Andy Biggs (JIVE)
Ian Browne (JBO)
Neal Jackson (JBO)
Tom York (JBO)
Martin Norbury (JBO)
John McKean (UC Davis)
Paul Phillips (JBO)
VLBA Project Code: BP076 (5-GHz data in figure),
BB141 (other VLBA+Ef data mentioned in text)
Observation Epoch: 20 December 2000 (BP076), 02 January 2002 (BB141)
Shown here are VLBA 5-GHz maps of the gravitational lens system CLASS B0128+437. This is formed by the deflection of the light from a distant (z=3.12) quasar by an intervening galaxy (the lens), resulting in four "images" of the quasar being seen from Earth (MERLIN map - middle). The high resolution of the VLBA reveals that the lensed source consists of at least 3 sub-components. The detection of so much structure in lensed images is relatively rare and will enable models of the mass distribution in the lens galaxy to be tightly constrained. Most striking is the absence of sub-components in image B, this most likely being a result of scatter-broadening in the interstellar medium (ISM) of the lensing galaxy.
CLASS B0128+437 has been the target of a concerted multi-frequency
VLBI campaign, having been observed to date at 2.3, 5 and 8.4GHz. The 5-GHz data (December 2000) were observed with the ten antennas of the VLBA whilst the 2.3 and 8.4GHz data (not shown) used in addition the 100-m Effelsberg telescope in Germany (in January 2002). Multiple-frequency data allow us to measure the brightness and size of the sub-components in each image as a function of frequency, greatly improving our understanding of this lens system.
For example, our observations have shown that whilst the brightness of the two sub-components marked with an 's' in the figure decline steeply with increasing frequency, that of the sub-component marked with an 'f' stays roughly constant. Therefore, the source consists of one flat-spectrum sub-component (probably the 'core' of the radio source and location of the black hole that powers the radio emission) and two steep-spectrum sub-components located in a one-sided jet that emanates from the core.
It is possible to model the distribution of mass in the lens galaxy using observations of multiple sub-components. This is done using the constraint that any model must be able to project a particular sub-component from each image to the same point in the source. In order to do this it is of course important to know which sub-components 'match up' in each image. With this accomplished in B0128+437 we have attempted to produce a mass model, but a reasonable fit to the data is only produced when the uncertainties on the image positions are unrealistically large. This may indicate substructure in the lensing galaxy, a detailed optical/infrared image of which is not yet available.
The scatter-broadening theory for the absence of the sub-components in image B is supported by the fact that we find image B to be larger and more diffuse than the other images. Gravitational lensing preserves the surface brightness of the source in each of the images, but this is demonstratively not the case in image B. Changing the weighting of the data from 'natural' (shown) to 'uniform' (higher resolution, poorer surface brightness sensitivity) makes little difference to the maps of images A, C and D whilst image B becomes extremely weak. The reduction in surface brightness of image B is most likely the result of scattering in a non-uniform ionised plasma in the lensing galaxy. Gravitational lenses can in this way be used to study the ISMs of high-redshift galaxies.
Biggs, A.D., Browne, I.W.A., Jackson, N.J., York, T., Norbury, M.N., McKean, J.P., Phillips, P.M., "Radio, optical and infrared observations of CLASS B0128+437", 2004, MNRAS, accepted (astro-ph/0402128)
Phillips, P.M., et al., "A New Quadruple Gravitational Lens System: CLASS B0128+437", 2000, MNRAS, 319, L7