Date of Graduation


Document Type


Degree Type



Eberly College of Arts and Sciences


Physics and Astronomy

Committee Chair

Mark E. Koepke.


Tornadoes often leave behind patterns of debris deposition, or "surface marks", which provide a direct signature of their near surface winds. The intent of this thesis is to investigate what can be learned about near-surface tornado structure and intensity through the properties of surface marks generated by simulated, debris-laden tornadoes. Earlier work showed through numerical simulations that the tornado's structure and intensity is highly sensitive to properties of the near-surface flow and can change rapidly in time for some conditions. The strongest winds often occur within tens of meters of the surface where the threat to human life and property is highest, and factors such as massive debris loadings and asymmetry of the main vortex have proven to be critical complications in some regimes. However, studying this portion of the flow in the field is problematic; while Doppler radar provides the best tornado wind field measurements, it cannot probe below about 20 m, and interpretation of Doppler data requires assumptions about tornado symmetry, steadiness in time, and correlation between scatterer and air velocities that are more uncertain near the surface.;As early as 1967, Fujita proposed estimating tornado wind speeds from analysis of aerial photography and ground documentation of surface marks. A handful of studies followed but were limited by difficulties in interpreting physical origins of the marks, and little scientific attention has been paid to them since. Here, Fujita's original idea is revisited in the context of three-dimensional, large-eddy simulations of tornadoes with fully-coupled debris.;In this thesis, the origins of the most prominent simulated marks are determined and compared with historical interpretations of real marks. The earlier hypothesis that cycloidal surface marks were directly correlated with the paths of individual vortices (either the main vortex or its secondary vortices, when present) is unsupported by the simulation results. Cycloids in the simulations arise from debris deposited beneath the central annular updraft that has converged from a much larger area and are modulated by turbulent fluctuations in debris amount. Other classes of marks noted in the literature such as "lineation" and "scalloping" are also reinterpreted. Variations in the shapes, sizes, and spacings of surface marks with the most critical dimensionless parameters characterizing near-surface and debris cloud structure are explored. Analysis techniques are presented to capture the geometric properties of marks in some regimes, and possibilities for inferring near-surface vortex flow scales from mark properties are discussed. The prospects are promising enough to warrant documentation of surface marks when available (likely through aerial photography), particularly for cases where useful Doppler measurements have been gathered.