Analysis of convectively generated cold pools and fronts from Mesonet data

 

Authors

Andrew Thomas Lesage — University of Utah
Steven K. Krueger — University of Utah

Category

Dynamics/Vertical Motion

Description

Example of frontal analysis based on Mesonet station time series. Red dots indicate triangles with strong convergence. Dark blue dots indicate triangles with strong divergence. Green squares are cold pool scores >3 and <5, while cyan squares are cold pool scores of > 5. (Radar image from the UCAR image archive.)
The Oklahoma Mesonet includes over 100 stations throughout Oklahoma and has been collecting observations for over 15 years at 5-minute intervals. Surface observations of temperature, pressure, and wind speed were used in this study to identify frontal passages and cold pools for the JJA 1997 period and the 15 April–31 May 2011 period, which coincided with the Midlatitude Continental Convective Clouds Experiment (MC3E). The Mesonet stations, minus those located in the panhandle, were used to construct a grid using the Delaunay Triangulation method. A unitless variable, termed a “cold pool score”, based on pressure rises and temperature falls at a station over 30-minute intervals, was used to determine the location of fronts associated with convectively generated cold pools as they progressed across the Mesonet. The frontal analysis was able to identify the fronts accurately in a variety of cases as well as to provide statistics for the frequency and strength of fronts. Western Oklahoma had the highest frequency of fronts during JJA 1997, and Eastern Oklahoma during the MC3E period. The temperature and pressure change values during the frontal passages had near-zero correlation in the JJA 1997 period and a correlation of -0.21 for the MC3E period. This suggests that there is high variability in the vertical structure of temperature perturbations in the cold pools. Strong convergence typically occurred ahead of a front, with strong divergence behind. Cold pools were identified by requiring a frontal passage across a Mesonet triangle to have occurred as well as strong divergence. This set of requirements appears to be capable of capturing convectively generated cold pools reasonably well. It also filters out the cases of frontal passages that do not have cold pools, such as the dry frontal passages. This method of identifying fronts and cold pools should prove to be useful for evaluating cumulus parameterizations. Additional research was performed to calculate an estimate of evaporation of precipitation assuming it was responsible for the entire pressure change during the frontal passage. These results may help constrain microphysical parameterizations of raindrop break-up.