The evolution of the magnetic inclination angle as an explanation of the long term red timing-noise of pulsars

We study the possibility that the long term red timing-noise in pulsars originates from the evolution of the magnetic inclination angle $chi$. The braking torque under consideration is a combination of the dipole radiation and the current loss. We find that the evolution of $chi$ can give rise to extra cubic and fourth-order polynomial terms in the timing residuals. These two terms are determined by the efficiency of the dipole radiation, the relative electric-current density in the pulsar tube and $chi$. The following observation facts can be explained with this model: a) young pulsars have positive $ddot{ u}$; b) old pulsars can have both positive and negative $ddot{ u}$; c) the absolute values of $ddot{ u}$ are proportional to $-dot{ u}$; d) the absolute values of the braking indices are proportional to the characteristic ages of pulsars. If the evolution of $chi$ is purely due to rotation kinematics, then it can not explain the pulsars with braking index less than 3, and thus the intrinsic change of the magnetic field is needed in this case. Comparing the model with observations, we conclude that the drift direction of $chi$ might oscillate many times during the lifetime of a pulsar. The evolution of $chi$ is not sufficient to explain the rotation behavior of the Crab pulsar, because the observed $chi$ and $dot{chi}$ are inconsistent with the values indicated from the timing residuals using this model.