Abstract: The characteristics of gas information are fundamental parameters for comprehending the various states of coal spontaneous combustion in the gob. Understanding the gas change law and its intrinsic correlation in a complex environment is crucial for establishing a coal spontaneous combustion prevention and control technological system. This study employs wavelet transform to analyze the multiscale evolution law of CO and O₂ volume fractions in different areas of the working face. The aim is to investigate the evolution process, periodic distribution, and correlation between the two kinds of gases in different regions. This analysis evaluates the sensitivity and correlation of different areas to coal spontaneous combustion. The findings suggest that the CO and O₂ volume fractions at the working face exhibit multiscale effects and demonstrate a certain periodicity at different time scales. The main period distribution of the two kinds of gases in the same area is similar at higher time scales, whereas a significant difference exists in the main period at smaller time scales. The oscillation characteristics of CO and O₂ volume fraction wavelet coefficients reflect the fluctuation process of the original volume fraction of the gases. The initial amplitude and initial phase difference of the wavelet coefficient fitting equation under the first principal period reflect the correlation between different regions and coal spontaneous combustion. The gas information in the gob has the highest correlation with coal spontaneous combustion, while the high extraction roadway exhibits the lowest correlation.Based on the correlation between the goaf and spontaneous coal combustion as the benchmark, the relative correlations of spontaneous coal combustion in the goaf, inside the bag wall at the upper corner, outside the bag wall at the upper corner, and in the high drainage roadway were 1, 0.16, 0.12 and 0.09, respectively. Furthermore, based on the correlation degree and the distribution of the "three zones" of coal spontaneous combustion, the study proposes the concept of accurately dividing the risk area of coal spontaneous combustion.