Earth Atmosphere Observatory at L2
What is the best mission design for a virtual-structure telescope to observe Earth's atmosphere from L2?
The Sun-Earth L2 point is a location from which a spacecraft would see Earth permanently eclipse the Sun, leaving only a thin ring of sunlight (called the solar annulus) surrounding the planet. It's a relatively stable position, thanks to the combined gravitational effects of the Sun and Earth. As Earth orbits the Sun, a spacecraft at L2 would remain about 1.5 million km (about one million miles) above Earth's nightside.
This configuration makes it a uniquely desirable place from which to study long-term changes in Earth's atmosphere. Spectrometers stationed at L2 would enable scientists to analyze the atmosphere by observing its effect on the sunlight shining through it.
Our task was to develop a concept for a space telescope mission to make a detailed study, from the L2 vantage point, of the atmosphere's constituents and dynamics over a span of 10 years. The hypothetical launch date was set at 2025 to 2030, to allow time for needed technology improvements.
The observatory would monitor the forcing and response of Earth's atmosphere. "Forcing" refers to the various factors that influence the atmosphere, both natural (such as volcano eruptions) and artificial (such as emissions from cars and industrial processes, deforestation, and replacement of natural ground cover with surfaces that reflect more sunlight).
The mission would reveal how the atmosphere varies at different altitudes and locations, and how it changes over time. Airborne chemicals would be sampled sequentially, via their spectroscopic signatures, to build up a profile. The data gathered would enable development of a model on which to base predictions of climate changes over both short and long terms.
The telescope would need to be capable of high-resolution observation of light wavelengths from 0.28 to 10.5 microns. This covers the range from ultraviolet through visible light to the near-infrared.
As Earth rotates on its axis "beneath" the observatory, new portions of the atmosphere would continually be brought to the solar annulus, where they could be analyzed by the telescope system. We determined that the telescope should sweep around the ring of sunlight once every 12 hours, enabling it to sample 1� slices of the atmosphere twice per day.
While L2-based observation enjoys unique advantages, it comes with a peculiar constraint resulting from the fact that Earth has a large moon that perturbs its orbit around the Sun. We commonly regard the Moon as orbiting the Earth, but the fact is that both bodies orbit their common center of mass, a point called the Earth-Moon barycenter.
L2 is actually located not quite on a straight line extending from the Sun to Earth, but rather on a line from the Sun to the Earth-Moon barycenter. So, as seen from L2, Earth and the Moon each trace an oscillating pattern as they orbit the Sun. Keeping Earth centered on the Sun so that the planet is constantly surrounded by a ring of sunlight requires that the telescope must move with the Earth as they travel around the Sun. Thus, the telescope must fly in a continuous powered elliptical orbit around the L2 point.
This need for a powered orbit places a serious constraint on the telescope's mass, especially since this mission is intended to last for ten years without refueling the thrusters. Excessive structure mass would require more thruster fuel, which itself would add mass and require even higher thrust levels with higher power to drive the Xenon ion electric propulsion system. The observatory design therefore evolved as a unique mass, thrust acceleration, and drive-power optimization problem.
Knowing the restriction on the telescope's mass, we designed it from the outset as a "virtual-structure" system of two spacecraft flying in tight formation. The Aperture Spacecraft contains the very large primary mirror. The Science Spacecraft contains the rest of the telescope. They would behave as though they were connected by a rigid structure through the use of very-high-precision formation flying, coordinated to within 1 cm. Both spacecraft would have sensors to help keep them in the proper orbit around L2, as well as the metrology they would need to fly in synch with each other.
The two spacecraft would use electric thrusters powered by radioisotopes. Photovoltaic cells would not be practical, since only 10-13% of the Sun would be visible around the occulting Earth. A photovoltaic array big enough to provide sufficient power would be flimsy and a source of vibration. On the science spacecraft, the array would obscure the mirror's view of Earth. On the aperture spacecraft, it would interfere with the thrusters.
Though the powered-orbit requirement limited the allowable mass, another constraint argued for a very large size. The required resolution (1 km at Earth over a broadband spectrum -- i.e., a diffraction limit of 67 microradians at 10.5 microns) at that distance dictated that our telescope would need a primary mirror with an unprecedented diameter of 19 meters. (For comparison, the James Webb Space Telescope, scheduled for launch in 2011, will have a mirror with a 6.5-meter diameter.)
A traditional mirror of that size would have an unwieldy amount of mass, and would be extremely difficult to build and maintain with the necessary optical precision.
We therefore designed the mirror as a membrane supported by an inflatable ring. Piezoelectric control elements embedded in the membrane would adjust the mirror's shape, based on feedback from optical sensors aboard the Science Spacecraft. Foreseeing the likelihood of distortion in the perimeter of the membrane where it attaches to the inner and outer rings, we included a margin of error and called for the membrane to have a total diameter of 25 meters.
In our initial design, we made the mirror quite flat, with a focal length of f/10 to minimize aberrations. It would have projected an image coming into focus some 250 meters away from the mirror, where the telescope's entrance aperture would have to be positioned. The focal image (all of which would need to be contained within the telescope) would have a diameter of 2.5 meters.
A telescope of these dimensions would have been far too large and massive to be practical, even with the technological improvements that might be expected by our hypothetical launch date of 2025 to 2030.
We solved the problem by increasing the mirror's curvature and thus shortening its focal length and reducing the focal image. The trade-off was a reduction in the quality of the image (more prone to spherical aberration and distortion, and less depth-of-field tolerance) and the need for more complexity in the optics design, but it was a necessary compromise. We found that the optimal balancing point was at f/5, halving both the focal length and the diameter of the telescope aboard the science spacecraft, and reducing the spacecraft's mass to just 1/10th of what it would have been at f/10.
Rotating the Telescope
The focal image would be a ring of sunlight shining through Earth's atmosphere. As mentioned, our plan was to make a complete scan around the ring twice per day, in 1� increments.
We opted to rotate the entire telescope rather than just the entrance aperture, so all the optics would remain aligned as they sample the ring sections. However, this solution to the sampling problem posed its own problems with regard to navigation of the spacecraft.
First, if the entire spacecraft were set rotating, its thrusters would rotate with it, adding enormously to the difficulty of keeping the science spacecraft flying in formation with the primary mirror, and also of flying the required powered orbit around L2. We solved this problem by keeping the science spacecraft's bus (where the thrusters are mounted) stationary. A motor in the bus would rotate the telescope section.
The second problem stems from the fact that a rotating telescope would behave like a gyroscope, resisting the movement of the thrusters. And as a prolate body (long and narrow) revolving about its long axis, it would try to reorient itself 90�, to rotate about its shorter axis. To avoid these problems, we designed a counter-rotating wheel into the system, which would bring the net angular momentum to zero.
Accomplishments and Challenges
In addition to developing the advanced spectrometry science and engineering requirements, and the overall architecture of the telescope and its operation, we designed an optimal method of transport from Earth and deployment at L2, and developed preliminary designs for guidance, navigation, and control.
The greatest challenge before us is how to make such a large mirror light enough (the target areal density is 1 kg/sq. meter), and yet capable of achieving and maintaining the necessary optical quality. It will take much more research to develop flight-ready technology, but the concept is based on realistic and technically feasible design approaches.