An analysis of the intrinsic cross-correlations between API and meteorological elements using DPCCA

• Reveal intrinsic cross-correlation between API and meteorological elements through DPCC.
• DPCC coefficients for various locations and seasons differ, but are explained by identical mechanisms.
• DCCA discloses cross-correlations between time series, but does not fully reflect interaction mechanism between them.

In order to reveal the intrinsic cross-correlations between air pollution index (API) records and synchronously meteorological elements data, the detrended partial cross-correlation (DPCC) coefficients are analyzed using a detrended partial cross-correlation analysis (DPCCA). DPCC coefficients for different spatial locations and seasons are calculated and compared. The results show that DPCCA can uncover intrinsic cross-correlations between API and meteorological elements, and most of their interactional mechanisms can be explained. DPCC coefficients are either positive or negative, and vary with spatial locations and seasons, with consistently interactional mechanisms. More remarkable, we find that detrended cross-correlation analysis can present the cross-correlations between the fluctuations in two nonstationary time series, but this cross-correlation does not always fully reflect the interactional mechanism for the original time series. Despite this, DPCCA is recommended as a comparatively reliable method for revealing intrinsic cross-correlations between API and meteorological elements, and it can also be useful for our understanding of their interactional mechanisms.