2019MNRAS.489.4589A

PSR J0108-1431 is an old pulsar where the X-ray emission is expected to have a thermal component from the polar cap and a non-thermal component from the magnetosphere. Although the phase-integrated spectra are fit best with a single non-thermal component modelled with a power law (PL) of photon index Γ = 2.9, the X-ray pulse profiles do show the presence of phase-separated thermal and non-thermal components. The spectrum extracted from half the rotational phase away from the X-ray peak fits well with either a single blackbody (BB) or a neutron star atmosphere (NA) model, whereas the spectrum from the rest of the phase range is dominated by a PL. From Bayesian analysis, the estimated BB area is smaller than the expected polar cap area for a dipolar magnetic field with a probability of 86 per cent, whereas the area estimate from the NA model is larger with a probability of 80 per cent. Due to the ambiguity in the thermal emission model, the polar cap area cannot be reliably estimated and hence cannot be used to understand the nature of the surface magnetic field. Instead, we can infer the presence of multipolar magnetic field from the misalignment between the pulsar's thermal X-ray peak and the radio emission peak. For J0108-1431, we estimated a phase-offset Δph > 0.1 between the thermal polar cap emission peak and the radio emission peak and argue that this is best explained by the presence of a multipolar surface magnetic field.