Presentation is loading. Please wait.

Presentation is loading. Please wait.

The stratospheric and tropospheric variability around the North Pole associated with the solar cycle *1,2 Yousuke Yamashita, 2 Hideharu Akiyoshi and 1.

Similar presentations


Presentation on theme: "The stratospheric and tropospheric variability around the North Pole associated with the solar cycle *1,2 Yousuke Yamashita, 2 Hideharu Akiyoshi and 1."— Presentation transcript:

1 The stratospheric and tropospheric variability around the North Pole associated with the solar cycle *1,2 Yousuke Yamashita, 2 Hideharu Akiyoshi and 1 Masaaki Takahashi 1 Center for the Climate System Research, University of Tokyo, Kashiwa, Japan 2 National Institute for Environmental Studies, Tsukuba, Japan Address: yousuke@ccsr.u-tokyo.ac.jp

2 The northern winter variability from climatology The prominent variability from the climatology in the northern hemisphere winter is known as the north-south dipole structure of pressure between the North Pole and mid-latitude. - Zonal Index [e.g. Rossby et al., 1939; Namias, 1950] : the change in strength of the polar jet - North Atlantic Oscillation (NAO) [e.g. Walker and Bliss, 1932]: The seesaw between the Icelandic low and Azores high - Arctic Oscillation (AO) [Thompson and Wallace, 1998, 1999]: The zonal symmetric seesaw between the North Pole and mid-latitude ring. The variability associated with the north-south dipole structure is effected by the solar cycle [e.g. Labitzke, 1987]

3 The Arctic Oscillation (AO) The primary mode (most prominent variability from climatology) of the empirical orthogonal function (EOF-1) for the sea level pressure (SLP) anomaly over 20-90 o N is characterized as the north-south seesaw pattern between the North Pole and mid-latitude ring. Thompson and Wallace [1998, 1999] call this seesaw pattern Arctic Oscillation (AO). The AO is also extracted from the zonal mean geopotential height anomaly over 40-90 o N, 1000-200hPa [Ogi et al., 2004]. The AO is not “periodic oscillation”, so the AO is sometimes called Northern Hemisphere annular mode (NAM). (hereafter AO/NAM) The similar phenomenon is seen between the South Pole and the mid- latitude of southern hemisphere (Antarctic Oscillation; AAO). The AO/NAM at 500 hPa (geopotential height) This anomalous pattern is corresponding to the positive phase of the AO/NAM (AO+). In negative phase (AO-), the sign is inverse. The westerly wind around 60 o N (polar jet) is strong by the thermal wind relationship. strong westerly wind 20N

4 The zonal mean feature of the AO/NAM is characterized as the strength of the polar jet. In the terms of the transformed Eulerian mean equation, the wave forcing term is important roll in the maintenance of the AO/NAM [e.g. Yamazaki and Shinya, 1999]. In winter, the stationary component of the wave forcing is larger than that of the transient component [e.g. DeWeaver and Nigam, 2000], and the wave activity over the Pacific and Atlantic storm tracks is important [Yamashita et al., 2005, SOLA] The maintenance of the AO/NAM Simplified TEM equation http://jisao.washington.edu/wallace/ncar_notes Schematic of temperature (shading), zonal wind (contours), residual meridional circulation (thick arrows), E-P flux of planetary wave (thin arrows) E-P flux wave forcing AO+AO- forcing by residual meridional circulation EQ NP

5 The vertical connection of the AO/NAM 90 day low-pass-filtered AO/NAM signature. Red corresponds to a weak warm polar vortex (AO-), while blue indicates strong, cold vortex (AO+). (Baldwin and Dunkerton, 1999)

6 (a) The stratospheric anomaly does not propagate to the troposphere, and tropospheric and stratospheric anomalies are individually seen. (b) Poleward and downward propagation is seen, and the tropospheric anomaly became same sign to the stratospheric anomaly. (Kodera and Kuroda, 2000) The vertical connection of the AO/NAM Composite of 35-day mean anomaly for (a) type T, and (b) type S events Time Connected type: the tropospheric anomaly is coupled to the stratospheric anomaly. Non connected type: the tropospheric and stratospheric anomalies are individually seen. What factor is important to these difference of the vertical connection ?  The vertical connection of the AO/NAM is changed with the solar cycle [Kodera and Kuroda, 2005]

