the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 29 Apr 2025
The Arctic Weather Satellite radiometer
Abstract. The Arctic Weather Satellite (AWS) is a project led by the European Space Agency (ESA) that has several novel aspects. From a technical perspective, it serves as a demonstrator of how to expand the network of operational satellite-based microwave sensors cost-effectively and acts as the proto-flight model for a suggested constellation of satellites, denoted as EUMETSAT Polar System (EPS) Sterna. The design philosophy has been to reduce complexity and instead focus the efforts on critical parts and characterise the instrument well before the launch. The single instrument onboard is a 19-channel microwave cross-track radiometer. There are 15 channels covering ranges around 54, 89 and 174 GHz. These are channels similar to ones found on existing sensors, however, thanks to the short development process, allowing use of more modern and recent technology, the performance and resolution of these channels on AWS exceed or match similar sensors, despite being a small satellite. Additionally, four channels around 325.15 GHz form a completely new frequency band for observations from space. The addition of these new channels aims to improve sensitivity to ice hydrometeors.
In this article, we outline the mission and describe the instrument to support the usage of radiances measured by AWS. The satellite was launched in August 2024, and the status towards the end of the commissioning phase is reflected here. For example, a characterisation of the noise performance is provided, showing that the target specifications have been met, for most channels with a margin. This is except for two channels identified to have technical issues already before the launch. If EPS-Sterna is selected by EUMETSAT, these and other identified problems will be corrected, but otherwise the constellation is expected to consist of recurrent models of AWS with minor modifications.
- Preprint
(5436 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2025-1769', Anonymous Referee #1, 02 May 2025
The manuscript entitled "The Arctic Weather Satellite Radiometer" provides a concise yet comprehensive overview of the AWS mission concept, its objectives, future developments (Sterna), and the functioning of the microwave radiometer. I particularly appreciated the completeness of the work—covering the description of the radiometer, the scan geometry, the antenna response, the data formats and the dissemination chain.
The manuscript also explains how the achievement of the SRF and NEDT targets was assessed, while highlighting and discussing the issues encountered and their causes (NEDT targets unmet for AWS18 and AWS42). I also appreciated the analysis of the Jacobians under all-sky conditions and the application to a real case (Cyclone Dikeledi), as these provide preliminary insight into the response at 325 GHz. While the analyses are simplified, I understand that a concise presentation is appropriate in this context.
In my opinion, the paper is well organized, of high quality, and should be published, subject only to minor/technical corrections.
Minor/technical corrections and suggestions:
- Line 321: The hyperlink/URL does not appear to be functional.
- Lines 80 and 538: The text refers to the "new space" philosophy/principles. It would be helpful to include a reference or add one or two sentences briefly explaining this concept for readers who may not be familiar with the term.
- Figures 8, 12, 13, 14, 15, and 16: Axis values are repeated and not easily interpretable. It would be beneficial to clarify what each axis represents and to indicate the units of measurement, where applicable. Additionally, the colorbar scale in Figure 15 should be corrected.
Minor textual corrections:
- Line 113: "as this set of channels offer a better basis" → "as this set of channels offers a better basis"
- Line 151: "is rectangular and allow for a" → "is rectangular and allows for a"
- Line 259: "The contribution of the far sidelobes are corrected" → "The contribution of the far sidelobes is corrected"
- Line 260: "A set of pitch and roll manoeuvres were performed" → "A set of pitch and roll manoeuvres was performed"
- Line 375: "the impact of clouds and precipitation (i.e. hydrometeors) are negligible" → "the impact of clouds and precipitation (i.e. hydrometeors) is negligible"
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC1 -
AC2: 'Reply on RC1', Patrick Eriksson, 08 May 2025
We thank you for the kind words about the manuscript.
Unfortunately, the NWP-SAF page with AWS SRFs has recently been moved. The new link is:
https://nwp-saf.eumetsat.int/downloads/rtcoef_info/mw_srf/rtcoef_aws_1_aws_srf.html
As explained in an author comment, something went wrong when converting our PDF manuscript to the preprint version. Thanks to your comment, we discovered this. The issues you brought up should now have been fixed. Please note that different colour scales are needed in Fig. 15 to clearly illustrate the features discussed in the text.
The remaining language points will be fixed in the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2025-1769-AC2 -
AC3: 'Reply on RC1', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC3-supplement.pdf
-
AC1: 'PDF of preprint corrected May 7!', Patrick Eriksson, 08 May 2025
A new version of the PDF was created on May 7. If you downloaded the preprint earlier than this date, please download it again. Your browser's cache may need to be cleaned to get the new version. You likely have the old, incorrect version if you notice labels and text missing in figures.
Background: An error occurred when the first version of the preprint was generated based on your PDF. Labels and some text of the figures were missing. We thank referee #1 for pointing this out. After contacting Copernicus, they have now generated a new version with the conversion errors corrected.
