Coupled Data Assimilation for Ocean-Biogeochemical Models Lars - PowerPoint PPT Presentation

Coupled Data Assimilation for Ocean-Biogeochemical Models Lars Nerger, Himansu Pradhan, Michael Goodliff Alfred Wegener Institute Helmholtz Center for Polar and Marine Research Bremerhaven, Germany ISDA 2019, Kobe, Japan, January 21 – 24, 2019 AWI

Coupled Ocean-Biogeochemical Models Ecosystem Physics Biogeochemical Model, … Ocean Circulation Model km coupling velocities Temperature Finite-Element Sea Ice R egulated Eco system 1.85 Ocean Model M odel – Version 2 FESOM REcoM2 Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models 1 1

Biogeochemical Process Models Wide variations of the model formulation Example: REcoM-2 R egulated Eco system M odel – Version 2 (Hauck et al., 2013 ) Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Satellite Ocean Color Observations This is not a photograph! Natural Color 3/16/2004 Spectral data at 5-8 wavelengths in visible part of spectrum spectral bands in ESA OC-CCI data • Satellite data is water leaving radiance or surface reflectance ➜ Data products are derived from this Picture source: Suomi-NPP/VIIRS, December 10, 2018 NASA (oceancolor.gsfc.nasa.gov) Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

<latexit sha1_base64="qnJs0B1X5/hxhzawdCAyXmdvxQ=">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</latexit> <latexit sha1_base64="qnJs0B1X5/hxhzawdCAyXmdvxQ=">ACYXicbZFNa9wEIZlN2STZM6TEXkSWwSyHYJdBeAqG5NBDWrJYL0xY+3YKyLRhqHLMZ/srdeukfqXbXhXwNCL16ZoaRXqWVkpbC8Lfnv1l7+259Y7O39X5750Owu3dly9oIHIlSleYmBYtKahyRJIU3lUEoUoX6d3ZIn9j8bKUl/SvMJAbmWmRADiXBgyrzpInCdnB2/j2BIT/hkIT8E49tXSNPIna2OHZKwo8H/6tUpzgyI5ucgVm7gFJImVTW2w/Yxyg2idiw2Mp/RsNtuZRL0w6NwGfyliDrRZ1cJMGveFqKukBNQoG14yisaNKAISkUtr24tliBuIMcx05qKNBOmqVDLT90ZMqz0riliS/p4GCmvnReoqC6CZfZ5bwNdy45qyr5NG6qom1GI1KsVp5Iv7OZTaVCQmjsBwkh3Vy5m4Gwj9yk9Z0L0/MkvxdXno8jpH8f902+dHRtsnx2wAYvYF3bKztkFGzHB/nhr3ra34/31N/3A31uV+l7X85E9CX/H7XntUo=</latexit> <latexit sha1_base64="qnJs0B1X5/hxhzawdCAyXmdvxQ=">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</latexit> <latexit sha1_base64="qnJs0B1X5/hxhzawdCAyXmdvxQ=">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</latexit> Satellite Chlorophyll Data (the most common product) Natural Color 3/16/2004 Chlorophyll Concentrations Chlorophyll computed as 4 th order polynomial of reflectance at two wavelengths: ✓ R ( λ blue ) 4 ◆◆ i ✓ X log 10 ( CHL a ) = a 0 + a i log 10 R ( λ green ) i =1 or combined with linear three-wavelength dependence (this is empirical! – derived from statistical analysis) Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models Figure: NASA � Visible Earth � , Image: SeaWiFS Project, NASA/GSFC & Orbimage

Example: Chlorophyll-a (SeaWiFS) mg/m 3 Daily gridded SeaWiFS chlorophyll data Ø gaps: satellite track, clouds, polar nights Ø 30% to 50% data coverage Ø irregular data availability Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models Nerger, L., and W.W. Gregg. J. Marine Systems 68 (2007) 237

