Ben Moat, Bablu Sinha, Penny Holliday and Gerard McCarthy D2.1 Model-observation and re-analyses comparison at key locations for heat transport to the Arctic Ben Moat, Bablu Sinha, Penny Holliday and Gerard McCarthy RAPID ARRAY OSNAP GSR EEL OVIDE 1) NEMO Ocean only 1/12° 1958 to 2015 1/12° 1958 to present day (ACSIS) 2) Three (HadGEM3) - Coupled simulations All are 1/12° Ocean with N512 Atmosphere 20 year control run b) 100 year control run fixed present day CO2 ( RUNNING NOC, UK ) c) 100 year “all-forced” run to present day and then a RCP scenario 1950 to 2050 (RUNNING at UK MET OFFICE ) 2.1 Assessment of key lower latitude influences on the Arctic and their simulation [Lead: NERC; Participants: UoS, NLeSC, MPI, HAV, GEOMAR, NIOZ, SAMS, MRI, MSS, UHAM, UNIRES, CNRS, DMI, UIB, NCAR, WHOI, MEOPAR] [Start: Month 1, End: Month 36; Deliverables associated to this task: D2.1, D2.2, D2.3, D2.4]
At key locations, Blue-Action *** will compare observations of heat and freshwater transport to state-of-the-art coupled climate models (HadGEM-Charisma, IPSL, CESM) and high resolution ocean-only models (e.g., NEMO (1/24°)). In-situ ocean data originating from transport mooring arrays (TMAs) and hydrographic GO-SHIP sections will be complimented with remote sensed data from Argo profiling floats, from underwater gliders and satellite data (including existing missions, and new missions data from Jason-3 and Sentinel-3, when they are operational). ***Correspondence, compensation or feedback between these ocean heat transport estimates and atmospheric heat transport will be examined in the coupled model simulations (D2.1) [NERC, MPI, UoS, NLeSC, GEOMAR, SAMS, NIOZ, MRI, HAV, MSS, UHAM, CNRS, UNIRES, UIB, NCAR, WHOI].
 RAPID 2004. OVIDE. Every 2 years since 2002. OSNAP: 2014 to present. Data released in the Autmn of 2017. EEL: 1975 to 1996. then EEL to present. Annual. M: 1948 to 2008. Mooring in place. 4 CTD to 100m a week. 1 2000m CTD weekly. ¼ degree eqivalents with an N216 atmosphere. Rcp 8.5 b) go6 and geo7 atmos (25km resolution) Qunatify different water massess across EEL, OVIDE ? volcanoes, solar, aerosols,  greenhouse gases (ALL FORCED) Baring strait – University of Washington Rebecca Woodgate ? 0.8Sv inflow and 26 Sv psu Station M – 1948 to 2009 daily CTD to 1000m, weekly CTD to 2200m. Mooring since 2010 (DNMI?) Testing ? Is the AMOC realistic ??? WP2 progress update. 29th September 2017

Observational Data Sets RAPID 26N ( 2004 to Feb 2017 ) AMOC time series release 2005 to 2017 ( October 2017 ) Heat transport time series 2004 to 2015 ( online ) Freshwater transport time series 2004 to 2015 available soon OVIDE (every 2 years since 2002) MOC and temperature transport Extended Ellet Line (EEL) (annually - 1975 to present) OSNAP Time series  release late 2017 ( Aug 2014 to April 2016 ) MOC, heat transport and fresh water transport

NEMO 1/12 (hindcast) OGCM NEMO-LIM2 ocean circulation/sea ice model (Marzocchi, 2015; Moat, 2016) 1/12 degree Horizontal with 75 levels in the vertical. Forced by the Drakkar Surface Forcing dataset (air temperature, winds, humidity, surface radiative heat fluxes and precipitation) Eddy resolving at low latitudes Run from 1958 to 2015 Surface current speed (m/s) latitude longitude

Temperature/Heat transports at 26.5°N 4 3 2 1 -1 -2 -3 (Moat et al. 2016, JGR-Oceans) RAPID thick line, model thin line ORCA083 (PW) RAPID (PW) FC 2.54±0.16 2.50±0.25 EKMAN 0.34±0.31 0.35±0.30 Mid- Ocean -2.02±0.27 -1.81±0.31 WBW 0.12±0.21 0.12±0.18 EDDY -0.02±0.06 0.08±0.03 Total heat transport 0.93±0.32 1.24±0.36 Temperature/Heat Transports (PW) RAPID: 17.20±4.60 Sv ORCA083: 15.18±4.03 Sv Extending to SALT and FW fluxes. 1/12° NEMO hindcast model overestimates the Southwards Mid-Ocean temperature transport. Model HT 0.3 PW lower than RAPID, but good agreement in the variability.

