A Statistical Analysis of Characteristics of Mesoscale Convective System Mountain Initiation Location Clusters in the Arkansas-Red River Basin

Abstract

Mesoscale Convective Systems (MCSs) are the focus of this analysis since it is the convective weather category which is smallest in number but produces the highest amount of precipitation. Being able to forecast these MCSs will make it easier to anticipate flooding events that can occur with these systems. The multi-sensor precipitation data, a combination of satellite, radar, and rain gage data, was used in Tucker and Li (2009). The MCSs initiating west of 104˚ W in the warm season (April - September) in the years 1996 to 2006 in the Arkansas-Red River Basin were used in this analysis. A cluster analysis was run on this data to group the MCSs to preferred locations. It has been shown that convective weather has preferred locations within the Rocky Mountain chain (Tucker and Crook 2005). The clusters containing 20 or more members are used in this analysis. Data for the surface and upper air variables was gathered from Iowa State's online database (Iowa State 2011) and data for the North American Regional Reanalysis (NARR) data was gathered NOMADS (National Climatic Data Center 2011). Once observations for all the variables were gathered for each MCS cluster, Multiple Linear Regressions (MLRs) and Principal Component Analyses (PCAs) were determined for the six hours prior through the three hours after initiation. The analysis of these model runs could help determine the characteristics needed for MCS mountain initiation within the cluster domain. The results from these analyses can be used to anticipate MCS mountain initiation if the conditions are known.