The complex, variable near infrared extinction towards the Nuclear Bulge

TL;DR

This study reveals complex, small-scale variations in near-infrared extinction towards the Nuclear Bulge, introducing a new method to accurately measure and account for these variations, which impacts previous extinction estimates and corrections.

Contribution

We developed the V-NICE method to measure variable near-infrared extinction law parameters on small scales, improving accuracy of reddening corrections in the Nuclear Bulge.

Findings

Extinction law parameter alpha varies significantly on 5 arcsec scales.

Steeper extinction law results in higher A_J and A_H values.

Extinction measurements depend on instrument filter transmission functions.

Abstract

Using deep J, H and Ks-band observations, we have studied the near-infrared (nIR) extinction of the Nuclear Bulge (NB) and we find significant, complex variations on small physical scales. We have applied a new variable nIR colour excess method, V-NICE, to measure the extinction; this method allows for variation in both the extinction law parameter alpha and the degree of absolute extinction on very small physical scales. We see significant variation in both these parameters on scales of 5 arcsec. In our observed fields, representing a random sample of sight lines to the NB, we measure alpha to be 2.64 +- 0.52, compared to the canonical "universal" value of 2. Our measured levels of A_Ks are similar to previously measured results (1 < A_Ks < 4.5); however, the steeper extinction law results in higher values for A_J (4.5 < A_J < 10) and A_H (1.5 < A_H < 6.5). Only when the extinction law is allowed to vary on the smallest scales can we recover self-consistent measures of the absolute extinction at each wavelength, allowing accurate reddening corrections for field star photometry in the NB. The steeper extinction law slope also suggests that previous conversions of nIR extinction to A_V may need to be reconsidered. Finally, we find that the measured values of extinction are significantly dependent on the filter transmission functions of the instrument used to obtain the data. This effect must be taken into account when combining or comparing data from different instruments.