Biogeochemistry in the Los Alamos National Laboratory (LANL) model Clara Deal (UAF) and Scott Elliott (LANL) Biogeochemistry: of or relating to the partitioning and cycling of chemical elements and compounds between living and nonliving parts of an ecosystem Our activities focus on marine organic C (i.e., primary production), and S (DMS). Also N and Si cycling, methane, and carbonate system.

Influence of Sea Ice on Arctic Marine Sulfur Biogeochemistry in the Community Climate System Model DOE EPSCoR-State/National Laboratory Partnership Grant PI: Clara Deal, Co-I: Meibing Jin International Arctic Research Center (IARC), University of Alaska Fairbanks (UAF) Scott Elliott, Elizabeth Hunke, Mathew Maltrud, Nicole Jeffery, and Adrian Turner Coupled Ocean Sea Ice Modeling Group (COSIM), Los Alamos National Laboratory

Talk outline Objectives Motivation and background 1. food web - DMS interactions 2. importance of ice algae 3. 1-D ice-ocean ecosystem modeling - map with sites - findings 3-D Model results 1. primary production 2. DMS 3. impact of decreasing sea ice on biogeochemistry Summary list of modeling activities

DOE EPSCoR Project Research Objectives Overall: Improve the treatment of arctic marine biogeochemistry in CCSM. Add sea ice algae and arctic DMS production and related biogeochemistry in POP coupled to CICE. Specifically: 1)Develop a state-of-the-art DMS ice-ocean ecosystem model 2)Assess and predict sea ice influences on DMS dynamics

Phytoplankton DMSP Fecal or detrital DMSP Assimilated S Grazer DMSP DMSP (Dissolved) DMS DMSO and Other products Methanethiol Organic sulfur Dissolved and Particulate Grazing DMS (g) SO 4 2- SO 2 & CH 3 SO 3 H “Sulfate” Aerosol CCN Sinking/Export Elimination Photolysis h Direct release Assimilation Demethylation/ demethiolation bacteria Feedbacks? ? Air Sea Radiative backscatter bacteria lysis or leakage Sinking/Export Ventilation Figure courtesy Ron Keine. DMS cycle is intimately linked to marine food web and plays important role in climate.

Why are sea ice algae important? important food source for benthic and pelagic herbivores timing of ice algal bloom is more important than the magnitude >50% primary production in central Arctic Ocean (Gosselin et al. 1997) between 4 to 25% primary production in arctic shelf seas (Legendre et al. 1992) Schematic representation of seasonal cycle of marine production

Ice algal biomass is highest in the bottom 2-3 cm of arctic FYI and MYI (Gradinger et al. 2009; and references therein), where S compounds accumulate. April through May, >90% ice algal biomass (chlorophyll a) observed in bottom of sea ice (3 cm layer) (Shin et al., 2003). Very high levels of total DMSP up to 15 μ M, were observed at Barrow, Alaska (Uzuka et al., 2003).

1-D Physical ice-ocean Ecosystem model (PhEcoM) applied at: land-fast ice zone, multi-year pack ice, and pack ice of SIZ. Findings from 1-D modeling studies: suggest “seeding” of phytoplankton bloom by ice algae (Jin, Deal, Wang, Alexander, Gradinger, et al. GRL 2007) shift in lower trophic level production and dominant phytoplankton type in response to climate regime shift (Jin, Deal, Wang, McRoy, JGR 2009) vertical mixing plays important role in determining microalgal composition and DMS sea-to-air flux (Jin, Deal, Wang, Tanaka, Ikeda, JGR 2006) (Deal [Jodwalis], Benner, Eslinger, JGR 2000) major controls sea ice algal production (Lee, Jin, Whitledge Polar Biol, 2010) (Jin, Deal, Wang, Tanaka, Shin, Lee, Whitledge, Gradinger, Annals Glaciol, 2006) 1-D DMS ice-ocean ecosystem (i.e., ice ecosystem-ocean ecosystem with DMS modules)

3-D Model Results

Simulated annual primary production within arctic sea ice (for 1992) reproduces observed large-scale patterns and seasonality. Deal, Jin, Elliott, Hunke, Maltrud, and Jeffery (2010) Large-scale modeling of primary production and ice algal biomass within arctic sea ice. J Geophys Res, minor rev CICE with ice ecosystem (from PhEcoM): (g C m -2 )

High ice algal production in the Bering Sea was due to high daily production rates, while the large sea ice area in the Canadian Archipelago/Baffin Bay and, in particular, the Arctic Ocean basins resulted in their considerable contribution to primary production.

