All‐Sky Assimilation of the MWHS‐2 Observations and Evaluation the Impacts on the Analyses and Forecasts of Binary Typhoons

Abstract

Satellite data assimilation is transitioning from clear‐sky to all‐sky approach at some operational forecasting centers; the all‐sky approach directly assimilates observations under clear, cloudy, and precipitating conditions and shows a positive impact on medium‐range forecasts. Although there has been progress in the all‐sky technique for the forecasting of high‐impact weather processes, such as typhoon, these studies mainly focused on a single typhoon case, and the influence of all‐sky assimilation on the forecasts of binary typhoons, which have tracks and intensities that are more complex, is less investigated. In this study, a binary typhoons case (i.e., Typhoons Haitang and Nesat), which occurred in the western Pacific in July 2017, is selected, and the Weather Research and Forecasting data assimilation system is applied to examine the impact of all‐sky assimilation of the Microwave Humidity Sounder‐2 data from FengYun‐3C on the analysis and forecasting of the binary typhoons for the first time. The results show that with more cloud‐ and precipitation‐affected data being assimilated, the vertical structures of temperature and humidity around the core areas of the two typhoons and the forecasts of wind, temperature, and humidity are improved significantly in the all‐sky assimilation. These improvements result in the simulated tracks being more similar to the observations, and an error reduction in intensity is achieved in the all‐sky assimilation. The heavy rainfall (>250 mm) caused by Nesat over Taiwan is also improved. These encouraging results suggest that all‐sky assimilation is able to improve the forecasts of binary typhoons.

Recent advancements for tropical cyclone data assimilation

“…The typical approach for all-sky radiance assimilation at most operational centers is to use observation error inflation. 137 A number of studies for TC all-sky MW radiance assimilation employing similar error model formulations [126][127][128][129] suggested positive impacts on TC predictions. A similar observation error model to account for the cloud amount for the direct assimilation of all-sky infrared radiances was developed as well.…”

“…Regional assimilation of all‐sky radiances generally presents a different challenge (e.g., domain size, model top, bias correction, etc.) compared to the global assimilation; however, several demonstration studies showed the feasibility of the all‐sky microwave assimilation in the regional models, and overall resulted in positive impacts on TC track, intensity, storm structure, and precipitation forecasts 126–129 …”

Assimilation of MWHS-2/FY-3C 183 GHz Channels Using a Dynamic Emissivity Retrieval and Its Impacts on Precipitation Forecasts: A Southwest Vortex Case

“…An apparently positive impact has been reflected in temperature and relative humidity in the AL and DY experiments (Figures 5(a ). e reasons for these negative impacts are not clearly, but some studies [35,36] have also addressed this situation. ey indicated that other advanced data assimilation approaches, such as hybrid (ensemble/3D-VAR) or 4D-VAR method, will deal with this problem as these advanced data assimilations can provide flow-dependent ensemble background covariance error which is not achieved in the 3D-VAR data assimilation.…”

The Potential Value of Combining Assimilation of Water Vapor Radiances and Cloud Liquid/Ice Water Path From ABI on Analyses and Forecasts for Hurricane Irma (2017)

“…A better description of the tropical cyclone thermodynamic fields under the all‐sky condition allows for reasonable deep convective triggering and development, thus leading to an improved representation of rapid intensification through the assimilation cycle (Kazumori & Kadowaki, 2017; Tong et al., 2020; Zhang et al., 2021). In addition to these impacts, all‐sky MW radiance assimilation also improves the tropospheric wind fields, resulting in better midlatitude atmospheric circulation and improves TC track prediction (Xian et al., 2019; Yang et al., 2016; Zhang et al., 2022).…”

“…Beyond the typical difficulties shared by the direct assimilation of radiances, such as large uncertainties and biases in the Radiative Transfer Model for cloud absorption and scattering (Li et al., 2016; Okamoto, 2017), the inconsistency between the simulated hydrometeors and observations (Barlakas et al., 2021), and high nonlinearity for the cloudy radiance (Errico et al., 2007), direct assimilation of cloud‐affected IR, near‐IR, and VIS radiances suffers from the significant challenge of the high sensitivity of Jacobian to clouds (Li et al., 2016; Vidot et al., 2015). In addition, most of the channels currently used for IR imager all‐sky assimilation are water vapor channels, making the thermodynamic information updated but not the cloud due to the lack of direct observations (e.g., Candy & Migliorini, 2021; Geer et al., 2019; Lee et al., 2020; Shahabadi & Buehner, 2021; Xian et al., 2019; Zhang et al., 2016).…”