Targeted Star Search

Selecting Target Stars

Life as we know it developed on a planet orbiting a G2 V star, the Sun. The cryptic “G2 V” designation is the Sun’s “spectral type.” Based on a star’s spectrum, astronomers group stars by temperature. From hottest to coolest, the spectral classes are O, B, A, F, G, K, and M. Each class is subdivided into ten and numbered from 0 to 9. The Sun is a G2 star. The “V” is the Roman numeral for five and designates the Sun’s luminosity class. Stars in luminosity classes I, II, and III are “giant” stars, very luminous and nearing the end of their life as a star. Class IV stars are “sub-giant” stars that are just entering “old age” and as the name implies, large but not giant. Stars in luminosity class V, like the Sun, burn only hydrogen in their cores and are relatively stable.

It is generally agreed that stars with spectral types from about F5 through K5 may be suitable hosts for habitable planets. Some recent studies indicate the some cooler stars, perhaps to spectral type M4, also may host habitable planets.

The HabCat Catalogs

In 2003, Margaret Turnbull and Jill Tarter published two lists of selected stars. The Nearby Habitable Systems (HabCat1) was created from the Hipparcos Catalogue by examining the information on distances, stellar variability, multiplicity, kinematics, and spectral classification for the 118,218 stars contained therein. They also made use of information from several other catalogs containing data for Hipparcos stars on X-ray luminosity, Ca II H and K activity, rotation, spectral types, kinematics, metallicity,

and Stromgren photometry. Combined with theoretical studies on habitable zones, evolutionary tracks, and third-body orbital stability, these data were used to remove unsuitable stars from HabCat, leaving a residue of stars that, to the best of our current knowledge, are potentially habitable hosts for complex life. The resulting HabCat1 catalog contains 17,133 well-selected “habstars”.

Since we need about one million target stars to fully utilize the capability of the ATA, a second catalog of stars was derived from the Tycho-2 Catalogue of 2.5 million stars. Unlike the Hipparcos stars, the Tycho stars did not have distance measurements. The approximately 250,000 stars of HabCat2 were selected primarily by their colors (brightness in blue and “visual” filters) and proper motion (motion across the sky).

You can download the HabCat1 list below. Here’s a peek at the first few entries and an explanation of the columns..













































B+19 5185
























B- 4 6001













HIP – Hipparcos Catalogue number

RA – Right Ascension (J2000) in hours, minutes, and seconds Dec – Declination (J2000) in degrees, minutes, and seconds v_mag – visual magnitude (apparent brightness) a star with v_mag less than 6 is visible to the eye with a dark sky parllx – parallax angle (in milliarcseconds) of motion of star due to Earth’s orbit around the Sun Distance in parsecs = 1000/parallax Distance in light years = 3262/parallax

B-V – Color Index, the difference in brightness measured in special Blue and Visual filters

HD – Henry Draper Catalogue number

BD – Bonner Durchmusterung Catalogue number

The HabCat1 list is in tab delimited text format.


SETI on the ATA: Galactic Center Survey

The region near the center of our galaxy contains the highest concentration of stars in the sky, yet has only been searched for ETI signals at a few “magic” frequencies. We will search a large area near the galactic center over the entire “waterhole” frequency range, from 1420 to 1720 MHz. This frequency range is defined by the spectral lines for the dissociation products of water, the sine qua non for life as we know it. This band also includes the minimum background noise due to synchrotron emission from the galaxy and the cosmic microwave background radiation. It is the “cosmic quiet zone” where we can best listen for a faint whisper across the interstellar expanse.

We will search for narrowband signals from ET in a 2 degree by 10 degree area including, but not centered on, the center of our galaxy. (For scale: the Moon is half a degree in diameter, so it would take “80 Moons” to cover the survey area.) This region near the galactic center contains perhaps 40 billion stars within a distance of about 25,000 light years of the Earth.