Citation: https://doi.org/10.5194/egusphere-2025-1769-AC1 -
RC2: 'Comment on egusphere-2025-1769', Anonymous Referee #2, 16 May 2025
General Comment:
This paper will be an important contribution to the AWS mission and payload that will tie together a collection of AWS papers and reports. I found the order of the sections to be awkward, and suggest following the order closer to the ESA SMOS paper referenced below. The new order brings the “why” up front where it’s now presently in the middle of the paper. Besides minor edits for the new flow, I see this mostly as cut-and-past of the present sections.
Present Contents order:
- Intro. 2. Radiometer 3. Mission 4. Pre-launch 5. Sim. 6. L1b quicklook 7. Conclusion
New Contents order:
- Intro 2. Science Obj. (previously the sim. Section) 3. Mission 4. Radiometer 5. Pre-launch 6. L1b quicklook 7. Conclusion
H. Kerr, P. Waldteufel, J. . -P. Wigneron, J. Martinuzzi, J. Font and M. Berger, "Soil moisture retrieval from space: the Soil Moisture and Ocean Salinity (SMOS) mission," in IEEE Transactions on Geoscience and Remote Sensing, vol. 39, no. 8, pp. 1729-1735, Aug. 2001, doi: 10.1109/36.942551
Specific Comment:
- The paper’s title should include the word Mission to go with Radiometer
- Merge radiometer background (Sect. 2.1) with intro section (Sect. 1.0). It wasn’t until 2.1 that I learned why Artic is in the name.
- With the transition of the Simulation section to Science Objectives, I’d move Table 1 to the new Sci. Obj. section to capture the targeted design specifications that meet the mission sci. obj. I see a statement on Line 368 of radiometric accuracy, but no target in Table 1.
- Per the ATBD, L1b is brightness temp., but the paper seems to consistently call it antenna temperature (traditionally ant. temp. is L1a, but AWS doesn’t seem to offer it).
- Sect. 4.0 Pre-launch Characterization: I didn’t see any discussion on non-linearity. The only thing I could find in the ATBD was a detector nonlinearity correction without mention on how it was derived. Also, it doesn’t look like AWS radiometer went through TVac calibration where an instrument-level non-linearity correct is derived? While I’m an advocate of the New Space approach, the non-linearity is very hard to derive on-orbit with its dependency on both scene brightness temperature and instrument kinetic temperature. What did the team do convince themselves that the non-linearity was marginal? While the detector NL can be the primary driver of system NL, how close the RF amplifiers are to the 1dB compression can be critical, too.
- Sect. 4.5: Can you include references or more details on this section? There seems to be all results with no information on how it was calculated.
- NEDT consistency (Sect. 2.3, 4.4, & 6.2): It doesn’t seem that the comparison of the various stated NEDTs are consistent. Just looking for more details and not re-analysis. 6 is the intrinsic NEDT equation, but there are other sources of NEDT contribution:
- S. Hersman and G. A. Poe, "Sensitivity of the Total Power Radiometer with Periodic Absolute Calibration," in IEEE Transactions on Microwave Theory and Techniques, vol. 29, no. 1, pp. 32-40, Jan. 1981, doi: 10.1109/TMTT.1981.1130283.
- Were they all convert to a 300K scene? What technique? Using rec. temp. or extrapolating NEDT at two other scene temps? Note that various estimates of NEDT are impacted by the calibration period (i.e., flicker noise).
- The OG NEDT: did it use an ambient target in the Earth view that was assumed to be 300K? Since it was in operational mode (i.e., flight calibration period), this would be a good estimate of the NEDT.
- The IO NEDT: I’d assume this estimate comes from the calibration sectors? Were these extrapolated to 300K? Note, this technique doesn’t typically have all of the other NEDT contributors.
- Sect. 2.6 Scan Sequence: The ATBD mentioned two potential cold sky sectors (ATBD Sect. 3.6.1) and that “the cold sky measurement depends on the orbit and occur before or after the earth scene.” What was the final result? I’m interested for a couple of reasons: a) can you use both b) SSO satellites usually have a sun shield as the solar beta angle is fairly constant off to one side of the spacecraft throughout the year. I’m wondering how that impacted calibration with potential illumination of the rotating reflector.
- Sect. 3.1 Platform:
- What, if any, thermal control of the radiometer is there?
- Can you add something on geolocation target accuracy and/or point to your sensitivity study in Table 2 of AWS-OMN-RP-0002 Issue C? This report seems to have more info. than the referenced AWS-SMHI-RP-0002 Issue A?
- What on-orbit verification (e.g., Coastline Inflection Point technique) will be used to tune the geolocation parameters?
- What pre-launch measurements were made to confirm pointing knowledge? The antenna pattern is a start, but doesn’t include the alignment/transform between the instrument and the spacecraft LVLH control. Is it just the close placement of the star trackers to the payload and use a post-launch empirical correction based on CIP?
