Last Topic - Speckle Interferometry Lecture Index Next Topic - Telescope Sites Next (working) Topic - Detectors: Introduction

ASTR 5110, Majewski [FALL 2017]. Lecture Notes

ASTR 5110 (Majewski) Lecture Notes

Seeing Control and Turbulence Compensation

Some References:

  • Chapters 9 and 11 of Bely, The Design and Construction of Large Optical Telescopes ; I shamelessly used this source liberally for much of this discussion here.

  • Pp. 144-146 of Gordon Walker, Astronomical Observations.

  • Section 4.4.4 of Lena, Observational Astrophysics.

  • Beckers 1993, ARAA, 31, 13 reviews adaptive optics as of 1993.

  • Schroeder, Astronomical Optics, Chapter 16.

"Fogge-Refraction" by the graphic artist Daniel Fogelqvist - By permission of the artist.

A. Local Seeing: Thermal Control

One of the major developments in the past few decades of observational astronomy is an understanding of and reduction in "local seeing", also known as "dome seeing".

  • Thermal instabilities created by the very presence of the telescope, enclosure, associated equipment, and even the astronomer can be a prominent source of image degradation.

  • These effects are worst near the heat-exchange surface, and create turbulence.

  • The most common sources of local convection are:

    • Large scale convection through entire dome from a floor warmer than the air, or telescope parts colder than the air.

    • Mirror seeing from differences in primary or secondary mirror and air.

    • Heat sources on telescope or in dome.

  • The outer scale of dome seeing turbulence is more like centimeters to meters (comparable to the size of the dome slit and equipment).

  • Obviously it is useful to maintain the temperature of all systems near the path of the light as close to the ambient air temperature as possible.

  • Maintaining thermal control is not only important for reducing local seeing but maintaining proper alignment/figure of optical surfaces, and, when working in the infrared, potentially reducing thermal background.

Thermal Control

Thermal control can be passive...

  • Using coatings, insulation, radiating surfaces, heat pipes to control external heat input or the dumping of internally created waste heat.

... or active:
  • Using heaters, coolants, and ventilation (e.g., fans and louvers).

Thermal Control of Optical Alignment/Figure

  • Nonuniform axial temperature gradients are a problem for mirror figure.

  • Recall that the use of low expansion or low thermal inertial glass compensates for this.

  • Meniscus mirrors with adaptive optics can correct for figure changes, even if high expansion glass is used.

  • Larger monolithic mirrors can be made with honeycomb cells that are ventilated actively.

    From Bely, The Design and Construction of Large Optical Telescopes.

  • Segmented mirrors generally account for errors between segments, but not thermally induced deformations in each segment.

    In this case low-expansion glass is important.

  • The telescope mirror will change focal length with temperature changes, and the tube/truss will change length as temperature changes.

    Use of low-expansion rods (like Invar or ceramic) can be used to keep the focal plane a fixed distance from the mirror cell.

    Alternatively, can measure the temperature of the struts and actively alter the focal plane distance according to temperature.

Radiative Cooling

An important concept to understand with regard to seeing, whether caused by the environment around the observatory or within the dome itself is the role played by radiative cooling.

  • This is the process by which a body cools down by emitting radiation.

  • For example, part of the way that the earth maintains its thermal balance is by reradiating absorbed short wavelength light from the sun in the infrared.

  • Other processes are also at play in the system, such as, of course, heat loss through convection (a primary source of turbulence) as well as transport of latent heat through water evaporation.

    However, radiative cooling is important at high altitudes, and becomes more important at night at lower altitudes.

  • At night, solid surfaces, like the ground around the observatory, the telescope parts, as well as the skin of the astronomer (!) will radiate heat into space.

    The heat exchange is in this direction because, generally speaking, these surfaces are at around 300 K whereas space is at 3 K.

    You may have noticed from personal experience that the effect is most pronounced on cloudless nights when you are directly exposed to the colder system of the dark cosmos than the more nearby, warmer clouds (which not only prevents the heat from escaping into space, but keeps it trapped in like an insulating blanket).