7 Solar difference between HS and LS mean fields The solar difference of the short wave radiation (ND). The LS mean data is subtracted from HS mean data. The ref1-00 of the CCM output is used. Short wave radiation 10 100 1000 1 NPEQSP 10 100 1000 1 strong UV radiation HS total radiation strong heating strong polar jet 10 100 1 LS total radiation NPEQSP 1000 weak polar jet weak heating weak UV radiation

8 In the early winter of the high solar (HS) period, the downward extension of zonal wind anomalies from the stratosphere and a formation of meridional dipole type anomalies in the troposphere are observed, while the downward extension is weak in the low solar (LS) period. The anomalous equatorward propagation of the waves appears and maintains the tropospheric anomaly in the HS. HS LS The solar cycle effect to the vertical connection of AO/NAM Kodera and Kuroda [2005] indicated that the vertical connection of the AO/NAM is changed with the solar cycle. [Kodera and Kuroda, 2005] The correlation maps to the zonal mean zonal wind at 10 hPa, 65 o N. Correlation to zonal mean zonal wind (contour), E-P flux (arrow) is shown. ・ Vertical connection of the AO/NAM in late winter. ・ Which periods of the wave are responsible for the vertical connection change?

9 Purpose of this study In the observational study, the connection between the stratosphere and troposphere is different with solar cycle. We attempt to know how the connection mechanism is changed with solar cycle. The strength of the winter mean polar vortex is different between high solar (HS) and low solar (LS) period with the change of the ozone heating. The dynamical difference in lower layer by the solar difference of the polar vortex only affects to the vertical connection of the AO/NAM? The transport of ozone is different though the change of the residual meridional circulation. The solar forcing may affect the residual meridional circulation with alteration of the wave forcing. In case of the AAO, the ozone plays the intensification of the vertical connection of the AAO [Kuroda et al., 2007]. Does the solar cycle difference of the ozone transport and its radiative forcing affect to the vertical connection of the AO/NAM? Since, we want to know the solar change of the chemical interaction of ozone (transport, chemical reaction and radiation of ozone), we use the AGCM include the chemical interaction with ozone (Chemistry Climate Model; CCM). To investigate the effect of the chemical interaction, we remove the chemical interaction from the CCM, and this model is compared to the original CCM.

10 Stratospheric NAM (S-NAM) index We divide the solar flux into high solar (HS) and low solar (LS) periods, depending on whether the solar flux is above or below of mean value. We perform the EOF decomposition for HS, LS individually. NCEP/NCAR reanalysis data (1979-2005) Zonal mean and Monthly mean geopotential height field The analysis domain is 40-90 o N, 10-200 hPa. We derive the principal component timeseries of the EOF-1. We call this index Stratospheric NAM (S-NAM) index. The analysis method using stratospheric index is similar to that of Kodera and Kuroda [2005]. HS LS 1980 2005 DJF mean 10.7cm solar radio flux

11 Contours are the regression coefficient, and shading indicates the correlation coefficient that exceeds the 90 % confidence level. Zonal wind regressed to the S-NAM index The amplitude of the S-NAM in late winter is larger (about 2 times) than that in early winter. The tropospheric structure is different between two maps. 10 100 1000 NPEQSP NPEQ NCEP/NCAR 10 100 1000 SP The S-NAM is different between the early and late winter

12 Contours are the regression coefficient, and shading indicates the correlation coefficient that exceeds the 90 % confidence level. Zonal wind regressed to the S-NAM index in early winter (ND) HS: The anomaly of high-latitude extends stratosphere to the troposphere. LS: Downward extension is weak. NCEP/NCAR 10 100 1000 10 100 1000 NPEQSPNPEQSP

13 Contours are the regression coefficient, and shading indicates the correlation coefficient that exceeds the 90 % confidence level. Zonal wind regressed to the S-NAM index in late winter (JF) HS: Downward extension is weak.  weak westerly acceleration is expected. LS: Downward extention is strong  strong westerly acceleration is expected. The vertical connection in late winter is inverse of that in early winter.. 10 100 1000 10 100 1000 NCEP/NCAR NPEQSPNPEQSP

14 Calculation for deviation fields We subtract daily climatology from daily mean data, and we get daily mean deviation data. Calculate the calendar day climatology from NCEP/NCAR reanalysis data (1979-2005) Remove noise with fourier analysis Original data (Z) - = Black line: simple average Red line: fourier analysis is applyed Daily climatology (Zc)Deviation data (Za)

15 Calculation for deviation fields The deviation data is divided to the transient ( 10day) component with fourier analysis. Deviation data (Za) =+ Transient (<10day) (Z H )Stationary (>10day) (Z L ) The E-P flux calculated from the deviation data

16 E-P flux regressed to the S-NAM index (ND) E-P flux regressed to the S-NAM index (JF) The wave forcing term (contour, shading), E-P flux (arrow) regressed to the S-NAM index is shown. wave forcing 60 NP 30 EQ 60 NP 30 EQ 60 NP 30 EQ 60 NP 30 EQ 10 100 1000 10 100 1000 10 100 1000 10 100 1000

17 E-P flux regressed to the S-NAM index (ND, HS) E-P flux regressed to the S-NAM index (JF, LS) transienttotalstationary total transientstationary

18 Model Description for CCM ・ CCSR/NIES Chemistry Climate Model (CCM) (based on version 5.4g of the CCSR/NIES AGCM) ・ T42L34 (top layer of the model is about 80km) ・ REF1 scenario of CCM Validation ( CCMVal-REF1 ) [Eyring et al., 2006] - density of halogen: Ab scenario ( base line: bust-guess scenario following the Beijin amendments, 1999 ) - density of greenhouse gas: IPCC-A1B scenario - QBO: nudging to observational data [Giorgetta and Bengtsson, 1999] - 11-year solar cycle: F10.7cm change [Lean et al.,1997] - surface area data of volcanic aerosol: SAGE-I, II, SAM-II - sea surface temperature: HadISST1 (UK Met. Office) ・ 3 ensemble (ref1-00, ref1-01, ref1-02) for 1980-2000 One ensemble run spends a month and 40-50GB memory. http://www.nies.go. jp/sisetu/ogata

19 Zonal mean zonal wind regressed to the S-NAM index. The regression coefficient is shown as contour. The shading indicates correlation coefficient which exceeds 90% confidence levels. The average of the 3 ensemble is used. Zonal wind regressed to the S-NAM index 10 200 1000 1 50 ND, HS JF, LS 10 200 1000 1 50 NCEP(ND, HS) NCEP(JF, LS)

20 ・ The vertical connection of the AO/NAM is analyzed with observational data. ・ In the HS during early winter, the anomaly of high-latitude extends stratosphere to the troposphere, while downward propagation is weak in the LS. This feature is similar to that in Kodera and Kuroda [2005]. In early winter, the transient wave plays important roll for the downward propagation during HS. ・ In late winter, the downward extension is strong during LS and weak during HS. This difference between the solar phases is corresponding to the difference of the stationary wave forcing in the troposphere. Results S-NAM ND, HS and JF, LS stratosph eric wave T-NAM troposphe ric wave S-NAM ND, LS and JF, HS stratosph eric wave

21 ・ The some output of CCM is not similar to the observation. We will consider the cause of these difference. ・ To investigate the effect of the chemical interaction, we remove the chemical interaction from the CCM, and this model is computing now. The results will compare to the results of the original CCM. Future works Results ・ The CCM output is used for analyzing the vertical connection of the AO/NAM. In the HS/LS during early/late winter, the vertical connection is similar to the observation.

22 Thank you for your attention

23 The difference between early (ND) and late (JF) winter The EOF-1 of the geopotential height anomaly over 40N-90N, 10-200hPa. The period is divided into max and min, then the index is calculated in each period.

24 Forcing by residual circulation regressed to the S-NAM index (ND) The forcing by residual meridional circulation (contour, shading) regressed to the S-NAM index is shown. 60 NP 30 EQ 10 100 1000 60 NP 30 EQ 10 1000 100 60 NP 30 EQ 10 100 1000 60 NP 30 EQ 10 100 1000 forcing by residual meridional circulation Forcing by residual circulation regressed to the S-NAM index (JF)

25 E-P flux regressed to the S-NAM index (ND, HS) E-P flux regressed to the S-NAM index (ND, LS) transienttotalstationary total transientstationary

26 E-P flux regressed to the S-NAM index (JF, LS) total transientstationary totaltransientstationary E-P flux regressed to the S-NAM index (JF, HS)

27 Mean and Deviation field associated with the solar difference () Similar manner, LS case can be derived. HS mean is calculated to following form.

28 Difference between HS and LS (NCEP)

29

30

31 The zonal mean structure of the S-NAM (CCM)

32 Zonal mean zonal wind regressed to the S-NAM index. The regression coefficient is shown as contour. The shading indicates correlation coefficient which exceeds 90% confidence levels. The average of the 3 ensemble is used. Zonal wind regressed to the S-NAM index 10 200 1000 1 50 ND, LS 10 200 1 50 1000 ND, HS 10 200 1000 1 50 JF, HS JF, LS 10 200 1000 1 50

33 Early winter (11-12), HS Zonal mean zonal wind regressed to the S-NAM index. The regression coefficient is shown as contour. The shading indicates correlation coefficient which exceeds 90% confidence levels. Zonal wind regressed to the S-NAM index (HD, HS) The amplitude of the AO/NAM is weaker than observation, while the downward extension is seen. It consists to the observation. REF1-00 10 200 1000 1 50 REF1-01 10 200 1000 1 50 REF1-02 10 200 1000 1 50 10 200 1000 1 50 REF1-avr

34 Zonal wind regressed to the S-NAM index (ND, LS) Early winter (11-12), LS Zonal mean zonal wind regressed to the S-NAM index. The regression coefficient is shown as contour. The shading indicates correlation coefficient which exceeds 90% confidence levels. REF1-00 10 200 1000 1 50 REF1-01 10 200 1000 1 50 REF1-02 10 200 1000 1 50 REF1-avr 10 200 1 50 1000 The AO/NAM anomaly is extend from the stratosphere to troposphere, although the downward extension is weak in observation. The amplitude is weaker than observation.

35 Zonal wind regressed to the S-NAM index (JF, HS) Late winter (1-2), HS Zonal mean zonal wind regressed to the S-NAM index. The regression coefficient is shown as contour. The shading indicates correlation coefficient which exceeds 90% confidence levels. The amplitude and vertical connection of S-NAM in ref1-02 is similar to that in observation. The S- NAM in ref1-00, 01 is much weaker than observation, and downward extension is seen. REF1-00 10 200 1000 1 50 REF1-01 10 200 1000 1 50 REF1-02 10 200 1000 1 50 REF1-avr 10 200 1000 1 50

36 Zonal wind regressed to the S-NAM index (JF, LS) Late winter (1-2), LS Zonal mean zonal wind regressed to the S-NAM index. The regression coefficient is shown as contour. The shading indicates correlation coefficient which exceeds 90% confidence levels. The vertical connection of S-NAM in ref1-01 and ref1-02 shows similar structure to that in observation. In contrast, in ref1-00 vertical connection is weak. REF1-00 10 200 1000 1 50 REF1-01 10 200 1000 1 50 REF1-02 10 200 1000 1 50 REF1-avr 10 200 1000 1 50

37 Solar difference between HS and LS (CCM)

38 Solar difference between HS and LS mean fields (ND) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-00 of the CCM output is used. Short wave radiationTemperature Zonal Wind

39 REF1-00REF1-01 Solar difference of the short wave radiation (ND, HS-LS) REF1-02 The solar difference of the short wave radiation.

40 Solar difference of the short wave radiation (JF, HS-LS) The solar difference of the short wave radiation. REF1-00REF1-01 REF1-02

41 Solar difference between HS and LS mean fields (ref1-00, ND) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-00 of the CCM output is used. Short wave radiationTemperature Zonal Wind

42 Solar difference between HS and LS mean fields (ref1-00, JF) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-00 of the CCM output is used. Short wave radiationTemperature Zonal Wind

43 Solar difference between HS and LS mean fields (ref1-01, ND) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-01 of the CCM output is used. Short wave radiation Zonal Wind Temperature

44 Solar difference between HS and LS mean fields (ref1-01, JF) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-01 of the CCM output is used. Short wave radiationTemperature Zonal Wind

45 Solar difference between HS and LS mean fields (ref1-02, ND) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-02 of the CCM output is used. Short wave radiationTemperature Zonal Wind

46 Solar difference between HS and LS mean fields (ref1-02, JF) The solar difference of the short wave radiation (upper left), temperature (upper right) and zonal wind (lower left). The LS mean data is subtracted from HS mean data. The ref1-02 of the CCM output is used. Short wave radiationTemperature Zonal Wind

47 Temperature (ND) REF1-00REF1-01 REF1-02

48 REF1-00REF1-01 REF1-02 Temperature (JF)

49 Zonal wind (ND) REF1-00REF1-01 REF1-02

50 Zonal wind (JF) REF1-00REF1-01 REF1-02

51 The S-NAM and T-NAM

52 Index is calculated, then the index is divided into max and min. The blue line indicates the S-NAM index, and red line indicated the T-NAM index. The EOF-1 of the geopotential height anomaly over 40N-90N, 10- 200hPa. The EOF-1 of the geopotential height anomaly over 40N-90N, 200- 1000hPa.

53 The period is divided into max and min, then the index is calculated in each period. The blue line indicates the S-NAM index, and red line indicated the T-NAM index. The EOF-1 of the geopotential height anomaly over 40N-90N, 10- 200hPa. The EOF-1 of the geopotential height anomaly over 40N-90N, 200- 1000hPa.

54 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

55 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

56 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

57 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

58 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

59 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

60 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

61 Index is calculated, then the index is divided into max and min The period is divided into max and min, then the index is calculated in each period

62 S-NAM and T-NAM (ND) T-NAM S-NAM

63 S-NAM and T-NAM (JF) S-NAM T-NAM

64 S-NAM and T-NAM (ND, HS) S-NAM S-NAM and T-NAM (ND, LS) S-NAM T-NAM

65 S-NAM and T-NAM (JF, HS) S-NAM and T-NAM (JF, LS) S-NAM T-NAM

66

67 correlation is not significant QBO: E QBO: W HS: warm, LS: cold NH Polar region temperature in late winter (Labitzke, 2005) warmwarm coldcold Monthly mean 30 hPa temperature in February HSLSHSLS

68 The solar influence between the QBO and the AO/NAM (early, late) (Naito and Hirota, 1997) Early, HS Early, LS Late, HS Late, LS Early winter: little difference between HS and LS. Late winter: in LS, every W winter has a stronger polar-night-jet than any in the E. in HS, jet is no stronger in the W than in the E. AO+ AO- AO+ AO-

69 The deviation from zonal mean

70 E-P flux regressed to the S-NAM index (ND) E-P flux regressed to the S-NAM index (JF) The wavc forcing term (contour, shading), E-P flux (arrow) regressed to the S-NAM index is shown. wave forcing

71 stationarytransient stationarytransient E-P flux regressed to the S-NAM index (ND) total

72 E-P flux regressed to the S-NAM index (JF) stationarytransient stationarytransient total

73 JFND The yearly timeseries of the meridional gradient of the potential vorticity is shown. The area average over 50-70N, 700-100 hPa is calculated. The meridional gradient of the PV

74

75

76 Time mean and Zonal mean Time meanZonal mean


Download ppt "The stratospheric and tropospheric variability around the North Pole associated with the solar cycle *1,2 Yousuke Yamashita, 2 Hideharu Akiyoshi and 1."

Similar presentations


Ads by Google