The major features of the East Asian subtropical westerly jet (EASWJ) in the upper troposphere simulated by the two versions of CCSR/NIES/FRCGC climate system model (MIROC_Hires and MIROC_Medres) are examined by analyzing the differences between the coupled model 20th century simulations and the NCEP/NCAR reanalysis, focusing on the evaluation of the model performances in reproducing the mean EASWJ structures, the seasonal evolution, interannual variability, and the relationship among the EASWJ seasonal evolution, the meridional temperature gradient and the diabatic heating in the upper troposphere. The mean EASWJ vertical and horizontal structures, the seasonal evolution, and the correspondence of the EASWJ location to the meridional temperature gradient in the upper troposphere are well simulated in the coupled models. The increase in model resolution can improve the simulation of the EASWJ structures, seasonal evolution and interannual variability. However, both coupled models overestimate the EASWJ intensity in winter, and underestimate the jet intensity in summer, relative to the NCEP/NCAR reanalysis. The biases in model EASWJ intensity are found to be associated with biases in meridional temperature gradients in the troposphere, and furthermore with the surface sensible heat flux in summer and convective condensation heating in winter as well as the meridional heat transport gradient. The coupled models simulate well the seasonal evolution of the diabatic heating averaged between 30°N-45°N, and its association with the westerly jet. However, the simulated maximum diabatic heating in summer is located eastward compared with the reanalyzed position, with a relatively weak diabatic heating intensity, especially in MIROC_Hires, while the MIROC_Medres model reproduces relatively strong diabatic heating near 130°E in winter. This study suggests that the condensation latent heating over the western Pacific in winter, the surface sensible heating over the northern side of the Tibetan Plateau in summer and the meridional heat transport gradient determine the EASWJ intensity, location and structure as well as its seasonal evolution. Thus the reasonable reproductions of the meridional heat transport gradient and the surface diabatic heating are the key points for improving the EASWJ simulation by the MIROC model.
We examined upper cloud variation over East Asia and the western North Pacific using 3-hourly Geostationary Meteorological Satellite infrared (GMS-IR) observations between 1988 and 1997. Of particular focus was the relationship to medium-scale tropopausal waves (MTWs) developed near the tropopause around mid-latitude westerlies. One- to two-day variation in upper clouds was significant throughout the year, except in summer because of eastward-traveling upper cloud systems developing in the southerly of MTWs. This finding confirms a previous theoretical prediction of southerly updraft appearance. The largest 1-day variation in upper cloudiness occurred over the Tibetan Plateau. Large variation also occurred over the China Plain to the east of the Tibetan Plateau and over the East China Sea, but less variation occurred over Japan and the western North Pacific to the east. Since appearance of cloud systems of MTWs near the eastern edge of the Tibetan Plateau is diurnally regulated, the phases of 1-day variation of cloud systems of MTWs over the China Plain are determined by the eastward phase speed of MTWs and the distance from the Tibetan Plateau. When the phase speed of MTWs is uniform, systematic phase delay of diurnal variation in upper cloudiness is significant over China Plain, which has been observed as a distinct phenomenon (Asai et al. 1998). The phase-locked appearance of MTW cloud systems can be ascribed to diurnal variation in upper-tropospheric circulation around the Plateau, which is related to updrafts that develop in the afternoon over the Plateau. This suggests that thermal forcing over the Tibetan Plateau may initiate MTWs and affect their behavior over East Asia, where the magnitude of 1-day variation is larger than that of MTWs in other mid-latitude regions.