  • In our discussions of atmospheric layers we mentioned how the ground cools much faster than the air after sunset. This is because of the effects of radiative cooling.

    During the day the ground heats because the incoming energy exceeds that radiated away. After the sun goes down the ground cools off as it continues to radiate heat away. From
  • You may have noticed that this radiative cooling often induces the formation of condensation or frost on surfaces like cars, house roofs, grass, etc., even if the air temperature is above freezing or the dew point.

Telescope Temperature

Traditionally (19th century), it was known that refractors can have better seeing than reflectors because of the stability of the air column inside closed tube.

  • In addition, exposing the refractor lens to the sky before observing helps it radiatively cool relative to the air, limiting vertical convection above it.

For large reflectors, it was realized that it is important to get stability of the air column in light path too.

The top of the telescope tube, which has a wider view of the sky, cools down radiatively faster than other parts.

  • Can create downflows of locally chilled air onto mirror.

  • Coat telescope parts with low-emissivity paint, or insulate with, e.g., aluminum foil.

  • Ventilation of telescope environment critical.

Mirror Seeing

Temperature differences between the mirror and the air create very thin (few mm) but very turbulent convective layer.

  • Ventilation (as shown happening below the mirror in above figure) helps.

  • Keep mirror as free as possible from surrounding structures.

  • Natural flushing by wind.

    • Wind flushing decreases temperature differences, but increases dynamical effects.

    • Thus there is an optimal wind speed.

      From Bely, The Design and Construction of Large Optical Telescopes.

  • Active heating and cooling of mirror.

B. Observatory Enclosure Considerations

Unexpected Turbulence, digital painting by Chalda Maloff. From page: artist_chalda_maloff.html.

Thermal Control

Important to keep all unnecessary heat-generating equipment away from telescope.

  • Now put everything possible, including observing room, labs, offices, motors, chillers, astronomers, etc. in other, insulated rooms or even other buildings.

  • Thermal barriers separating rooms.

  • Locate air exhausts away from enclosure and downwind. Secondary exhaust in case wind direction changes.

Important to reject solar heat during the day by insulating and cooling.

If the enclosure protects telescope from Sun during day, it can also be a source of degradation at night if it maintains temperature differences with ambient air.

  • Many domes/enclosures have been painted with white titanium dioxide paint.

    • Low solar absorptivity, reduces daytime heating.

    • However, has high thermal emissivity and quickly cools by radiating to sky at night.

    • Air passing over white paint is cooled and pockets of cold air can fall into dome opening, creating thermal turbulence.

      From Bely, The Design and Construction of Large Optical Telescopes.

  • Now philosophy is to make skin of dome unpainted aluminum or cover with aluminum Mylar tape.

    • Shiny metal surfaces also have low absorptivity (which reduces the daytime heating), but also have low thermal emissivities (so they get hotter than white domes in the daytime).

    • MUST have good ventilation or cooling in day because aluminum highly heat conductive.

    • But these metallic enclosures track the ambient air temperature better at night because of their lower thermal emissivity.

  • Keep daytime telescope room slightly overpressured and rest of building negative pressure to keep daytime airflow out from telescope.

  • Other observatories actively chill the floor with cooling coils filled with something like glycol (e.g., KPNO 4-m). (By the way, such surfaces often condense water and one should be careful about slipping on them.)

    From Bely, The Design and Construction of Large Optical Telescopes.

    Creates a stratified thermal inversion that can be maintained at night in low wind.

    Works well for older observatories with large thermal inertia.

    But have to guess what the night temperature will be.

  • At Fan Mountain we actively air condition the 40-inch dome during the day.

  • Another solution, and cheaper, is to have low thermal inertia mirrors and open telescope structures and ventilate aggressively.

    Includes even having fan-forced ventilation.

    This philosophy drives the current design of modern telescope enclosures.

From Bely, The Design and Construction of Large Optical Telescopes.

In this new philosophy, important then to have a well-flushed enclosure.

  • Pockets of air at different temperature cause turbulence.

  • Opening up the enclosure as much as possible allows wind flow through enclosure.

    • Not bad if isothermal.

    • Is bad if wind so high that telescope shakes.

    • Need to compromise with variable openings to adjust degree and direction of natural flushing.

  • When enclosure open, wind should flow smoothly so as not to excite high frequency modes of telescope.

  • Implicit in all of this is that when you observe, important to:

    • Open the dome, dome slit, doors, louvers, mirror/lens cover etc. after sunset (in twilight) and before observing to equilibrate your equipment with the ambient air as soon as possible.

    • Be cognizant of thermal sources in the dome and airflow through dome.

Dome Size and Shape

Recent trends are to make the telescope enclosure as small as possible.

From Bely, The Design and Construction of Large Optical Telescopes.

  • Cheaper.

  • Easier to flush.

    • A uniform air-flow of 1 m/s flushes a 30 m enclosure 120 times per hour.

Three main types of enclosure:

From Bely, The Design and Construction of Large Optical Telescopes.

  • Traditional dome (e.g., McCormick, Fan Mountain, KPNO/CTIO 4-m, etc.).

    • Dome clears telescope in all directions.

    • Can rotate dome separate from telescope (often useful).

    • But can foster stagnant air pockets and internal vortices depending on wind angle of attack.

      From Bely, The Design and Construction of Large Optical Telescopes.

      Many traditional domes now have louvers and fans inserted to fix this problem.

    • Better shape for minimizing snow/ice loads.

  • Corotating building (e.g., MMT, LBT).

    • Can tuck stuff closer to telescope, which corotates with it (don't need large "clear space" for telescope to swing through).

    • Smaller building possible, but need to move much more mass.

    • Can create more pockets and air funneling.

      From Bely, The Design and Construction of Large Optical Telescopes.

    • All electrical lines and fluid pipes become complicated to deal with "wrap-up".

  • Roll-off roof/hangar or retractable enclosure (e.g., McCormick "doghouse", Sloan telescope).

    (Left) Exposed Sloan 2.5-m telescope with enclosure rolled off to left. From (Middle) A view of the wind baffles on the 2.5-m telescope, which make up the boxy metal structure you see in this photo. This box is mounted independently from the rest of the telescope. (Right) The telescope shown without the wind baffle. The latter two images are from

    • Should roll-off far enough to away from scope and on downwind side to prevent wakes.

    • Often wind baffles installed to minimize shaking.

    • Difficult for large telescopes because wind baffles need to be enormous and movable -- difficult engineering.

Dome shapes have been well studied in wind tunnels.

  • The best shapes for flushing actually are found to be the two on the right, below.

    From Bely, The Design and Construction of Large Optical Telescopes.

    (But some shapes are better for minimizing ice/snow accumulation, while non-spherical shapes easier to build.)

  • A view of Kitt Peak, showing a variety of domes used.


  • The Bok telescope (foreground of above picture) has the second type of dome.

  • The octagonal shape was used for WIYN (background of KPNO picture, and shown below):


  • In all cases, the use of louvers, openings and windscreens is critical.

    From Bely, The Design and Construction of Large Optical Telescopes.

    • Openings in the walls, with adjustable louvers, can control wind flow and direction.

      Louvers in the side of the KPNO 0.9-m dome. From 0.9m/anight.html.

    • Windscreens are generally a set of panels or canvas covers that are raised along the lower or upper parts of the dome and can control airflow and prevent high winds from rocking the telescope.

      View of the CTIO 1.5-m telescope from outside showing elevated windscreen. From page: fwalter/ctiopics.html.

    • An "up and over" shutter can sometimes be used as a windscreen.

      From Bely, The Design and Construction of Large Optical Telescopes.

      From Bely, The Design and Construction of Large Optical Telescopes.

    • Ventilation on flat surfaces better than curved (air flowing around the curved surfaces creates negative pressure that prevents inward flow).

Dome Siting

How the dome is situated on the mountain is also important.

