11:15 AM - 11:30 AM
[AAS02-03] Objective analysis of typhoon intensity by assimilation of satellite-derived cloud-tracking winds in the eye: Case of Typhoon Trami (2018)
Keywords:Data assimilation, Numerical experiment, Geostationary meteorological satellite
Tropical-cyclone (TC) intensity estimation at TC warning centers is mainly performed by a subjective analysis of the Dvorak technique with cloud patterns from satellite imagery. After new-generation geostationary meteorological satellites such as Himawari-8/9 and GOES-R were launched, high-frequency observations (with 1- or 2-minute intervals) over limited areas including typhoons were available. Recently, some studies proposed new techniques to track the horizontal motions of cloud tops at the mesoscale using continuous images from high-frequency observation. The cloud tracking quantitatively estimates spatiotemporal distributions of the horizontal wind at the cloud-top height. Particularly, after the eye formed in the mature typhoon, low-level rapid rotation visualized by shallow cumuli within the eye can be estimated. On the other hand, cloud tracking cannot directly derive the maximum wind speed in the eyewall cloud. When observations in meteorology are limited, data assimilation has been used to obtain the best estimation of the atmospheric state.
In this study, by data assimilation of satellite-derived low-level winds in the eye, we examine the feasibility of a quantitative estimation of the maximum wind speed in the eyewall of the typhoon (i.e., an assimilation-based objective analysis of typhoon intensity). We used a nonhydrostatic atmosphere model and data-assimilation system (SCALE-LETKF) with 30 ensemble members to perform observing system simulation experiments (OSSEs) for Typhoon Trami in 2018 after the eye was formed in the mature stage. Virtual observation data for the cloud-tracking winds are produced by a nature run based on the SCALE model. We focus on the low-level cloud-tracking winds in the typhoon eye (i.e., inside the radius of maximum wind speed, RMW). Two OSSEs are designed by assimilating two types of cloud-tracking winds, separately. One OSSE (called TH20) examines the assimilation of axisymmetric tangential winds in a specified layer at every 3 km radius from the typhoon center (Tsukada and Horinouchi 2020). The other OSSE (called H23) examines the assimilation of zonal and meridional winds with a horizontal resolution of 3 km in the eye (Horinouchi et al. 2023). To estimate the impacts of the data assimilation, we design one experiment without any assimilations (i.e., free-run experiment).
The typhoon centers among the ensemble members in the free run gradually spread from the storm center of the nature run with time. Unlike the free run, the typhoon centers in TH20 concentrated on the storm center of the nature run as the assimilation progressed. The axisymmetric tangential winds in the ensemble-averaged typhoon reproduced the rotating speed of the storm in the nature run with small errors in the eye, by the assimilation. On the other hand, the axisymmetric tangential winds had large errors (> 10 m s-1) at around RMW which is no observation area. The ensemble spread for the typhoon centers in H23 was smaller than that in TH20. In the ensemble-averaged storm in H23, the axisymmetric tangential winds were small errors (~ 3-4 m s-1) within a radius of 200 km (including the RMW). The results suggest that the assimilation of the low-level horizontal winds in the eye can effectively modify the storm structure in the background atmosphere even if the cloud-tracking winds do not have observations at RMW. We conclude that the objective analysis of the typhoon intensity is feasible by assimilation of the satellite-based cloud-tracking winds.
References:
Horinouchi et al. (2023): https://doi.org/10.1175/MWR-D-22-0179.1
Tsukada and Horinouchi (2020): https://doi.org/10.1029/2020GL087637
In this study, by data assimilation of satellite-derived low-level winds in the eye, we examine the feasibility of a quantitative estimation of the maximum wind speed in the eyewall of the typhoon (i.e., an assimilation-based objective analysis of typhoon intensity). We used a nonhydrostatic atmosphere model and data-assimilation system (SCALE-LETKF) with 30 ensemble members to perform observing system simulation experiments (OSSEs) for Typhoon Trami in 2018 after the eye was formed in the mature stage. Virtual observation data for the cloud-tracking winds are produced by a nature run based on the SCALE model. We focus on the low-level cloud-tracking winds in the typhoon eye (i.e., inside the radius of maximum wind speed, RMW). Two OSSEs are designed by assimilating two types of cloud-tracking winds, separately. One OSSE (called TH20) examines the assimilation of axisymmetric tangential winds in a specified layer at every 3 km radius from the typhoon center (Tsukada and Horinouchi 2020). The other OSSE (called H23) examines the assimilation of zonal and meridional winds with a horizontal resolution of 3 km in the eye (Horinouchi et al. 2023). To estimate the impacts of the data assimilation, we design one experiment without any assimilations (i.e., free-run experiment).
The typhoon centers among the ensemble members in the free run gradually spread from the storm center of the nature run with time. Unlike the free run, the typhoon centers in TH20 concentrated on the storm center of the nature run as the assimilation progressed. The axisymmetric tangential winds in the ensemble-averaged typhoon reproduced the rotating speed of the storm in the nature run with small errors in the eye, by the assimilation. On the other hand, the axisymmetric tangential winds had large errors (> 10 m s-1) at around RMW which is no observation area. The ensemble spread for the typhoon centers in H23 was smaller than that in TH20. In the ensemble-averaged storm in H23, the axisymmetric tangential winds were small errors (~ 3-4 m s-1) within a radius of 200 km (including the RMW). The results suggest that the assimilation of the low-level horizontal winds in the eye can effectively modify the storm structure in the background atmosphere even if the cloud-tracking winds do not have observations at RMW. We conclude that the objective analysis of the typhoon intensity is feasible by assimilation of the satellite-based cloud-tracking winds.
References:
Horinouchi et al. (2023): https://doi.org/10.1175/MWR-D-22-0179.1
Tsukada and Horinouchi (2020): https://doi.org/10.1029/2020GL087637