A Re-examination of the Distribution of Galactic Free Electrons

TL;DR

This paper refines the galactic electron distribution model using updated pulsar data, revealing a two-component disk structure and analyzing deviations linked to spiral arms and distance uncertainties.

Contribution

It introduces an improved two-disk model of galactic electrons based on new pulsar distance measurements, comparing different functional forms and examining deviations.

Findings

Two exponential components with specific densities and scale heights fit the data well.

The sech^2 model provides a similar fit without a midplane cusp.

Deviations correlate with spiral arms and distance uncertainties.

Abstract

We present a list of 109 pulsars with independent distance information compiled from the literature. Since the compilation of Frail & Weisberg, there are 35 pulsars with new distance estimate and 25 pulsars for which the distance or distance uncertainty have been revised. We used this data to fit a smooth, axisymmetric, two disk model of the distribution of galactic electrons. The two exponential model components have mean local midplane densities at the solar circle of 2.03e-2 cm^-3 and 0.71e-2 cm^-3, and scale heights of 1.07 and 0.053 kpc. The thick component shows very little radial variation, while the second has a radial scale length of only a few kiloparsecs. We also examined a model which varies as sech^2(x), rather than exp(-x), in both the radial and vertical direction. We prefer this model with no midplane cusp, but find that the fit parameters essentially describe the same electron distribution. The distances predicted by this distribution have a similar scatter as the more complex model of Taylor & Cordes. We examine the pulsars that deviate strongly from this model. There are two regions of enhanced dispersion measure, one of which correlates well with the Sagittarius-Carina spiral arm. We find that the scatter of the observed dispersion measure from the model is not fit well by either a normal or log-normal distribution of lump sizes, but may be caused instead by the uncertainties in the distances.