The survey area will be covered by pointing the array at 80 positions and steering the phased array beams within the central one half degree square portion of the FOV. Figure 1 shows the FOV (at 1420 MHz) and two beams, observing two different positions at the same frequency at the same time to help exclude radio interference. The white squares show the half degree squares enclosing about half of the pointing positions in the survey. The background image is a radio picture taken at 300 MHz using the Very Large Array. It demonstrates the potential for commensal radio astronomy imaging at other frequencies with the ATA. A more familiar view of the Milky Way is shown in Figure 2.

figure 1 Figure 1. The white squares are 0.5 degree on a side and show half the area to be covered in the Galactic Center Search.. The yellow circle is the field of view of the array at 1420 MHz. The yellow ovals show the size of the phased array beams. The Moon is shown for scale. The background image shows radio emission from our galaxy. (Kassim, LaRosa, Lazio, and Hyman; ASP Conference Series, Vol. 186, 1999).figure 2 Figure 2. The red rectangle shows the approximate location of the search area in this optical image. The constellation of Sagittarius is visible in the evening sky in July. (Credit: W. Keel, U. Alabama, Tuscaloosa and Cerro Tololo, Chile.)

SETI Observations

The question “are we alone in the universe?” has fascinated people for centuries.  In 1959 two Cornell University physicists described how we might answer that question through interstellar communication.  Philip Morrison and Guiseppe Cocconi analyzed how two civilizations separated by many light years of space might use electromagnetic radiation to communicate.

Electromagnetic radiation is the general term for the more familiar light and radio waves, but also includes gamma rays, x rays, ultraviolet rays, and infrared.  All of these forms of electromagnetic rays or waves travel at the speed of light, the fastest possible speed.  The only difference between these types of electromagnetic waves is the length or the wave or the “wavelength.”  Since the waves all travel at the speed of light, another way to characterize the wave is by the number of waves received per second.  This is the frequency of the wave. 

Morrison and Cocconi considered how well the types of electromagnetic waves passed through interstellar space.  While space is usually described as a vacuum, it does contain some gas and dust.  Over great distances that gas and dust absorb many types of radiation but radio waves pass through nearly unimpeded.  For interstellar communication, a particular range of radio frequencies, “microwaves” from 1 GHz to 10 GHz, are particularly good choices.  At lower frequencies our galaxy emits prodigious amounts of radio waves creating a loud background of noise.  At higher frequencies the Earth’s atmosphere, and presumably the atmosphere of other Earth-like planets, absorbs and emits broad ranges of radio frequencies.  The result is a quiet “Microwave Window” through which efficient radio communication is possible.

figure 1

The Microwave Window has another interesting feature to recommend it as a place for interstellar communication: the “Water Hole.”  Some atoms and molecules in space emit radio waves at particular frequencies.  Hydrogen atoms emit at 1420 MHz (a wavelength of 21 cm).  Hydroxyl molecules, composed of one atom of hydrogen and one atom of oxygen (OH), emit at four specific radio frequencies ranging from 1612 MHz to 1720 MHz.  When a hydrogen atom combines with a hydroxyl molecule it forms a molecule of water, the most essential molecule for life as we know it.  Thus, the range of frequencies from 1420 to 1720 MHz is called the Water Hole.  It has been a popular frequency range for many SETI programs.

While the Microwave Window is a “cosmic quiet zone” the existence of the radio emission lines marking the Water Hole illustrates a problem: distinguishing communication signals from natural astrophysical emission.  Most cosmic radio sources as “very broadband.”  They “broadcast” over a very wide range of frequencies.  For example, the radio emission from a quasar would span most of the diagram above and the amount of power would change very slowly over the frequency range.  As mentioned above, hydrogen and hydroxyl emit radio waves at particular frequencies.  The diagram below shows the power emitted by a cloud of hydroxyl (OH) molecules near a red giant star.  The cloud rotates around the star so some of the molecules are moving toward us and some are moving away.  The Doppler Effect shifts the frequency of the radio waves to higher (moving toward us) and lower (moving away) frequencies.  The resulting spectrum (power vs. frequency) shows two peaks.  These features are very narrow compared to most astrophysical sources.

