The Near-Real-Time SCALE-LETKF System: A Case of the September 2015 Kanto-Tohoku Heavy Rainfall

Lien, Guo-Yuan; Miyoshi, Takemasa; Nishizawa, Seiya; Yoshida, Ryuji; Yashiro, Hisashi; Adachi, Sachiho A.; Yamaura, Tsuyoshi; Tomita, Hirofumi

doi:10.2151/sola.2017-001

Cited by 42 publications

(54 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our simulation applies the Scalable Computing for Advanced Library and Environment Localized Ensemble Transform Kalman Filter (SCALE‐LETKF) DA system (Lien et al ., ). The SCALE‐LETKF system combines the open source Scalable Computing for Advanced Library and Environment Regional Model (SCALE‐RM: version 5.1.2; see Nishizawa et al ., ; Sato et al ., ; Nishizawa and Kitamura, ) and a Localized Ensemble Transform Kalman Filter (LETKF: Hunt et al ., ).…”

Section: Experimental Setup and Methodsmentioning

confidence: 97%

A convective‐scale 1,000‐member ensemble simulation and potential applications

Necker

Geiss²,

Weißmann

et al. 2020

Quart J Royal Meteoro Soc

Self Cite

View full text Add to dashboard Cite

This study presents the first convective-scale 1,000-member ensemble simulation over central Europe, which provides a unique data set for various applications. A comparison with the operational regional 40-member ensemble of Deutscher Wetterdienst shows that the 1,000-member simulation exhibits realistic spread properties overall. Based on this, we discuss two potential applications. First, we quantify the sampling error of spatial covariances of smaller subsets compared with the 1,000-member simulation. Knowledge about sampling errors and their dependence on ensemble size is crucial for ensemble and hybrid data assimilation and for developing better approaches for localization in this context. Secondly, we present an approach for estimating the relative potential impact of different observable quantities using ensemble sensitivity analysis. This will provide the basis for consecutive studies developing future observation and data assimilation strategies. Sensitivity studies on the ensemble size indicate that about 200 ensemble members are required to estimate the potential impact of observable quantities with respect to precipitation forecasts. K E Y W O R D Sconvective-scale, covariance, data assimilation, ensemble sensitivity analysis, localization, observing system, sampling error

show abstract

Section: Experimental Setup and Methodsmentioning

confidence: 97%

A convective‐scale 1,000‐member ensemble simulation and potential applications

Necker

Geiss²,

Weißmann

et al. 2020

Quart J Royal Meteoro Soc

Self Cite

View full text Add to dashboard Cite

show abstract

“…(2016) takes a similar approach where, for each local analysis, the horizontal localization length‐scale is adjusted so that the number of locally assimilated observations becomes constant (roughly double the ensemble size) at every analyzed grid point. Guo‐Yuan Lien et al . (2017; private communication) applied a similar method to assimilation of phased array weather radar (PAWR) data by their LETKF (Lien et al ., 2017) and confirmed that limiting the number of locally assimilated observations to a few times the ensemble size leads to significantly better forecasts than assimilating all data or applying a traditional thinning method before assimilating them.…”

Section: Discussion: Interpretation Of Results From the Literature Inmentioning

confidence: 99%

Why does EnKF suffer from analysis overconfidence? An insight into exploiting the ever‐increasing volume of observations

