Abstract: Taking the fault gas carbon dioxide observation data from Tianshui Seismic Station from 2022 to 2025 as the research object, db4 wavelet was adopted for analytical processing, and short-term low-frequency and high-frequency anomalies were investigated respectively. The results show that the wavelet analysis method boasts strong capability in identifying information of different frequencies. It facilitates the separation of trend variations and local fluctuations of data, effectively recognizes and eliminates interference factors, and presents seismic precursor anomalies more intuitively and prominently, serving as an effective technical approach for seismic precursor data processing. Earthquake case verification was conducted on the Ms6.2 Jishishan earthquake on December 18, 2023 and the Ms5.6 Longxi earthquake on September 27, 2025 in Gansu Province. It is found that jump anomalies of fault gas carbon dioxide emerge 3 to 6 months prior to earthquakes with magnitude 5.0 and above within the 300-kilometer range around the station, characterized by carbon dioxide concentration variations exceeding twice the mean square deviation.