessary to further develop both physical and ecosystem models.