Lightning Data Assimilation Scheme in a 4DVAR System and Its Impact on Very Short-Term Convective Forecasting

Abstract

A proof-of-concept method for the assimilation of total lightning observations in the 4DVAR framework is proposed and implemented into the Variational Doppler Radar Analysis System (VDRAS). Its performance is evaluated for the very-short-term precipitation forecasts of a localized convective event over northeastern China. The lightning DA scheme assimilated pseudo observations for vertical velocity fields derived from observed total lightning rates and statistically computed vertical velocity profile from VDRAS analysis data. To reduce representative errors of the derived vertical velocity, a distance-weighted horizontal interpolation is applied to the input data prior to the DA. The case study reveals that although 0–2 hour precipitation nowcasts are improved by assimilating lightning data alone compared to CTRL (no radar or lightning) and RAD (radar only), better results are obtained when the lightning data are assimilated with radar data simultaneously. The assimilation of both data sources results in improved dynamical consistency with enhanced updraft and latent heat as well as improved moisture distributions. Additional experiments are conducted to evaluate the sensitivity of the combined DA scheme to varied vertical velocity profiles, radii of horizontal interpolation, binning time intervals, and relationships used to estimate the maximum vertical velocity from lightning flash rates. It is shown that the scheme is robust to these variations with both radar and lightning assimilated data.

A Study on Local‐Scale Thermal and Dynamical Mechanisms in the Initiation of a Squall Line Under Weak Forcing

“…VDRAS provides convection‐permitting dynamical and thermodynamical fields (e.g., 3‐D wind, temperature, shear, divergence, and CAPE (convective available potential energy)/CIN (convective inhibition)) with frequent updating of ∼10 min, which are not attainable directly from observation instruments (e.g., radars and the auto weather stations [AWSs]). VDRAS can assimilate data from Doppler radar network, surface mesonet (X. C. Chen et al., 2016), and lightning network (Xiao et al., 2021). Since 2001, high‐resolution analyses of convective‐scale meteorological fields obtained by the 4DVAR system have been used for various purposes, such as case studies of convective storms (e.g., Friedrich et al., 2016; Gochis et al., 2015; Sun, 2005; Sun & Zhang, 2008; Tai et al., 2011; Xiao et al., 2019, 2017; F. Zhang et al., 2020; L. Zhang et al., 2021), initialization of high‐resolution numerical models for model‐based nowcasting (e.g., Sun, 2005; Sun & Zhang, 2008; Tai et al., 2011), operational nowcasting in several forecasting offices (Crook & Sun, 2002, 2004; Sun et al., 2010) and quality control of radar observations (He et al., 2019; Lim & Sun, 2010).…”

“…Its operational potential for nowcasting thunderstorms has recently been demonstrated (Chang et al, 2016;. The aforementioned literature (Crook & Sun, 2002;Sun et al, 2010;Tai et al, 2011;Xiao et al, 2021) includes additional details about VDRAS.…”

An observation operator for geostationary lightning imager data assimilation in the French storm-scale numerical weather prediction system AROME

“…Nudging methods have been employed by Marchand and Fuelberg (2014) and Fierro et al (2012) and methods using an ensemble Kalman filter have been developed by Mansell (2014) and Allen et al (2016). Variational approaches were investigated in more recent studies by Fierro et al (2016Fierro et al ( , 2019; Zhang et al (2017); Hu et al (2020); Xiao et al (2021). A general improvement was observed in accumulated precipitation for short term forecasts (≤3h) when lightning data is assimilated.…”

Assimilation of the Maximum Vertical Velocity Converted From Total Lightning Data Through the EnSRF Method

“…Xiao et al. (2021) assimilated pseudo vertical velocity fields derived from observation total rates and statistically computed vertical velocity profile. Wang, Chen, et al.…”

“…Chen, Sun, et al (2020) proposed a method to update model kinematic states by assimilating satellite-observed total lightning data to improve the convective forecasting. Xiao et al (2021) assimilated pseudo vertical velocity fields derived from observation total rates and statistically computed vertical velocity profile. Wang, Chen, et al (2020) assimilated vertical velocity by a nudging approach based on the Global/Regional Assimilation and Prediction System in mesoscale model.…”