  • Want to make sure surface layer of air does not enter into enclosure or flow over telescope enclosure.

    • Dome should be elevated above ground layer and the enclosure should not interact with it.

    • Dome shape and support can change surface layer flow; some designs "lift" layer over dome:

      From Bely, The Design and Construction of Large Optical Telescopes.

    • Ideal mountain shape is an isolated conical peak.

      • Impinging airflow tends to divide and flow to either side of peak, rather than up and over.

      • Ideal slope angle on windward side is 7-18 degrees.

    • If peak is flat, observatory should be placed as close to windward ridge as possible to sit in unperturbed flow.

      • WIYN telescope on Kitt Peak has best seeing on mountain, partly for this reason.

    • Multiple telescopes should be laid out perpendicular to wind to avoid interference and wakes.

      From Bely, The Design and Construction of Large Optical Telescopes.

    • Ridges not as good as single peaks (disturb air-flow, tend to push it up-slope).

  • Same siting considerations will also minimize dust getting into dome.

  • Finally, how telescope attached to mountain is important for minimizing vibrations to telescope:

    From Bely, The Design and Construction of Large Optical Telescopes.

    • Generally, concrete pier attached to bedrock, but isolated from rest of structure.

    • Damping layers (sand, lava cinder, loose soil) helpful.

    • Fractured bedrock more prone to vibration, so minimize stress on rock during construction.

C. Active and Adaptive Optics

Refraction/Waves by Michael Burges, 2003, Oil on Wood, 100 x 70 cm. From


Active optics is typically the term used for the removal of global, low-order Zernike polynomial effects at low frequency (< few Hertz):

  • For example, image wander effects from wind buffeting or optical misalignment or figure distortion from thermal or gravitational loading. Also, low order, tip-tilt from seeing.

  • Typically through single tip-tilt tertiary or wobbling secondary mirror.

  • Also, piston motion for secondary (e.g., corrects defocus).

  • Can also control the actuators on the primary mirror.

From Bely, The Design and Construction of Large Optical Telescopes.

Adaptive optics is typically the term used for the correction of high frequency (few to 1000 Hz) wavefront disturbances by atmospheric turbulence.

  • Feedback removal of wavefront distortions by countering them with movable/shapeable optics.

  • Use of deformable mirrors to counteract corrugations in wavefront coupled in feedback loop with detailed/complex wavefront monitoring.

Basic Optical Configurations

Active optics example:

From Bely, The Design and Construction of Large Optical Telescopes.

  • Wavefront errors measured using a bright star off-axis by several arcmin, so not to interfere with science target.

  • Compensated by moving secondary and moving the primary actuators.

    The actuators typically fairly broadly spaced (~D/12 in the above case).

  • Reference star typically not in isoplanatic patch, in which case need to average over seeing effects by relatively long exposures (~ seconds).

    Allows fairly faint stars to be used.

Adaptive optics:

From Hecht, Optics.

Brief History

Horace Babcock (Carnegie Observatories) first suggested the idea of adaptive optics.

In remarkably prescient paper (1953) he:

  • describes the overall concept;

  • suggests a way of measuring the atmospheric wavefront distortions (by observing the Schlieren pattern);

  • proposes a concept for an adaptive mirror (using an oil film for which the thickness was controlled by electrical charges);

  • discusses the small relative size of the isoplanatic patch (few arcsec);

  • discusses the need for high time resolution and consequent limitation of wavefront sensing to bright (V < 6.3) stars.

From Babcock (1953, PASP, 65, 229)

Unfortunately, the technology was just not ready (for several decades) to implement these ideas.

Following Babcock, adaptive optics was pursued in parallel (but independently) by astronomers (starting early to mid-1970s) and U.S. Department of Defense (latter starting in 1973):

  • Imaging of and from satellites through atmosphere gives DOD same problem as astronomers.

  • Budgets for astronomy MUCH less than for DOD, and this is VERY expensive technology.

    So DOD got there much faster (by 1977 had succeeded with method).

