The Simultaneous Influence of Thermodynamics and Aerosols on Deep Convection and Lightning
November 20, 2015
Hosted by Steve Rutledge (advisor), Jeff Pierce, Sue van den Heever, Steven Reising (Electrical and Computer Engineering)
The dissertation consists of a multi-scale investigation of the relative contributions of thermodynamics and aerosols to the observed variability of deep convective clouds in the Tropics. First, estimates of thermodynamic quantities and cloud-condensation nuclei (CCN) in the environment are attributed to convective features (CFs) observed by the Tropical Rainfall Measuring Mission (TRMM) satellite for eight years (2004-2011) between 36°S-36°N across all longitudes. The collection of simultaneous observations was analyzed in order to assess the relevance of thermodynamic and aerosol hypotheses for explaining the spatial and temporal variability of the characteristics of deep convective clouds. Specifically, the impacts of normalized convective available potential energy (NCAPE) and warm cloud depth (WCD) as well as CCN concentrations (D ≥ 40 nm) on total lightning density (TLD), average height of 30 dBZ echoes (AVGHT30), and vertical profiles of radar reflectivity (VPRR) within individual CFs are the subject of initial curiosity.
The results show that TLD increased by up to 600% and AVGHT30 increased by up to 2-3 km with increasing NCAPE and CCN for fixed WCD on the global scale. The partial sensitivity of TLD/AVGHT30 to NCAPE and CCN individually are found to be comparable in magnitude, but each independent variable accounts for a fraction of the total range of variability observed in the response (i.e., when the influences of NCAPE and CCN are considered simultaneously). Both TLD and AVGHT30 vary inversely with WCD such that maxima of TLD and AVGHT30 are found for the combination of high NCAPE, high CCN, and shallower WCD. The relationship between lightning and radar reflectivity is shown to vary as a function of CCN for a fixed thermodynamic environment. Analysis of VPRRs shows that reflectivity in the mixed phase region (altitudes where temperatures are between 0°C and -40°C) is up to 5.0-5.6 dB greater for CFs in polluted environments compared to CFs in pristine environments (holding thermodynamics fixed).
A statistical decomposition of the relative contributions of NCAPE, CCN, and WCD to the variability of convective intensity proxies is undertaken. Linear model approximations of convective intensity proxies are studied in order to estimate relative weighting for each predictor in a multiple linear regression framework. Simple linear models of TLD/AVGHT30 based on the predictor set composed of NCAPE, CCN, and WCD account for appreciable portions of the variability in convective intensity (R2 ≈ 0.3-0.8) over the global domain, continents, oceans, and select regions. Furthermore, the results from the statistical analysis suggest that the simultaneous contributions from NCAPE, CCN, and WCD to the variability of convective intensity are often comparable in magnitude. There was evidence for similar relationships over even finer-scale regions [O(106 km2)], but differences in the relative prognostic ability and stability of individual regression parameters between regions/seasons were apparent. These results highlight the need to investigate the connection between statistical behavior and local meteorological variability within individual regions.
Following the global and regional analyses, data from Dynamics of the Madden-Julian Oscillation (DYNAMO) field campaign (2011-2012; central equatorial Indian Ocean (CIO)) and other sources was used to assess the relative impact of aerosols on deep convective clouds within a fine-scale environment with spatially homogeneous thermodynamics and variable aerosols in a pristine background over the CIO (CCN ~50-100 cm-3, on average; NCAPE and WCD are hypothesized to be approximately constant, spatially). The experiment was designed to compare differences in the convective cloud population developing in more-polluted and pristine regions, north and south of the equator, respectively. Analysis of the covariability of rainfall, cold cloud frequency, CCN, NCAPE, and lightning/radar reflectivity in deep convective clouds over multiple (> 20) episodes of the Madden-Julian Oscillation (MJO) leads to a hypothesis for a potential bi-directional interaction between aerosols and convective clouds that develop in association with the MJO. Close scrutiny of the results from climatology leads to the conclusion that thermodynamics and aerosols both influence deep convective cloud behavior over the CIO in a manner similar to that observed on the global scale, but the possibility that other factors are required to reproduce the full range of variability of deep convective clouds on fine-scales is acknowledged.
The research presented in this dissertation constitutes one of the first efforts to link the documented variability of radar reflectivity and lightning within convective features observed by the TRMM satellite to their environment using novel representations of thermodynamic and aerosol quantities from reanalysis and a chemical transport model, respectively. The independent variables studied here (i.e., NCAPE, CCN, and WCD) were chosen specifically to address preeminent hypotheses in the literature and the results from this investigation suggest that NCAPE, CCN, and WCD each contribute significantly to the variability of deep convective clouds throughout the Tropics and Subtropics (and perhaps seasonally). Implications of the findings from the current investigations and the relevance of these results to future studies are discussed.