Design and Implementation of a GSI-Based Convection-Allowing Ensemble-Based Data Assimilation and Forecast System for the PECAN Field Experiment. Part II: Overview and Evaluation of a Real-Time System

Abstract

Multiscale ensemble-based data assimilation and forecasts were performed in real time during the Plains Elevated Convection At Night (PECAN) field experiment. A 20-member ensemble of forecasts at 4-km grid spacing was initialized daily at both 1300 and 1900 UTC, together with a deterministic forecast at 1-km grid spacing initialized at 1300 UTC. The configuration of the GSI-based data assimilation and forecast system was guided by results presented in Part I of this two-part study. The present paper describes the implementation of the real-time system and the extensive forecast products that were generated to support the unique interests of PECAN researchers. Subjective and objective verification of the real-time forecasts from 1 June through 15 July 2015 is conducted, with an emphasis on nocturnal mesoscale convective systems (MCSs), nocturnal convective initiation (CI), nocturnal low-level jets (LLJs), and bores on the nocturnal stable layer. Verification of nocturnal precipitation during overnight hours, a proxy for MCSs, shows both greater skill and spread for the 1300 UTC forecasts than the 1900 UTC forecasts. Verification against observed soundings reveals that the forecast LLJs systematically peak, veer, and dissipate several hours before the observations. Comparisons with bores that passed over an Atmospheric Emitted Radiance Interferometer reveal an ability to predict borelike features that is greatly improved at 1-km, compared with 4-km, grid spacing. Objective verification of forecast CI timing reveals strong sensitivity to the PBL scheme but an overall unbiased ensemble.

Nocturnal Convection Initiation during PECAN 2015

“…The OU MAP 4-km forecast and verification domains are unlikely to fully resolve NCI events related to bores and pristine NCI events. Previous results showed that 1-km grid spacing and finer are necessary to fully capture the lifting properties, such as the slope of the ascent or the erosion of inhibition, resulting from bores and atmospheric waves (Johnson et al 2017;Johnson and Wang 2019). We suggest investigating the benefits of higher-resolution simulations, particularly for the improvement of bore/ density current and pristine NCI forecasts.…”

“…The ensemble forecasts produced in real time by the University of Oklahoma (OU) Multiscale data Assimilation and Predictability (MAP) laboratory (Johnson et al 2016) were verified for the 49 NCI events identified in Table 1. The 20-member, convection-permitting ensemble forecasts at a 4-km horizontal grid spacing were initialized daily at 1300 UTC using the Advanced Research version of WRF (ARW), version 3.6.1.…”

“…The 20-member, convection-permitting ensemble forecasts at a 4-km horizontal grid spacing were initialized daily at 1300 UTC using the Advanced Research version of WRF (ARW), version 3.6.1. The details of the configuration of the real-time system, data assimilation and its evaluation can be found in Johnson et al (2017) and Johnson and Wang (2017). T he OU M A P ver ification method for the NCI event at 0300 UTC 26 June 2015 is illustrated in Fig.…”

The 2015 Plains Elevated Convection at Night Field Project

“…A webbased PHI tool, developed originally for the NOAA Hazardous Weather Testbed Spring Experiments and used for MCS and CI forecasting, was extended to predict bores as well in PECAN (Karstens et al 2015;Haghi et al 2015). The data driving this tool in PECAN were obtained from a 1-km WRF model developed and run in real time by the University of Oklahoma MAP group (Johnson et al 2015). The PHI tool user selects two points: one in the ambient air and one in the denser, cooler air of the convective outflow.…”

Evaluation of Forecasts of a Convectively Generated Bore Using an Intensively Observed Case Study from PECAN

“…The experiments in this study use the multiscale GSIbased EnKF and ensemble forecast system, where both the in situ and convective-scale radar data are assimilated (Johnson et al 2015;Johnson and Wang 2017;Johnson et al 2017). The forecast model is the Advanced Research version of the Weather Research and Forecasting (WRF-ARW; Skamarock et al 2005) Model version 3.6.1.…”