The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
Lightning and Its Relationship with the Large-Scale Environment
Abstract
Lightning can have many societal and environmental impacts, and the ability of a global climate model (GCM) to correctly parameterize lightning has become of utmost importance as lightning has recently been named an essential climate variable. Utilizing relationships deduced from the large-scale environment as the basis of a lightning parameterization can be fruitful as it limits the variability between models compared to using ice and precipitation variables. There is further benefit to exploring different mathematical frameworks to create the most robust parameterization to predict lightning in GCMs. This dissertation employs satellite observations of lightning, atmospheric fields from reanalyses, advanced statistical methods, and GCM output toward better understanding lightning production in present and future climates. An additional goal of improving student comprehension of difficult atmospheric science concepts through a variety active learning techniques will also be addressed.
To quantify the relationship between the large-scale environment and lightning, seven atmospheric variables (CAPE, nCAPE, LCL, column saturation fraction, 700-hPa omega, wind shear from 900 to 700 hPa, and wind shear from 900 to 300 hPa) from MERRA-2 were compared to lightning occurrence and intensity from TRMM LIS observations from 1998 to 2013. All environmental variables show a significant shift toward larger values when lightning is present except for shear. Deep shear decreases when lightning is present, while low-level shear shows little mean change. However, strong geographical differences exist in the relationship between the environ-mental variables and lightning occurrence, particularly between land and ocean and the tropics and subtropics.
Consistent with the idea that lightning intensity and/or occurrence is dependent on variations in the large-scale circulation, the Quasi-Biennial Oscillation (QBO) influence on TRMM LIS lightning occurrence and intensity were explored using ERA-Interim proxies for tropopause temperature and cross-tropopause shear from 1998 to 2013, encompassing seven full QBO events. QBO west (east) phase flash rates increase most in MAM and JJA (DJF), and appear to be more strongly driven by relatively colder tropopause temperatures during those respective seasons. Lightning occurrence is most strongly impacted by the QBO in MAM and SON, especially when strong ENSO months are removed, with opposite hemispheric signals between phases and weaker cross-troposphere shear playing a more equal role.
These relationships were further exploited in a GCM lightning parameterization developed using a logistic regression, which predicts lightning occurrence correctly 84.6% of the time using only CAPE, LCL, and r as predictors. The parameterization was retrained using ISS LIS observations for more latitudinal extent and evaluated in CAM5 for present day and end-of-century scenarios. The prediction shows a varied response, where both increases and decreases in lightning occurrence were found over land, unlike previous studies of flash rates. Advanced statistical models (a zero-inflated Poisson (ZIP) regression and a Log-Gaussian Cox Process (LGCP) model) were used to estimate both lightning intensity and occurrence for a daily case study and seasonal analysis over Brazil. The ZIP regression only performs well when there is sufficient lightning in the domain while the LGCP model shows the best prediction fidelity but is computationally-intensive to implement and may be difficult to apply directly to a GCM lightning parameterization. All statistical models show a similar pattern for end-century predictions over Brazil, with decreases in the north and increases in the south.
Lastly, active learning is investigated in an introductory meteorology class through four teaching methods: enhanced lecture, think-pair-share (TPS), game-based (GB) learning, and role-playing (RP). All four methods improved learning, and were preferred by students over the traditional lecture-based teaching of the class. However, the topic for which they were taught often varied the results, and although active learning for the class should be considered, its approach is not one-size-fits-all.
* An edited portion and version of this chapter was published by AGU as “Evaluating the relationship between lightning and the large-scale environment and its use for lightning prediction in global climate models” in Journal of Geophysical Research: Atmospheres by Etten-Bohm, Montana, J. Yang, C. Schumacher, and M. Jun, 2021. Copyright (2021) American Geophysical Union.
Citation
Etten-Bohm, Montana (2022). Lightning and Its Relationship with the Large-Scale Environment. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /197230.