RAPID MODEL comparison at 26.5°N Climatological Seasonal Cycle 3 2 1 -1 -2 FC TOTAL EKMAN INTERIOR (Qmo+Qwbw+Qeddy) RAPID NEMO 1/12 HINDCAST MHT (PW) MHT (PW) MOC (Sv) Temperature/Heat transport (PW) Overturning Gyre J F M A M J J A S O N D RAPID: 17.20±4.60 Sv ORCA083: 15.18±4.03 Sv Extending to SALT and FW fluxes. Model underestimates the MOC (2004 to 2012) RAPID: 17.20±4.60 Sv ORCA083: 15.18±4.03 Sv MOC (Sv) RAPID Gyre : 0.14±0.07 PW ORCA083 Gyre : 0.14±0.06 PW Model reproduces the Gyre, overturning circulation and seasonal cycle

Heat transports at 26.5°N 2 COUPLED 1/12 CONTROL (25km atmosphere) COUPLED 1/12 CONTROL (25km atmosphere) COUPLED, R2=0.92 HINDCAST, R2=0.84 Johns 2011, R2=0.94 Heat transport (PW) 1.1 ± 0.26 PW 0.94 ± 0.22 PW Mean over RAPID period = 1.24±0.36 PW NEMO 1/12 HINDCAST MOC (Sv) 0 30 Larger variability in 20year coupled model RAPID: 17.20±4.60 Sv ORCA083: 15.18±4.03 Sv COUPLED : 15.3 ± 3.6 Sv Rsquare COUPLED = 0.92 Rsquare hinf = 0.84 10 year coupled mean HT 1.0823 +/- 0.25746 PW 10 year ocean only mean HT 0.93641 +/- 0.21831 PW Johns 0.079*MOC-0.12 Extending to SALT and FW fluxes. 25Km resolution atmosphere 8.3 km, becoming finer at higher latitudes such that at 608N/S it becomes 4.6 km. Model info : Hewitt et al. (2016) Roberts et al. (2016)

Fresh water transports at 26.5°N OBSERVATION (McDonagh et al., 2015) NEMO 1/12 HINDCAST Fresh water transport (Sv) Bering Straits: 0.8Sv at 32.5 psu 26 Sv psu

OVIDE SECTION (1/12 degree hindcast) Observations Northwards 16.6±2.5 Sv Northwards 0.45±0.07 PW OSNAP time series out in October 2017 Northwards 7.07±1.74 Sv psu

Model results: OSNAP Section NEMO 1/12 hindcast HT ST Northwards OSNAP time series from the array will be released late 2017. OSNAP time series out in October 2017

OSNAP Challenge 2017 (BLUE-ACTION won) 2014 2015 2016 Model Group RMSE (Sv) Correlation ( r ) 1 Ben Moat, (NOC, UK) 3.3 0.59 2 Charlene Feucher (Universiry of Alberta) 5.87 0.50 3 Laura Jakson (Met office, UK) 6.34 -0.14 4 Andrea Storto (CMMC, Italy) 9.86 0.33 Mean of predictions OSNAP MODEL OSNAP time series out in October 2017 http://www.o-snap.org/news-events/osnap-challenge/

Plans Extending analysis to Salt and Freshwater transports on all the sections Greenland Scotland ridge and Extended Ellet Line Extend the analysis to the two high resolution coupled runs ( 1/12 ocean with 25km (N512) atmosphere ) Meeting with PRIMEVERA ( Malcolm Roberts ) early October. Bee Berx committed .

WP2 Discussion points Greenland Scotland ridge Update on data availability – time series across whole section Interaction with D2.4 – Synthesis and Dissemination of ocean and atmosphere heat transport to the Arctic. Presentations Modelling Workshop Evaluating climate and Earth System models at the process level, 23-24 May 2017, Brussels. Understanding Change and Variability in the North Atlantic Climate System, ACSIS - OSNAP – RAPID Joint Science Meeting, 19-21 September 2017, Oxford, UK.

The Blue-Action project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 727852