15.5 The Structure, Evolution, and Dynamics of a Nocturnal Convective System Simulated Using the WRF-ARW Model

Thursday, 6 August 2015: 11:30 AM
Republic Ballroom AB (Sheraton Boston )
Ben Blake, University of Oklahoma, Norman, OK; and D. Parsons and K. R. Haghi

Previous studies have documented a nocturnal maximum in thunderstorm frequency across the central United States. Forecast skill for nocturnal convection remains relatively low, and one reason is the greater occurrence of elevated convection at night. This study makes use of the WRF-ARW model Version 3.6.1 to reproduce a nocturnal, elevated, mesoscale convective system (MCS) event that occurred over Oklahoma on June 3-4, 2013. The purpose of this study is to evaluate the performance of the model in the nocturnal environment and to advance the knowledge of the dynamics, structure, and evolution of nocturnal convection.

The NOAA Rapid Refresh (RAP) model data available from the Nomads NCDC server was used for atmospheric data and the North American Land Data Assimilation System (NLDAS-2) Noah Land Surface Model was used for soil data. Both of these data sources allow for hourly analysis updates. The model is run for 22 hours from 18 UTC on June 3rd to 16 UTC on June 4th. Three domains are used with two-way nesting. 9 km, 3 km, and 1 km are the respective grid spacing for the outer, middle, and inner domains. 100 vertical levels are used, and there is a higher concentration of eta levels below 1500 meters in order to more accurately represent the evolution of boundary layer processes.

Convection initiates around 23 UTC on June 3rd and continues until approximately 16 UTC on June 4th. The nocturnal MCS was located south of a quasi-stationary frontal boundary, which is contrary to a 1993 study that noted the favored position for organized convective development is to the north of a quasi-stationary frontal boundary (Trier & Parsons 1993). A low-level jet (LLJ) is evident from wind profiler data and it is known to provide an elevated source of moist, unstable air by which storms can be sustained. The model produces a strong horizontal gradient in CAPE and CIN that is co-located with the narrow corridor of high mixing ratios associated with the LLJ. Once the convection becomes elevated, the source of inflowing parcels with CAPE is almost entirely above 1 km, which agrees with previous studies.

The LLJ can also be thought of as a wave ducting mechanism that traps wave energy below a certain height. Wave ducts prevent the upward leakage of wave energy and allow for horizontal wave propagation as opposed to solely vertical propagation. The model also produces wave-like features, including gravity currents, bores, and gravity waves. These waves were also evident in the radar observations and in the time series data from the Oklahoma Mesonet. As convection transitions from surface-based to elevated, studies have observed that the mechanism responsible for lifting inflow parcels evolved from a cold pool to a bore.

To investigate the effects of these wave features on the nocturnal environment, comparisons to hydraulic theory were made to diagnose the flow regime that is occurring. The depth of the gravity current is calculated using a hydrostatic approximation based on the density, pressure, and virtual potential temperature differences between the ambient air and the storm outflow. The ratio of the depth of the gravity current to the strength of the inversion of the ambient air is known as the non-dimensional height. The flow relative Froude number, defined as the ratio of the adjusted inflow speed to the speed of the long period gravity wave, is also calculated. The non-dimensional height and Froude number are then used to determine the flow regime. According to hydraulic theory (Koch et.al. 1991), bores can form in the partially blocked or complete blocking flow regimes. For the four analyzed regions corresponding to different wave propagation angles, a partially blocked flow regime is occurring. This type of flow regime is also evident in the vertical cross sections of potential temperature containing feature relative winds. Data analysis is still in progress at this time.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner