3.8  Wavelengths

Although nominally CDS has a fixed wavelength range, small changes in the zero point of the pixel-wavelength relation can occur, primarily from on board temperature changes caused by varying illumination of the structure by the Sun. The wavelength array returned by the gt_xxxx functions or the pix2wave function is only an average determined from sample spectra. For accurate wavelength analyses, the zero point should be determined from each spectrum individually.

The calibration appropriate to a particular dataset is loaded when the data file is read. If you want to know what the currently loaded coefficients are, use the routine

            IDL> qwavecal 

The slant of the two channels, and the tilt of the lines complicates the wavelength determination.

The NIS wavelength calibration also has some peculiar aspects. The wavelength calibration requires a `scan mirror correction', which appears to be different for different observations. Then, the NIS wavelength calibration has been found to be a function of the CDS optical bench temperature (D. Pike, priv. comm.) and its history, which continuously changes depending on the solar illumination. It has been observed to cause shifts of up to a pixel (0.1 Å), which for a strong line such as O V 629 Å corresponds to a velocity of 56 km/s.

In conclusion, NIS does not have an accurate `standard' wavelength calibration, and therefore, in the absence of any reference wavelength scale, most absolute wavelength measurements are subject to various uncertainties. Also, the NIS shows a series of geometrical effects that are not well understood and are difficult to explain, and can only be partially corrected for.

In practice, a simple way to extract an array of wavelengths `w' for each NIS window is:

; get the array w  of wavelengths for the first (0) window,
; corresponding to the first spatial position (xix=0,yix=0)

dummy=gt_spectrum (qlds,WINDOW=0,xix=0,yix=0,lambda= w)

;once obtained, it might be useful to plot an averaged spectrum:

; average over the Solar X,Y:

sp=average(average(data, 2, missing=-100),2, missing=-100) 

window,0 & plot, w,sp,psym=10

3.8.1  Wavelength shift with scan mirror position

For details, see:

http://orpheus.nascom.nasa.gov/~thompson/slit_inten/mirror_position.html

Figure 12: [Figure from Bill Thompson] The wavelength shift (in pixels) seen in the Mg IX 368 A line (diamonds) as a function of scan mirror position. The solid line is a polynomial fit to the SYNOP_F data. Also shown as a dashed line is the same results from Fe XVI 361 A.

Figure 13: [Figure from Bill Thompson] The wavelength shift (in pixels) derived from the NIS-2 lines He I 584 A (pluses) and O V 630 A (diamonds). The solid line is a polynomial fit to the average of the two lines.

The pixel shifts derived by Bill Thompson from the above data are:

DPIX = POLY(MIR_POS-128, [0,-0.0030423011,-3.3262600e-5,-2.0959131e-7]) ;NIS-1
DPIX = POLY(MIR_POS-128, [0,-0.0018515786, 2.3390659e-5,-4.0578093e-8]) ;NIS-2

3.8.2  Wavelength shift with temperature

An early demonstration of the effect of CDS optical bench temperature on the NIS wavelength zeropoint. The temperature varies as a result of different illumination conditions caused by changing the CDS pointing. These results are taken from analysis of the OV 630A line in the daily synoptic scans. The dots show the recorded optical bench temperature and the open circles the pixel centroid of the OV line. Prior to 12-Jun-96 a reasonable correlation can be seen. On that date the stabilisation temperature was raised to 23 Deg. C ie it was held above the peak of its natural level. Since then the temperature and the wavelength zeropoint have been very stable, although some analysis hints that a small temperature dependence still remains. See SYNOP_STAB_DEMO. The sharp drop in temperature around 28-Jun was caused by closure of the CDS doors for station keeping manouvers.

Figure 14: [Figure from CDS user guide]