Hiromu Seko

A Cloud-Resolving 4DVAR Assimilation Experiment for a Local Heavy Rainfall Event in the Tokyo Metropolitan Area

AbstractA cloud-resolving nonhydrostatic four-dimensional variational data assimilation system (NHM-4DVAR) was modified to directly assimilate radar reflectivity and applied to a data assimilation experiment using actual observations of a heavy rainfall event. Modifications included development of an adjoint model of the warm rain process, extension of control variables, and development of an observation operator for radar reflectivity.The responses of the modified NHM-4DVAR were confirmed by single-observation assimilation experiments for an isolated deep convection, using pseudo-observations of rainwater at the initial and end times of the data assimilation window. The results showed that the intensity of convection could be adjusted by assimilating appropriate observations of rainwater near the convection and that undesirable convection could be suppressed by assimilating small or no reflectivity.An assimilation experiment using actual observations of a local heavy rainfall in the Tokyo, Japan, metropo...

Spatial-temporal fractions verification for high-resolution ensemble forecasts

Experiments with two ensemble systems of resolutions 10 km (MF10km) and 2 km (MF2km) were designed to examine the value of cloud-resolving ensemble forecast in predicting precipitation on small spatio-temporal scales. Since the verification was performed on short-term precipitation at high resolution, uncertainties from small-scale processes caused the traditional verification methods to be inconsistent with the subjective evaluation. An extended verification method based on the Fractions Skill Score (FSS) was introduced to account for these uncertainties. The main idea is to extend the concept of spatial neighbourhood in FSS to the time and ensemble dimension. The extension was carried out by recognising that even if ensemble forecast is used, small-scale variability still exists in forecasts and influences verification results. In addition to FSS, the neighbourhood concept was also incorporated into reliability diagrams and relative operating characteristics to verify the reliability and resolution of two systems. The extension of FSS in time dimension demonstrates the important role of temporal scales in short-term precipitation verification at small spatial scales. The extension of FSS in ensemble space is called the ensemble FSS, which is a good representative of FSS for ensemble forecast in comparison with the FSS of ensemble mean. The verification results show that MF2km outperforms MF10km in heavy rain forecasts. In contrast, MF10km was slightly better than MF2km in predicting light rains, suggesting that the horizontal resolution of 2 km is not necessarily enough to completely resolve convective cells.

Three-dimensional distribution of water vapor estimated from tropospheric delay of GPS data in a mesoscale precipitation system of the Baiu front

Three-dimensional distributions of water vapor in a mesoscale precipitation system, which developed on 7 July 1996 in the Baiu front, were estimated directly from the GPS data of ‘GPS Earth Observation Network’ (GEONET) of the Geographical Survey Institute. In estimating the three-dimensional distribution of water vapor from the GPS data, we used the tropospheric delays along each path from GPS satellites to GPS receivers on the ground. The result showed that the moist air extended up to the height of 6 km in the precipitation region in the early stage of the precipitation system and that a dry air intruded into the precipitation system from the northwest in the middle-level (from 3 km to 5 km altitude) in the later stage. This dry air intrusion in the middle-level was supported by numerical simulations.

Effect of lateral boundary perturbations on the breeding method and the local ensemble transform Kalman filter for mesoscale ensemble prediction

Abstract The effect of lateral boundary perturbations (LBPs) on the mesoscale breeding (MBD) method and the local ensemble transform Kalman filter (LETKF) as the initial perturbations generators for mesoscale ensemble prediction systems (EPSs) was examined. A LBPs method using the Japan Meteorological Agencys (JMAs) operational one-week global ensemble prediction was developed and applied to the mesoscale EPS of the Meteorological Research Institute for the World Weather Research Programme, Beijing 2008 Olympics Research and Development Project. The amplitude of the LBPs was adjusted based on the ensemble spread statistics considering the difference of the forecast times of the JMAs one-week EPS and the associated breeding/ensemble Kalman filter (EnKF) cycles. LBPs in the ensemble forecast increase the ensemble spread and improve the accuracy of the ensemble mean forecast. In the MBD method, if LBPs were introduced in its breeding cycles, the growth rate of the generated bred vectors is increased, and the ensemble spread and the root mean square errors (RMSEs) of the ensemble mean are further improved in the ensemble forecast. With LBPs in the breeding cycles, positional correspondences to the meteorological disturbances and the orthogonality of the bred vectors are improved. Brier Skill Scores (BSSs) also showed a remarkable effect of LBPs in the breeding cycles. LBPs showed a similar effect with the LETKF. If LBPs were introduced in the EnKF data assimilation cycles, the ensemble spread, ensemble mean accuracy, and BSSs for precipitation were improved, although the relative advantage of LETKF as the initial perturbations generator against MDB was not necessarily clear. LBPs in the EnKF cycles contribute not to the orthogonalisation but to prevent the underestimation of the forecast error near the lateral boundary. The accuracy of the LETKF analyses was compared with that of the mesoscale 4D-VAR analyses. With LBPs in the LETKF cycles, the RMSEs of the forecasts from the LETKF analysis were improved and some of them became comparable to those of the mesoscale 4D-VAR analyses based on the JMAs operational data assimilation system. These results show the importance of LBPs in the MBD method and LETKF. LBPs are critical not only to ameliorate the underestimation of the ensemble spread in the ensemble forecast but also to produce better initial perturbations and to improve the LETKF analysis.

