Large portions of the electromagnetic spectrum are observable only from the cold vacuum of space. Even the visible light and the radio waves that do make it through the atmosphere can be masked by the terrestrial environment. Optical astronomy is hampered by light pollution from cities and radio astronomy is hampered by interference from TVs, radios, and microwave ovens. Most infrared radiation is absorbed a few kilometers above the ground by water vapor in the lower atmosphere. Ultraviolet radiation and all shorter wavelengths are extinguished higher up in the ozone layer. We must go into space to observe most of the infrared and microwave regions and all of the ultraviolet radiation, X-rays, and gamma rays.
After the Second World War, American scientists used captured German V2 rockets to send instruments high into the atmosphere. (Wernher von Braun, who built the V2 rockets, later came to work for NASA and was the architect of the Saturn V rocket that took astronauts to the Moon.) Since then, beginning in the early 1970s, a series of grander and more expensive satellites have pried open the short-wavelength sky and given us a view of a violent, high-energy universe. In the past three decades, scientists have made brief sorties into space with small telescopes mounted on balloons and rockets, but the best data comes from satellites launched into Earth orbit and beyond. Because space astronomy is expensive, many missions are joint ventures between two or more countries. Government agencies cooperate on the funding, and scientists from the countries collaborate on interpreting the data.
The dramatic gains possible from space astronomy can be illustrated in the mid-to-far infrared part of the electromagnetic spectrum (i.e. the longest infrared wavelengths). Working from the ground has two problems. First, water vapor in the atmosphere absorbs most infrared wavelengths so effectively that we cannot observe the entire infrared spectrum even from a high, mountaintop observatory like Mauna Kea. Second, the Earth and its atmosphere are at a temperature of about 300 Kelvin and emit thermal radiation in the infrared band. All astronomical sources are therefore seen against a bright background of thermal infrared radiation. The background is a million times darker in space. In infrared astronomy, the difference between working on the ground and working in space is like the difference in optical astronomy between observing in the day and at night! As amazing as this may sound, the results are even more dramatic in the X-ray and gamma ray parts of the electromagnetic spectrum, where we had previously been unable to make any ground-based observations.
Space astronomy offers three distinct advantages. It opens up spectral regions that cannot be observed from the ground. It offers an environment where the optical and infrared noise is very low. It allows telescopes to operate outside the blurring effects of the Earth's atmosphere, where in principle they can achieve their resolution limit. These advantages are set against the much greater cost and limited lifetime of an observatory in space. The Hubble Space Telescope is just one of many orbiting observatories that together map out the entirety of the electromagnetic spectrum with the exception of radio. In the 1990s, NASA created an ambitious plan to span the electromagnetic spectrum with four Great Observatories. Hubble Space Telescope was one, and the first to be launched in 1990. A year later the Compton Gamma Ray Observatory was launched. Next came the Chandra X-ray Observatory in 1999. The last to go up was the Spitzer Space Telescope in 2003. Each of these was a multi-billion dollar investment and each of them has returned excellent science.
Space astronomy is expensive! A large ground-based telescope of 8-10 meters aperture might cost $100-200 million. However, a complex telescope in space might cost $1-2 billion, even though it is only 2-3 meters in aperture. What makes the large difference? Space is an extreme environment and launch puts enormous stresses on scientific equipment. To make sure a space observatory will not fail in orbit, the components must be rigorously tested and often duplicates have to be built. NASA has recently experimented with more modest missions with a price tag of $200-400 million. These telescopes have a simpler design and can not tackle such a wide range of astronomy. However, the cost of a single failure is not punitive. As large as all these numbers are, they are dwarfed by the projected cost of sending astronauts to the Moon or Mars.