Signals produced by technology for communication can be much narrower than any known astrophysical source.  The bottom half of the diagram shows the spectrum of the signal from the Pioneer 10 spacecraft.  The frequency scale has been expanded by nearly a factor of 10.  The narrow signal component on the left is further expanded by a factor of 100 to show the very narrow frequency components of the data in the signal.  This illustrates a clear way to distinguish extraterrestrial communication signals for astrophysical signals: look for very narrow frequency components.  Most SETI programs take this approach, breaking the radio spectrum into many millions of very narrow frequency channels

figure 2

Where to Look

Searching for extraterrestrial signals can be likened to searching for a gold nugget buried in a field.  We have identified the field that we will explore: the Microwave Window in the radio spectrum.  We have a way to distinguish the gold nugget for other pieces of metal: look for very narrow bandwidth signals.  But, with limited resources (telescope time, electronics, etc.) what is the best strategy to search through the field? 

There are two basic strategies.  We could examine the entire field but only sift through the top few centimeters of soil.  If a gold nugget is near the surface anywhere in the field, we’ll have a good chance of finding it.  We could study the chemistry and geology of the surface and select particulars spots where we will dig deep looking for gold.  If the gold is buried well below the surface, this strategy has the best chance.

In SETI, there are two basic search strategies.  Sky surveys sweep a telescope over large areas of the sky looking for strong signals that may come from any direction.  Targeted searches point a telescope at the direction of selected stars.  By dwelling on a star for long periods of time, a targeted search is sensitive to weaker signals.

We will pursue both strategies with SETI on the ATA.

Technical Overview

I. Introduction

The completed Allen Telescope Array (ATA) is intended to consist of approximately 350 6.1 meter offset Gregorian dishes at the Hat Creek Radio Observatory site in northern California. Given the number of antennas and a very wide field-of-view (2.45° at 21cm wavelength), this array will have an unprecedented amount of flexibility in observing. Several individual users may simultaneously use the array to observe a different part of the sky at an independent frequency, or image the sky at one or more frequencies.

II. Overall Architecture

The ATA has four main conceptual systems:

(1) the antenna collects the radiation from space;
(2) the signal path brings the radiation from the feed (which is located at the antenna focus) back to the user;
(3) the monitor and command systems allow the dishes to be accurately moved, and the signal path controlled. It also monitors the health and status of the array and all of its components;
(4) the site includes the overall antenna configuration, as well as other infrastructure. The ATA will permit remote users to access and use the instrument via a secure Internet connection.

Figure 1 displays the conceptual framework of the ATA and introduces the overall architecture and terminology.

figure 1

Physically, the ATA consists of many elements (350 when fully built out), which are composed of an antenna and all of the associated mechanical and electrical systems to create the signal path and to monitor and command the array. This build-out number of antennas yields approximately one hectare (10,000 m2) of geometric collecting area. Each element has a pair of analog radio frequency (RF) modulated fiber optic cables for the signal path and additional digital fiber for the monitor and command back to the converter room. The monitor and command fiber is routed directly to the control room, while the analog fiber is processed in the converter and processor rooms before ultimately getting to the user either in the control room or elsewhere via a secure Internet link. In addition, there is an antenna assembly tent for the actual fabrication, and eventually plans for the Myhrvold Development Laboratory, and a visitor's center.

Note that the elements are units highly integrated in the physical structure of the antenna. All of the major components are present at each element. The appearance of the ATA will be dominated by the large number of such elements.

III. Antenna

The ATA antenna is a 6.1-meter (approximately 20 foot) offset Gregorian antenna, with a 2.4-meter (8 foot) secondary. These reflecting elements were hydroformed from aluminum in a proprietary process under a contract with Andersen Manufacturing, in Idaho Falls, ID. A metal shroud connects the bottom half of the sub-reflector to the bottom of the primary to deflect ground spillover. The inclusion of the shroud is important in achieving lower system temperatures.

