Abstract In this paper, we analyze the principle of tidal and atmospheric pressure effects of the water level, focusing on solving the problem of multicollinearity between the input variables. The Partial Least Squares Regression (PLS) model is used to eliminate the tidal and atmospheric pressure effects, and the results show that the response coefficients can correctly reflect the physical processes of tidal force and atmospheric pressure fluctuations. Both the magnitude of the tidal and atmospheric pressure coefficients, and the hysteresis and delay constants of atmospheric pressure, are closely related to the parameters of the aquifer (i.e., storage, porosity, permeability, bulk modulus, etc.),the crustal stress, and strain state around the borehole. Through continuous 60 days (Step size is 9 days) calculation of Chuan 03# well, the seismic and co-seismic responses, the adjustment processes after earthquakes, and even precursor anomalies are clearly identified in the tidal and atmospheric pressure response coefficients.