    Astronomers did not get there until late 1980s (and in some cases with some borrowed technology from DOD).

  • First experiments were a bit discouraging:

    • Showed how expensive this would be.

    • Application to areas only around bright reference stars seemed rather limiting.

But a resurgence in interest due to several factors:

  • Idea of synthetic, laser reference stars proposed by military in 1982, prototypes in 1984, declassified in 1991.

    Until the early 1990s, the Starfire Optical Range at Kirtland Air Force Base in Albuquerque, New Mexico, was one of the U.S. Air Force's most closely guarded secrets. Here, a laser beams probe the atmosphere above the system's 1.5-m telescope, allowing minute computer-controlled changes to be made to the mirror surface thousands of times each second. (b) The improvement in seeing produced by such systems can be dramatic, as can be seen in these images acquired at another military observatory atop Mount Haleakala in Maui, Hawaii, employing similar technology. The uncorrected image (left) of the double star Castor is a blur spread over several arc seconds, with little hint of its binary nature. With adaptive compensation applied (right), the resolution is improved to a mere 0.10 and the two components are clearly resolved. (R. Ressmeyer; MIT Lincoln Laboratory). Image and caption from

    Laser guide stars independently proposed by Labeyrie and Foy in 1985.

  • Better detectors (CCDs), increasing sensitivity.

  • Realization that many complexities and limitations of AO reduced or disappear in the infrared.

    • Recall that r 0 ∝ λ 6/5.

      Means that number of adaptive elements decreases.

      Means that temporal control frequencies decrease.

    • Array detectors in IR developed.

    • Diffraction limit approachable at NIR wavelengths (Strehl ratios approaching 1).

  • Better seeing sites and reduction of local seeing decrease complexities even more.

Now many observatories implementing adaptive optics systems.

Wavefront Correctors

Tip-Tilt Mirrors:

  • Up to the fifth Zernike polynomial can be corrected out with only tip/tilt and piston motion of, e.g., secondary mirror.

  • Many secondaries on infrared telescopes already move for chopping observations.

  • Fine (or fast) steering mirrors are smaller, easier to control (smaller inertia) optics at the telescope output.

Deforming primary:

  • Beyond tip-tilt correction requires deforming an optical element.

  • Deforming the primary involves either applying forces or moments with actuators.

    From Bely, The Design and Construction of Large Optical Telescopes.

  • The more actuators, the more "correctable" is the wavefront.

Dedicated, deformable, phase-correcting mirror:

  • For compensation of atmospheric turbulence, requires much faster response than can get from primary mirror actuators.

  • Deformable mirrors made of these plates actuated by piezo-electric mechanisms.

    Two main types:

    • Piston motion actuators by piezoelectric stacks.

    • Bimorph mirrors which are made from a pair of piezoelectric wafers glued together, with localized control of each.

      Local bending by stretching one wafer and contracting other at same point.

    From Bely, The Design and Construction of Large Optical Telescopes.

  • On MMT and LBT, the secondaries can be deformable (MMT a 24 inch secondary with 336 actuators, LBT a 36 inch secondary with 672 actuators).

    (Left) Side view of the MMT adaptive secondary, showing wires to actuators that are adjusted electromagnetically. (Right) The 2mm thick deformable mirror with 336 permanent magnets. From page: adaptive_optics.htm.

Wavefront Sensors

The Shack-Hartmann sensor is an evolved version of the Hartmann test.

  • Telescope entrance pupil reimaged onto a lenslet array (instead of Hartmann mask).

  • Each lenslet creates an image of the star.

    Principle of a Shack-Hartmann lenslet array. From

  • The position of the centroid of each lenslet image yields the slope of the wavefront tilt at the pupil position represented by the lenslet.

  • Wavefront tilts are generally achromatic enough that one can use photons of a broad wavelength range to improve S/N (often use optical sensing to correct even NIR images).

    From Hecht, Optics.

    Some Shack-Hartmann spot array patterns and their associated wave aberrations. The contour lines in the plots represent one wavelength (0.55 microns) of wavefront height. From

Isoplanatic Angle

If want to correct a field of view on the sky, need to have a reference star in the same isoplanatic angle.

  • A good approximation of the isoplanatic angle is

    θ0 ~ k r0 / H,

    where r0 is the Fried parameter, H is the height of the turbulent seeing layer, and k is a constant.

    The value of k we used before was 0.6 (but this depends on how one defines the level of isoplanaticity; Beckers adopts 0.3 for this constant).

  • The temporal timescale on which changes need to be made in that isoplanatic angle are given approximately by

    τ0 ~ k r0 / Vwind,

    where Vwind is the wind velocity in the turbulent layer.

  • Recall that r0 has a λ 6/5 dependence, and so is larger at longer wavelengths.

  • Using these equations (and adopting the more strict k=0.3 and H=5 km), one finds the following variation in Fried parameter (r0), temporal timescale (τ0) and isoplanatic angle θ0 with wavelength:

    From Beckers 1993, ARAA, 31, 13.

  • Note how small the isoplanatic angle and temporal timescale is in the optical and how the idea of adaptive optics really is best applied at infrared wavelengths.

Natural Guide Stars

To undertake adaptive optics requires a reference star in the isoplanatic angle.

  • The above table shows how quickly one needs to read out a detector in order to sample the changing wavefront (τdet) a few times per change.

  • This readout rate for a detector translates directly to a stellar magnitude limit, given a detector with a given readout noise and quantum efficiency and a desired S/N (as shown in table footnote).

    As may be seen, the brightness of a star that can be used for Shack-Hartmann sensing and adaptive optics becomes increasingly bright at shorter wavelengths.

  • The last column in the table above shows what fraction of the sky has a star available at this magnitude within an isoplanatic angle, and therefore, in principle, can be corrected with adaptive optics.

  • Obviously this is rather limiting for high resolution imaging of interesting sources.

    For example:

    • The sky coverage is tiny for using normal stars for references for most things.

    • But the density of faint high redshift galaxies is high enough that there is a reasonable number that will fall near a bright enough reference star even at optical wavelengths (see fourth line in table below).

    • The density of QSOs, however, is lower, and only a reasonable number can be imaged with adaptive optics when working in the infrared (see line 8 in table below).

    • Of course, in both cases, one has to find the extragalactic objects near the reference stars!

Artificial Guide Stars

The solution to the limiting sky coverage and improved correctability (i.e. brighter reference sources) is to create an artificial guide star wherever it is needed.

The current method to do this is using laser guide stars.

Two methods are used:

  • Rayleigh scattering involves focusing a powerful laser to a point 10-20 km up, above most, but not all, turbulence.

    Only backscattered photons from the focused height contribute to the wavefront estimate.

  • Probably more common is to make a sodium laser star.

    • Use the NaD line at 5890 Å.

    • There is an ~11.5 km thick, enhanced neutral sodium layer in the mesosphere at about 90 km up, well above the 10 km "jet stream" layer and most turbulence.

      The 90 km enhanced sodium layer likely comes from meteoritic dust.

      Optical thickness of the layer is about 0.05 at sodium line center.

      From Beckers 1993, ARAA, 31, 13.

    • Shoot NaD laser at this layer.

      Those Na atoms excited by laser re-emit by spontaneous emission or by stimulated emission.

      Those emitted back to telescope can be used for wavefront sensing.

      A sodium dye laser beam pierces the sky over Lick Observatory on July 22, 2003. The laser is the final piece of the laser guide star adaptive optics system that allows twinkle-free viewing of the entire nighttime sky. The beam, which reaches 60 miles into the upper atmosphere, is visible in scattered light for several kilometers. The yellowish cast of the dome is due to the streetlights of nearby San Jose. Photo: Marshall Perrin/UC Berkeley. From page: 03-04/03-01/images.html.
    • To create a 50 cm, ~1 arcsec spot need ~5 kW laser pulse.

