Noise of the Earth Atmosphere (10 GHz)

Joachim Köppen, Strasbourg, 2011/12

Objective: While the observation of a celestial source is affected by the noise background from the receiver as well as from the earth atmosphere, the measurement of the calibrator is only subjected to the receiver noise. Since the sky noise depends on elevation angle in a known way, but the receiver noise is constant, measurement of the noise of the 'empty' sky at several elevation angles allows to separate the two contributions, and a more correct value for the antenna temperature is obtained.

Observations

• Goto Calibrator and Start Observe & Record; measure here for about 5 min
• At the same azimuth, move the telescope manually to at least four elevations: 10°, (15°), 20°, (25°), 30°, (40°), 60°. At each position measure for about 5 min. The signal level should decrease slightly with elevation.
• Finish by Goto Calibrator, stay another 5min, and then Stop & Finish the observations

Interpretation (more details):

• Import the text file with the data into Excel
• Convert all the dBµV values into linear units (P = 10^(dB/10))
• For each elevation, take the average value of all the measurements
• plot these data against 1/sin(EL) - the value of airmass in a planar atmosphere
• Fit a straight line to the data:
• Its intersection with the y-axis gives the noise level of the receiver (plus the cosmic microwave background) = P0
• The slope gives the atmospheric contribution to the background noise = slope
• Application: With these parameters, the true antenna temperature of a celestial source (sun, moon) can be determined:
Tant = 290K * (Psource - Psky(EL))/(Pcal  P0)
where the sky background Psky(EL) at the source's elevation EL is either measured or computed from the fit formula
Psky(EL) = P0 + slope/sin(EL)
In the plot above, I simply did the linear fit on the dBµV values instead of converting them (more accurately) to linear powers ...

• Using the flux calibrator measurements (Pcal), and neglecting the contribution from the cosmic microwave background, we can express the measurements in Kelvin above the constant background:
T = 290K * (P - P0)/(Pcal  P0)
• [if we had a second calibration source, of a much lower temperature, we could also separate the receiver noise from the CMB radiation ]

last update: Sept. 2011 J.Köppen