A 26oN update and U.S. AMOC Effort www.clivar.us.org A new interagency program with a focus on AMOC monitoring and prediction capability Presented at the Office of Climate Observations Annual Meeting on September 3, 2008 on behalf of the U.S.AMOC Science Team by Molly Baringer In 2007, the U.S. Joint Subcommittee on Ocean Science and Technology identified as a near-term priority in the Ocean Research Priorities Plan the improved understanding of the mechanisms behind fluctuations of the MOC, which will lead to new capabilities for monitoring and making predictions of the MOC changes. NASA Earth Science Division Satellite data analyses, modeling and space-based observations NOAA Climate Program Office Observing systems, monitoring, climate modeling NSF Geosciences program Process studies, models, and observations

The National Ocean Research Priorities Plan and Implementation Strategy presents research priorities that focus on the most compelling issues in key areas of interaction between society and the ocean. The US is committed to protecting the ocean through responsible stewardship and scientifically sound management. Understanding society’s impact on the ocean and the ocean’s impact on society forms the basus for ensuring a clean, healthy an stable ocean environment.

Unanswered questions surrounding the AMOC include: What is the current state of the AMOC? How has the AMOC varied in the past on interannual to centennial time scales? What governs AMOC changes? Is the AMOC predictable on 10-100 year timescales? What are the impacts of AMOC variability and change? October 2007: Implementation Strategy for a JSOST Near-Term Priority: Assessing Meridional Overturning Circulation Variability: Implication for Rapid Climate Change

Western Boundary Time Series (WBTS) U.S. PI’s: M. O. Baringer1, C. S. Meinen1, S. L. Garzoli1 1 NOAA-Atlantic Oceanographic and Meteorological Laboratory U.S. Collaborators: B. Johns2, L. Beal2 (MOCHA/NSF) 2 RSMAS, University of Miami, Miami FL International Collaborators: H. Bryden3, S. Cunningham3, T. Kanzow3, J. Marotzke4, J. Hirschi3 (RAPID/NERC) 3 National Oceanography Centre, Southampton, U.K. 4 Max Planck Institut, Hamburg, Germany 27N, monitor components of MOC near western boundary of NA, within Florida Current,with submarine cable, direct velocity measurements and hydrography: transport of upper limb since 1982, water mass of upper and lower since 1984, transport of lower limb since 2004

An Observing System for Meridional Heat Transport Variability in the Subtropical Atlantic Funding Agency: NSF (Funded period 2004 – 2014) U.S. PI’s: B. Johns1, M. Baringer2, L. Beal1, C. Meinen2 1 RSMAS, University of Miami, Miami FL 2 NOAA/AOML, Miami, FL International Collaborators’s: H. Bryden3, S. Cunningham3, T. Kanzow3, J. Marotzke4 (RAPID/NERC) 3 National Oceanography Centre, Southampton, U.K. 4 Max Planck Institut, Hamburg, Germany U.S. Collaborators: S. Garzoli2 (WBTS/NOAA) Goal: “To set in place a system for continuous observation of the meridional overturning circulation and northward heat transport in the Atlantic Ocean, with which to document its variability and its relationship to observed climate fluctuations, and to assess climate model predictions.” Specific Objectives: Determine the “present day” mean MOC & MHT at 26°N and year-to-year variability Determine the spectrum of MOC variability, and related mechanisms, to help optimize MOC observing systems Provide a benchmark of MOC strength and variability for climate and ocean synthesis models Louise Bell / Neil White, CSIRO

Driving Florida current Variability with wind stress curl (WSC) variations DiNezio et al, 2008, JPO, in press. Meinen et al, 2008, JGR, in press Correlation between WSC and the NAO

Influence of the RAPID-MOC and Gulf Stream cable measurements on the ECCO-GODAE global state estimate Ocean state estimation provides a dynamicaly consistent estimate of the ocean circulation, providing important Information on integragted dynamical quantities such as heat flux or fresh water transport. However, these Models have to rely on the assimilation of a multitude of observations - but mainly sparsly distributed interior Observations. There are very few observations of integrated quantities. However, the rapid array provides Continuous observations of the meridional mass transports. Assimilation of GS-> increase in Gulf Stream by 3Sv Assimilation of RAPID -> NADW increase of about 2 Sv Assimilation of GS and RAPID : 1 Sv MOC increase and about 0.05 PW heat flux increase over A wide range of latitudes. J. Baehr, 2008, JPO, submitted

Vertical structure of the Florida Current 87% Variance explained by the first EOF mode Using more than 100 insitu sections across the Florida Current to develop a technique to estimate the vertical structure of the current given only the total transport.

Meridional Heat Transport: Qnet = ∫∫ ρcp vθ dx dz Qnet = QFC + QEK + QWB + QINT QFC → Cable voltage calibrated for temperature transport, (Shoosmith et al., 2005) r = 0.94, σ = 0.1 PW QEK → QuickScat wind stress (daily) • Reynolds SST (weekly) QWB → Directly calculated from moored CM’s/thermistors in Abaco WB array QINT → Zonally-averaged interior transport profile from endpoint geostrophic moorings • Seasonally-averaged interior hyrdographic climatology (Hydrobase, R. Curry) Missing: Contribution to QINT by spatially correlated v,T variability across interior: “Gyre/eddy” heat transport = ∫∫ ρcp v'θ' dx dz

Annual Mean Heat Transport Quantity Mean Value Std. error Bias error Qarray 1.26* ± 0.06 ± 0.08 (?) Qgyre/eddy 0.10 ± 0.03 - __________ ______________ _____________ 1.36 ± 0.07 ± 0.08 * deseasonalized → Qnet = 1.36 ± 0.11 PW Recent Estimates at 24-26°N MHT Ganachaud and Wunsch (2003) 1.27 ± 0.15 Lumpkin and Speer (2007) 1.24 ± 0.25 Lavin et al. (1998) 1.27 ± 0.26 Fillenbaum et al. (1997) 1.44 ± 0.33 Molinari et al. (1990) 1.21 ± 0.34 Trenberth and Caron (2001) 1.1 (NCEP) RAPID/MOCHA XBT estimates

Part One Conclusions Annual mean (2004-2006) MHT across 26°N = 1.36 ± 0.11 PW. Consistent with previous direct estimates (within errors), but at upper end. Annual mean MOC transport = 18.8 ± 5 Sv is also slightly larger than estimates from WOCE period (16-18 Sv). Short term MHT variability is large. Range -> 0.1 – 2.5 PW, Std. Dev = 0.41 PW and. About half is due to Ekman transport variability, remainder due to geostrophic variability. Range of variability is consistent with eddy-permitting/eddy-resolving models, but geostrophic variability may play a bigger role than previously suggested by models (→ Hirschi et al., 2007) Seasonal cycle emerging… in agreement w/ prior climatological estimates and model results. MOC dominated by FC and interior ocean while the MHT dominated by Ekman annual cycle. Mean Volume and MHT estimates from the RAPID-MOC array should provide one of the best constrained benchmarks for indirect estimates of the ocean transports (from flux climatologies, TOA radiation, etc.), and for comparison with numerical models.

U.S. AMOC Scientific Objectives The design and implementation of an AMOC monitoring system An assessment of AMOC’s role in the global climate An assessment of AMOC predictability Recommended Activities Develop an AMOC state estimate or “fingerprint” Monitor AMOC transports Evaluate coherence and connectivity of AMOC circulation and transports Assess AMOC observing systems with ocean models Reconstruct AMOC variability and associated property fields Model the ocean state during the instrumental period Develop longer-term proxies for AMOC variability Diagnose mechanisms of AMOC variability and change Assess AMOC predictability Determine impact and feedback of AMOC variability Assess role of AMOC in producing observed changes

Evaluation of Meridional Transport of Water and Heat in the Atlantic Ocean Using Satellite Data PI: W. Timothy Liu, Co-I: Xiaosu Xie 1 Jet Propulsion Laboratory, NASA, Pasadena, California Comparing annual mean Atlantic meridional heat transport. Red curve is calculated from four components of satellite surface heat flux. The green curve is computed from ECCO data. Various symbols are from past studies, from surface heat flux or hydrographic data.

