Back to search

Michigan Technological University

Investigating Microphysical Variability and Entrainment in Clouds Using Airborne Digital Holography, Machine Learning, and Large Eddy Simulations

Abstract

dc:description.abstract

Advancing our understanding of cloud microphysics—the processes governing cloud particles' formation, growth, and interactions—is essential for accurate weather and climate modeling. In this work, we investigate how droplet size distributions vary in a cloud and explore how microphysical properties such as droplet size and number concentration respond to entrainment and mixing. In the first part of the work, a novel algorithm is developed combining hypothesis testing with density-based clustering to identify characteristic cloud droplet size distribution types in marine stratocumulus clouds. Applied to data from the airborne Holographic Detector for Clouds (HOLODEC), the algorithm successfully identifies a relatively small set of distributions that cumulatively make up the average distribution shape at one level in the cloud. We find that locally, size distributions are narrower than the average distributions typically used in Global Climate Models, with similar droplet populations clustering in defined regions. We show how these are relevant to calculating process rates in climate models. As a follow-on, we examine Large Eddy Simulations of stratocumulus clouds to determine if these high-resolution cloud models can capture these characteristic distribution types. While characteristic distribution types are present in simulations, they lack the kilometer-scale structure and variability observed for real clouds. In the second part of the work, we investigate how cloud droplets respond to dry air mixing into cumulus clouds. The study uses HOLODEC and other in situ data from flights through warm-based cumulus clouds. Using two dimensionless parameters, one for droplet evaporation and one for humidity adjustment during turbulent mixing, we look at the evolution and microphysical response to entrainment. A turbulent diffusion model representing the mixing of conserved variables is developed to simulate the ensemble-mean behavior of droplet evaporation during the mixing process. The model shows trajectories through the 2D dimensionless space that closely mimic the field observations plotted in the same space. This provides a simple framework for understanding the evolution of the mixing process in clouds. We find for both the field data and the model, inhomogeneous mixing--- where a subset of droplets at the cloud boundary evaporate completely --- to be the primary entrainment mechanism. These findings have implications for improving cloud representation in models.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy in Physics (PhD)
Level thesis:degree_level
Open Access Dissertation
Discipline thesis:degree_discipline
Department of Physics

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Allwayin, Nithin
Committee members dc:contributor.committeemember
  • Will H. Cantrell
  • Michael L. Larsen
  • Claudio Mazzoleni
Contributors dc:contributor
  • Raymond A. Shaw

Subjects

dc:subject × 8

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Michigan Technological University
Base URL
digitalcommons.mtu.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Allwayin, Nithin. Investigating Microphysical Variability and Entrainment in Clouds Using Airborne Digital Holography, Machine Learning, and Large Eddy Simulations. Open Access Dissertation thesis, https://doi.org/10.37099/mtu.dc.etdr/1835