NAME =  Yoshihiro Tachibana  
ORGANIZATION =  JAMSTEC / Tokai University  
ADDRESS =  Liberal Arts Education Center, Tokai Univ. Hiratsuka 2591292  
COUNTRY =  Japan  
PHONE =  463-58-1211  
FAX =  463-59-4047  
E-MAIL =  tachi@rh.u-tokai.ac.jp  
POSTER_ONLY =  no  
THEME =  T3  
DATE =  09-Aug-04-16:36:51  
ABSID = T3YT09Aug04163651 
TITLE = Nonlinear influence of diurnal variations upon time-mean fields 
AUTHOR_1 = Yoshihiro Tachibana 
INSTITUTION_1 = JAMSTEC / Tokai University 
PRESENTER = AUTHOR_1 
AUTHOR_2 = Eisho Azuma 
INSTITUTION_2 = Tokai University 
AUTHOR_3 = Shin-Ya Ogino 
INSTITUTION_3 = Kobe University 
AUTHOR_4 = Akira Watanabe 
INSTITUTION_4 = Fukushima University 
ABSTRACT =   It is widely known that diurnal variations of temperature, humidity and wind in the tropics are large through the troposphere. We evaluated influence of these diurnal variations upon the atmospheric time-mean thermodynamic fields by using GPS radiosonde special observation network data covering Indochina Peninsula in 1998. GPS radiosondes launched every six hours in three observatories for about two weeks in rainy and dry seasons. By using these data, daily mean horizontal advections of temperature and moisture were evaluated through the troposphere. The daily mean horizontal advection term can be divided into two sub processes; daily-mean temperature advection by daily mean wind and daily mean temperature, and daily mean temperature advection by perturbation wind and perturbation temperature. Each perturbation field is approximated by sinusoidal cycle with a twenty-four-hour period by using the least squares methods based on the 4-times-a-day GPS observation data. This approximation mathematically enables us to evaluate the nonlinear influence of the pure diurnal time-scale terms on the time-mean fields. The contribution of the nonlinear terms to the daily mean temperature tendency in both dry and rainy seasons were large through the troposphere especially in the lower and middle troposphere. The results suggest that objective analysis data and numerical forecast models do not accurately portray the time-mean fields if they cannot properly represent the diurnal variations.