Line W: A sustained measurement program sampling the North Atlantic Deep Western Boundary Current and Gulf Stream at 39°N U.S. PI’s: J. Toole1, R. Curry1, T. Joyce1, M. McCartney1 and W. Smethie, Jr.2 1 Woods Hole Oceanographic Institution, Woods Hole, MA 02543 2 Lamont Doherty Earth Observatory, Palisades, NY 10964 International PI’s: J. Smith3 3 Bedford Institute of Oceanography, Dartmouth, Nova Scotia B2Y 4T3 Canada Line W: Volume transport time series May 2004, full line deployed, May 2008 9th reoccupation of the hydrographic section, April 2006, first two years of moored array, compared to IfM-BIO Grand Banks array

Export Pathways from the Subpolar North Atlantic Funding Agency: NSF (2007 – 2008) PI’s: Amy Bower1 and Susan Lozier2 1 Dept. of Physical Oceanography, Woods Hole Oceanographic Institution, Woods Hole, MA 2 Earth and Ocean Sciences, Nicholas School of the Environment, Duke University, Durham, NC Goal: To obtain a better understanding and description of the pathways of waters that constitute the lower limb of the Atlantic Meridional Overturning Circulation, with a focus on the Labrador Sea Waters Specific Objectives: Conduct a field program with RAFOS floats seeded in the Labrador Sea to directly assess pathways Conduct data and modeling studies to assess dynamics controlling these pathways and their temporal variability 16

Pathways of meridional circulation in the North Atlantic Ocean U.S. PI’s: P.B. Rhines1, S. Hakkinen2 1 Dept of Oceanography, U. Washington, Seattle, WA 2 NASA Goddard Space Flight Center, Greenbelt, MD Surface drifter tracks show significant changes since year 2000 in the path of the surface currents associated with the North Atlantic Current. Surface drifter tracks entering (cyan) and leaving (magenta) the region (30-50W, 35-45N): during 1991-2000, top, and during 2001-2006, bottom.

Assessing Meridional Transports in the North Atlantic Ocean U.S. PI’s: K. A. Kelly1, L. Thompson 1 Applied Physics Lab, University of Washington , Seattle, WA 2 School of Oceanography, U. Washington, Seattle, WA NAC LC subtropical 15 years of surface currents shows that the North Atlantic Current is stronger when the subtropical gyre is weaker (and vice-versa) strong NAC weak subtropical

Decadal Climate Predictability and Predictions – Focus on the Atlantic U.S. PI’s: T.L. Delworth, A.J. Rosati Geophysical Fluid Dynamics Laboratory/NOAA Princeton, NJ, USA Goal: Determine role of AMOC in producing observed climatic variability Difference in simulated summer rainfall when the AMOC is weak (cold North Atlantic) versus when it is strong (warm North Atlantic). Units are cm per day. Blue indicates less rain when AMOC is weak. Results from GFDL CM2.1 model.

Towards An Understanding Of The Role of The Atlantic Thermohaline and Wind- Driven Circulation In Tropical Atlantic Variability (TAV) U.S. PI’s: P. Chang1, R. Saravanan2 and R. Zhang3 Dept of Oceanography, Texas A&M University, College Station, TX Dept of Atmospheric Sciences, Texas A&M University, College Station, TX Geophysical Fluid Dynamics Laboratory, Princeton, NJ Fig. 1: Equatorial Atlantic ocean temperature change in response to North Atlantic fresh water input in GFDL CM2.1. Fig. 2: Simulated changes in monsoon rainfall over West Africa. (After Chang et al. 2008)

New U.S. AMOC Projects starting soon: Recommended Activities Develop an AMOC state estimate or “fingerprint” Monitor AMOC transports Evaluate coherence and connectivity of AMOC circulation and transports Assess AMOC observing systems with ocean models Reconstruct AMOC variability and associated property fields Model the ocean state during the instrumental period Develop longer-term proxies for AMOC variability Diagnose mechanisms of AMOC variability and change Assess AMOC predictability Determine impact and feedback of AMOC variability Assess role of AMOC in producing observed changes New U.S. AMOC Projects starting soon: Variability and Forcing Mechanisms of the Atlantic Meridional Overturning Circulation - Tong Lee/ Jet Propulsion Laboratory with Harvard University, University of Hamburg, and Laboratoire de Physique des Oceans Atlantic MOC Observing System Studies Using Adjoint Models - Carl Wunsch/ MIT with AER, Inc. A NOPP Partnership for AMOC: Focused Analysis of Satellite Data Sets - Peter Minnett/ Miami with Remote Sensing Systems, Inc. Observing System Simulation Experiments for the Atlantic Meridional Overturning Circulation - George Halliwell/Miami with NOAA AOML

Summary U.S. effort is spinning up; some early tangible results Continued partnership with international collaborators is essential to the community-wide goals of AMOC monitoring and prediction Particular need for collaboration is for transatlantic measurements; the monitoring of subpolar North Atlantic and subtropical South Atlantic are key priorities.