Comparison of initial perturbation methods for the mesoscale ensemble prediction system of the Meteorological Research Institute for the WWRP Beijing 2008 Olympics Research and Development Project (B08RDP)

Different initial perturbation methods for the mesoscale ensemble prediction were compared by the Meteorological Research Institute (MRI) as a part of the intercomparison of mesoscale ensemble prediction systems (EPSs) of the WorldWeather Research Programme (WWRP) Beijing 2008 Olympics Research and Development Project (B08RDP). Five initial perturbation methods for mesoscale ensemble prediction were developed for B08RDP and compared at MRI: (1) a downscaling method of the Japan Meteorological Agency (JMA)’s operational one-week EPS (WEP), (2) a targeted global model singular vector (GSV) method, (3) a mesoscale model singular vector (MSV) method based on the adjoint model of the JMA non-hydrostatic model (NHM), (4) a mesoscale breeding growing mode (MBD) method based on the NHM forecast and (5) a local ensemble transform (LET) method based on the local ensemble transform Kalman filter (LETKF) using NHM. These perturbation methods were applied to the preliminary experiments of the B08RDP Tier-1 mesoscale ensemble prediction with a horizontal resolution of 15 km. To make the comparison easier, the same horizontal resolution (40 km) was employed for the three mesoscale model-based initial perturbation methods (MSV, MBD and LET). The GSV method completely outperformed the WEP method, confirming the advantage of targeting in mesoscale EPS. The GSV method generally performed well with regard to root mean square errors of the ensemble mean, large growth rates of ensemble spreads throughout the 36-h forecast period, and high detection rates and high Brier skill scores (BSSs) for weak rains. On the other hand, the mesoscale model-based initial perturbation methods showed good detection rates and BSSs for intense rains. The MSV method showed a rapid growth in the ensemble spread of precipitation up to a forecast time of 6 h, which suggests suitability of the mesoscale SV for short-range EPSs, but the initial large growth of the perturbation did not last long. The performance of the MBD method was good for ensemble prediction of intense rain with a relatively small computing cost. The LET method showed similar characteristics to the MBD method, but the spread and growth rate were slightly smaller and the relative operating characteristic area skill score and BSS did not surpass those of MBD. These characteristic features of the five methods were confirmed by checking the evolution of the total energy norms and their growth rates. Characteristics of the initial perturbations obtained by four methods (GSV, MSV, MBD and LET) were examined for the case of a synoptic low-pressure system passing over eastern China. With GSV and MSV, the regions of large spread were near the low-pressure system, but with MSV, the distribution was more concentrated on the mesoscale disturbance. On the other hand, large-spread areas were observed southwest of the disturbance in MBD and LET. The horizontal pattern of LET perturbation was similar to that of MBD, but the amplitude of the LET perturbation reflected the observation density.

An Overview of the Beijing 2008 Olympics Research and Development Project (B08RDP)

The Beijing 2008 Olympics Research and Development Project (B08RDP), initiated in 2004 under the World Meteorological Organization (WMO) World Weather Research Programme (WWRP), undertook the research and development of mesoscale ensemble prediction systems (MEPSs) and their application to weather forecast support during the Beijing Olympic Games. Six MEPSs from six countries, representing the state-of-the-art regional EPSs with near-real-time capabilities and emphasizing on the 6–36-h forecast lead times, participated in the project. The background, objectives, and implementation of B08RDP, as well as the six MEPSs, are reviewed. The accomplishments are summarized, which include 1) providing value-added service to the Olympic Games, 2) advancing MEPS-related research, 3) accelerating the transition from research to operations, and 4) training forecasters in utilizing forecast uncertainty products. The B08RDP has fulfilled its research (MEPS development) and demonstration (value-added service) purposes. T...

“Big Data Assimilation” Revolutionizing Severe Weather Prediction

AbstractSudden local severe weather is a threat, and we explore what the highest-end supercomputing and sensing technologies can do to address this challenge. Here we show that using the Japanese flagship “K” supercomputer, we can synergistically integrate “big simulations” of 100 parallel simulations of a convective weather system at 100-m grid spacing and “big data” from the next-generation phased array weather radar that produces a high-resolution 3-dimensional rain distribution every 30 s—two orders of magnitude more data than the currently used parabolic-antenna radar. This “big data assimilation” system refreshes 30-min forecasts every 30 s, 120 times more rapidly than the typical hourly updated systems operated at the world’s weather prediction centers. A real high-impact weather case study shows encouraging results of the 30-s-update big data assimilation system.

