Abstarct: Signal analysis is being used in many fields such as geophysics. It has prominent role in seismic data interpretation. Because of non stationary property of seismic signals, conventional representation methods in time domain and Fourier domain cannot show signal information simultaneously. So time- frequency methods are widely used. Time- frequency methods such as short time Fourier transform (STDT), wavelet transform, S transform (ST), Wigner-Ville transform, deconvolutive short time Fourier transform (DSTFT) and etc are used in order to spectral decomposition. Spectral decomposition method generates high volume of spectral components which contain useful information about structural trends and sedimentary deposits. This high volume overloads limited space of computer and makes interpretation difficult. There are different solutions for this problem such as single frequency slice representation method and color stacking method. Although these methods obviate some drawbacks but frequency choice is not clear. So we propose using principal component analysis. Principal components are constructed by mapping single frequency slices to eigen-vectors. Because three largest principal components contribute big part of spectral variance existing in spectral decomposition data, using color stacking method, we can show this information in just one picture. In this thesis, firstly, we apply spectral decomposition using STFT, ST and DSTFT methods on land 3D seismic data in order to buried channel detection. Then we perform interpretation using single frequency volume and single frequency slice methods and show their limitations. Then we produce different types of RGB images and show their dependency to frequency choice. Finally, we improve seismic data interpretation using principal component analysis method and show the independency of produced images to frequency choice. Although this method does not lead to quantitative information such as porosity and permeability, but it can be used as a suitable tool for qualitative interpretation.