the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 May 2026
On the calibration and monitoring of dual-polarization radar receivers operating at C-band using solar flux reference at S-band
Abstract. The solar radio noise accurately measured by terrestrial solar observatories has been proved to be an effective reference for checking the quality of dual-polarization weather radar receivers. The longest, most complete and accurate record of solar spectral irradiance ("flux", for the sake of brevity) values exists thanks to the Dominion Radio Astrophysical Observatory (DRAO) of the National Research Council (NRC) of Canada, located in British Columbia. Solar flux measurements acquired by DRAO at 2.8 GHz represent a valuable and useful basis for the calibration of radar receivers. In Europe the large majority of weather radars operates between 5.4 and 5.6 GHz. To be rigorous, one needs measurements of solar radio noise at the same frequency of the weather radar. If one is prepared to accept some uncertainty, it is possible to extrapolate solar flux values from S- to C-band. The formula proposed by Tapping (2001) is widely used in Europe: it is based on a constant scaling factor to predict the flux at the higher frequency from the measured flux at 2.8 GHz, after having subtracted the quiet component at both frequencies. We have analyzed 240 quality-checked, 1-hour lasting solar flux measurements, simultaneously acquired at S- and C-band by the Expanded Owens Valley Solar Array (EOVSA) observatory during the current XXV solar cycle and empirically derived the optimal value of the scaling factor for the conversion from S- to C-band. We found that there is a clear non-linear dependence between this variable scaling factor and the slowly varying solar component. Thanks to the new conversion formula, which is based on the variable scaling factor, the agreement between C-band radar observations in Europe and DRAO-converted reference values improved both in relative and in absolute terms. Not much can be done to compensate the shift in time between measurements in Europe vs DRAO (noon at 20 UTC). Our recommendation is that of developing and installing an optimized C-band radio telescope in Europe tailored to radar calibration purposes.
- Preprint
(1544 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-2097', Anonymous Referee #1, 21 Jun 2026
- AC1: 'Reply on RC1', Andrea Battaglia, 07 Aug 2026
-
RC2: 'Comment on egusphere-2026-2097', Hidde Leijnse, 07 Jul 2026
This paper investiagtes the relation between the solar spectral irradiance at S-band to that at other bands. This is very relevant for weather radar calibration, especially for radars operating at bands other than S-band because the most widely used reference (DRAO) is at S-band. The authors propose a useful adaptation to the relation used to convert DRAO flux data to C-band flux estimates that is shown to be more accurate than the original. This adaptation is based on a thorough analysis of EOVSA spectral data of sun measurements. The autors demonstrate the usefulness of these data based on weather radar data, both in "sun-check" and "sun-track" modes. I also appreciate that the authors attempted to use a nearby C-band reference, and that the results led them to conclude that we need a dedicated nearby reference at C-band for proper C-band weather radar calibration. I think the paper is scientifically sound, and is on the whole well-written. It could use some clarifications in some places. These are listed below.
Specific remarks:
- Section 2.1: What happens if there is rain over the observatory? How will the attenuation caused by that rain (much more pronounced at C-band) affect the results? Or is there a correction in place?
- Section 2.2, I suggest also introducing the Finnish ANJ radar here.
- Fig. 2: the spectra show some very local peaks, but also some local dips. How could these be explained, and what do they tell us about the behavior of the spectrum? These excursions are much larger than the difference between S- and C-bands, so it would be good to know about their relevance.
- Eq. (2): should the p-variable be called pC instead of pλ? This because on the right-hand side of the equation C-band is indicated.
- In some cases the order in which things are explained could be modified. For example, when introducing the normalization constant of 141.2, it would have been more intuitive for me to have the explanation on lines 214-220 before Eq. (3) on line 210.
- Fig. 3a: can parentheses be added for the log functions in the legend, so that the reader can see what the argument of the logarithm is?
- lines 231-236: repetition of the value of 87.9%. Mentioning this value only once would be sufficient for me.
- Section 5.1 is confusing to me. After reading it several times, and going back to the equations and figures to which are referred, I understood what is actually being compared. I think it would help to explain in words that Eq.(1) refers to the Tapping formula, Eq.(3) refers to the proposed formula, and that Fig. 5 refers to the relation between the Swiss weather radar and the reference. If that is more clearly stated, it is much easier to understand. In terms of what this section shows, I think it is more a proof that the assumption that the weather radar calibration did not change than that it shows the value of the new formula (I can conclude that the old one is also fine for this case).
- line 383: "we have thought of replacing". Was the missing date replaced by the 5 March 2026 data, or did you only think of it? It's not clear from the text.
- Section 5.3: given this explanation, it seems like the behavior depends largely on the distribution of magnetic fields. Is there information about typical distributions available that can help explain the results?
- line 457: it now reads like "ANJ" is the abbreviation for "Albis". Please rephrase this sentence for clarity.
- lines 473-475: consider adding something about that the processing of the reference data should be optimized in terms of integration time and scanning strategy that maximizes the Sun-to-noise ratio.
Typos that I noticed:
- line 20: "tool" should be "tools"
- line 25 "fluctuations" should be "fluctuation"
- line 366: "2006" should be "2026".
- line 385: "tof such referemce" should be "of this reference" or something similar.
- line 412: "whhose" should be "whose"
Citation: https://doi.org/10.5194/egusphere-2026-2097-RC2 - AC2: 'Reply on RC2', Andrea Battaglia, 07 Aug 2026
Video supplement
The slowly varying solar component measured in microwave observations with EOVSA Andrea Francesco Battaglia https://doi.org/10.5446/73084
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 168 | 49 | 17 | 234 | 11 | 11 |
- HTML: 168
- PDF: 49
- XML: 17
- Total: 234
- BibTeX: 11
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Solar flux measurements are very important for weather radar networks and the (dualpol) data quality. So to revisit parameterizations, that relate flux measurements at a given frequency (S-Band) to the weather radar frequency are really important. Especially if it is well known, that there are "problems" when the sun is active. The authors provide a clear and mostly well written analysis. Even though the overall improvement of the new parameterization is rather small, it highlights the necessity for high quality solar flux observation especially in Europe, where a comparable site like DRAO is missing. A question to the authors: do you apply this new formulation now as a standard to the Suisse radars?
I've made my comments in the attached manuscript. I'd suggest to check English from a native speaker (though the article is in general well written) , or to re-read it carefully. I made some suggestions along the manuscript.