IV. Signal Path

Figure 2 shows the signal path from sky to user. Recall that although just two data paths from one antenna are shown, there would be approximately 700 paths in all antennas for a built-out array. The analog fiber links carry both polarizations of the full 500 MHz - 10 GHz bandwidth back to the RF converter, where it is converted to up to four independent 1 GHz wide dual-polarization channels at L-band. It is then digitized and passed through to the processor.

figure 2

Each of the up-to-four independent tunings is further split into four synthesized beams, which may be steered independently to different parts of the sky. The outputs are sent to either a correlator (which takes all of the signals and makes an image) or one of several phased-array back-ends (PABE’s), or SETI detectors.

figure 3 and figure 4

V. Monitor and Command

Figure 5 shows a rough schematic of the monitor and command system, a.k.a. MonCom. Recall that although only one element is shown, there are in fact a large number of elements. Standard networking tools are used, making each antenna essentially an “internet appliance.”

figure 5 and figure 6

VI. Conclusion

The Allen Telescope Array is the first implementation of a large-N, small-D radio telescope. It is a promising candidate for at least a portion of the Square Kilometer Array (SKA).

The ATA is a scalable architecture. It began operation in October 2007 with 42 elements. The rapid increase in computing capability (Moore’s Law) will drive ever-increasing functionality and performance in the digital hardware as the array is built out.

A Phased Array

A parabolic antenna brings radio waves from a particular direction in space (along the axis of the parabola) to a focus as in Figure 1. The radio waves from the direction along the axis arrive at the focus “in phase,” i.e., they have all traveled exactly the same distance. The radio waves are collected at the focus by a “feed” antenna where they are converted to electrical signals and amplified. These signals are then sent to other electronics for further processing.

A dish antenna is most sensitive to radio waves from a narrow range of directions near its axis. This angular region of sensitivity, corresponding to a small spot on the sky, is called the antenna’s beam. The width of the beam depends on the diameter of the antenna and the wavelength of the radio waves observed. The beam width (in radians) is roughly the wavelength of the radio wave divided by the diameter of the antenna. For example, a 100 meter antenna observing at a wavelength of 21 centimeters (a frequency of 1420 MHz) has a beam width of 0.0025 radians or 0.14 degrees (about one third the angular size of the Moon).

figure 1

The ATA is an array of many small dishes that can be combined to form the equivalent of a single large dish antenna. How this works is illustrated in the following diagrams based on originals by Dr. Ron Ekers.

We can replace the solid surface of a large dish by a set of small dishes positioned along the parabolic surface as in Figure 2. Each small dish brings radio waves to a focus and converts them to electrical signals. These signals then travel by cables to the focus of the large dish. This replicates the effect of the large single dish. The cables simply replicate the path that the reflected radio waves would have taken in Figure 1.

figure 2
figure 2

But there is no reason to send the cables up to the original focus position. As long as the lengths of the cables remain the same, they can instead go down to the ground and be collected by electronics there, as in Figure 3.

figure 3
figure 3

There is also no reason to suspend the small dishes above the ground. As long as we know where they would have been on the surface of the original large dish, we can put them on the ground and add an appropriate amount of cable to delay the signals as shown in Figure 4. The dish that is closest to the original position (the dotted line) will need the most delay. Combining the signals collected by the individual dishes with the appropriate delays synthesizes the effect of the large single dish, resulting in a synthesized beam equivalent to that of the large dish.

figure 4
figure 4

This arrangement of small dishes is called a phased array since the cables and electronics ensure that the radio waves from a particular direction are “in phase.” By carefully changing the lengths of the cables and tweaking the electronics, we could bring radio waves from a different direction into phase. In other words, we could steer the phased array without moving the dishes. If we replicate the electronics and cables, we can point in two directions at the same time as shown in Figure 5.

figure 5
figure 5

In practice, we use one set of cables with separate sets of electronics to change the relative delays between the dishes, and thereby synthesize multiple beams. The synthesized beams may be steered anywhere on the sky, but we usually limit them to an area corresponding to the beam width of an individual dish. (The sensitivity of the synthesized beam decreases dramatically outside of this area.) This area is called the field of view (FOV) of the array. For the ATA, with 6.1 meter dishes, the FOV at a wavelength of 21 cm is 0.04 radians or 2.5 degrees (five times the angular width of the Moon).

