Investigating 3D and 4D Variational Rapid-Update-Cycling Assimilation of Weather Radar Reflectivity for a Flash Flood Event in Central Italy

Abstract

Abstract. The precipitation forecast over the Mediterranean basin is still a challenge because of the complex orographic region which amplifies the need for local observation to correctly initialize the forecast. In this context the data assimilation techniques play a key role in improving the initial conditions and consequently the timing and position of precipitation pattern. For the first time, the ability of a cycling 4D-Var to reproduce a severe weather event in central Italy, as well as to provide a comparison with the largely used cycling 3D-Var, is evaluated in this study. The radar reflectivity measured by the Italian ground radar network is assimilated in the WRF model to simulate an event occurred on May 3, 2018 in central Italy. In order to evaluate the impact of data assimilation, several simulations are objectively compared by means of a Fraction Skill Score (FSS), which is calculated for several threshold values, and a Receiver Operating Characteristic (ROC) curve. The results suggest that both assimilation methods in cycling mode improve the 1, 3 and 6-hourly quantitative precipitation estimation. More specifically, the cycling 4D-Var with a warm start initialization shows the highest FSS values in the first hours of simulation both with light and heavy precipitation. Finally, the ROC curve confirms the benefit of 4D-Var: the area under the curve is 0.91 compared to the 0.88 of control experiment without data assimilation.

Assimilating X- and S-Band Radar Data for a Heavy Precipitation Event in Italy

“…It is also worth mentioning [42] who used the default matrix provided by the WRFDA system, which is produced with global data and its use for regional cases is sometimes discouraged [65]. Additionally, more recent works (see [96]) used a time period shorter than one month to compute the covariance matrices of the background errors. However, we acknowledge that further investigations are needed to assess the impact of a long-term application of the NMC method to build a high-level, quality B matrix and obtain better QPF data.…”

Comment on nhess-2020-406

“…The 3D-Var method has the advantage of being computationally cheap even if it misses the time dependency, hence all the observations that are acquired during the assimilation window are considered at its central time. The WRF 4D-Var (Huang et al, 2009) take the time variable into account using a numerical weather forecast as dynamical constraint. More specifically, the method computes the model trajectory that reduce the misfit with the observations distributed in the assimilation window.…”

“…Nowadays, the large availability of high frequency (both in space and time) meteorological data, remote sensing observations and in situ measurements, has encouraged many operational centres to use data assimilation techniques for improving the accuracy of initial state. More specifically, the assimilation of ground radar reflectivity and radial velocity with three-dimensional variational (3D-Var) method proves good results in terms of quantitative precipitation forecast (QPF) for several case study in United States and Korea (Xiao and Sun, 2007;Lee et al, 2010;Ha et al, 2011). The assimilation of radar data with Weather Research Forecast (WRF) 3D-Var confirms positive results also in Europe by using WRF model, for the simulation of flash flood events in central (Maiello et al, 2014; and northern Italy (Lagasio et al, 2019), as well as by using Advanced Regional Prediction System (ARPS) and Application of Research to Operations at Mesoscale (AROME) models for two heavy rainfall cases in Croatia and France (Stanesic and Brewster, 2016;Caumont et al 2009), respectively.…”

Impact of Radar Reflectivity and Lightning Data Assimilation on the Rainfall Forecast and Predictability of a Summer Convective Thunderstorm in Southern Italy

“…The Mediterranean area is often struck by severe weather events [4] which are determined by specific combinations of factors related to three main ingredients: the warm sea, the orography of the basin and the synoptic-scale meteorology. Deep convective events over the Mediterranean develop mainly in fall and winter [4][5][6][7][8][9][10][11][12], but they can also occur in other seasons. The storms are often associated with mid-latitude cyclonic systems or wave troughs, with or without secondary cyclogenesis, or with deep moist convection development produced by mesoscale convective systems (MCSs) [13][14][15][16]; when an MCS is located over the same area for several hours, large amounts of precipitation can accumulate in less than a day [10,11,17,18].…”

“…For example, the Constant Altitude Plane Position Indicator (CAPPI), used for the assimilation in this study, has a spatial resolution of 1 km and a temporal resolution of five minutes; this gives the possibility to assimilate radar data frequently and at high horizontal resolution, which is an important aspect of radar data assimilation (RDA) for convective events. Focusing on the complex Italian territory, RDA has shown an important impact on quantitative precipitation forecasts [9,28,52,[60][61][62] for several convective events. A significant advantage of radar data compared to lightning is that radar observations are also available for light to moderate precipitation, while lightning data are scarce in these conditions; furthermore, in the first stages of developing convection, flashes are absent or few, delaying the issuance of severe weather warnings.…”