Convection and precipitation under various stability and shear conditions: numerical experiments for mesoscale convective systems

The structure and intensity of mesoscale convective systems (MCSs) are strongly regulated by their environmental conditions such as horizontal wind, temperature, and moisture. These environmental conditions are determined by synoptic-scale and/or larger-scale meteorological settings that develop in various climates ranging from the Tropics, sub-Tropics, mid-latitude regions, and arid/semi-arid regions. Because of the diversity of climates over the globe, the meteorological settings that host mesoscale phenomena have a wide variety represented by the horizontal and vertical distribution of wind, temperature, and moisture. Although a large number of studies have investigated the various aspects of MCSs in a certain specific region of the world, much less number of studies compared the characteristics of MCSs that develop in different climate regions.

A significant difference between tropical and midlatitude convection was found for vertical velocities. Zipser and LeMone (1980) showed that the size and intensity of updrafts and downdrafts are significantly weaker in the tropical system than in the midlatitude one by comparing their observations with the data from the Thunderstorm Project. Lucas et al. (1994) suggested that the weaker vertical velocities in the Tropics are due to smaller buoyancy that is attributed to more effective water loading and entrainment, and they identified the midlatitude soundings as large buoyancy and the tropical sounding as small buoyancy. The importance of buoyancy profiles has been pointed out by a number of studies.

The purpose of this study is to investigate the sensitivity of the structure and intensity of MCSs to the vertical profiles of temperature and moisture that represent tropical, oceanic and midlatitude, continental environments. A special focus is given to the vertical distribution of convective available potential energy (CAPE) that is determined by the temperature and moisture profiles. For the present purpose, we use the Weather Research and Forecasting (WRF) model/the Advanced Research WRF (ARW) to perform a set of numerical experiments for MCSs under various temperature, moisture, and shear conditions in an idealized model setup. The base-state atmosphere is horizontally homogeneous, and uni-directional shear is assumed.

The grid resolutions of the experiments are 500-m spacing in the horizontal directions and 116 vertical levels in the computational domain of 300 km (east-west) by 60 km (north-south) by 23 km (vertical). The physics processes included in the present idealized experiments are only cloud-microphysics (Goddard warm-rain and ice-phase parameterization) and turbulent mixing (Deardorff TKE scheme). The lateral boundary conditions are periodic at the north and south boundaries and open at the east and west boundaries. The lower boundary is free slip, while the upper boundary is rigid with a Rayleigh-type damping layer.

The thermodynamic profiles are given by a tropical sounding for a squall line over the tropical western Pacific (Trier et al. 1996) as a tropical, oceanic environment (TROPICS), and by an analytical sounding for mid-latitude supercells (Weisman and Klemp 1982) as a mid-latitude, continental environment (MIDLAT). The MIDLAT conditions are further divided as a moist type (the original Weisman-Klemp humidity profile, referred to as MIDLATM) and a dry type (reduced humidity from the Weisman-Klemp profile, referred to as MIDLATD). These thermodynamic conditions are characterized by the same CAPE value for the surface air parcel but by different CAPE distribution for elevated parcels above the surface level. On the other hand, five types of wind shear are examined: 5 m/s difference in the 0-2.5-km layer or in the 2.5-5-km layer; 15 m/s difference in the 0-2.5-km layer or in the 2.5-5-km layer; and 15 m/s difference in the 0-5-km layer. For each thermodynamic profile (TROPICS, MIDLATM, MIDLATD), these five shear profiles are defined. Furthermore, the initial disturbances are given by both warm and cold linear-type thermal aligned in the north-south direction. The set of experiments initialized by the warm line thermal is referred to as W-type, while that initialized by the linearly oriented cold pool is as C-type. Taking into account the different initialization, we perform five sets of experiments (TROPICS-W, TROPICS-C, MIDLATD-W, MIDLATD-C, MIDLATM-W) with five different shear profiles. Thus, the total number of experiments is 25. The integration time period for each experiment is 6 hours.

We compared the structure and intensity of the simulated MCSs in different conditions in terms of means and variability of precipitation intensity, vertical velocities, updraft fractional areas, and the relevant properties. The temporal and spatial mean precipitation intensity (which is the same as the accumulated precipitation) generally increases with the increase in temperature lapse rate. CAPE for the surface air parcel distinguishes the intensity of MCSs if the lapse rates among the different conditions are similar. A critical impact on the MCS intensity is related to the vertical distribution of CAPE for elevated parcels above the surface (mainly within the PBL depth). The CAPE distribution is closely tied to the lapse rate in the troposphere as well as the moisture content in the boundary layer. The role of the shear profile in determining the MCS intensity becomes more significant in drier conditions by counteracting cold pools through the Rotunno-Klemp-Weisman mechanism. The sensitivity to the initial disturbances is less significant in moister environments. An interesting result is found for the maximum instantaneous precipitation intensity: the maxima increases with the decrease in lapse rate. There may be an optimal lapse rate for developing precipitation maxima.

We conclude that temperature lapse rate in the convectively unstable troposphere is useful in comparing the characteristics of MCSs that occur in various stability and shear conditions. An implication of this study is that precipitation characteristics in a future climate under global warming mayl be explained by the static stability. It is suggested that stability diagnosis for the climate simulation data should be useful for examining the variation of precipitation characteristics in the simulated climates.