Chinese Journal of Atmospheric Sciences Vol131 No16 Nov , , , SAMIL, R42L26,,,,, 700 hpa, Tiedtke,, 2,,,,

Size: px

Start display at page:

Download "Chinese Journal of Atmospheric Sciences Vol131 No16 Nov , , , SAMIL, R42L26,,,,, 700 hpa, Tiedtke,, 2,,,,"

切穷方
5 years ago
Views:

1 Vol131 No16 Nov , , , SAMIL, R42L26,,,,, 700 hpa, Tiedtke,, 2,,,, (2007) P426 A The Impacts of the Cumulus Convective Parameterization on the Atmospheric Water2Content and Ra infall Simulation in SAMIL L IU Yi2Min 1, L IU Kun 1,2, and WU Guo2Xiong 1 1 S tate Key L aboratory of N umerical Modeling f or A tmospheric Science and Geophysical Flui d Dy namics, Institute of A tmos2 p heric Physics, Chinese A cadem y of S ciences, Bei j ing Graduate Universit y of Chinese A cadem y of S ciences, Bei j ing Abstract The new version of the State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geo2 physical Fluid Dynamics/ Institute of Atmospheric Physics ( L ASG/ IA P) spectral atmospheric model SAMIL _ R42L26 has been developed and evaluated since The model shows it s ability in representing the global climate, but it also suffers from some problems in simulating precipitation in the tropics, such as the double ITCZ in the cen2 tral2western Pacific, less precipitation in the equator area and in the mid2latitudes of the Northern Hemisphere, and more in Latin America. These biases seem to result mainly f rom the treatment of subgrid scale convection, which is parameterized with the Tiedtkeπs mass flux scheme. A sensitivity experiment shows that these systematic errors are linked each other, and owe their existence mainly to the dry bias of the lower troposphere. In view of these results, several modifications are made to the Tiedtkeπs mass flux scheme used in SAMIL, and the results show apparent im2 provements in simulating the climate compared with the control experiment. In particular, the double ITCZ is e2 liminated and the modeled precipitation in the northern tropical area is more realistic. The modified scheme can also , CB403607,, ,, 1965,,, E2mail iap. ac. cn 3 E2mail : iap. ac. cn

2 Vol1 31 reduce both cold and dry biases in the middle troposphere. Result s f rom the compared experiment show that the im2 provement s in the model performance mainly result f rom the improved shallow convection due to the introduction of the modified convective parameterization. Key words general circulation model, cumulus convective parameterization, mass flux convection scheme, shallow convection 1 SAMIL2R42L26, 42, ( ) ( ), 2p, 26, hpa, [1, 2 ] Edwards2Slingo [3 ] Slingo [4, 5 ], [6, 7 ], [8 ], [9 ], Tiedtke, [ 10 ] 2005, SAMIL2R42L26,,, ( ITCZ),, ( PBL) [11, 12 ] [13 15 ] [16 18 ], [19, 20 ] SAMIL2R42L26 Tiedt ke [21 ], Nordeng [22 ] Tidet ke, Tiedt ke ( ECHAM510 BAM3. 0 ECWMF ),,,,, Wang [ ] Li [26 ], Tidet ke SAMIL2 R42L26, 2 SAMIL ; 3 SH1 ; Tiedt ke 4, 5 2 SAMIL2R42L26, SAMIL2R42L26 14, 12, ERA240, b ( TCTL) DJ F ( ),, 10 S (SPCZ),,

3 6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric Xie2Arkin (a c) TCTL (b d) : (a b) ; (c d) : 4 mm/ d (10 mm/ d ) 6 mm/ d (10 mm/ d ) ; : > 6 mm/ d Fig. 1 The precipitation distributions derived from Xie2Arkin observation (a, c) and t he control experiment TCTL of long2term inte2 gration (b, d) : (a, b) Boreal winter (DJ F) ; (c, d) boreal summer (JJ A). Isoline intervals : 4 mm/ d (below 10 mm/ d), 6 mm/ d (a2 bove 10 mm/ d) ; shading : > 6 mm/ d Xie2 Arkin [27 ] ( 1a),, 10 N, 2 mm/ d, JJ A (6 7 8 ) ( 1d), ( 1c), 2 mm/ d, 2. 2 (DJ F) ERA2 40 [28 ] ( 2a) 2b 10 %, 850 hpa ( 20 % ) ;, 10 % 15 % ; 600 hpa, 10 % 2c d 2b,,, ( 20 % ) ; 850 hpa 700 hpa, 0 10 N,, 15 % ERA240 ( 2e) ( 2f) :, - 4 K, 2 K ,, SH1, ERA240, qobs, q [ - (1/ ) ( q - qobs) ] : 5q 5t + v q + w 5q 5z = - ( c - e) - 1 ( q - qobs ), (1)

4 Vol ERA240 (a c) ERA240 (b d e f) : (a b) ( %) ; (c d) ( %) ; (e) ; (f) : K Fig. 2 Distributions of t he zonal mean fields derived from ERA240 reanalysis data (a, c) and t he differences between t he zonal mean fields derived from t he TCTL experiment and ERA240 reanalysis data (b, d, e, f) : Relative humidity ( %) in DJ F (a, b) and JJ A (c, d) ; temperature ( K) in DJ F (e) and JJ A (f), q, c e,, r,, r = robs t + qsat t- t qsat t q = qsat r, (2) r t- t - robs t- t e - t, (3), qsat, t (10 ), = 15, 15, ERA ( ) 6 SH1 3, 3

6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric... 1205 3 SH1 ERA240 : (a) (c) ( %) ; (b) (d) ( : K) Fig.

5 6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric SH1 ERA240 : (a) (c) ( %) ; (b) (d) ( : K) Fig. 3 The differences between t he zonal mean fields derived from t he SH1 experiment and ERA240 reanalysis data : Relative humidity ( %) in DJ F (a) and JJ A (c) ; temperature ( K) in DJ F (b) and JJ A (d) 3. 2 SH1 3a SH1 30 N 30 S, 10 %, SH1 10 % 500 hpa,, 10 % 15 % ( 3c), SH1 25 S, 850 hpa ( 20 % ) SH1 3b d, SH1 : 700 hpa, 2 K ; - 4 KSH1-2 K,, - 2 K, 2 4 K SH1 0 2 K,,, ITCZ( 4) 4 Tiedtke Tiedt ke, ( R H)

1206 31 Vol1 31 4 (a c) SH1 (b d) ( : mm/ d) : (a b) ; (c d) : 4 mm/ d (10 mm/ d ) 6 mm/ d (10 mm/ d ) ; : 6 mm/ d Fig.

6 Vol (a c) SH1 (b d) ( : mm/ d) : (a b) ; (c d) : 4 mm/ d (10 mm/ d ) 6 mm/ d (10 mm/ d ) ; : 6 mm/ d Fig. 4 The precipitation distributions derived from t he TCTL experiment (a, c) and t he SH1 experiment (b, d) in DJ F (a, b) and JJ A (c, d). Isoline intervals : 4 mm/ d (below 10 mm/ d), 6 mm/ d (above 10 mm/ d) ; shading : > 6 mm/ d 5 3, Tiedt ke ERA240 Fig. 5 Same as Fig. 3, but for t he differences between t he SH2 experiment and ERA240 reanalysis data

7 6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric ,, 4. 1 Wang [ ] Li [26 ], Tiedt ke : (1) m m - 1, [29 ] ; (2) s s - 1,, Tiedt ke SAMIL2R42L26 14, 12 SH2 SH1, SH2 R H ( 5a), 30, 10 % 500 hpa ( 5c), ( 20 %) 850 hpa, R H, 15 %20 % ( 5b) ( 5d) SH1-4 K SH2-2 K,,, 2 K 0 2 K SH2 SH1, Tiedt ke, SH1 R H,, SH2 Tiedt ke :,,, 6 ERA240 SH2 850 hpa (SPCZ), TCTL R H, 50 % ; SH2 60 %, 850 hpa SH2 R H 10 %, SH2 R H 60 %, ERA240, 700 hpa, SH2 R H ( ), SH2, 20 N 20 S, SH2 850 hpa, 700 hpa,, 7 SH2, 925 hpa 400 hpa, SH2,, 700 hpa, 700 hpa, 850 hpa, ; SH2, 850 hpa,,,, 8 SH2, < 0 SH2,, Hadley,

1208 31 Vol1 31 6 ERA240 (a d), (b e) SH2 (c f) 850 hpa ( %) : (a b c) ; (d e f) : 10 % Fig.

8 Vol ERA240 (a d), (b e) SH2 (c f) 850 hpa ( %) : (a b c) ; (d e f) : 10 % Fig. 6 The 850 hpa relative humidity ( %) distributions derived from ERA240 reanalysis data (a, d), t he TCTL experiment (b, e), t he SH2 experiment (c, f) : (a, b, c) DJ F ; (d, e, f) JJ A. The isoline interval is 10 % 7 (a c) SH2 (b d) ( %) : (a b) ; (c d) Fig. 7 The zonal mean distributions of cloud fraction derived from t he TCTL experiment (a, c), t he SH2 experiment (b, d) : (a, b) DJ F ; (c, d) JJ A

6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric... 1209 8 (a d) SH2 (b e) (c f) ( : hpa/ d) : (a b c) ; (d e f) < 0 Fig.

(d, e, f) JJ A. Shading : < 0 9 (a c) SH2 (b d) : (a b) ; (c d) : 4 mm/ d (10 mm/ d ) 6 mm/ d (10 mm/ d ) ; : > 6 mm/ d Fig.

9 6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric (a d) SH2 (b e) (c f) ( : hpa/ d) : (a b c) ; (d e f) < 0 Fig. 8 The zonal mean distributions of vertical velocity derived from t he TCTL experiment (a, d), t he SH2 experiment (b, e), and t heir differences (c, f) (unit s : hpa/ d) : (a, b, c) DJ F ; (d, e, f) JJ A. Shading : < 0 9 (a c) SH2 (b d) : (a b) ; (c d) : 4 mm/ d (10 mm/ d ) 6 mm/ d (10 mm/ d ) ; : > 6 mm/ d Fig. 9 The precipitation distributions derived from t he TCTL experiment (a, c), t he SH2 experiment (b, d) : (a, b) DJ F ; (c, d) JJ A. I2 soline intervals : 4 mm/ d (below 10 mm/ d), 6 mm/ d (above 10 mm/ d) ; shading : > 6 mm/ d, 9 SH2, SH2, 10 N, ; 2 mm/ d 8 mm/ d,

10 Vol1 31 ( 1a c), SH2, ( 9 ),,,,, 5, SAMIL2R42L26,,,,,,,, Tiedt ke SAMIL2R42L26 :,,,,, :,,,, SH2,, 90 Tiedt ke, ERA240 ECMWF ( European Centre for Medium2Range Weather Forecasts, http :/ / www. ecmwf. int),! ( References) [ 1 ] Bonan G B. The land surface climatology of t he NCAR land surface model coupled to t he NCAR Community Climate Model. J. Cli mate, 1998, 11 : [ 2 ],,,. 2 2 I :., 2005, 21 : Wang Z Z, Wu G X, Liu P, et al. The development of GOALS/ LASG A GCM and its global climatological features in climate simulation. I : Influence of horizontal resolution. J ournal of T ropical Meteorology (in Chinese), 2005, 21 : [ 3 ] Edwards J M, Slingo A. Studies wit h a flexible new radiation code. I : Choosing a configuration for a large2scale model. Quart. J. Roy. Meteor. S oc., 1996, 122 : [ 4 ] Slingo J M. A cloud parameterization scheme derived from GA TE data for use with a numerical model. Quart. J. Roy. Meteor. S oc., 1980, 106 : [ 5 ] Slingo J M. The development and verification of a cloud pre2 diction scheme for the ECMWF model. Quart. J. teor. S oc., 1987, 113 : Roy. Me2 [ 6 ]. LASG F GCM20., Dai Fushan. Impacts of low2level cloud over the eastern Pa2 cific on t he Double ITCZ in an ocean2at mosphere coupled model Diagnostic analyses based on LASG F GCM20. Ph. D. dissertation (in Chinese), Institute of Atmospheric Physics, Chinese Academy of Sciences, [ 7 ] Teixeira J, Hogan T F. Boundary layer clouds in a global at2 mospheric model : Simple cloud cover parameterizations. J. Climate, 2002, 15 : [ 8 ],,,. L ASG/ IAP., 2006, 30 : Bao Qing, Liu Yimin, Zhou Tianjun, et al. The sensitivity of the spectral atmospheric general circulation model of LASG/ IAP to the land process. Chinese J ournals of A tmospheric Sciences (in Chinese), 2006, 30 : [ 9 ] Palmer T N, Shutt s G J, Swinbank R. Alleviation of a sys2 tematic westerly bias in general circulation and numerical weat her prediction models t hrough an orographic gravity wave drag parameterization. Quart. J. Roy. Meteor. S oc., 1986, 112 : [10 ]..,

11 6 No1 6 : L IU Yi2Min et al. The Impact s of the Cumulus Convective Parameterization on the Atmospheric Song Xiaoliang. The evaluation analysis of two kinds of mass2 flux cumulus parameterizations in climate simulation. Ph. D. dissertation (in Chinese), Institute of Atmospheric Physics, Chinese Academy of Sciences, [ 11 ] Riehl H, Yeh T C, Malkus J S, et al. The nort heast trade of the Pacific Ocean. Quart. J. Roy. Meteor. S oc., 1951, 83 : [ 12 ] Nitta T, Esbensen S. Heat and moisture budget analyses u2 sing BOMEX data. Mon. Wea. Rev., 1974, 102 : [ 13 ] Holland J Z, Rasmusson E M. Measurement s of at mospheric mass, energy and momentum budgets over a 5002kilometer square of t ropical ocean. Mon. Wea. Rev., 1973, 101 : [ 14 ] Albrecht B A. Parameterization of trade2cumulus cloud a2 mount. J. A tmos. S ci., 1981, 38 : [ 15 ] Raga G B, Jensen J B, Baker M B. Characteristics of cumu2 lus band clouds off the coast of Hawaii. J. A tmos. Sci., 1990, 47 : [ 16 ] Siebesma A P, Cuijpers W M. Evaluation of parametric as2 sumptions for shallow cumulus convection. J. A tmos. Sci., 1995, 52 : [ 17 ] Stevens B, Ackerman A S, Albrecht B A, et al. Simulation of trade wind cumuli under a strong inversion. J. A tmos. S ci., 2001, 58 : [ 18 ] Brown A R, Cederwall R T, Chlond A, et al. Larger2eddy simulation of the diurnal cycle of shallow cumulus convection over land. Quart. J. Roy. Meteor. Soc., 2002, 128 : [ 19 ] Riehl H, Malkus J S. On t he heat balance and maintenance of circulation in the trades. Quart. J. Roy. Meteor. S oc., 1957, 77 : [ 20 ] Simpson J. Global circulation and tropical cloud activity. In : Theon J S, Matsuno T, Skata T, et al ( Eds). The global role of tropical rainfall. A Deepak Publishing, Hampton, [ 21 ] Tiedt ke M. A comprehensive mass flux scheme for cumulus parameterization in large2scale models. Mon. Wea. Rev., 1989, 117 : [ 22 ] Nordeng T E. Extended versions of t he convective parame2 terization scheme at ECMWF and t heir impact on the mean and transient activity of t he model in the tropics. Tech. Memo., 206, ECWMF, Reading, U K, 1994 [ 23 ] Wang Y, Sen O L, Wang B. A highly resolved regional cli2 mate model ( IPRC2RegCM) and its simulation of the 1998 se2 vere precipitation event over China. Part I : Model description and verification of simulation. J. Cli mate, 2003, 16 : [ 24 ] Wang Y, Xie S2P, Xu H, et al. Regional model simulations of marine boundary layer clouds over t he Sout heast Pacific off South America. Part I : Control experiment. Mon. Wea. Rev., 2004, 132 : [ 25 ] Wang Y, Xu H, Xie S2P. Regional model simulations of ma2 rine boundary layer clouds over t he Sout heast Pacific off South America. Part II : Sensitivity experiments. Mon. Wea. Rev., : [ 26 ] Li L2J, Wang B, Wang Y, et al. Improvement s in climate simulation with modifications to the Tiedtke convective pa2 rameterization in the grid2point atmospheric model of IAP/ LASG ( GAMIL). A dv. A tmos. Sci., 2007, 24 : [ 27 ] Xie P, Arkin P A. A 172year mont hly analysis based on gauge observation, satellite estimates, and numerical model output s. B ull. A mer. Meteor. S oc., 1997, 78 : [ 28 ] Uppala S M, K llberg P W, Simmons A J, et al. The ERA2 40 re2analysis. Quart. J. Roy. Meteor. S oc., 2005, 131 : doi : / qj [ 29 ] Siebesma A P, Holt slag A A M. Model impact s of ent rain2 ment and detrainment rates in shallow cumulus convection. J. A t mos. Sci., 1996, 53 :

Vol1 34 feedback. Finally the eastward extension of precipitation bias along the SPCZ ( Southern Pacific Convergence Zone) can be effectively r

Vol1 34 feedback. Finally the eastward extension of precipitation bias along the SPCZ ( Southern Pacific Convergence Zone) can be effectively r 34 1 2010 1 Vol134 No11 Jan. 2010,,. 2010. SAMIL Tiedt ke [J ]., 34 (1) : 163-174. Liu Kun, Liu Yimin, Wu Guoxiong. 2010. The impacts of the modified Tiedtke cumulus convective parameterization scheme