Numerical simulation of the 16-19 October 1994 southeast Texas heavy rain event: precipitation results and diagnosis of the lifting mechanism

Abstract

During the period 16-19 October 1994, a mesoscale convective system (MCS) developed within a weakly forced large-scale environment over relatively flat terrain. This resulted in extreme rainfall totals and subsequent widespread flash flooding in southeast Texas. Liberty, Texas recorded 775 mm of rain and an area greater than 78,000 km² received in excess of 254 mm. Events like this, with weak synoptic scale forcing and where initial convective development occurs between sparsely distributed observing stations, are a particular challenge for numerical models. A meteorological overview of this event is presented along with a comparison to the Maddox et al. (1979) heavy rain conceptual model. This event was fairly consistent with the conceptual model, including adequate moisture and location in the western side of the large-scale ridge. The NMC surface analyses did not indicate a front after 0600 UTC 16 October. This important feature of the frontal-type model was not analyzed during the actual case. A numerical model simulation is conducted to evaluate the ability of the PSU/NCAR Mesoscale Model (MM5) to reproduce this complex event and to diagnose the mechanism for initiation. Using only operational surface and upper-air observations in a data nudging technique, very high resolution simulations (i.e., 36, 12, and 4 km meshes) reproduced well the timing, evolution, intensity, geographical location, areal coverage, and maximum amount of rainfall. Next, the model is used to determine the lifting mechanism that was not apparent during the actual event. Several diagnostic techniques were used to assess vertical motion. These techniques included positive vorticity advection, positive isothermal vorticity advection, and quasi-geostrophic fields. Additionally, frontogenetic forcing and stability of the troposphere were computed. The diagnostic simulation suggests that the initiation of the rain event was a result of low-level frontogenetic forcing that lifted the air into a very unstable environment. The upper-level forcing mechanism was important only for destabilizing and moistening the middle troposphere. Northeastern Texas was capped by an inversion that inhibited convection while a surface moisture ridge extended northwestward from the Gulf of Mexico. These two features positioned the area for convective initiation.