In this dissertation، two innovative approaches for the inversion of spectral induced polarization (SIP) tomography data are introduced، each providing new perspectives in describing subsurface features from methodological and application viewpoints. The first approach combines continuous Homotopy and Bayesian inference، aiming to enhance the evaluation of Cole-Cole model (CCM) parameters and provide a comprehensive uncertainty analysis. In this approach، continuous Homotopy is first used for SIP data inversion to estimate the complex resistivity، followed by the extraction of spectral parameters of the CCM using the Markov Chain Monte Carlo (McMC) algorithm. This method has demonstrated high accuracy and efficiency in analyzing synthetic and real data، proving its ability to deliver reliable crosssections of subsurface structures.The second approach analyzes the dependency and correlation of CCM parameters using a Bayesian framework and a developed 2.5D inversion code. Through synthetic modeling، evaluation of McMC chains، and statistical tools such as corner plots، the relationships between spectral parameters are investigated، revealing the impact of these dependencies on estimating subsurface electrical properties. This approach not only contributes to a deeper understanding of complex geological structures but also advances geophysical data interpretation.Overall، the results of this research indicate that the proposed approaches can improve the accuracy and reliability of SIP data inversion and subsurface feature analysis، representing a significant step forward in developing advanced methods in earth sciences and geophysics.The first achievement of this dissertation is the development of an advanced inversion code for SIP data by combining continuous Homotopy and Bayesian inference، enablingmore precise and comprehensive analysisofsubsurfaceelectricalproperties. Thesecondachievementistheintroductionofanovelmethod for analyzing the dependency and correlation of CCMparameters، which enhances the understanding of the complex relationships among these parameters and improves geophysical data interpretation.