Land surface sensitivity of mesoscale convective systems
Date
2016
Authors
Tournay, Robert C., author
Schumacher, Russ, advisor
Vonder Haar, Thomas, advisor
van den Heever, Susan, committee member
Nelson, Peter, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Mesoscale convective systems (MCSs) are important contributors to the hydrologic cycle in many regions of the world as well as major sources of severe weather. MCSs continue to challenge forecasters and researchers alike, arising from difficulties in understanding system initiation, propagation, and demise. One distinct type of MCS is that formed from individual convective cells initiated primarily by daytime heating over high terrain. This work is aimed at improving our understanding of the land surface sensitivity of this class of MCS in the contiguous United States. First, a climatology of mesoscale convective systems originating in the Rocky Mountains and adjacent high plains from Wyoming southward to New Mexico is developed through a combination of objective and subjective methods. This class of MCS is most important, in terms of total warm season precipitation, in the 500 to 1300m elevations of the Great Plains (GP) to the east in eastern Colorado to central Nebraska and northwest Kansas. Examining MCSs by longevity, short lasting MCSs (<12 hrs), medium (12-15 hrs) and long lasting MCSs (>15 hrs) reveals that longer lasting systems tend to form further south and have a longer track with a more southerly track. The environment into which the MCS is moving showed differences across commonly used variables in convection forecasting, with some variables showing more favorable conditions throughout (convective inhibition, 0-6 km shear and 250 hPa wind speed) ahead of longer lasting MCSs. Other variables, such as convective available potential energy, showed improving conditions through time for longer lasting MCSs. Some variables showed no difference across longevity of MCS (precipitable water and large-scale vertical motion). From subsets of this MCS climatology, three regions of origin were chosen based on the presence of ridgelines extending eastward from the Rocky Mountains known to be foci for convection initiation and subsequent MCS formation: Southern Wyoming (Cheyenne Ridge), Colorado (Palmer divide) and northern New Mexico (Raton Mesa). Composite initial and boundary conditions were developed from reanalysis data, from which control runs of regional MCSs were made as well a series of idealized experiments with imposed large scale soil moisture (SM) anomalies to study to impact to each regional MCS on SM variations in initiation region as well down stream in the GP. Another idealized experiment was made to study the impact of varying the planetary boundary layer (PBL) parameterization in the context of the idealized SM variations. While the distribution of SM has a major impact on CAPE and the location and magnitude of CI, also important is the differences in shear driven by the differences in large scale SM, playing a major, and varying depending on where the regional MCSs interact with the shear anomalies. Utilizing a different PBL parameterization impacts the timing and amount of initial CI, impacting the total precipitation produced by the MCSs, but not nearly the magnitude of alteration to the MCS as varying the SM distribution. A climatology of CI in the Rocky Mountains and adjacent high plains is made using a high resolution observational dataset. From this climatology, the sensitivity of CI to land surface variables, including SM and vegetation is studied. It was found that the timing of CI had a stronger relationship with SM, with earlier CI over wetter than average soils, with the greatest difference in May in the north of the domain, nearly all statistical significant values across regions from north to south in June and July with little difference in August in the northern regions. Outside of May, which showed a strong relationship of earlier CI over less vegetated regions, the relationship was similar, but weaker than, that between SM and CI timing. Examining the CAPE, CIN and PW at CI and null points reveal that the values are generally more conducive to CI over wet soils and anomalously vegetated areas at both CI and null points, with stronger difference in the high plains in the east of regions. Examining the covariance of SM and vegetation at CI points revealed that July and August showed expected covariance relationships with concurrently measured convective variables (i.e., high SM/vegetation associated with high CAPE and vice versa for low SM/vegetation) while May and June higher CAPE and CIN over low vegetation anomalies. A climatology of elevated mixed layers in the central GP was conducted, revealing that the greatest number of EMLS occurred in the northern GP. Back trajectories (BT) were conducted from the radiosonde point of detection for 18 and 36 hours, revealing that the BT point mean for days with severe weather were further west and south from the origin point. The SM and vegetation was sampled at the BT point, revealing a negative, significant correlation with EML depth when pooling the northern stations in 18-hr BTs, and a significant, negative correlation with EVI when pooling the southern sites. A modeling case study was conducted in which an idealized SM anomaly was imposed over the EML origin region. Experiments were also conducted to test the sensitivity of ML formation and EML transport using different PBL parameterizations. While the YSU PBL parameterization produced the deeper PBL over anonymously dry soils in the EML origin region, the EML was not transported to the east as it was in those experiments using the MYNN parameterization, impacting the timing and extent of precipitation in the model runs.
Description
Rights Access
Subject
elevated mixed layer
planetary boundary layer
mesoscale convective system
convection initiation