Spring rainfall in Taiwan can be either enhanced or suppressed by an El Niño event, revealing an asymmetric relationship. This observational study aims at examining this asymmetric relationship and associated large-scale dynamic processes. Analysis results disclose four major El Niño/Southern Oscillation (ENSO)-spring rainfall relationship types during 1950-2003: El Niño-anomalous wet (EN-w) type, La Niña-anomalous dry (LN-d) type, El Niño-anomalous dry (EN-d) type, and La Niña-anomalous wet (LN-w) type. The EN-w and LN-d (EN-d and LN-w) types exhibit a positive (negative) correlation between the spring rainfall anomaly and the ENSO-related sea surface temperature anomaly (SSTA). The overall ENSO-spring rainfall relationship is dominated by the positive correlation. The cause of the asymmetric ENSO impacts between the positive- and negative-correlation groups is attributed to a connection between ENSO and the Indian Ocean (IO) SSTA and associated large-scale atmospheric circulation. The positive-correlation types tend to concur with an evident ENSO-IO connection, featuring significant in-phase SSTA centers in the tropical eastern Pacific and eastern IO. During the El Niño (La Niña) event, these SST anomalies force a descending (ascending) branch over the western Pacific and help initiate and maintain a lower-level anticyclone (cyclone) anomaly in the Philippine Sea (southeast of Taiwan). Flows west of this anomalous anticyclone (cyclone) enhance (suppress) moisture transport from the South China Sea into Taiwan, resulting in increased (decreased) spring rainfall. For the negative-correlation types, the ENSO-IO connection tends to be weak or broken. The significant SSTA appears only in the tropical eastern Pacific, which induces a major vertical motion branch over the maritime continent. As a result, an anomalous lower-level anticyclone (cyclone) occurs over the Asian continent (west of Taiwan) during the El Niño (La Niña) event. Flows east of this anomalous anticyclone (cyclone) weaken (strengthen) moisture transport from the South China Sea into Taiwan, leading to suppressed (enhanced) spring rainfall. It is also noted that the variability of the Pacific subtropical high (PSH) over the western Pacific is closely linked to these four relationship types. The EN-d (LN-w) type concurs with a moderate westward expansion (eastward retreat) in the western-Pacific sector, while the EN-w (LN-d) type is concurrent with a southward (great eastward) displacement.
The 18-day continuous heavy rainfall occurrence from July 31 to August 17, 1998, over southern Korea is investigated to understand the synoptic-scale characteristics and development mechanisms of the long-lasting heavy rainfall. An elongated monsoon front is maintained over central China throughout the Korean Peninsula for about 20 days under strong summer time baroclinic development between the continental polar lows north of Korea and the northwestward extended Northwestern Pacific high. The remarkable features of the synoptic-scale environment that resulted in heavy rainfall events are the quasi-stationary location of the upper-level jet and the low-level jets. Further, there is a sufficient moisture supply from the low-level jets that are intensified by the typhoons in landfall on southern China and over the East China Sea. The daily heavy-rainfall event over southern Korea is strongly tied to the intensity and the north-south oscillation of the upper-level jet that locates just to the north of the heavy-rainfall area, as well as the diabatic heating process that well corresponds to the upward motion in the area of the heavy rainfall according to the solution of the Sawyer-Eliassen equation. The periods of 2-3 days and 1 day appearing in the variations of 200 hPa v-component wind and u-component wind, respectively, correspond well to those of about 2 days and 1 day appearing in the spectrum of hourly rainfall data. The 1 day period in the variation of 200 hPa v-wind fields demonstrates that the heavy rainfall over Korea was enhanced by the secondary circulation around the upper-level jet. The period also appears approximately in the variations of heat source, moisture sink, and vertical motion over Korea.
Possible feedbacks between the midlatitude atmosphere and ocean relevant to maintain the mean climatological state in January are examined by using a coupled model with intermediate complexity (ICM), which consists of a 2-layer atmospheric general circulation model (AGCM) and a 2.5-layer shallow water ocean model covering the Pacific Ocean. Two series of the numerical experiments including the seasonal cycle were conducted: the experiments using the coupled model (ICM run) and the experiments using the AGCM (AGCM run). In each series of the experiments, the global mountain height was systematically changed; a parameter α that is multiplied to the surface elevation was varied from 0 to 1.4 with the interval of 0.2. The ICM run with α = 1 is regarded as the control run. Comparison of the January climatology between the ICM and the AGCMshows that the upper-level wind and low-level baroclinicity over the western Pacific Ocean increase linearly with α in both models while the sensitivity to change in α is higher in the ICM. This appears to be explained by the following positive feedback; in the coupling system, orographically forced stationary waves at larger α spin up ocean gyres, transporting the warm water near the western boundary that results in an enhancement of local evaporation and precipitation. The latent heat release works to intensify the stationary waves which are responsible to maintain the low-level baroclinicity. The low-level wind, SST and the storm track activity do not change linearly with increasing α. In particular, the storm track strengthens with mountain uplift until α = 0.6 while weakening afterwards. This weakened storm track decelerates the low-level wind downstream, which reduces the wind-driven circulation and the increase in SST around the gyre boundary at the heigher mountain.
