HR: 1340h
AN: SM13A-0313 [Abstracts]
TI: Can Wavelet Analysis Provide an Automated Technique to Create the Dst Index?
AU: * Jach, A
EM: ajach@cc.usu.edu
AF: Mathematics and Statistics Department, Utah State University,
3900 Old Main Hill, Logan, UT 84322-3900
United States
AU: Kokoszka, P
EM: Piotr.Kokoszka@usu.edu
AF: Mathematics and Statistics Department, Utah State University,
3900 Old Main Hill, Logan, UT 84322-3900
United States
AU: Zhu, L
EM: zhu@cc.usu.edu
AF: Center for Atmospheric and Space Sciences, Utah State University,
4405 Old Main Hill, Logan, UT 84322-4405
United States
AU: Sojka, J J
EM: fasojka@sojka.cass.usu.edu
AF: Center for Atmospheric and Space Sciences, Utah State University,
4405 Old Main Hill, Logan, UT 84322-4405
United States
AB:
The primary geomagnetic storm indicator is the Dst index. This index has a well established ``recipe" by which ground-based
magnetometer observations are processed to produce this single planetary index. Typically four ground stations are used in
this process. Assuming the calibration, data gaps, and measurement uncertainties are known, the major step in creating the
Dst is the subtraction of the daily Sq variation. This variation is associated with ionospheric current systems resulting
from upper atmospheric winds that drive electric currents in the E region plasma. The Sq is primarily a daytime phenomena and
it exhibits significant day-to-day variability. Determining a quiet day Sq variation, which has seasonal dependence as well
as shorter term ones, is the ``art" in this analysis procedure. Schemes for selecting this can be as quantitative as
averaging the three quietest days in the month, i.e., a new Sq variation is inferred each month. The question addressed by
this study is whether it is feasible for an automated technique to remove the Sq without operator intervention? Our study is
based on wavelet analysis, specifically the {mathematical specifications}. An algorithm was developed for four ground
stations for two months of measurements, (March and April 2001). Note a number of procedural differences exist between this
wavelet technique and the normal method of calculating the Dst. The automated technique does not require selecting specific
days that are defined to be quiet days. The results of applying the procedure to a year of data and then comparing these with
the Kyoto Dst are presented. Additional comparisons with the Kyoto Dst are presented in which the automated analysis uses
only three of the four magnetometer data sets.
DE: 2778 Ring current
DE: 2788 Magnetic storms and substorms (7954)
DE: 3270 Time series analysis (1872, 4277, 4475)
DE: 3280 Wavelet transform (3255, 4455)
SC: SPA-Magnetospheric Physics [SM]
MN: Fall Meeting 2005