Understanding and Predicting Nocturnal Convection Initiation using an Ensemble-based Multi-scale Data Assimilation System

Nocturnal convection has long been known to occur most frequently after sunset over the Great Plains (Wallace 1975). However, initiation of these storms is notoriously difficult to forecast during the evening hours (e.g., Weisman et al. 2008) in numerical models due to the storms often involving interactions across many scales. While convection initiation (CI) in well-mixed daytime boundary layers often occurs off of boundaries of surface convergence and is well-understood, nocturnal CI is relatively unexplored (Wilson and Roberts 2006). Assimilation of synoptic and other in-situ mesoscale observations can help to improve analyses of any features which may aid in nocturnal CI such as low-level jets, density currents, or gravity waves. Additionally, many previous studies (e.g., Wheatley et al. 2014) have shown the benefits of radar data assimilation (DA) through improved forecasts in the structure and evolution of mesoscale convective systems (MCSs). In this study, a multi-scale, GSI-based EnKF forecast system (Johnson et al. 2015) is applied to cases from the Plains Elevated at Night Experiment (PECAN) to address the following: (1) the impact of radar and in-situ DA for nocturnal CI forecasts; (2) the optimal model and DA configuration for nocturnal CI forecasting; (3) the understanding of mechanisms that lead to nocturnal CI.

A multi-scale WRF-ARW (v3.5.1) simulation is performed with hopes to forecast the CI in northwestern Kansas. On a 12 km outer domain, synoptic and mesoscale observations are assimilated every three hours to better analyze the pre-storm environment. The inner convection permitting domain assimilates radar observations (reflectivity and radial velocity) every five minutes and conventional observations every 30 minutes. Results from a June, 2013 case study are positive and have shown the impact of DA on nocturnal CI forecasts. While a control forecast with no DA had little to no CI in the area of interest, ensemble forecasts with DA showed strong agreement in CI. The addition of conventional observation assimilation on the inner domain improved the analysis of the surface outflow and associated cold pool of a weak MCS in southwestern Kansas of which the convection formed off of.

Since CI is known to be sensitive to both planetary boundary layer (Kain et al. 2008) and microphysics parameterization schemes (Burghardt et al. 2014), correct tuning of DA parameters such as localization for different scales and observation types are important in order to improve forecasts. Additionally, since nocturnal CI is likely influenced by gravity waves and other small-scale processes, such as bores, outflows, and solitary waves, smaller grid spacing may be needed to accurately represent the physics processes that produce CI. Initial results to test the impact of these different model configurations and parameterizations on CI forecasting will be presented for cases from three of the CI intensive observation periods (IOPs) during PECAN. Additionally, results will be presented that show the impact of assimilating new observations recorded during the project, such as mobile soundings and both fixed and mobile profiling units.