imagepol.rotationmeasure

Find Rotation Measure (traditional approach)

Synopsis
rotationmeasure(rm, rmerr, pa0, pa0err, nturns, chisq, sigma, rmfg, rmmax, maxpaerr, plotter, nx, ny)

Arguments

rm	in	Output Rotation Measure image file name
		Allowed:	String
		Default:	Unset
rmerr	in	Output Rotation Measure error image file name
		Allowed:	String
		Default:	Unset
pa0	in	Output position angle (degrees) at zero wavelength image file name
		Allowed:	String
		Default:	Unset
pa0err	in	Output position angle (degrees) at zero wavelength error image file name
		Allowed:	String
		Default:	Unset
nturns	in	Output number of turns image file name
		Allowed:	String
		Default:	Unset
chisq	in	Output reduced chi squared image file name
		Allowed:	String
		Default:	Unset
sigma	in	Estimate of the thermal noise
		Allowed:	Float
		Default:	Auto
rmfg	in	Foreground Rotation Measure (rad/m/m) to subtract
		Allowed:	Float
		Default:	0
rmmax	in	Maximum Rotation Measure (rad/m/m) to solve for
		Allowed:	Float
		Default:	0
maxpaerr	in	Maximum input position angle error (degrees) to allow
		Allowed:	Float
		Default:	None
plotter	in	Name of plotter
		Allowed:	String
		Default:	None
nx	in	Number of plots in x direction
		Allowed:	Integer
		Default:	5
ny	in	Number of plots in y direction
		Allowed:	Integer
		Default:	5

Returns
T or fail

Description

This function (short-hand name rm) generates the Rotation Measure image from a collection of different frequencies. It will only work if the Imagepol tool was constructed from an image containing Stokes Q and U, and a frequency axis (regular or irregular) with at least 2 pixels. It will work out the position angle images for you.

See also the fourierrotationmeasure function for a new Fourier-based approach.

Rotation Measure algorithms that work robustly are not common. The main problem is in trying to account for the n - $\pi$ ambiguity (see Leahy et al, Astronomy & Astrophysics, 156, 234 or Killeen et al; http://www.atnf.csiro.au/~nkilleen/rm.ps).

The algorithm that this function uses is that of Leahy et al. Please refer to their paper for details (their Appendix A.1). But as in all these algorithms, the basic process is that for each spatial pixel, a vector of position angles (i.e. at the different frequencies) is fit to determine the Rotation Measure and the position angle at zero wavelength (and their errors). You can write out an image containing the number of n - - $\pi$ turns that were added to the data at each spatial pixel and for which the best fit was found. You can also write out the reduced chi-squared image for the fits.

Note that as yet no assessment of curvature (i.e. deviation from the simple linear position angle - $\lambda^{2}_{}$ functional form) is made.

You can write out any or all of the output images.

The argument sigma gives the themal noise in Stokes Q and U. By default it is worked out for you from the image data. But if it proves to be inaccurate (maybe not many signal-free pixels), then you can input it here. This is used for working out the error in the position angles (propagation of Gaussian errors).

The argument maxpaerr specifies the maximum allowable error in the position angle that is acceptable. The default is an infinite value. From the standard propagation of errors, the error in the linearly polarized position angle is determined from the Stokes Q and U images (at each spatial pixel for each frequency). At each spatial pixel we do a fit to the position angle vector (i.e. at the different frequencies) to determine the Rotation Measure. If the position angle error for any pixel in the vector exceeds the specified value, it is dropped from the fit. The process generates an error for the fit and this is used to compute the errors in the output images.

Note that maxpaerr is not used to specify that any pixel for which the output position angle error exceeds this value should be masked out.

The argument rmfg is used to specify a foreground RM value. For example, you may know the mean RM in some direction out of the Galaxy, then including this can aid the algorithm (reduces ambiguity).

The argument rmmax specifies the maximum absolute RM that should be solved for. This quite an important parameter. If you leave it at the default, zero, basically, no ambiguity handling will be invoked. So you must have some idea of what you are looking for - this is the basic problem with Rotation Measure algorithms.

Plotting can be done with a PGPLOT device via argument plotter. The syntax is plotter=name/type. For example plotter='plot1.ps/ps' (disk postscript file) or plotter='1/xs' (X-windows device) or plotter='plot/glish' (Glish PGplotter). The plots show the used position angle data (after ambiguity adjustments), position angle errors, and fit for each profile. The pixel location of each profile and reduced chi-square of the fit are written in the title.

Example

- im := imageconcat(infiles="im.f1 im.f2 im.f3 im.f4 im.f5", axis=4)
- p := imagepol(im)
- ok := p.rm(rm='rm', rmerr='rmerr', rmmax=800, maxpaerr=10)

Say we have 5 images, each with axes RA, DEC, Stokes, and Frequency in that order. We make an Image tool which concatenates these images along the frequency axis - you have ordered them in increasing or decreasing frequency order. Note that the Image tool is virtual - it is not written to an output file. This just means that the data are read from the input images as needed. The Imagepol tool is then made from this virtual Image tool. We then compute the Rotation Measure and Rotation Measure error images with the traditional method and write them out to disk.

Next: imagepol.sigma - Function Up: imagepol - Tool Previous: imagepol.pol - Function Contents Index

	Package	general
	Module	images
	Tool	imagepol

imagepol.rotationmeasure - Function