Next: 9.5 Current phase correction
Up: 9. Atmospheric Fluctuations
Previous: 9.3 Statistical properties of
These methods were originally developed for meteorology and control of industrial
emissions.
- 1.
- LIDAR: compares laser back scattering or transmission. DIAL: differential
absorption LIDAR, works at two frequencies (on and off the atmospheric line of
interest), detects 0.01 g/m3 water vapor. Disadvantage: works best from aircraft,
expensive equipment.
- 2.
- SODAR: Remote sounding with sound waves. Detects turbulence, but gives little
quantitative results.
- 3.
- IR window: H2O line absorption in front of a strong continuum source (Sun,
Moon, Jupiter). Disadvantage: Directions of observing and monitoring beam differ.
The phase correction degrades as a function of the separation angle and the distance
of the dominant turbulent layer.
- 4.
- Radiometric: Uses the atmospheric emission. Dedicated monitors operate mostly
near the 22 GHz or 183 GHz lines (several spectral channels). The inter-line regions
of the 1mm and 3mm windows are also sensitive enough, but make it difficult to
remove cloud emission.
For the radiometric approach, it is useful to study the sensitivity as a function of
frequency, i.e. by how much the sky emission changes for a fixed fluctuation of
water vapor, which corresponds to a fixed wet path fluctuation. Fig.9.4
shows what change in Tsky one must measure under conditions of various
humidity.
There are two reasons to use the 22.2 GHz line: Clouds are easier corrected at this
frequency, and receiver components are less expensive.
One notices that the 84-116 GHz window is 1-2 times as sensitive as the 22.2 GHz
line, and the 210-248 GHz window 4.5-8.3 times. A dedicated receiver near the
183 GHz water line would have the highest sensitivity, but can suffer from
temperature dependent saturation effects. It is better adapted for sites where the
total amount of water above the instrument is typically less than 3 mm.
Next: 9.5 Current phase correction
Up: 9. Atmospheric Fluctuations
Previous: 9.3 Statistical properties of
S.Guilloteau
2000-01-19