Data Assimilation Issues Model Much higher error than in physics • Skill Only fraction of fields observed • Complexity Observations Fields are less constrained • Data gaps • Data error level 15 - 30% � representation error • Empiric algorithms Assimilation • Approx. log-normal Need to transform concentrations � representation error • Diurnal variability Unknown, but expected to be high • Representation errors Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Example 1 Assimilation of total chlorophyll to constrain 2 phytoplankton groups Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Example: Global Chlorophyll Assimilation Global configuration MITgcm 80 o N - 80 o S, 30 layers General ocean circulation model Resolution: of MIT ( Marshall et al., 1997 ). lon : 2 deg lat : 2 deg in North up to 0.38 deg in South REcoM-2 R egulated Eco system M odel – Version 2 (Hauck et al., 2013 ) Assimilate with PDAF (http://pdaf.awi.de) Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Assimilation of Total Chlorophyll Assimilated : Total chlorophyll from ESA OC-CCI Total chlorophyll (5 day composite) mg/m 3 Assimilation : • Assimilate satellite total chlorophyll (ESA Ocean color - climate change initiative): Chl TOT = Chl DIA + Chl PHY • Handle logarithmic concentrations log(Chl TOT ), log(Chl DIA ), log(Chl PHY ) • Multivariate update through, e.g. logarithmic observation errors Cov(log(Chl TOT ), log(Chl DIA )) • How are both phytoplankton groups influenced? • Validate with satellite and in situ data Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Assimilation of Total Chlorophyll Assimilated : Verification : Phytoplankton group data Total chlorophyll from ESA OC-CCI SynSenPFT (Losa et al. 2018) mg/m 3 Total chlorophyll (5 day composite) Small phytoplankton mg/m 3 mg/m 3 Diatoms logarithmic observation errors Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Effect on Chlorophyll in Phytoplankton Groups logarithmic RMS errors (southern regions) • Assimilation improves groups Small phytoplankton Diatoms individually through cross- covariances • Stronger error-reductions for Diatoms • In situ data comparison: RMSe Free Assim. Diatoms 1.3 0.91 Small Phyto. 0.53 0.45 (bias and correlation also improved) Current work • Asses impact of assimilating chlorophyll group data (much lower errors for diatoms) Pradhan et al., J. Geophy. Res. Oceans, in press , doi:10.1029/2018JC014329 Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Ensemble-estimated Cross-correlations Cross correlations between total and group chlorophyll • Significantly different correlations for small phytoplankton and diatoms • Negative correlations exist (despite Chl TOT = Chl DIA + Chl PHY ) Pradhan et al., J. Geophy. Res. Oceans, in press , doi:10.1029/2018JC014329 Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Example 2 Weakly- and Strongly Coupled Assimilation Constrain Biogeochemistry with Temperature Data Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Example: weakly- and strongly coupled assimilation HBM (Hiromb-BOOS Model) – operationally used at Germany Federal Maritime and Hydrographic Agency 5 km km North Sea Baltic Sea 900 m Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models 1 1

Biogeochemical model: ERGOM N 2 O 2 Atmosphere O 2 N 2 Nutrients Zooplankton Phytoplankton 3- PO 4 Micro- Cyanobacteria zooplankton - NO 3 Flagellates + NH 4 Diatoms Meso- zooplankton Si Detritus Si Ocean Detritus N Sediment Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Observations – Sea Surface Temperature (SST) NOAA/AVHRR Satellite data 10 April 2012 25 May 2012 • 12-hour composites • Vastly varying data coverage (due to clouds) • Effect on biogeochemistry? • Assimilation using assimilation framework PDAF Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Weakly & strongly coupled effect on biogeochemistry Oxygen mean for May 2012 (as mmol O / m 3 ) Free run Assimilation WEAK Free run Free – Assimilation WEAK Changes up to 8% (slight error reductions) l Larger in Baltic than North Sea l Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models

Weakly & strongly coupled effect on biogeochemistry Oxygen mean for May 2012 (as mmol O / m 3 ) Free run Assimilation WEAK Free run Free – Assimilation WEAK Assimilation STRONG Free – Assimilation STRONG Strongly coupled l slightly larger changes l Strongly coupled DA further improves oxygen l Used actual (linear) concentrations Lars Nerger et al. – Coupled DA for Ocean Biogeochemical Models