Equatorial Atlantic Circulation and Tropical ClimateVariability Peter Brandt

GEOMAR, Kiel, Germany

Equatorial Atlantic Circulation and Tropical Climate Variability

With contributions from:

Richard Greatbatch1, Jurgen Fischer1, Sven-Helge Didwischuss1, Andreas Funk2, Alexis Tantet1,3, William Johns4

1GEOMAR Helmholtz-Zentrum fur Ozeanforschung Kiel, Germany2WTD 71/FWG, Forschungsbereich fur Wasserschall und Geophysik, Kiel, Germany3now at Institute for Marine and Atmospheric Research, Utrecht University, The Netherlands4RSMAS/MPO, University of Miami, USA 2

OutlineIntroduction

• ITCZ and tropical Atlantic variability (TAV)

• TACE observing system

Data & MethodsEUC TransportEUC-TAV Relation

• EUC during warm/cold events

• Shear variability

Equatorial Deep Jets• Equatorial basin modes

• Interaction with EUCSummaryOutlook

Sahel rainfall climatology

MA-Position

JJA-Position

Sahel

Guinea

Guinea rainfall climatology

Atlantic Marine ITCZ ComplexITCZ position and

rainfall intensity affect densely populated regions in West Africa

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Rainfall and SST annual cycleIntroduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets

Summary Outlook

Mechanisms of Tropical Atlantic Variability

Mechanisms influencing Variability of Tropical Atlantic SST

Chang et al., 2006

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Tropical Atlantic Variability (TAV) modes

Zonal mode (Atlantic Nino)Meridional mode (gradient mode)ENSO influenceNAO influence

MERIDIONAL MODE

ZONAL MODE

Strong seasonality of Tropical Atlantic Variability makes understanding and prediction of tropical Atlantic variability a challenge.

Sutton et al. 2000

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Meridional Mode (March-April) During spring the

meridional SST gradient dominates TAV

Underlying mechanism is the Wind-Evaporation-SST (WES) Feedback Mechanism (Saravanan and Chang, 2004)

Kushnir et al. 2006

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Zonal Mode (June-August) Zonal Mode is

associated with rainfall variability, onset and strength of African Monsoon (Caniaux et al. 2011, Brandt et al. 2011)

Underlying mechanism is the Bjerknes feedback that is strong during boreal spring/summer (Keenlyside and Latif 2007)

Kushnir et al. 2006

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Equatorial Atlantic Cold Tongue

Cold tongue develops during boreal summer

Interannual variability of ATL3 SST index (3°S–3°N, 20°W–0°) much smaller than seasonal cycle

10

Brandt et al. 2011

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Onset of Atlantic Cold Tongue and West African Monsoon

WAM onset follows the ACT onset by some weeks.

Significant correlation of ACT and WAM onsets

11

WAM onset – northward migration of rainfall (10°W-10°E.) (Fontaine and Louvet, 2006)ACT onset – surface area (with T<25°C) threshold Caniaux et al. 2011, Brandt et al. 2011

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Regression of SST and Wind onto

12

WAMOnsetSignificantcorrelation with cold tongue SST (zonal mode) andSST in the tropical NE Atlantic(meridional mode)

ACTOnset

Cold tongue

SST;Wind

forcing in the

western equatorial Atlantic

(zonal mode)

Brandt et al. 2011

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

SST Errors in Coupled Climate Models

Jungclaus et al. 2006

Dark gray model too warm

Large errors in the eastern tropical Atlantic

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

2006-2011 Tropical Atlantic Climate Experiment

A focused observational and modeling effort in the tropical Atlantic to advance the predictability of climate variability in the surrounding region and to provide a basis for assessment and improvement of coupled models.

TACE was envisioned as a program of enhanced observations and modeling studies spanning a period of approximately 6 years. The results of TACE were expected to contribute to the design of a sustained observing system for the tropical Atlantic.

TACE focuses on the eastern equatorial Atlantic as it is badly represented in coupled and uncoupled climate models and is a source of low prediction skill on seasonal to interannual time scales. 14

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

TACE observational network

15 Observing system during the TACE period including different process studies, like e.g. the 23°W equatorial moorings

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Equatorial Mooring Array at 23°W

single mooring from June 2005

3 mooringsfrom June 2006 to May 2011

16

Ship Section Mean

Brandt, et al. 2013, submitted

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

EUC from Shipboard Measurements

20 shipboard velocity sections are used to calculate the dominant variability pattern in terms of Hilbert EOFs

Sorted with respect to the seasonal cycle 17

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Reconstruction of Zonal Velocity Sections

Dominant variability pattern from ship sections

Pattern are regressed onto moored time series

Method validation by using the ship sections itself

Alternative: optimal width method

18

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Validation of EUC Transport Calculation using Ship Sections

19

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Eastward EUC TransportGeneral

agreement between different methods

20

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

EUC TransportYears with strong and weak annual cycleShip sections alone are hardly conclusive

about seasonal cycle

21

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Pacific EUC TransportMean EUC

Transport (solid) and EUC transport for strong El Niños (dashed)

Strongly reduced EUC transport during El Niños. EUC disappeared during 1982/83 El Niño (Firing et al. 1983)

22

Johnson et al. 2002

What is the relation between Atlantic EUC transport and tropical Atlantic variability?