Model results show strong seasonality of ice algal bloom. CICE with ice ecosystem (from PhEcoM): Deal, Jin, Elliott, Hunke, Maltrud, and Jeffery (2010) Large-scale modeling of primary production and ice algal biomass within arctic sea ice. J Geophys Res, minor rev

DMS(P) production in bottom ice may add substantially to a maximum of reduced S that dominates the marginal ice environment. Baseline simulation DMS (log 10 nM) injected into the ocean product layer May Ice edges are defined by the 15% contour (white), and CICE thicknesses are superimposed in meters (black). S. Elliott, C. Deal, G. Humphries, E. Hunke, N. Jeffery, M. Jin, M. Levasseur, and J. Stefels, Pan- Arctic simulation of coupled nutrient-sulfur cycling due to sea ice biology, in prep. CICE with DMS ice ecosystem:

Modeled pan-Arctic annual primary production averaged over in a) sea ice, and b) ocean upper 100m. Coupled CICE-POP with ice-ocean ecosystem (ocean ecosystem; Moore et al. 2004): (g C m -2 ) Jin, M., C. Deal, S. Lee, S. Elliott, E. Hunke, M. Maltrud and N. Jeffery. Modeling study of the Arctic sea ice and ocean primary production and model validation in the western Arctic, Deep Sea Res Part-II (submitted) g C m -2 yr -1 observed Chukchi shelf (Lee et al. 2007)

Time series of modeled annual upper ocean 100m integrated primary production within Arctic Circle compared to mean estimated using remotely sensed Chl a and algorithm developed for the Arctic by Pabi et al. (2008). Coupled CICE-POP with ice-ocean ecosystem: Jin, M., C. Deal, S. Lee, S. Elliott, E. Hunke, M. Maltrud and N. Jeffery. Modeling study of the Arctic sea ice and ocean primary production and model validation in the western Arctic, Deep Sea Res Part-II (submitted).

Modeled pan-Arctic annual primary production difference (2007 minus 1992) in a) sea ice, and b) ocean upper 100m. Jin, M., C. Deal, S. Lee, S. Elliott, E. Hunke, M. Maltrud and N. Jeffery. Modeling study of the Arctic sea ice and ocean primary production and model validation in the western Arctic, Deep Sea Res Part-II (submitted). Coupled CICE-POP with ice-ocean ecosystem: (g C m -2 )

Normalized time series of modeled sea ice primary production within the Arctic Circle shows lack of correlation with ice area. Coupled CICE-POP with ice-ocean ecosystem: Jin, M., C. Deal, S. Lee, S. Elliott, E. Hunke, M. Maltrud and N. Jeffery. Modeling study of the Arctic sea ice and ocean primary production and model validation in the western Arctic, Deep Sea Res Part-II (submitted).

Coupled CICE-POP with DMS ice-ocean ecosystem: Simulated surface seawater DMS concentrations, in general, agree with observations; few nM under-ice, higher in SIZ, and highest near ice edge.

Summary of activities Focus on marine organic C (i.e., primary production) and DMS cycle. 1-D DMS ice-ocean ecosystem modeling (i.e., ice ecosystem-ocean ecosystem with ice DMS module and ocean DMS module) Stand-alone CICE with DMS ice ecosystem modeling Coupled CICE-POP DMS ice-ocean ecosystem modeling A major challenge is the sparseness and heterogeneity of the available biological and biogeochemical data for 3-D model initialization and validation. Thank you for your time and attention!

 Recent climate models (Gunson et al. 2006): - 50% reduction of ocean DMS emission: radiative forcing: +3 W/m 2 air temperature: +1.6 °C - doubling of ocean DMS emission: radiative forcing: -2 W/m 2 air temperature: -0.9 °C  Use of a climate model to force ocean DMS model in Barents Sea (Gabric et al. 2005): - By the time of equivalent CO 2 tripling (2080) zonal annual DMS flux increase: >80% zonal radiative forcing: -7.4 W/m 2 What is the impact of DMS on climate?