      Laser is pulsed to shutter the camera to avoid scattered light.

      Pulses about 0.001 to 0.02 duty cycles.

    • Typical brightness about V = 12-14.

      Sodium laser guide star and the star η UMa in the V band. Both images were taken on the same night. The LGS is shown on the left and has brightness V=9.5 mag. η UMa is on the right. The two pictures were normalized by their peak brightness and put side-by-side to compare the spot sizes. From page: chaos/sac_peak_laser.html.

  • Some limitations of LGS:

    • If laser off to side of telescope, get elongated star.

      For 10-m separation, ~3 arcsec elongation.

    • Laser guide stars do not correct for atmospheric tip-tilt, because the outgoing and returning beam follow the same path through air "wedge" (the tip-tilt is canceled).

      From Bely, The Design and Construction of Large Optical Telescopes.

      Have to use a natural guide star for this correction (but can use a fainter one).

    • The path of the LGS is slightly different than source at infinity.

      The LGS projects a conical (not parallel) beam through atmospheric turbulence layers.

      From Bely, The Design and Construction of Large Optical Telescopes.

      Thus the wavefront perturbation they sense is not the same as that of target at infinity.

      This is called the cone effect or focus anisoplanatism and reduces Strehl ratio by about 50% at 1 micron (and more at shorter wavelengths).

      Can be overcome with multiple laser stars.

    • Upgoing beam affected by turbulence too, but the result is a net blurring equally across all Shack-Hartmann sensors (and a net lowering of S/N).

Comparison of laser vs. natural guide star corrected images of the Galactic center (in L band at 3.8 microns) taken with the Keck telescope. The LGS image is 8 mins long, the NGS image is 150 minutes long. From page: ~jlu/gc/pictures/lgs.shtml.


From Beckers 1993, ARAA, 31, 13.

An animation showing a set of images of a star observed through turbulent atmosphere without any correction, with only tip-tilt correction with a fast steering mirror, and closed loop adaptive optics with a deformable mirror. From the Max Planck Institute for Astrophysics webpage: AO/INTRO/AOWFSintro.html.

An image made by summing together the frames from the first third of the above animation (no seeing compensation) and showing the "long exposure" image. By Guillermo Damke.


Multi-Conjugate Adaptive Optics

Multi-conjugate adaptive optics (MCAO), proposed by Beckers in 1988, uses both multiple wavefront sensors and multiple deformable mirrors.

The complex beam path of a proposed AO instrument for Gemini telescopes as of 1999, with three deformable mirrors to conjugate at 8, 4 and 0 km. From the Gemini Newsletter, Issue #19, December, 1999.
Several advantages:

  • Since turbulence happens in multiple layers, can improve the inversion of the turbulence profile by sensing each layer and conjugating them each individually.

    Equivalent to atmospheric tomography to get 3-D structure of turbulence layer.

    For example, with two guide stars and sensors, MCAO-like spectroscopy:

    • Two wavefront sensors look at two different guide stars separated by an angle θ.

    • The two beams have increasing shear, θh, with height, h.

      From the Gemini Newsletter, Issue #19, December, 1999.
    • A phase feature (e.g., marked by the "plus" feature in the figure below) at height h0 seen by the two sensors will be seen with a spatial shift θh0.

      With θ known, h0 can be derived for that phase feature.

    • Once column distribution of turbulence determined, can use multiple deformable mirrors to correct.

  • MCAO removes cone effect given the multiple beams (as shown in the cone effect figure above).

  • Net effect is substantially larger corrected field of view (even 1-2 arcmin) with uniform PSF.

    From the Gemini Newsletter, Issue #19, December, 1999.

    Last Topic - Speckle Interferometry Lecture Index Next Topic - Telescope Sites Next (working) Topic - Detectors: Introduction

    Unless otherwise attributed, material copyright © 2005,2009,2011,2013,2017 Steven R. Majewski. All rights reserved. These notes are intended for the private, noncommercial use of students enrolled in Astronomy 511 and Astronomy 5110 at the University of Virginia.