The Tropical Rainfall Measuring Mission (TRMM) blended rainfall product (3B42-V6 data set) has been used for inter-comparison with observations from a high density rain-gauge network over the western state of Maharashtra, India. The state of Maharashtra was chosen for the inter-comparison exercise primarily because of its diverse rainfall regime. The rain distribution over Maharashtra, characterized by a pronounced longitudinal gradient is closely linked to the north-south oriented orography of the region known as the Western Ghats. The validation exercise was carried out for 7 monsoon seasons from June to September from 1998 to 2004 at various space and time scales. The results are consistent at daily, monthly and seasonal timescales. The comparisons among stations and over geographically similar climatic zones demonstrated that the performance of 3B42-V6 product varies over the different climatic regimes. In terms of spatial reproductivity, the existence of the west-east rainfall gradient along the west coast is captured by the satellite product, but the orographic effect (rainfall maxima is over the Western Ghats as captured by the rain-gauge) is not reflected by 3B42-V6 product. The 3B42V6 product shows rainfall maxima at the coast. The satellite estimates of rainfall amounts over the state were found to be most accurate over regions of moderate rainfall and mainly inaccurate in regions of sharp rainfall gradient. In terms of magnitude of the rainfall amounts, over the windward side of the Western Ghats the 3B42V6 product was unable to resolve the heavy orographic rainfall amounts and over the leeward side the rainfall amounts in the immediate rain-shadow region were overestimated. One of the key results obtained from the daily rainfall intercomparison exercise is the ability of the 3B42-V6 estimates to detect the wet and dry phases of monsoon over most parts of the state (except the leeward side). Though the rainfall amounts estimated by the satellite product were sometimes under/over estimated, the timing of the rain events as estimated by the satellite product was generally coincident with the gauge observations over most of the regions except in the immediate rain-shadow region of the state. The TRMM 3B42-V6 estimates therefore could have tremendous potential to be used for intraseasonal studies over most regions of the state.
On 7-8 April, 2008, a sub-synoptic scale cyclone traveled eastward over the sea south of the Japan Islands. Heavy precipitation and high winds caused severe damages in the regions along the passage of the cyclone. In order to describe the structure of the cyclone and understand processes underlying its evolution, operationally observed meteorological data and numerical model output are analyzed. The result reveals that the cyclone fits a class of subtropical low (abbreviated as SL). The SL is basically a vortex that is characterized by the following features: It is accompanied by heavy precipitation and typhoon force winds (wind speed exceeding 17 m s-1), is on the meso-α scale, has contours of nearly axisymmetric surface isobars without a cold front nor a warm front, has potential temperature higher than the surrounding atmosphere around the vortex center and exhibits a cloud-free “eye”. The sea surface temperature is below 26°C. The SL under study was initiated beneath the exit region of an upper-level straight jet that became collocated with the region marked by increasing diffluence in the 300 hPa geopotential height field immediately downstream of the trough axis. While propagating over the sea south of the Japan Islands, the incipient low evolved into a meso-α scale, typhoon-like vortex. It is speculated that the intensification of the vortex was caused by latent heat released in deep convections. In terms of vorticity at low levels, the vortex reached its peak activity about 30 hours after it was born. Since then, the vortex started loosing its axisymmetric structure. A region of high vorticity began to develop at the east side of the vortex, corresponding to the development of a bent-back warm front. The T-born type frontal structure similar to the third stage of evolution in the Shapiro and Kayser’s midlatitude cyclone model eventually emerged. In the decaying stage of evolution, the spiral pattern of a potential vorticity streamer was observed, followed by formation of the spiral cloud pattern. Thus, as a distinct weather system, the concept of the SL may be broadened to include not only an intense subsynoptic-scale vortex but also a frontal system that develops at the later stage.
Three-dimensional structures of 70 tropical cyclones (TCs) that made landfall in the southern part of the main islands of Japan during 1979-2004 were examined by using the Japanese 25-year Reanalysis (JRA-25) dataset. These 70 TCs had been classified into 5 clusters C1-C5 in a previous study by means of the fuzzy c-means method based on the surface wind fields. A cyclone phase space analysis using three indicators exhibited that the average TC in each cluster has a thermal structure at its particular stage of the structural change in midlatitudes. The average C1 and C5 TCs, both of which were characterized by the area of strong winds concentrated near the cyclone center, were a relatively intense TC at the early stage of the extratropical transition (ET) and a weakening TC with a thermally symmetric structure, respectively. The C2 and C3 TCs, accompanied by the area of strong winds in the right semicircle with respect to the direction of the storm motion, were a mature TC and a thermally asymmetric, weak TC at the late ET stage, respectively. The C4 TC accompanied by strong winds on both right and left sides was a large, mature TC in August or, alternatively, a TC undergoing ET with a strong warm core in the months except for August. Composite analyses indicated that the average TC structure of each cluster is related to the environment characterized by other features such as a trough in the midlatitude westerly, the subtropical high and another TC.
To confirm the structure of convective circulation in the atmospheric boundary layer (ABL) over the northwestern Pacific Ocean around the Southwest Islands of Japan under a subtropical high, we perform intensive observations using radiosonde and Aerosonde in August 2002 and high-resolution three-dimensional numerical simulations. A well-mixed subcloud layer exists below 0.7 km during the observation period. The absolute values of the correlation coefficient between anomalies in potential temperature and mixing ratio of water vapor are mainly small (less than 0.3) in the lower subcloud layer. The frequency of a positive correlation coefficient at a height of 0.1 km is only 46%. This suggests that fewer warm moist thermals exist in the lower subcloud layer. The simulation results using 100-m horizontal grid resolution show a moist air mass in updraft regions. These thermals are driven by positive buoyant flux, and this positive buoyancy is contributed by moisture, i.e., density anomalies in water vapor and dry air, instead of heat. This should be attributed to the supply of small sensible and abundant latent heat fluxes from the sea surface under the small air-sea temperature difference.
A comparison is made of the stratospheric mean-meridional circulations, Brewer-Dobson (B-D) circulations, diagnosed from the reanalyses, JRA-25, ERA-40, ERA-Interim, NCEP/NCAR and NCEP/DOE. The reanalyses coincidently exhibit seasonality of B-D circulation, although considerable discrepancy among the reanalyses is found particularly in low-latitudes. Meridional overturning circulation at 100 hPa in the northern-hemisphere is maximal in winter, while that in the southern hemisphere is maximal in fall and significantly smaller than the northern hemispheric one. Interannual variability of B-D circulation in winter is coincident among the reanalyses, because they may reasonably represent wave-mean flow interactions of planetary waves which drive mean-meridional circulation. Yearly trends are not reliably observed due to large diversity among the reanalyses. Zonal mean vertical velocity becomes very noisy owing to inconsistency between the observation and global numerical weather prediction (NWP) model used in assimilation, except for JRA-25 and ERA-Interim. Further efforts are desired to improve reanalyses mainly through reduction of systematic errors of NWP model and implementation of advanced data assimilation schemes.