HR: 17:35h
AN: NG54A-06    [Abstracts]
TI: Detection of Intermittent Turbulence in Stable Boundary Layer Using Empirical Mode Decomposition
AU: * Gilliam, X
EM: kathleen.gilliam@ttu.edu
AF: Texas Tech University, Department of Mathematics and Statistics, MS 41042, Lubbock, TX 79409, United States
AU: Ho, C
EM: christina.ho@ttu.edu
AF: Texas Tech University, Department of Mathematics and Statistics, MS 41042, Lubbock, TX 79409, United States
AU: Basu, S
EM: sukanta.basu@ttu.edu
AF: Texas Tech University, Department of Geosciences, MS 42101, Lubbock, TX 79409, United States
AB: Detecting and characterizing intermittent turbulence in stable boundary layer is important for both understanding the meteorological processes and for structural damage mitigation. Since such intermittent turbulence events are often short in duration with high amplitudes, traditional stationary or linear signal processing often fail to catch the physical phenomena. Our analysis seeks these localized phenomena with the empirical mode decomposition (EMD) method. EMD has recently gained more and more popularity for analyzing nonlinear and non-stationary signals. In particular, the method extracts concentrated structures without leaving the time domain, making event identification possible. In this research, we first apply EMD method to the data to extract the so- called intrinsic mode functions (IMF) in the time domain. During this filtering process, our first several IMFs will contain the shortest oscillations of the data with high amplitudes. Thus, the potential intermittent turbulent events embedded in the data can be identified. To further characterize the intermittence, a rigorous statistical test, based on the oldest methods in nonparametric statistics, the development of a randomized reference distribution is explored. Based on this statistical test, a threshold is defined to provide information in the data that is statistically significant, in our case, the intermittent turbulence events. An extensive collection of field observation data from various campaigns is used to corroborate the potential of our coherent structure detection methodology.
DE: 3307 Boundary layer processes
DE: 3322 Land/atmosphere interactions (1218, 1631, 1843)
DE: 3349 Polar meteorology
DE: 3379 Turbulence (4490)
DE: 4490 Turbulence (3379, 4568, 7863)
SC: Nonlinear Geophysics [NG]
MN: 2007 Joint Assembly