TOKYO METROPOLITAN AREA CONVECTION STUDY FOR EXTREME WEATHER RESILIENT CITIES

The present paper describes background, mission, research topics, and preliminary results of the research project “Tokyo Metropolitan Area Convection Study for Extreme Weather Resilient Cities (TOMACS)”. TOMACS is one of the research projects of “Social System Reformation Program for Adaption to Climate Change” which has been started since July 2010 under the “Special Coordination Funds for Promoting Science and Technology” of the Ministry of Education, Culture, Sports, Science and Technology (MEXT). TOMACS aims to understand the processes and mechanisms of extreme weather, using dense meteorological observation networks designed in the Tokyo metropolitan district, to develop a monitoring and predicting system of extreme phenomena (MPSEP), and to implement social experiments on extreme weather resilient cities in collaboration with related government institutions, local governments, private companies, and residents. More than 25 organizations and over 100 people participate in the present research projects. One of unique features of TOMACS is utilization of dense meteorological instruments in the Tokyo Metropolitan area which is one of the most urbanized areas in the world. The field campaign in the Tokyo metropolitan area, using research instruments and operational meteorological networks is planned by MRI and thirteen groups in the summers of 2011-2013 to target the tropospheric environment, boundary layer, initiation of convections and lifecycle of thunderstorms. Observation on environmental conditions of convections are carried out using radio sonde, wind profiler, GPS network, unmanned air viecle, and network of automated weather stations. Generation and development of convective precipitations are investigated by observations using Doppler lidar, rapid scan geostationary satellite, Kuband polarimetric radar, X-band polarimetric radar network (X-NET) and C-band research polarimetric radar and C-band operational Doppler radars. Several thunderstorms were captured by the dense meteorological network during 2011 campaign observations. The present paper shows preliminary results of the analysis. Social experiments on extreme weather resilient city using radar networks are also presented.

Cloud-Resolving 4D-Var Assimilation of Doppler Wind Lidar Data on a Meso-Gamma-Scale Convective System

AbstractThe authors evaluated the effects of assimilating three-dimensional Doppler wind lidar (DWL) data on the forecast of the heavy rainfall event of 5 July 2010 in Japan, produced by an isolated mesoscale convective system (MCS) at a meso-gamma scale in a system consisting of only warm rain clouds. Several impact experiments using the nonhydrostatic four-dimensional variational data assimilation system (NHM-4DVAR) and the Japan Meteorological Agency nonhydrostatic model with a 2-km horizontal grid spacing were conducted in which 1) no observations were assimilated (NODA), 2) radar reflectivity and radial velocity determined by Doppler radar and precipitable water vapor determined by GPS satellite observations were assimilated (CTL), and 3) radial velocity determined by DWL were added to the CTL experiment (LDR) and five data denial and two observational error sensitivity experiments. Although both NODA and CTL simulated an MCS, only LDR captured the intensity, location, and horizontal scale of the obs...

Semi-diurnal and diurnal variation of errors in GPS precipitable water vapor at Tsukuba, Japan caused by site displacement due to ocean tidal loading

Simultaneous GPS and water vapor radiometer (WVR) observations were carried out at Tsukuba, Japan from May 1 to June 30, 1998. The precise point positioning method of the GIPSY/OASIS-II software package (GIPSY) was used to retrieve precipitable water vapor (GPS_PWV) from GPS data, which was then compared with precipitable water vapor observed by WVR (WVR_PWV). They agreed quite well with the root mean square difference of less than 1.5 mm. However, periodic variations were found in the difference between GPS_PWV and WVR_PWV (dPWV). It was also found that semi-diurnal or diurnal components of these variations had a positive correlation with site displacement due to Ocean Tidal Loading (OTL). Two months of dPWV data were decomposed by the period of a component of OTL, and then composite time series data with a period equal to that of the component were made. This process was performed for K1, O1, M2, and S1 components of OTL. In each component, a periodic variation in dPWV appeared which was similar to those of the simulated GPS_PWV errors from OTL effects calculated with ‘GOTIC’ (Sato and Hanada, 1984), a program for the computation of OTL effect. Inclusion of OTL effects into GIPSY analysis reduced dPWV. Inthe M2 component, the amplitude of the dPWV was reducedby about 80%. This suggests that the OTL components calculated by the GOTIC succeeded in simulating the actual site displacement by OTL effects in Japan. On the other hand, in K1 components, the amplitude of dPWV without OTL in GIPSY is 1.5 times larger than the simulated GPS PWV error, with considerable error remaining even in the case of GIPSY analysis with OTL. The error may be due to multi-path effect, temperature dependency on conversion from Zenith Wet Delay to PWV, or instrument dependency of WVR on temperature. Analysis utilizing much longer data periods than the present two months is required to overcome these difficulties.