Idealized numerical simulations of an unsaturated, conditionally unstable flow over a two-dimensional mountain ridge were used to study the effects of the moist Froude number (Fw) and the orographic aspect ratio of mountain height to width (h/a) on the propagation, cloud type, and pattern and amount of rainfall of orographically induced precipitation systems. For low Fw, the flow belonged to an upstream propagating flow regime (i.e., Regime I) and was insensitive to h/a. For large Fw, both Fw and h/a dictated the precipitation pattern and the nature of the convection. When Fw was fixed, the flow shifted toward a downstream propagating regime as h/a increased. This dependence on h/a at higher wind speeds appears to be linked to the advection time and cloud growth time. A slightly larger Fw was required for the regime transition to occur when the fixed aspect ratio was small. We also found that although the local maximum rainfall was not directly controlled by Fw, the total domain rainfall was sensitive to Fw, in particular for Regimes I and II (i.e., flow with a long-lasting orographic convective system over the mountain peak, upslope or lee slope) when the mountain half-width was fixed. On the other hand, for the unstable flow studied here, the total domain accumulated rainfall was not sensitive to h/a when Fw was fixed. It is suggested that flash floods may occur when quasi-stationary convective precipitation systems stay over the mountain area for flow Regime II or when abundant moisture is supplied for a significant period of time due to a low-level jet for flow Regimes III and IV, which are defined as flow with an orographic convective precipitation system over the mountain and a downstream propagating convective system and flow with an orographic stratiform precipitation system over the mountain and possibly a downstream propagating cloud system, respectively. In addition, local orographic rainfall from stratiform precipitation systems (i.e., Regime IV) can be as heavy as that from convective or mixed type precipitation systems belonging to Regime III.
As a continuation of a previous study on Niigata-Fukushima heavy rainfall (HR), results from numerical experiments of Fukui HR, which also occurred in July 2004, are presented. As in the previous study, Fukui HR is discussed with an emphasis on the importance of latent instability. It is shown that although Fukui HR is not simulated with global analysis (GANAL) data of JMA, slight modification of the moisture field gives rise to rainfall patterns which are somewhat similar to those indicated by Radar-AMeDAS data. The structure of the convective system is discussed, and it is suggested that the so-called back-building mechanism played an important role in Fukui HR, which occurred under the upper-level westerly jet and the strong low-level vertical shear.
An intercomparison study of ground-based Mie-scattering lidar systems was held at Gosan, Korea during ABC-EAREX 2005 in order to ensure the correctness and accuracy of the lidar retrieval algorithms as well as the aerosol extinction profiles. We compared the vertical profiles of the range-corrected raw lidar signals as well as the aerosol extinction coefficients determined from different types of Mie-scattering lidar systems. Three different aerosol loading cases were selected; in these cases, the mean deviations of the range-corrected raw lidar signals were estimated to be 4.5%, 8.0%, and 7.3%. An intercomparison of the instruments by an identical data-handling procedure (Fernald approach) demonstrated that the aerosol extinction profiles for the three selected cases were in agreement within a bias difference of 0.0051 km-1. The mean deviation of the signals for the aerosol layer between 0.6 and 3.0 km was 3.0%. The systems were intercompared for two lidar inversion algorithms by using simultaneous lidar measurement data sets as input; these input data sets showed agreement with their aerosol extinction profiles within a bias difference of 0.0086 km-1 and a mean deviation of 11.6%.
In this note, we consider the opposition of two mechanisms for blocking-onset in a somewhat different way from Dong and Colucci (2007). We assume a region of localized confluence and diffluence, which is caused by a disturbance. The disturbance consists of a localized potential vorticity pair in a basic westerly flow. We introduce a unit vector P, which lies in the direction from negative to positive potential vorticity center of the pair. The westerly flow is accelerated (or decelerated) by two forcing terms, advection term and deformation term. The former is the advection of disturbance potential vorticity gradient vector by the westerly basic flow. The latter is the deformation of planetary vorticity gradient vector by the disturbance deformation matrix. In which direction these two terms accelerate or decelerate the westerly flow, it is dependent on the orientation of P. In any case, the advection and deformation terms are anti-parallel to each other and tend to cancel each other. The advection term dominates over the deformation term for a disturbance of relatively short scale. On the other hand, the deformation term dominates over the advection term for a disturbance of relatively long scale.
This note describes a method to improve the current Tropical Rainfall Measuring Mission (TRMM) passive microwave rain rate estimates by adopting a modified probability density function (PDF) of rain rates in Bayesian retrieval algorithms. Unlike the original database that was generated from cloud model simulations mostly for more "eventful" cloud/rain cases, the modified PDF includes the probability of zero rain rates derived from actual observations. Based on analysis of one year TRMM data, this simple remedy lowers the passive microwave rain rate estimates by 3% and reduces the discrepancy between radiometer and radar estimates by 6%.