Snell's Law describes the bending of light as it passes from one medium into another. The most common visual example of this is a pencil partially submerged in a glass of water. The pencil appears bent because the light changes direction as it passes from the water into air. Because the index of refraction of air is dependent on the temperature of the air, pockets of warm and cold air bend light just as it does going from water to air.
This is easy to visualize on a macroscopic level with this picture of a jet. The heat from the jet's engines creates a hot, turbulent cloud of air behind the jet. The prevailing winds are passing from the left to the right, taking the hot air with it. The background is indiscernible through the hot cloud because the light passing through it gets distorted.
Atmospheric distortion can easily be seen above the jets wings
For smaller changes in temperature, this phenomenon is visible to the naked eye with stars. White light coming from a star passes through numerous pockets of warm and cold air in the atmosphere. As this light is refracted, the spot of light becomes a spectrum of colors. Since the air is constantly moving and churning above us, the spectrum moves randomly a little bit every second. At any given moment, the light from a star that hits our eye can be red, or blue, or white, depending on which pockets of air the light travels through at that moment. This result is often referred to as "twinkling."
The atmosphere is the most turbulent near the ground where physical obstructions create a complicated path. This is also where the air is likely to have the largest temperature variation over a short distance. After the sun sets, cool air passes over objects still warm from the day's sunlight, transferring that heat to the air. The combination of a heat source and non-steady air motion distorts the light from the stars, making them appear as small fuzzy dots instead of point sources when viewed through a telescope.
An often neglected aspect of observatory design is "dome seeing," a term encompassing how the design and construction of the dome (and the surrounding structure) affects the delivered image quality. During nighttime viewing, the air is usually cooler than any structures of significant mass. This causes the air inside and immediately surrounding the dome to be warmer than the ambient air temperature. When the dome shutter is opened, a stack effect (or chimney effect) causes the warmer air in the dome to vent out of the shutter, turbulent flow that mixes warm and cool air. Because of the shutter's limited area, the air distortion is concentrated directly in the line-of-sight of the telescope.
Uncooled, unventilated dome
Dome with active ventilation
The telescope image production depends upon the general atmospheric optical turbulence ("Seeing") and any atmospheric optical turbulence within the dome. Ideally the air temperature inside of the observing chamber is at the same temperature as the general air mass outside of the observatory. This is because the index of refraction of air changes with temperature. If there are various volumes of air in the optical path at different temperatures, then the star light will be bent slightly within each volume (often called "seeing cells"). With any turbulence or layering of the air, these changes in the index of refraction will cause differential effects that cause blurring of the image. These effects within the dome are called "Dome Seeing".
During the daytime, solar radiation causes heating of the dome and the observatory walls. There is also some convection heating of the outside surfaces of the building and the dome due to the air temperature rising during the day. The temperature of the inside skin of the observatory and the telescope should be at the expected evening ambient temperature which is typically slightly warmer than the morning temperature. The heat input during the day soaks into the mass of the building and the telescope raising the temperature of these components. This heat is released during the night when the air temperature within the observing chamber becomes cooler than the structure. Depending upon the thermal mass of the building and the coupling between the structure and the air, this heat can be released over many hours. In this case the air within the observing chamber will always be warmer than the outside ambient air temperature.
On a previous job, an array of temperature and humidity sensors were placed both inside and outside the dome. A few days worth of temperature data was fed into a data acquisition system and the results analyzed:
Every evening, the temperature inside the dome, and especially the mirror temperature, remained above the outside ambient temperature. This data illustrates that the mirror temperature remains 3-4 degrees C above the ambient temperature, ensuring that warm air coming off the mirror passes through the optical path throughout the night. In addition, the thermal time constant of the mirror is so long that the mirror doesn't finish cooling off from the first day by the time the sun comes up the next day. The average temperature of the mirror continues to increase, even when the daily average temperature remains about the same.
At this site, when the ventilation fans were activated, the upper part of the dome cooled off quickly and remained at a cooler temperature after the fans were deactivated. The dome itself was a small mass, and it responded quickly to convective cooling. The lower part of the dome, near the warm floor and pier, took longer to cool down. In addition, the air in the lower part of the dome rapidly warmed back up after deactivating the fans because the heat sources weren't eliminated.
The structure of the observatory building stores heat by increasing its temperature. For a given heat input, the rate of temperature change depends upon the specific heat of the material and the mass of the material. The heat loss from the material depends upon the same factors but includes how the heat is transferred into the air within the observing chamber. The heat is transferred from the material to the air by convection. The specific heat of the material, the mass of the material, and the heat transfer coefficient determines the "Time Constant" for the cooling.
For optimal observing, a low thermal mass and a short time constant building is desired. This allows the observing chamber temperature and the rest of the building to follow the ambient air temperature changes with a minimum difference in temperature.
The effective thermal mass can be reduced by using insulation. The insulation reduces the heating of the material during the day time and reduces the heat flow back into the observing chamber during the night time when we are observing so the temperature within the observing chamber better follows the changes in the outside air temperature.
It is also possible to increase the heat transfer between the material and the air significantly by increasing the air flow velocity (ventilation). Ventilation can significantly reduce the temperature rise of the air (by dilution) and decrease the time constant.
The thickness of the material and its thermal conductivity has a small effect on the heat transfer for metals, but is significant for non metals such as concrete. A typical concrete beam with no ventilation can have a time constant of 12 hours – the opposite of what is desired.
The ideal wall (or floor) construction uses a minimum mass of material and insulates between the outside and inside skins. Concrete or sheet rock is the wrong material. Wood construction would be good as this material is a fair insulator, has a low mass, and has a low specific heat. Typical steel building construction can also be good. Undesired characteristics of steel construction include thermal conduction, radiation, and convection heat transfer between the outside skin and the structure. This heat transfer problem can be reduced by minimizing heat transfer between the outside skin and steel structure by using solutions such as insulation or a sun shield. Steel buildings often use an outside skin of metal while the inside wall covering is usually specified to be sheet rock. For an observatory, the inside skin needs to be a low thermal mass material such as steel, aluminum, fiber glass, or wood. Absolutely NO sheet rock should be used in the building with its large thermal mass and resulting large heat-transfer time-constant.
DFM Engineering has decades of experience working with numerous observatories. In that time, DFM has improved the seeing at many observatories and has always believed numerous design improvements are possible at any site. To address existing problems, DFM offers Observatory Dome and Pier recommendations, to implement improvements including layout and ventilation utilizing a combination of insulation, ventilation, and forced convection. By taking into account the prevailing winds, dome layout, and various heat sources and their time constants, a proper solution can be chosen to significantly reduce the dome seeing issues and dramatically improve the seeing ability of your telescope and its Delivered Image Quality (DIQ).
If you are planning a new observatory, DFM Engineering has designed and built the ideal telescope observatory, available in three different sizes. All our designs feature integrated mirror handling equipment to provide for safe and easy servicing of telescope optics.
Learn more about DFM's revolutionary telescope observatory design: