In 9804 and 9806 cruises, the crystal oscillator of the frequency generator in the 
CTD fish SBE9-plus was broken, and more noises and spikes were seen than usual in 
the raw data. Further, noises and spikes occur only momentarily in general but they 
appeared for a long time in these cruises. The summary of the standard deviations 
of the raw data while the CTD fish stops still near the surface at each cast in 9804, 
9806, 9904 and 9906 cruises is shown as follows. In consequence, the standard 
deviations of salinity in 9804 and 9806 cruises were considerably greater than other 
cruises.

  Standard deviations in 9804 cruise
　     Temperature　　　　　　　　　　　　　Salinity
    maximum value ：0.0860              maximum value ：0.0836　　
    mean value    ：0.0087　            mean value    ：0.0058
  9806 cruise
　     Temperature　　　　　　　　　　　　　Salinity
    maximum value ：0.0830              maximum value ：0.4051　　
    mean value    ：0.0073　            mean value    ：0.0282
  9904 cruise
 　    Temperature　　　　　　　　　　　　　Salinity
    maximum value ：0.0793              maximum value ：0.0289　　
    mean value    ：0.0080　            mean value    ：0.0022
  9906 cruise
 　    Temperature　　　　　　　　　　　　　Salinity
    maximum value ：0.0396              maximum value ：0.0181　　
    mean value    ：0.0038　            mean value    ：0.0026

  From these things, we tried to smooth the raw data by a different data process. 
The SNOISE.exe and SSPIKE.exe programs that remove noises and spikes were executed 
with the range expanded from WOCE standards. ALIGNCTD.exe that correct the time-lag 
between temperature sensor and conductivity sensor response was executed to minimize 
the noise by changing the interval of time constant. 

  Interval of time constant in 9804 cruise
NC3912：0.15   NC3913：0.15   NC3914：0.15   NC3915：0.20
NC3916：0.25   NC3919：0.10   NC3923：0.15   NC3924：0.10
NC3925：0.05   NC3926：0.15   NC3927：0.50   NC3928：0.20
NC3937：0.15   NC3938：0.10   NC3939：0.10   NC3940：0.15
NC3941：0.15   NC3943：0.20   NC3944：0.20   NC3945：0.15
NC3946：0.20   NC3947：0.20   NC3948：0.15   NC3949：0.20
NC3952：0.20   NC3953：0.15   NC3954：0.10   NC3955：0.25
NC3956：0.10   NC3957：0.10   NC3958：0.10   NC3959：0.30
  9806 cruise
NC3967：0.05   NC3968：0.05   NC3970：0.20  NC3974：0.10
NC3975：0.45   NC3978：0.35　 NC3979：0.25  NC3980：0.25
NC3981：0.25　 NC3982：0.30   NC3983：0.10  NC3984：0.25
NC3985：0.10   NC3986：0.10   NC3987：0.10  NC3989：0.15
NC3990：0.15   NC3991：0.25　 NC3992：0.20  NC3994：0.15
NC3995：0.15   NC3996：0.15   NC3997：0.15 

With the exclusion of above stations, we used -0.025

  However, we could not smooth out the raw data for NC3914-3916, 3922, 3923, 3926-3930 
in 9804 cruise and NC3966-3970, 3975, 3978-3982, 3991, 3992 in 9806 cruise in 
particular. So doubtful flags 3 were set on all temperature and salinity data in 9804 
and 9806 cruises. Be careful in using the CTD data.