15B.2 STATISTICAL ANALYSIS OF SURAFCE WIND DISTRIBUTION OF TYPHOONS ON WESTERN NORTH PACIFIC OBSERVED BY SCATTEROMETER FOR 9 YEARS Kotaro Bessho*, Tetsuo Nakazawa, and Mitsuru Ueno Meteorological Research Institute, Tsukuba, Japan 1. INTRODUCTION Surface wind distribution of typhoons over the western North Pacific is most important information for the safety of ships. For example, the Regional Specialized Meteorological Center (RSMC) Tokyo Typhoon Center in Japan Meteoro- logical Agency (JMA) issues the advisories of analysis and forecasts of typhoons including maximum sustained wind speed and wind radii of over 50- and 30-knot wind areas. Such wind information is based on the hypothesis in which typhoons have almost axisymmetric surface wind distribution. On the other hand, from the observations by scatterometer on the low orbital satellite of QuikSCAT, it has been empiri- cally understood that there are sometimes non-axisymmetric wind structures in typhoons. In this study, typical surface wind distributions within ty- phoons over the ocean are revealed statistically for the first time by using the whole QuikSCAT observational data for typhoons from the start of its operation in 1999 to 2007. Es- pecially, the asymmetry of the wind horizontal distribution is emphasized in this paper. 2. DATA AND METHOD Sea surface wind data of tropical cyclones observed by QuikSCAT are archived in the web site of Remote Sensing Systems (RSS). The SeaWinds on QuikSCAT launched in June 1999 is a Ku-band (a frequency near 14 GHz) radar which works as a scatterometer to transmit a radar pulse down to the sea surface and measures the power scattered back from sea (Katsaros et al. 2001). The power of the backscattered radiation is proportional to the ocean surface roughness correlated with the near-surface wind speed and direction. It is considered that QuikSCAT can easily detect a TC as a closed circulation in the surface winds on the ocean. In the RSS database, QuikSCAT sea surface wind was re- trieved from the measurements of the power of the back- scattered radiation by a geophysical model function called as Ku-2001 (Wentz et al. 2001). The wind speed and direction data are almost equivalent to those with averaging period of 8-10 minutes at a height of 10 meters. In this study, ver- sion-03a QuikSCAT surface wind swath data set was used. First, QuikSCAT sea surface wind data which include ty- phoon observations were downloaded. Then the swath data were clipped within the square 1,000 km on a side centered on typhoon positions which were interpolated from JMA RSMC typhoon best track data adjusted to the QuikSCAT observational time. Finally, the clipped data were composited into the 1,000 km x 1,000 km horizontal 2-D with the resolu- tion of 25 km relative to motion direction of typhoons in ac- cordance with the purpose of each analysis. The motion speed of typhoons was not subtracted from the surface wind speed. The snapshots, which do not cover more than 50 % of the circle with radius 400 km from the center of typhoon, were eliminated from the analyses. In this study, there are 1,599 QuikSCAT observational cases on western North Pacific from 1999 to 2007 (Figure 1). 3. RESULTS 3.1 COMPOSITE IN ALL CASES Figure 2 shows sea surface wind distribution composited in all 1,599 cases of QuikSCAT observation on western North Pacific. The averaged motion speed of typhoons is 2.9 m/s. The wind speed distribution is non-axisymmetric, and shows wave number 1. It is found the typhoons have basically asymmetric surface wind distribution. Maximum wind speed is more then 18 m/s, and is located on the right side of the center. This wind distribution is coincident with the fact of so-called dangerous semicircle, and reflects the adding from the motion speed of typhoons. 3.2 CLASSIFICATION BY MOTION SPEED OF TYPHOON Figure 3 shows composited sea surface wind distributions classified by the motion speed of typhoon. In JMA, the speed of typhoon is divided into three classes which are stationary (Fig. 3 (a), STR, less than 1.5 m/s), slow (Fig. 3 (b), SLW, less than 2.6 m/s) and others (Fig. 3 (c), OTH, more than 2.6 m/s). The wind distributions in SLW and OTH is similar to those composted in all cases and show strong asymmetry (Fig. 1). The distribution in STR is almost axisymmetric, but the small and clear peak is found in the first quadrant of the dis- tribution. Because it is considered that the motion speed of typhoon has no effect on the surface wind distribution in STR cases, there is possibility of influences of another physical parameters. At this time, it is not clear what kind of parame- ters affect this asymmetry of the wind distribution in STR. *Corresponding auther address: Kotaro Bessho, Meteoro- logical Research Institute, Tsukuba, Japan, 305-0052; e-mail: