The goal of the present study is to develop a method to assimilate Microwave Imager (MWI) brightness temperatures (TBs) into Cloud-Resolving Models (CRMs). To address the non-linear relationship of TBs to the state variables of CRM and the flow-dependency of the CRM forecast error covariance, we adopted an ensemble-based variational data assimilation method. However, there often exist large-scale displacement errors of rainy areas between the observation and CRM forecasts. In such cases, ensemble-based data assimilation can give erroneous analysis, particularly for observed rain areas without forecasted rain. In order to solve this problem, we propose ensemble-based assimilation that uses ensemble forecast error covariance with displacement error correction. Based on this idea, we developed a data assimilation method that incorporates the MWI TBs into the CRM developed by the Japan Meteorological Agency (JMANHM). This method consists of a displacement error correction scheme and an ensemble-based variational assimilation scheme. In the displacement error correction scheme, we obtained the optimum displacement that maximized the conditional probability of TB observation given the displaced CRM variables. In the assimilation scheme, we derived a cost function in the displaced ensemble forecast error subspace. Then, we obtained analyses of CRM variables by non-linear minimization of the cost function. In order to see the impact of the above MWI TB assimilation method on CRM analyses and forecasts, we performed assimilation experiments to incorporate TMI (TRMM Microwave Imager) low-frequency TBs (10, 19, and 21 GHz with vertical polarization) into the CRM for a typhoon case around Okinawa (9th June 2004). The results of the experiments show that the assimilation of TMI TBs alleviated the large-scale displacement errors and improved CRM forecasts. The displacement error correction also avoided misinterpretation of MWI TB increments due to precipitation displacements as those from other variables in the assimilation scheme.
An aquaplanet atmospheric general circulation model with prescribed sea surface temperatures (SST) is used to provide insight on tropical intraseasonal variability. Previous work showed that a zonally asymmetric SST distribution that resembles observations but with the meridional SST gradient reduced to one-quarter of its observed value poleward of 10° could produce a robust Madden-Julian oscillation (MJO) with 50-day periodicity and amplitude greater than observations in this model. Wind-evaporation feedback was shown to be necessary to destabilize the model MJO. This present work extends these aquaplanet sensitivity experiments to examine the impact of other SST basic states on intraseasonal variability, and also determines whether cloud-radiative and moisture-radiative feedbacks are important for destabilizing the model MJO. A zonally symmetric model with reduced meridional SST gradient produces dominant intraseasonal power at too short of timescale (~30 days) and reduced amplitude relative to the zonally asymmetric simulation, suggesting that basic state zonal asymmetries and surface mean westerlies are important for producing a realistic MJO. The intraseasonal mode in the zonally symmetric model resembles the WISHE mode of linear theory, with wind speed and latent heat flux anomaly maxima occurring in surface easterly anomalies to the east of maximum convection, unlike the zonally asymmetric model in which latent heat fluxes maximize near and to the west of convection like in observations. An experiment with in which longwave radiative heating is prescribed to be near climatology suggests that both radiative and wind-evaporation feedbacks may help to destabilize the model MJO.
The quality of tropical cyclone (TC) best track data has increasingly received more attention in climate research in recent years because of the rising interest in the possible effects of global warming on TC activity. In this paper, three best track datasets for the western North Pacific TCs were compared in a seasonal context. Firstly, the statistical characteristics were examined based on their distributions. Then, the potential impact of particular features within the datasets was evaluated by using contrastive cluster analyses. The mean amplitude of the annual cycle in the Japan Meteorological Agency (JMA) data is largest among the three datasets. The smallest interannual variability helped the Shanghai Typhoon Institute of China Meteorological Administration (CMA-STI) dataset to resist the negative effect of outliers. An inhomogeneity around 2003–2004 was identified in the Joint Typhoon Warning Center (JTWC) data by the outliers and was compared with the JMA and CMA-STI data. By using contrastive cluster analyses, the differences among the datasets and some effects related to particular issues, including the representation of the annual cycle, the outliers, and wind conversion were evaluated. The JTWC dataset is affected by outliers and wind conversion among the different average times, whereas the representation of the annual cycle was important for the CMA-STI data. After excluding all the outliers, the results indicated more consistent classifications of the TC annual cycle between the JMA and JTWC data.
The relationships between near-core surface wind asymmetries in typhoons and both the environmental vertical wind shear and typhoon movement are examined using the historical QuikSCAT (NASA Quick Scatterometer) wind data, putting stress on the azimuthal location of the tangential wind maximum. The results clearly show that the highest winds occur preferentially to the left of shear as well as to the right of storm motion when their effects on the wind fields are separately evaluated using a composite approach. A further composite analysis reveals that the magnitude of left to right asymmetry with respect to storm motion or shear depends largely on the directional difference between shear and storm motion with the largest asymmetry being generated when their directions are opposite to each other. In cases where shear and storm motion vectors point toward a similar direction under relatively strong shear conditions, the left to right asymmetry is very small and a significant percentage of the cases exhibit a left-of-motion maximum. These results are highly consistent with previous ones which were obtained from the JMA (Japan Meteorological Agency) operational mesoscale analysis wind data. In the present study, the effect of inner-core rainfall asymmetry on the azimuthal location of the tangential wind maximum is also examined. The results show that the relation between shear and surface wind asymmetry can be described in terms of rainfall asymmetry in place of shear, as expected from the increasingly well-documented strong connection between shear and rainfall asymmetry, suggesting that the asymmetric features of over-ocean wind fields in typhoons could be estimated to some degree using information on typhoon motion and rainfall asymmetry.
The individual contributions of changes in sea surface temperature (SST), vertical wind shear and tropical cyclone (TC) tracks to the interannual TC intensity change in the western North Pacific (WNP) basin are examined based on the selected 7 warm years and 7 cold years during the period 1970–2007. The selected warm and cold years are defined by the Nino-3.4 SST anomalies index, and correspond to El Niño and La Niña events, respectively. The intensity model used in this study can simulate the spatial distribution and differences of TC intensity when the model is integrated along the observed TC tracks in the warm and cold years. It is found that the change of TC tracks plays a dominant role in the observed TC intensity difference between warm and cold years. During the warm years, TC formation is enhanced in the southeast quadrant, and more TCs take a northwestward track during the warm years than during the cold years because of the interannual change in the large-scale steering flows. As a result, TCs have a longer time for intensification and develop into intense TCs during the warm years when compared to the cold years.
A tracer advection scheme for a three-dimensional icosahedral grid model is developed. Then, its tracer transport performance is evaluated using realistic flow fields. The consistency with continuity (CWC) scheme is newly introduced to a three-dimensionally monotonic scheme using existing techniques of an upwind-biased scheme and the intermediate density method. The CWC guarantees mass conservation and Lagrangian conservation, both of which are important properties for tracer transport. In addition, the scheme is positive definite. This transport scheme is implemented with the nonhydrostatic icosahedral atmospheric model and transport simulations are performed. A comparison with radon observations reveals that the proposed method reproduces high-concentration events more realistically than the original method does. Furthermore, results show that the proposed method has less numerical diffusion and is superior to the original method in terms of a sharper gradient.
In many river basins in the Japan Sea side of Japan, a drastic change in the river discharge occurring each April was noted between the 1980s and the 1990s. We conducted a climate simulation with 5 km grid spacing using a regional climate model to investigate the climatic change in snow water equivalent (SWE), which was closely related to the river discharge in spring. The decrease in the SWE from the 1980s to the 1990s is simulated for the entire Japan Sea side area from winter to early spring corresponding to the temperature increase of around 1 K. In the Agano River basin, the river discharge was more significantly influenced by the difference in the April SWE between the 1980s and the 1990s than by other factors such as snowfall, rainfall, and evaporation. The large reduction in the SWE in April was primarily caused by the large amount of snowmelt in winter in the 1990s. The increase in warm air advection frequency from the south may have caused the large amount of snowmelt. However, simulation data analysis indicated that the snowmelt was more greatly influenced by the large-scale temperature increase than by the increase in the frequency of warm air advection from the south. The surface heat and radiation budget under the snowmelt conditions showed that the accumulated shortwave radiation energy in the 1990s was greater than that in the 1980s, which contributed to the large amount of snowmelt because the snowmelt periods in the 1990s were much longer than those in the 1980s. Therefore, the temperature increase caused the long snowmelt period, and the increased snowmelt was promoted by the larger amount of shortwave radiation energy in the 1990s.
In mid-January 2009, sudden changes in circulation occurred in the tropical troposphere and stratosphere. Convective activity situated over the equatorial Maritime Continent showed an abrupt weakening, whereas that over the South American to African sectors became stronger. Changes also occurred in the latitudinal structure; convective activity in the Northern Hemisphere became weaker, whereas that in the Southern Hemisphere became stronger. The change in convective activity took place in association with a change in tropical circulation, from east–west to north–south type (i.e., from Walker- to Hadley-type circulation). Almost simultaneously with these events in the troposphere, a change in meridional circulation occurred in the stratosphere during a record-breaking stratospheric sudden warming event in January 2009. Stratospheric tropical temperature showed a decrease in response to a strengthening of the hemispherical meridional circulation. In the present study, we show how the stratospheric and tropospheric circulation changes are dynamically coupled.