Hotta

Ota

2021

Quart J Royal Meteoro Soc

View full text Add to dashboard Cite

Ensemble Kalman filters (EnKF) are empirically known to suffer from insufficient posterior spread and this issue is aggravated when assimilating a large volume of observations. This problem, commonly called analysis underdispersion or analysis overconfidence, has been well recognized, but why it occurs seems to be rather poorly understood. Inspired by the theory of the degrees of freedom for signal, this article investigates this problem by analyzing the trace of the matrix HK, where H and K represent, respectively, the observation operator and the gain matrix. A simple mathematical argument shows that tr HK for EnKF is bounded from above by the ensemble size, which entails that assimilating many more observations than the ensemble size leads automatically to tr HK underestimation, as long as the observations are of accuracy comparable to the background. Since tr HK can be interpreted as the squared spread of the posterior ensemble measured in the normalized observation space, underestimated tr HK implies overconfidence in the analysis spread, which, in a cycled context, requires covariance inflation to be applied. The theory is then extended to cases where covariance localization schemes (either B‐localization or R‐localization) are applied to show how they alleviate the analysis underdispersion. These findings from the mathematical argument are demonstrated with a simple one‐dimensional covariance model. Finally, the findings described above are used to form speculative arguments about how to interpret several puzzling features of the local ensemble transform Kalman filter (LETKF) previously reported in the literature, such as why using fewer observations can lead to better performance, when optimal localization scales tend to occur, and why covariance inflation methods based on the relaxation to prior information approach are particularly successful when observations are distributed inhomogeneously.

show abstract

“…Regarding the analysis and predictability of stationary linear convective systems and local torrential rain, the possibility of improving the numerical forecast by assimilating Himawari-8 rapid-scan atmospheric motion vectors (Kunii et al 2016) and using a sophisticated atmosphere data assimilation system (Lien et al 2017;Honda et al 2018) has been reported. Lien et al (2017) pointed out that a high-resolution atmospheric model with a horizontal resolution of a few kilometers is needed to improve the forecast. However, the effect Fig.…”

Section: Introductionmentioning

confidence: 99%

Air-Sea Coupled Data Assimilation Experiment for Typhoons Kilo, Etau and the September 2015 Kanto-Tohoku Heavy Rainfall with the Advanced Microwave Scanning Radiometer 2 Sea Surface Temperature

Wada¹,

Tsuguti

Okamoto³

et al. 2019

Journal of the Meteorological Society of Japan

View full text Add to dashboard Cite

The September 2015 Kanto-Tohoku heavy rainfall event occurred in a stationary linear convective system between Typhoons Kilo and Etau. We investigated the influence of sea surface temperature (SST) on the local heavy rainfall event using a regional air-sea strongly coupled data assimilation system based on the local ensemble transform Kalman filter (LETKF) and a nonhydrostatic atmosphere model (NHM) coupled with an oceansurface wave model and a multilayer ocean model with an Advanced Microwave Scanning Radiometer 2 (AMSR2) level 2 (L2) SST product. From the validation of SST analyzed by the coupled data assimilation system with the Japanese geostationary multi-functional transport satellite 2 hourly SST product and in-situ observations at a moored buoy, we demonstrated that the coupled system with the AMSR2 L2 SST led to an improvement in the SST analysis. Based on the verification using radiosonde observations and radar-rain gauge rainfall analysis, the analysis of the lower-atmospheric components was improved by the air-sea coupled NHM-LETKF. The local torrential rainfall event that occurred around 37°N in the Tochigi prefecture was embedded in a stationary linear convective system. The location of the linear convective system corresponded to the synoptic-scale convergence area between the cyclonic circulation associated with Etau and easterly lower-tropospheric winds. Strong southerly winds associated with Etau caused a periodic enhancement of local convection along the convergence area on the upwind side of the linear convective system and resulted in a wave-like train of the total water content around an altitude of 4-8 km on the leeward side. The improvement of SST analysis could not only change the transition of Etau to the extratropical cyclone but also the lower-tropospheric wind field and thereby

show abstract

The Near-Real-Time SCALE-LETKF System: A Case of the September 2015 Kanto-Tohoku Heavy Rainfall

Cited by 42 publications

References 40 publications

A convective‐scale 1,000‐member ensemble simulation and potential applications

A convective‐scale 1,000‐member ensemble simulation and potential applications

Why does EnKF suffer from analysis overconfidence? An insight into exploiting the ever‐increasing volume of observations

Air-Sea Coupled Data Assimilation Experiment for Typhoons Kilo, Etau and the September 2015 Kanto-Tohoku Heavy Rainfall with the Advanced Microwave Scanning Radiometer 2 Sea Surface Temperature

Contact Info

Product

Resources

About