SETI observations use three beams to observe three stars (or positions on a sky survey grid) at the same frequency at the same time in order to eliminate signals caused by terrestrial interference. If a signal appears in two or more beams, it is almost certainly from our own equipment or from manmade interference. If a signal appears in only one beam, it is worthy of further scrutiny.

General Overview

The Allen Telescope Array is a response to one of the most enticing sirens to beckon the SETI community: a major telescope that can be dedicated to the search. Despite the seductiveness of this idea, construction of an instrument designed to meet the requirements of full-time SETI has always foundered on the large costs.

That situation has changed. Thanks to the far-sighted benevolence of many donors, including technologistsPaul Allen (co-founder of Microsoft) and Nathan Myhrvold (former Chief Technology Officer for Microsoft), the first 42 elements of the ATA are conducting SETI searches every day of the week.

ata location The instrument, called the Allen Telescope Array, is situated at the Hat Creek Observatory, located in the Cascade Mountains just north of Lassen Peak, in California.

Radio SETI experiments have historically relied on existing radio astronomy telescopes. While this allows such searches to be conducted on quite large instruments (for example, the 305 m Arecibo dish, in Puerto Rico), the amount of telescope time available for the search is necessarily restricted. Project Phoenix, for example, took control of the Arecibo telescope for approximately three weeks in the spring and a similar block of time in the fall. Since observations were only made at night, this amounted to a total of only three weeks of full-time observing annually.

During the period from September 1998 through March 2004, Project Phoenix observed for a total of 100 days at Arecibo. That’s only 5% of the available time. The Allen Telescope Array offers SETI scientists access to an instrument seven days a week, and permits the search of several different targets (primarily exoplanet systems) simultaneously. As a result, the Allen Telescope Array is speeding up SETI targeted searching by a factor of at least 100.

Because of its ability to study many areas on the sky at once, and with greatly improved access to a telescope, the ATA will allow an expansion from Project Phoenix’s stellar reconnaissance of 1,000 stars to a million or more nearby stars over the course of the next two decades.

The fundamental idea behind the Allen Telescope Array was generated during a series of workshops held in 1997 - 1999 in which a group of scientists, engineers, and technologists considered how best to pursue SETI in the coming two decades (the SETI Science and Technology Workshops). The favored scheme was an array of relatively small dishes (antennas), with a pseudo-random arrangement on the ground extending over about 1 km. This provides a very high quality beam shape (the spot in the sky to which the telescope is most sensitive), and thanks to the relatively large number of antennas, also minimizes (unwanted) sensitivity outside the primary beam.

ata project phoenix compared The Allen Telescope Array is optimized to cover frequencies between 500 and 10,000 MHz, which is more than five times the range searched in Project Phoenix. Thanks to a generous donation from Franklin Antonio (co-founder and Chief Scientist at Qualcomm) the ATA receivers are being replaced with new systems that will cover 1,000 to 15,000 MHz, and offer improved sensitivity and greater reliability.

Software for the search, dubbed SonATA (SETI on the ATA) is a software-only search system implemented with enterprise servers donated by Dell and Intel. Detailed descriptions of the SonATA system that began operation in 2010 can be found here.

