Last December, we looked at how monitoring the phases of an extrasolar planet can permit a crude longitudinally-resolved map of an extrasolar planet in a tight synchronous orbit of its star, (here). Specifically, we focoused on the well-studied transiting hot Jupiter HD 189733 b. This map was what you would call a one-dimensional map of the planet because it was only longitudinally resolved. Assumptions about atmospheric circulation were put into it to make it appear realistic, latitudinally, because it looks more representative of what is probably the case. In the case of the 8µm map for HD 189733 b, the polar regions of a planet are not likely to be as warm as the equator for example, so these modifications are probably not too far off. But for the sake of illustration, let’s take a comparative look at a one-dimensional and two-dimensional map of a planet that we’re a bit more familiar with.
The production of a longitudinally resolved, one-dimensional map for an extrasolar planet does not require that planet to be tidally locked. Observing the brightness of a planet over an entire rotation period will also permit such a map to be made. Above is an example of what such a map would produce for Earth if it were observed like an extrasolar planet might be. In the one-dimensional, we are permitted only longitudinal information about surface features. We can tell there are two major land masses, which are of course the Americas and the Eurasian+African continents. Other than that, we can’t tell much else. Adding latitudinal information lets us distinguish between North and South America, for example.
It is possible to extract latitudinal information about the distribution of surface features on an extrasolar planet if the planet undergoes secondary eclipses and has a non-zero impact parameter through very careful monitoring of the ingress and egress of the eclipse, as described by Rauscher et al (2006). It’s really rather clever. The limb of the eclipsing star is in an opposite orientation for the egress and ingress. Surface features on the planet that contribute to the light curve may therefore by eclipsed asymmetrically.
In the above diagram, a planet with a surface feature, say a cloud system for example, passes behind its eclipsing host star. During the ingress, the cloud system is eclipsed about half way through the ingress due to the geometry of the stellar limb and the surface of the planet. During the egress, the cloud system is one of the first features to emerge. An “ingress map” can be made charting where on the surface the features contributing to the light curve may have been during ingress. Clearly this would not be generally known, except for the constraint that a surface feature must have been somewhere on the planet along a line curved to the stellar limb. The same can be done for an “egress map.” Combing the two can permit one to form a two-dimensional map of the eclipsed planet’s sun-facing hemisphere.
Looking through the daily postings on arXiv the other day, I was delighted to find that Majeau et al have done just that, taking Spitzer 8µm data of HD 189733 b to produce a new map of the planet, this time, one in two dimensions. The title of their paper says it all, this is the first two dimensional map of an extrasolar planet, and a huge congratulations to them for the achievement.
The hottest spot on the planet is again found to be displaced from the sub-stellar point by a few degrees in longitude, in agreement with previous work. They also find that the hottest point is on or very near the equator, which isn’t exactly surprising, but it does let us confidently rule out any significant axial tilt for the planet’s rotation.