The Committee on Radio Astronomy Frequencies (CRAF) is a committee of
the European Science Foundation (ESF).
The issue of the frequency choice for cloud profiling radar has been discussed extensively within the radio astronomy community, e.g. during the October 1996 meeting of ITU-R WP7D (8-16 October 1996) in Geneva ( IUCAF report of ITU-R WP7D meeting, 8-16 October 1996).
In relation to this subject the following observations can be made:
[1] ITU-R Doc.9D/41-E states that a 1 GHz allocation at 95 GHz for space based active earth sensors is necessary for this application. After discussions with space agencies and remote sensing colleagues it became clear that instead of 1 GHz one should read 100 MHz. There is no need at all for more bandwidth.
[2] Virtually all of the microwave spectral lines of molecules present in the interstellar medium are at frequencies above 20 GHz and access to them for research is limited only by their intrinsic intensities and the absorption of the terrestrial atmosphere. The band considered is in this spectral domain and located in a "forest" of spectral lines (of which each adds in its own specific way to the story which physics is explaining to mankind).
[3] Astronomers are fortunate in the availability of wideband receiving systems spanning almost the entire atmospheric spectral windows. For instance, all current mm-wave antennas are outfitted with sensitive systems commonly used Superconductor-Insulator-Superconductor (SIS) mixers) operating in the range 70-116 GHz, approximately. This provides astronomers access to a large number of molecular spectral lines - a 1992 compilation lists detections of about 30-40 lines/GHz near 95 GHz, and the numbers are now much higher. Many spectral lines now considered of astrophysical importance by astronomers are outside bands in which the Radio Astronomy Service has a primary allocation (such as 86-92 GHz, 98.77-98.08 GHz and 105-116 GHz): e.g. five spectral lines of methanol at frequencies between 94.405 and 94.542 GHz. But there has been no opportunity to obtain appropriate protection in the Radio Regulations since 1979 (quoted from ITU-R Doc.7/86-E).
[4] Emissions from spaceborne or airborne stations can be particularly serious sources of interference to the Radio Astronomy Service. Allocations of a band near 95 GHz for use by space-based radar systems, would produce several undesirable consequences for 3-mm RA observations using currently installed receivers:
a. If a subband near 95 GHz is allocated to active sensors, RA observations could no longer be conducted near frequencies of satellite transmitters, and interference at levels exceeding threshold limits considered detrimental for RA observations, as listed in Recommendation ITU-R RA.769, might be experienced in the 86-92 GHz, 98.77-98.08 GHz and 105-116 GHz bands, whenever satellites were above the horizon of mm-wave observations.
b. SIS receivers can saturate at very low input leveles, and active sensor transmissions could affect the operation of the receivers over their total frequency band. In addition, calculations suggest that in the event of a satellite transmission into the mean beam of a mm-wave antenna, the SIS receiver would be destroyed.
Additional comments:
Depending on the installation, SIS mixers are destroyed if 0.01 mW to 10 mW of power is input. The saturation level is typically of the order of -110 dBW.
A damage to the SIS receiver could occur when a RA receiver is within the cloud profiling radar main beam and the RA antenna is pointing directly to the satellite of very close to it (< 3 degrees), hence to the zenith.
A saturation could occur when: (a) the RA receiver is within a +/- 14 degree vertical cone centered on the satellite with RA antenna is pointing directly to the satellite (+/- 0.01 degree). The validity of this result is conditional, depending on the cloud profiling radar and radio astronomy antenna patterns. Hence, updates will be necessary as better knowledge on the antenna patterns become available.
Remaining interference possibilities are:
i) a possible saturation when the RA antenna exactly points to the distant satellite, which could be eliminated by avoiding the known satellite path at known time (orbit information needed at RA site).
ii) signal interferences during the horizon-to-horizon appearances of the satellite, at a level which depends on the relative orientations of the respective antennas and the frequency separation between the cloud profiling radar transmit signal and the RA observation band.
Potential techniques to minimize these interferences without appreciably reducing the signal integration time (which is certainly not a trivial undertaking), have to be developed.
Adding appropriate filters to RA receivers is difficult or is not feasible, because:
a. Currently, there is no adequately low-loss 3-mm wavelength filter technology available and the prospects for such technology becoming available are at present uncertain.
b. Addition of a notch filter into the RA receiver band would permanently prevent RA observations at frequencies covered by the filter.
ITU-R WP7D suggested several possible solutions to this problem, such as: - operate the space-based active Earth sensors in the 78 GHz band already allocated to this application. However, operation of a cloud profiling radar in a higher frequency bands (e.g. 120-130 GHz) may have less impact on RA operations. Compared with operations at 95 GHz, the decreased scattering losses would result in a 5 dB gain in sensitivity, and the atmospheric attenuation is comparable.
- regarding the current proposal, suggestions that satellite systems could switch off transmissions, or that RA observations could cease, for the short period that a satellite is in the line-of-sight of mm-wave observatories might be a possible solution if only one cloud profiling radar satellite was involved, but appears untenable if several systems of satellites were to use the allocated band.
ITU-R WP7D suggested furthermore that in the event that an allocation is made at WRC97 in the frequency band 94-95 GHz for cloud profiling radars, the allocation should cover only the 94.0-94.1 GHZz band, which appears to have relatively few spectral lines. Furthermore, WP7D suggests that to prevent the proliferation of space-based radars, such an allocation be restricted to cloud profiling radars, possibly through a footnote to the Radio Regulations.
- Finally, concerning the transmissions of such systems, a pulsed signal would have lower impact on RA observations than a chirp-type transmission.
An other astronomical argument in favour of the subband 94.0-94.1 GHz which was not mentioned in ITU-R Doc.7/86-E is that the band 94.9-95.0 GHz subband is close to a strong CH3OH transition at 95.169 GHz which can be observed in extragalactic objects. Usually extragalactic sources are red- shifted (frequencies become lower due to the Doppler effect) by more than 100-200 MHz. By using e.g. VLBI measurements we can use the CH3OH maser line to measure distances of such extragalactic sources (difficult to obtain by other means). Measurement of distance is one of the most fundamental studies in astronomy. Around 94.0-94.1 GHz no such strong maser lines exist.
Conclusion:
In case an allocation of 100 MHz for cloud profiling radars near 94-95 GHz has to be made, the ESF Committee on Radio Astronomy Frequencies, CRAF, informs you that the 94.0-94.1 GHz subband is preferred on the basis of both astronomical arguments and technical radio astronomical arguments as given above.
In addition, CRAF strongly supports a solution (mentioned above) that cloud profiling radar transmitter is shut down whenever a mm-wave RA- telescope is within +/- 14 degrees from nadir of the satellite. This would eliminate the possibilities of damaging the SIS receiver and a largest part of saturation occurrences.