Vertical wind distribution in the tropical upper troposphere observed by Equatorial Atmospheric Radar

Vertical wind (W) distribution in and around cirriform clouds in the tropical upper troposphere was investigated with using VHF radar named EAR (Equatorial Atmospheric Radar) wind data. EAR is a wind profiler operated with VHF frequency (47 MHz), and is installed at West Sumatra (0.20S, 100.30E, 865m above sea level). A VHF wind profiler can observe a vertical wind even in heavy precipitation event by receiving an echo from atmospheric turbulence. However, a received echo from atmospheric turbulence is weak in the upper atmosphere due to the decrease of air density. Further, an expected amplitude of W in and around the cirriform clouds is weak (10-20cm/s) in the upper troposphere. To overcome these difficulties, we introduced special observational mode for W observation and manual quality control for EAR Doppler data in November 2003. We could get continuous data up to 13km in most of the period with time resolution of 12 minutes, height resolution of 150m and the accuracy of 5cm/s. The resolution and accuracy is enough to resolve the fine structure inside the cloud and they could produce the following results.

(1) The high frequency variability with a vertically standing structure and frequency a little less than Brunt Vaisala frequency was detected in most of the observational period in middle and upper troposphere. The magnitude of the variability frequently exceeds 30cm/s; this variability is the most prominent variability except convective regions in the cloud systems.

(2) We could observe the passage of several isolated typical mesoscale cloud clusters. The vertical profile of W in the stratiform rain part has the following two peculiar characteristics.

* Just after the passage of convective part, the upper two to three kilometers inside the cirriform clouds has zero or negative vertical motion (6Nov and 8Nov cases). This profile is much different from the "intermittent" or "transition" region observed in previous studies for tropical and midlatitude line shape systems.

* In the later two or three hours of stratiform rain period, the gentle upward motion with small variability was observed (6, 8 and 20Nov cases). In about three or four kilometers from middle to upper troposphere, the vertical wind has weak positive 0-40cm/s value continuously and hardly has upward motion more than 40cm/s or any downward motion. In this region, the graivity wave activity is small and there is no significant convective motion. We propose that this gentle upward motion region is identified as a subregion in the stratiform rain region.

(3)The daily mean W in the later November (13-30Nov) in the upper troposphere has positive value continuously. Daily mean value in 1-9 November (Period-I) is about 0cm/s, which is the same value as normal condition, whereas it changed to upward motion with 5cm/s in 13-30 November (Period-II). The curious thing is that the upward motion was observed even in the cloudless or very optically thin cirrus condition. The cumulus activity around the EAR site, which can be a candidate for making diabatic upward motion, did not change between both periods; it cannot explain the change in W. Furthermore, the large scale inclination of isentropic surface calculated with global analysis cannot explain the production of upward motion in Period-II. However, the large scale east-west contrast of cumulus activity has largely changed between first and second halves. Accordingly easterly wind shear strengthened and vapor in the upper troposphere has increased in Period-II. These changes in large scale feature may be result in that of vertical motion.