Fourier analysis allows to separate different frequency bands in the data. The need to do this arises in many fields of work. It applies for example to low or high tones in music and speech, or to the different types of vibrations in earthquakes. Here, a "windowed" Fourier analysis is applied to separate the high frequency "wiggles" from the box-shaped perturbations (compare original data in Analysis Tool window). In this particular example, Fourier analysis introduces spurious noise in both the low-frequency and high-frequency panels. The nature of the data makes wavelet filtering the preferred tool. Please try and compare!