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Interannual Variability: SST ATL3 and Wind West Atlantic

Richter et al. (2012): canonical events have strong/weak winds prior to cold/warm events

Canonical cold event: 2005

Canonical warm event: 2008

Noncanonical cold event: 2009 (warmest spring with weak winds, but coldest SST in August)

23

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Interannual Variability: SST ATL3 and EUC Transport

Canonical cold/warm events are associated with strong/weak EUC

EUC during 2009 was weak and shows no variation during the strong cooling from May to July

24

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Interannual Variability: SST ATL3 and April/May 2009 Anomalies

According to Richter et al.(2012) noncanonical events are driven by advection from northern hemisphere during strong meridional mode events

SST and wind anomalies during April/May 2009 (Foltz et al. 2012)

25

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Regression MapsStrong June EUC associated with anomalous cold

Cold Tongue and southerly wind anomalies in the northern hemisphere early onset of the West African Monsoon

26Brandt, et al. 2013, submitted

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

June EUC – Wind/SST Relation

27

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

June EUC – Wind/SST Relation

28

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

June EUC – Wind/SST Relation

29 Regression maps reflect a canonical behavior according to Richter et al. (2012)

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Monthly Regressions of Zonal Velocity onto EUC Transport

During all months: strengthening of the eastward EUC associated with strengthening of westward surface flow (strongest shear enhancement in June)

February: weak near surface flow variability, stronger changes in the south

30

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Seasonal Cycle of Upper Ocean Diapycnal Heat Flux

Strongest shear (1/s2) and diapycnal heat flux (W/m2) during June

31Hummels et al. 2013

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Deep Velocity Observations along 23°W

Equatorial Deep Jets are a dominant flow feature below the Equatorial Undercurrent and oscillate with a period of about 4.5 years (Johnson and Zhang 2003, Brandt et al. 2011)

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Equatorial Deep Jets and Basin Mode Oscillations

Downward phase and upward energy propagation

EDJ are excited at depth and propagate toward the surface

update from Brandt et al. 2011

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Equatorial Deep Jets

Excitation of equatorial basin modes (Cane and Moore, 1981)

Vertical Mode DecompositionEquatorial Deep Jets

Harmonic analysis

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Equatorial Deep JetsGreatbatch et al. (2012): EDJ can be

described by high-baroclinic, equatorial basin modes.

How are the Jets forced?1. Inertial Instability (Hua et al. 1997, d’Orgeville et

al. 2004, Eden and Dengler 2008)2. Destabilization of Rossby-gravity waves (Ascani

et al. 2006, d’Orgeville et al. 2007, Hua et al. 2008, Ménesguen et al. 2009)

Upward energy propagation toward the surface hindered by the EUC (e.g. McPhaden et al. 1986) or tunneling through the shear zone (Brown & Sutherland 2007)? 35

Deep Ocean Dynamics | Introduction Equatorial Deep Jets

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Surface Geostrophic Velocity4.5-year cycle of the geostrophic equatorial

zonal surface velocity (from sea level anomalies 15°W-35°W)

Corresponding signal of the ATL3 SST index (3°S–3°N, 20°W–0°)

36

Eastward surface flow anomaly corresponds to warm eastern equatorial Atlantic.

Brandt et al. 2011

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

EDJ interaction with the EUC?Consistent downward phase propagation below the

EUC4.5-year cycle also North, South and above the EUC

core Phases suggest meridional displacement of the EUC

core with the EDJ cycle

37

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

EDJ interaction with the EUC?Consistent downward phase propagation below the

EUC4.5-year cycle also North, South and above the EUC

core Phases suggest meridional displacement of the EUC

core with the EDJ cycle

38

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

SummaryInterannual EUC transport variability largely in agreement with zonal mode variabilityThere are noncanonical events likely associated with meridional mode events during boreal spring4.5-yr EDJ oscillations dominate depth range below the EUC: high-baroclinic, equatorial basin modesPossible interaction of basin mode and EUC (time series are hardly long enough) Improved numerical simulations are required for the understanding of physical processes responsible for EDJ affecting SST and TAV

39

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Summer (JJA) Sea Surface temperature bias pattern for CMIP5White stipples indicate where models are consistently wrong

Persistent errors in climate models with little sign of reduction

Toniazzo and Woolnough, 2013

Despite improved process understanding, model errors remained large resulting in poor TA climate prediction.

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Climate Modelling/PredictionState-of-the-art climate models still show

large errors in the SE AtlanticPossible sources: atmospheric convection,

clouds, aerosols, but similarly oceanic processes (Xu et al. 2013) like:• Advection from equatorial region, too weak stratification

• Not resolved coastal upwelling processesSeveral initiatives to improve ocean data

base in the SE Atlantic and to reduce model bias• EU PREFACE (PI Noel Keenlyside) • German SACUS (PI Peter Brandt)• NSF Proposal (PI Ping Chang)

41

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Closing knowledge gaps – enhanced observationsGulf of Guinea and Eastern Boundary Upwelling regions

Glider campaigns and cruises in 2014, 2015, and 2016, various seasons

Enhanced ARGO floats in Eastern Atlantic

8E6S, PIRATA mooring

Current meter at 0E,eq

Mooring 20S

Introduction Data & Methods EUC Transport EUC-TAV Relation Equatorial Deep Jets Summary Outlook

Current meter mooring array was deployed at 11°S off Angola during Meteor cruise in July 2013

AcknowledgementsThis study was supported by the German

Federal Ministry of Education and Research as part of the co-operative projects “NORDATLANTIK” and “RACE” and by the German Science Foundation (DFG) as part of the Sonderforschungsbereich 754 “Climate-Biogeochemistry Interactions in the Tropical Ocean”.

Moored velocity observations were acquired in cooperation with the PIRATA project.

43