Transit of Venus

Twice (or rarely, three times) every 120-odd years, Venus passes across the solar disk as seen from the Earth. Such an event occurred in June 2004, providing the first opportunity to observe the transit of a planet with an atmosphere across the Sun since the advent of quantitative spectroscopy. This was thus an opportunity to demonstrate in a graphic way some of the techniques that we would like to appy to much more distant planets, and in addition, it offered the possibility of learning something new about Venus's atmosphere. Accordingly, we undertook observations of the June 2004 transit, using the Vacuum Tower Telescope (VTT) of the Kiepenheuer Institute for Solar Physics (KIS) on Tenerife.

Venus's atmosphere consists mostly of CO2, which has strong vibration-rotation absorption bands in the near IR; we chose to observe at a small number of wavelengths near 1.5 ¡m, where in a short wavelength range one can find lines that become opaque to tangential rays at all heights between the cloud tops (at about 65 km above the solid surface) and the mesopause near 110 km.

A great advantage of the Sun's proximity is that it can be spatially resolved, so that one can ignore the vast majority of its light that does not pass near Venus's limb. The geometry of the transit observations is illustrated in Figure 2. We placed the slit of the KIS spectrograph at a fixed place on the Sun, and allowed Venus's motion to carry it across the slit. The pixel size and slit width were both about 0.4 arcsec, or roughly 80 km projected onto Venus's disk. We used the VTT adaptive optics system, which held the effective seeing width to about 1 arcsec throughout the five hours occupied by the transit. The spectral resolution was about R = X/SX = 200,000. A consequence of the high spectral resolution and the small (256 x 256 pixels) detector format was that only a small wavelength range could be observed at one time. Since there was some uncertainty in our estimations of CO2 line strengths, we took observations in four neighboring wavelength regions with varying combinations of line strengths.

At any instant we recorded a spectrum such as that shown in Figure 3. At either end of the slit one sees only the solar spectrum, modified by absorption by CO2 in the Earth's atmosphere. Where Venus's disk crosses the slit, the observed intensity is much smaller; it is not essentially zero only because of scattered light in the Earth's atmosphere and in the telescope and spectrograph optics. The excitement occurs where Venus's limb crosses

Figure 2. Slit-jaw image of Venus on the solar disk as observed at the KIS vacuum tower telescope on Tenerife, showing the vertical spectrograph slit nearly tangent to Venus's disk. The prominent circular boundary on the right-hand side is the edge of the adaptive-optics aperture, while the limb of the much larger solar disk cuts diagonally through the top of the image.

Figure 2. Slit-jaw image of Venus on the solar disk as observed at the KIS vacuum tower telescope on Tenerife, showing the vertical spectrograph slit nearly tangent to Venus's disk. The prominent circular boundary on the right-hand side is the edge of the adaptive-optics aperture, while the limb of the much larger solar disk cuts diagonally through the top of the image.

Figure 3. (Left panel) Typical spectrum of the Sun and Venus, with distance along the slit x increasing from left to right, and wavelength A increasing from bottom to top. (Right panel) The difference between the spectrum at left and a simple model of it as the product of a function of x and a function of A, as described in the text.

Figure 3. (Left panel) Typical spectrum of the Sun and Venus, with distance along the slit x increasing from left to right, and wavelength A increasing from bottom to top. (Right panel) The difference between the spectrum at left and a simple model of it as the product of a function of x and a function of A, as described in the text.

the spectrograph slit—here a portion of the observed light has passed through Venus's atmosphere as well as our own. By comparing the spectra from these regions to those of unadulterated sunlight, one can infer Venus's contribution to the absorption.

Venus Limb Spectrum

1.08

1.07

o cc

1.03

1.5976 1.5978 1.5980 1.5982

Wavelength (micron)

Figure 4. Transmission spectrum of Venus's atmosphere, averaged over a few hundred spectra taken at different latitudes. The dot-dashed line shows the absorption spectrum of the Earth's atmosphere from Livingston & Wallace (1991), on an arbitrary vertical scale. Lines resulting from 12CO2 and 13CO2 are identified.

The spectrum in the left panel of Figure 3 shows two strong spectrum lines; both are CO2 lines, but the lower one is underlain by an atomic line originating in the solar photosphere. The right panel of the figure shows a difference spectrum, constructed from the left-hand image as follows: We created a model of the observed spectrum by forming averages along both A and x, and then forming the two-dimensional product of these two one-dimensional marginal functions. We then subtracted a suitably scaled version of this model image from the observed one; the right-hand panel displays this difference.

The difference image reveals a fringe structure that originates in the instrument; at present our ability to retrieve scientific information from these images is limited by our ability to model this fringing successfully. The purely telluric absorption line almost disappears from the difference image, showing that it is accurately represented as the product of an average line profile and the average intensity along the slit. The solar line shows significant residuals, however, arising from local variations in velocity and intensity in the solar photosphere. These are caused mostly by the solar granulation. Finally, near Venus's limb, one can see localized dark spots (two are indicated by arrows). These are the signature of absorption by Venus's atmosphere.

We have constructed an average spectrum of the Venusian absorption by averaging together several hundred individual spectra like that shown in Figure 3, with due allowance for the changing position of Venus's limbs on different spectra. The result for one of our four wavelength bands (a different one than illustrated in Figure 3) is shown in Figure 4. Also shown in the figure is a spectrum of the telluric absorption as seen at one airmass, taken from the high-resolution IR spectral atlas by Livingston & Wallace (1991).

As indicated in the figure, the observed lines can be identified with two different isotopomers, namely 12CO2 and 13CO2. A curious feature of the spectrum is that the strengths of the 13 C lines are similar to or greater than those of the 12 C lines, even though 12 C is by far the more abundant isotope. In fact, some of the 13CO2 lines in the Venusian spectrum are significantly stronger, relative to the 12CO2 lines, than in the telluric spectrum. This behavior occurs because the lower levels of the two sets of lines are dissimilar in energy: for the 13CO2 lines the lower level lies 200 cm-1 above the ground energy, while for the 12CO2 lines it is more than 700 cm-1 higher. At the cold

Venus Limb Spectrum

1.07

o cc

Figure 4. Transmission spectrum of Venus's atmosphere, averaged over a few hundred spectra taken at different latitudes. The dot-dashed line shows the absorption spectrum of the Earth's atmosphere from Livingston & Wallace (1991), on an arbitrary vertical scale. Lines resulting from 12CO2 and 13CO2 are identified.

temperatures prevailing near Venus's mesopause (about 200 K), the pile-up of molecules in lower-lying states is enough to compensate for the low relative abundance of 13CO2.

Work is continuing to resolve issues related to the detector fringing, in hopes of measuring accurate Doppler shifts of the lines seen in Figure 4. If this can be done, it will be possible to measure the cross-terminator wind speed as a function of Venusian latitude and (by measuring lines of differing strength) height above the cloud deck. Such measurements could be very useful for understanding the way in which Venus's atmosphere makes the transition from super-rotation (near the top of the cloud deck) to the dayside-to-nightside flow that dominates at much higher altitudes.

Was this article helpful?

0 0

Post a comment