By building the new telescope as an array, several major advantages can be realized. To begin with, many “pixels” can be generated on the sky at once. Rather than looking at only one star at a time, as the Arecibo telescope and its kin are constrained to do, several stars can be examined simultaneously. This again speeds up the process of stellar reconnaissance. In some cases, it may be desirable to sacrifice spectral resolution (typically 1 Hz) in order to gain additional pixels. In other words, one can trade amount of sky covered for sensitivity to very narrow-band signals. Depending on the type of signal we expect, this might be a judicious trade-off.

In addition, it is easy to expand an array by merely buying additional antennas and connecting them into the system. Single, large dish antennas are not amenable to such simple improvement. The bottom line is compelling. Because of its ability to study many areas on the sky at once, with more channels and every day of the week, the Allen Telescope Array will be able to check out a truly significant sample of the cosmic haystack. This is not an incremental step forward: the Allen Telescope Array will increase the stellar reconnaissance by orders of magnitude. It is a very large step for SETI research.

The design for the Allen Telescope Array’s antennas – so clearly different from the type of “tripod” arrangement found in many backyard satellite dishes – incorporates so-called offset optics because sometimes, as in football, going to the side can reduce interference. The antennas use what’s known as an offset Gregorian system. A secondary mirror bounces incoming radio signals collected by the large (6.1 meter diameter) primary reflector back to the feed antenna (hidden from view by a fabric shroud) where they are amplified and sent on their way to the control buildings.

By introducing a secondary mirror and a surrounding shroud, the antenna is less likely to pick up noisy radiation from the (relatively hot) ground surrounding the telescope. Moving the reflector assembly off-center minimizes the chance of terrestrial signals bouncing off the antenna structures and interfering with our study of cosmic emissions. This offset design has also been used for the 100 m Robert C. Byrd Telescope, which is in operation in West Virginia.

ATA: How It Works

ata antenna

The ATA is part of a new trend in radio astronomy. Rather than a single large dish, it is an array of a large number of small dishes (LNSD). Other countries (Australia, South Africa) are building LNSD arrays and an international consortium is planning a Square Kilometer Array that will use this new approach to radio telescope design.

Large single dishes are expensive, one-of-a kind development projects with every aspect optimized to get the best result for that investment. Once built, improving their performance is difficult and expensive. The ATA represents an entirely different approach, for it uses commercial technology wherever possible. The dish components are manufactured through a process developed for the commercial television market, and are relatively inexpensive.  The sensitivity of the array is easily increased by simply adding more dishes, an approach that is clearly impractical for large, single-dish antennas. The ATA also takes advantage of receiver and cryogenic technologies originally developed for radio communication and cell phones. The instrument employs programmable chips and software for signal processing, which allows an increase in capability as new computer technology becomes available.

Over time, and with sufficient funding, the ATA will grow to 350 dishes. It will then have a collecting area equivalent to a single dish 114 meters in diameter, and the angular resolution of a dish 700 meters across. The ATA-350 will have point source sensitivity comparable to the National Radio Astronomy Observatory’s Robert C. Byrd Telescope and the Very Large Array, while being far faster and superior for imaging surveys. 

The Allen Telescope Array

The Allen Telescope Array (ATA) is a “Large Number of Small Dishes” (LNSD) array designed to be highly effective for simultaneous surveys undertaken for SETI projects (Search for Extraterrestrial Intelligence) at centimeter wavelengths.

The initiative for building the ATA emerged from a series of workshops convened by the SETI Institute beginning in 1997.  These workshops were charged with defining a path for future development of SETI technologies and search strategies. The relentless advance of computer and communications technologies made it clear that LNSD arrays were more efficient and less expensive than the large antennas traditionally constructed for radio astronomy and SETI. The final report of the workshop, “SETI 2020,” recommended the construction of a so-called One Hectare Telescope, having a collecting area commensurate with its name.  

The SETI Institute sought private funds for such an instrument, and in 2001 Paul Allen (co-founder of Microsoft) agreed to fund the technology development and first phase of implementation, culminating in the construction of 42 antennas. In October 2007 the array began commissioning tests and initial observations. 

Detail level page

This is a project detail page.


Subscribe to SETI Institute RSS