地磁数据分析的一些思考

通过若干实例，本文讨论了地磁数据分析中的一些方法.这些方法包括时间序列的多项式拟合，相关性分析，频谱分析和小波分析等.在Matlab和AutoSignal软件下实现了以上方法，并基于这些方法研究了地磁场形态和地磁现象的频谱特征.我们分析了Surlari观测台（余纬45.3°，经度26.3°）的地磁时间序列，并与Ottawa（余纬44.6°，〖JP2〗经度284.5°），Canberra（余纬125.3°，经度149.3°）, Kakioka（余纬53.8°，经度140.2°），Vernadsky〖JP〗（余纬155.3°，经度295.7°）等台站的结果进行了对比分析，讨论了Fourier变换时间窗和小波基函数选取的影响，有限长时间序列引起的边缘效应，以及小波尺度和Fourier频率的关系等问题.我们采用谱分析和小波分析处理了2007年两个台站的地磁数据，研究了地磁场形态和地磁现象的频谱特征，并讨论了Fourier变换和小波变换的优缺点.

CONSIDERATIONS ON GEOMAGNETIC DATA ANALYSIS

This article aims to make a few remarks, examples and results about methods for analysis of geomagnetic data. In our study we used several methods and algorithms about numerical derivatives depending on time, polynomial regression, correlation factor, spectral analysis and wavelet analysis. With these algorithms were developed several programs in MatLab and AutoSignal software to study geomagnetic field morphology and to determine the spectrum of geomagnetic phenomena in different time intervals. Considering some period of time geomagnetic data recorded at the Observatory Surlari (colatitude 45.3°, longitude 26.3°) and comparing them with data recorded at other observatories, such as Ottawa (colatitude 44.6°, longitude 284.5°), Canberra (colatitude 125.3°, longitude 149.3°), Kakioka (colatitude 53.8°, longitude 140.2°) and Vernadsky (colatitude 155.3°, longitude 295.7°), including comparison to the windowed Fourier transform, the choice of an appropriate wavelet basis function, edge effects due to finite-length time series, and the relationship between wavelet scale and Fourier frequency. From geomagnetic data analysis (spectral and wavelet) on definitive data in year 2007 recorded at two observatories to show that disadvantages of applying only the spectral analysis are related to less capacity of locating frequencies, amplitudes and phases in time. Also, the advantages and disadvantages of Fourier Transform and Wavelet Transform are discussed.