xie

Slide1: 

Central America: A conduit for Atlantic-to-Pacific influences Shang-Ping Xie Timmermann, Y. Okumura*, S. de Szoeke, H. Xu, J. Small, T. Miyama+, Y. Wang International Pacific Research Center, University of Hawaii *UCAR; +FRCGC, JAMSTEC, Japan Water hosing experiments (GCMs) Regional model results AR4 CMIP Orographic effects

Slide2: 

Pan-Atlantic Pattern SST & wind NAO  subtropical wind Tropical meridional mode  Shift in ITCZ Xie & Tanimoto (1998, GRL) A model for interpreting paleo variability? (Chiang 2004, Hadley circulation book) WES feedback

Slide3: 

Cariaco Basin north of South America Wet tropical South America Warm Greenland Peterson et al. (2000, Science) Close connection between the subpolar and tropical North Atlantic Chiang (2004, Hadley circulation book)

Slide4: 

Annual mean response (Yr 81-100) NCAR CCSM2 ECHAM5/MPI-OM GFDL_CM2.1 HadCM3 SST & wind stress A. Timmermann, Y Okumura, et al. 2006 Meridional (WES) mode in ATL Cross-Central American winds Reduced NS asymmetry in EP

Slide5: 

January Wind (QSCAT) h (km) Use Regional Models to resolve the narrow and mountainous Central America

Slide6: 

IPRC Regional Ocean-Atmosphere Model (iROAM) on Earth Simulator Atmosphere: IPRC-RAM 0.5°×0.5°, L 28 GFDL Modular Ocean Model 2 0.5°×0.5°, L 35 Interactive Ocean forced by NCEP reanalysis Prescribed OISST Land surface model

Slide7: 

Annual-mean climatology SST, precipitation & surface wind Obs: TRMM, OISST & QSCAT 25 iROAM 25 Northward-displaced ITCZ and equatorial cold tongue

Slide8: 

Precip (gray shade), surface wind & SST (contours) in 125-95W TRMM obs IROAM Double ITCZ for a brief period of Mar-Apr ITCZ’s meridional migration Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov 20N 10N EQ 10S 20N

Slide9: 

Mesoscale Features Sfc wind, Z20(m) & orography February: Gap wind jets < 40 m

Slide10: 

Response to a 2oC cooling in the North Atlantic Large equatorial cooling in Jan-Apr  Reduced annual cycle Bjerknes feedback, instead of WES, is triggered. Implications for water-hosing experiments of relevance to Younger Dryas Xie et al. (2007, JC, in press)

Slide11: 

-2oC -2oC -2oC Dec Jan Feb Seasonality Cooling begins in the Gulf of Panama in response to an intensified Panama jet. Panama cooling coincides with the seasonal shoaling of the thermocline there. The cooling intensifies and extends to the equator along the thermocline ridge.

Slide12: 

January - March July - September Atmospheric adjustment mostly through the Panama gap. The thermocline ridge in the Gulf of Panama helps spread the SST cooling to the equator, triggering the Bjerknes feedback. Wind anomalies are larger, but with a smaller SST response.  Deepened thermocline in the Gulf of Panama. -2oC -2oC SST & Wind difference Response to a 2oC cooling in the North Atlantic

Slide13: 

Sfc wind, Z20(m) & orography February August < 40 m > 60 m Seasonality of the Pacific Response 1: Mean state Why is the coupled response favor the warm than the cold season? Mean northeasterlies in Feb  increased wind speed in response to Atl cooling  increased evaporation (wind speed & dry advection) & mixing. Shoaling thermocline in Feb (by > 20 m)  Stronger air-sea coupling in Feb than in July.

Slide14: 

Seasonality of the Pacific Response (2) Pressure Wind, Pressure & Precip (July) 1000 mb 850 mb du & q at 100W, Eq During the cool season (Jun-Oct), strong easterlies on the equator at 850 mb, but not at the surface  seasonal SST cooling and stable ABL? Anomalies patterns similar to the observed mid-summer drought: suppressed convection & intensified cross-Central American wind (Small et al. 2006). cf. Rowan’s talk

Slide15: 

Factors for strong equatorial response during the cool season (Jan-May) Pros Mean NE wind Shoaling thermocline Strong vertical coupling in wind Con Stronger atmospheric response in boreal summer Feb

Slide16: 

iROAM Common among GCMs: Reduced annual cycle cooling north/on the equator during Jan-May + LBM-POP (Axel et al.) FOAM (L. Wu et al.)

Slide17: 

Equator Wind-Evaporation-SST (WES) Feedback Why does the meridional mode develop subsequently in some models?  Prolonged southern/double ITCZ in some GCMs  stronger WES feedback involving convective heating

Slide18: 

Intercomparison of AR4 models cf. Mechoso et al. (1995), Devay et al. (2002) by Simon de Szoeke Meridional asymmetry Role of the eastern Pacific warm pool

Slide19: 

EP climate in AR4 models SST, rain & eq wind latitude

Slide20: 

NE trades over the EP warm pool vs. meridional asymmetry GCM #c

Slide21: 

0 2 4 6 8 -4 -2 0 2 4 1 2 3 4 5 6 7 8 9 a b c d e f 0 r=-0.87 FMA v EQ (m s -1 ) DJF CA wind speed ( m s-1) NE wind off Central America vs. Meridional asymmetry

Westward and disconnected equatorial cold tongue bias: 

Westward and disconnected equatorial cold tongue bias West Longitude Latitude SST GCM #5 Obs

Niño 1+2 meridional wind cools SST: 

Niño 1+2 meridional wind cools SST Nino 1+2: 80-90° W, 0-10°S

Slide24: 

Orographic Effects by Haiming Xu, Justin Small EP warm pool convection Moisture transport from the Atlantic to Pacific Basin-scale climate

Slide25: 

Winter SST & wind TMI SST & precip; QuikSCAT wind (Jan-Feb) ITCZ displaced on the south edge of EP warm pool

Slide26: 

0.25o orog January Wind (QSCAT) T42 orog 0.323 0.309 Cross-Central American moisture transport Hypothesized to hold the key to the preference for AMOC h (km) Mexico (<27N) – Panama (Jan-Feb ‘02) Xu et al. (2005, JC)

Slide27: 

Summary Tropical response to a MOC shutdown Robust response in the tropical Atlantic: the WES feedback and a meridional dipole; Weakened meridional asymmetry in the eastern Pacific in some models but the zonal mode triggered in other; The equatorial annual cycle weakens, and cooling north/on the equator during Jan-May. Response during May-Dec varies among models. AR4 CMIP & CGCM run without mountains: NE wind through Panama during Dec-Feb influences the subsequent development of meridional asymmetry (ITCZ displacement) Tropical Pacific response may be sensitive to the treatment of narrow and mountainous Central America. How can paleo data help constrain models?

Slide28: 

MOM2 OISST iROAM Upon coupling, the model SST tracks observations closely.  Air-sea feedback Ocean spin-up Coupled Nino3 SST

Slide29: 

Role of internal air-sea feedback Double ITCZ ITCZ moves back and forth across the equator Suppressed cloud radiative effect south of the equator Precip (shade), SST & sfc wind Jan Mar May Jul Sep Nov 20N 10N EQ 10S 20N

Slide30: 

T42 orog Mexico Panama q (g/kg) Wind speed (m/s) 0.25o orog