SETI searching habitable zone KOIs

03 Feb 2011
Posted by colby

Kepler Objects of Interest, or KOIs, are planets. Well, maybe they are planets. Each KOI references a star and the fact that a planet may be in orbit around that star. As part of a data release to the public on 01 Feb 2011, the Kepler science team published over 1200 KOIs, 54 appear to be planets that are within the habitable zone of their respective star, i.e. an orbit that could support the lukewarm temperatures for liquid water, and therefore of immediate interest to anyone Searching for Extra-Terrestrial Intelligence.

Whenever I speak to a stranger about SETI science programs, I usually spend some conversational moments trying to figure out exactly what he or she thinks about SETI. The majority of the time, SETI is lumped in with all sorts of non-scientific lore, from UFOs to alien abductions, and usually in a good humored way. Rarely do I find someone referring to SETI as a science.

While working closely with researchers from the SETI Institute, I've come to an understanding of their endeavors as a full-fledged scientific subject. The work we do together has lead to the exquisitely designed Allen Telescope Array. I think it's important to publicize the sophisticated and scientific work put into the Allen Array.

The SETI Institute is directly involved with the Kepler project and as such was able to get a list of the 54 KOIs appearing in habitable zone orbits earlier than the general public, and we have been doing regular observations of 50 stars for the last few weeks. From the summary of results: out of 126,530 potential signals detected, those were whittled down quickly to 5 pulsed candidates, which were then further refined down to 2. Narrow band, pulsing emissions are surmised to be a hallmark of intelligent life and are not produced by natural phenomena, as pulsars produce wide band emissions. Repeated observations of those 2 stars did not show a repeat of the signals, leaving us with a tantalizing possible detection that is more likely to have been some Earth-local radio interference (either from the ground or satellites orbiting the Earth).

Unfortunately, the Allen Telescope Array may not be able to continue operations beyond mid-April of 2011 unless we receive a much larger infusion of funding. The ATA is an incredible instrument already: the collecting area of a 40m radio telescope, the resolving power of a 300m telescope, the ability to simultaneously scan 4 independent bands across 10GHz of the EM spectrum (other radio telescope designs depend on switching physical receivers to cover that much frequency space in terms of octaves) and a field of view that is up to 5 moon diameters across. If we can finish building the array (from the current 42 antennas to 350), it will have the collecting area of a 112m telescope (like the Green Bank Telescope or GBT), the resolving power of a 2km wide telescope and receivers that operate across 15GHz of bandwidth, retaining a field of view at up to 5 moon diameters across (almost 350 times bigger than the GBT and 17 times bigger than the VLA) and with receiver performance matching that of the best radio receivers in the world.

Let me put this another way via shiny pictures!

Here are two images of Cassiopeia A, the super-nova remnant of a star that exploded about 300 years ago.
The image on the left was created using data acquired by at the Ryle Telescope (previously known as the 5-km Telesscope) in Cambridge, UK and the Very Large Array (VLA) in New Mexico, USA. The image is composed of 3 frequencies, 1.4GHz, 5Ghz and 8.4GHz. The total telescope time used was over 700 hours over the course of 16 years, during which the VLA antennas were physically moved into different configurations on the ground 30 times.

The efforts required to create the image on the left were not because the technology was limited. It physically takes that much time and effort to collect the data using a radio interferometer with the resources of the VLA (27 antennas, each 25 meters) and the Ryle (5 antennas also 25 meters).

The image on the right is a simulation of ATA-350 imagery. It is based on the original data used to create the image on the left. That is, it's a simulation firmly grounded in reality and shows exactly what you could see if the ATA were built out to 350 antennas. Instead of over 700 hours of telescope time and 16 years of moving antennas around on the ground, the data on the right could be captured in 7 minutes at just one of the frequencies used in the image on the left (1.4GHz). The primary draw back is that the ATA image would not have as much resolution as the VLA image.

The reasons behind the faster performance of an ATA-350 are its field of view or FOV and the number of total antennas.

Field of view can be computed as FOV = λ/D, where λ is the wavelength of the light being observed and D is the diameter of the antenna. e.g.

FOVvla 1400/25 -> 56
FOVata 1400/6 -> 233

The units in this case are not important as we're comparing the relative field of view for each antenna design. By taking these unitless numbers and plugging them into the equation for computing the area of a circle, a 6 meter dish the size of an ATA antenna has over 17 times the field of view on the sky as an individual VLA antenna. This means the ATA, when all of its antennas are combined, retain most of that larger field of view, capturing a larger portion of the sky all at once. To create the image on the left, the VLA required hundreds of separate pointings to store a mosaic of data that were later combined to produce one single image, not unlike stitching together individual camera shots for a panoramic photo.

An important limiting factor of radio interferometers is how all of the signals from the antennas are combined together. Each signal from an antenna is combined with the signal from all other antennas. The total number of combinations can be represented as n(n-1)/2 where n is the number of antennas. These combinations are referred to as baselines, so:

Baselinesvla 27(27-1)/2 -> 351
Baselinesata 350(350-1)/2 -> 61075

This is the primary reason why the ATA with 350 antennas could be used to create the image on the the right with 7 minutes of data taking rather than hundreds of hours. It is also the primary reason why the ATA-350 will be an incredible survey instrument, doing both SETI science and radio astronomy science.

The full cost of building out will end up being 1/3rd of what it cost to build the GBT and 1/4th of the cost of the VLA (in adjusted dollars from 1972 to 2010, $78 million -> $413 million). It will be a huge loss if we let the ATA wither and die on the vine, not just for SETI science but for Astronomy as a long term human endeavor.