- Sect. 3.4 L1b:
- Consider adding a less detailed version of the ATBD Fig. 6 “Overview of the AWS instr. signal proc. chain” be included that allows the activities in the Pre-launch section be tied to the calibration algorithm?
- Regarding Line 260 starting with “The final L1b…”: Is this 2.5 times the channel’s FWHM projection on the surface?
- Sect 4.1 SRF: on line 307 “should be minimal” Is there any reference for this statement? The fact that the Front-end and back-ends were measured separately and combined can have SRF inaccuracies if the VSWR (i.e., match) of the back-end isn’t sufficient. The same VSWR issue can arise between the antenna system and Front-end. You can point to your SRF impact section to support that any residual inaccuracies from combining the separate SRFs would be minimal as a majority of the inaccuracies are captured by the SRF incorporated into RTTOV.
- Sect. 5.3 Impact of SRF:
- Please add more info. on the “five distinct atmospheres” or a reference.
- Regarding AWS14’s passband crossing over the absorption line, what spectral sampling did you use? Did you have a data point right on the absorption line? How did the RTTOV folks handle this?
- I’m not clear on the take away or point of the last paragraph. Because AWS measured the SRF and made them available to the community (e.g., RTTOV), you’re mitigating the impact of not having stringent bandpass requirements (definitely in line with “New Space” approach of replacing knowledge of passband instead of strict controlling of it). I think you should emphasize that you’re saying that any residual SRF uncertainties are marginal. Using the ATMS has an example isn’t the same in my opinion because they had the strict requirements in place, so the SRF was fairly close to the boxcar.
- Sect. 5.4 Impact of SRF: Is the impact in Table 4 the difference between zeroing out the 03 or using the climatology mean O3 profile? That is, are the numbers in Table 4 the residual error of using the mean O3 profile?
Technical Corrections:
None at this time.
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC2 -
AC4: 'Reply on RC2', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC4-supplement.pdf
-
RC3: 'Comment on egusphere-2025-1769', Anonymous Referee #3, 20 May 2025
This paper describes on the AWS microwave sounding instrument and the associated spaceflight mission to demonstrate its capabilities and performance. It is generally well-written and provides many engineering details on the design, construction, and testing of the instrument, and gives some insights into the decisions behind design choices and descopes relative to currently operational microwave sounders in the context of cost savings and suitability for future constellation implementation.
I will start with a few "big picture" comments. There has been considerable recent interest and activity in the area of earth observing small satellites that typically require sacrifices in performance, capability, and/or reliability in order to reduce costs and facilitate their implementation in an effective and expeditious manner. This paper does not adequately mention or reference prior art or previous work in this area (small satellite microwave radiometers/missions, in particular) to provide context or background for this work. As an example, here is an article that summarizes some of the recent missions and aspirations in this field: https://www.science.org/doi/pdf/10.1126/science.adr3312 - there are ongoing and upcoming small satellite weather sensing projects from the public and private sectors that are described herein. To what extent does the present work draw inspiration from, build upon, and surpass this prior work? Just a few sentences or a small paragraph on this would substantially improve the contextual narrative and set the stage for the impressive work to be described in the paper.
Regarding the technical content of the paper, many details are given in the paper, but I think there are a few basic pieces of information that are missing that would be of keen interest to the readership. For example, what is the mass, volume, power consumption, and data rate of the instrument (and even better, simple comparisons to what is flying now or is planned to fly)? There is some of this kind of information presented for the satellite bus, but instrument parameters would be more meaningful.
The instrument does not include "traditional" channels near 24 and 31 GHz due to the size of the reflector that would be needed. This is a reasonable design trade, but those are very important channels for the retrieval of total precipitable water - it would be useful to simply discuss the impact of this omission and how it might be mitigated.
Another design choice appears to be to fly the satellite at a lower altitude than current operational sensors (600 km versus 817 km). This of course yields better spatial resolution for a given reflector size, but at the cost of substantial footprint broadening at larger instrument scan angles. The spatial resolution at the scan edges (55 degrees) could be much too coarse for effective operational use. Some discussion of this point would be helpful - what are the AWS scan-edge resolutions and how do these compare with present systems, for example? And how does this impact the planned approach of spatially combining/aligning the footprints for the various bands?
Another compromise is the choice (reasonably) of a constant scan velocity versus and non-constant scan velocity (whereby the scan is slower over the earth and faster away from the earth, so that the integration time for earth-viewing footprints is longer, thus noise is lower). The penalty paid for this is approximately sqrt(2) in noise amplification (assuming scan accelerations consistent with current operational sounders). Again, some discussion of the regret of this would be useful, especially in light of the profound impact of the radiance noise in numerical weather prediction applications.
The terms "inter-pixel error" and "orbital stability" are used without definition - what are these and how were they quantitatively assessed?
I believe the L-band satellite communications transmitter frequency (1.7 GHz) falls within the IF bands of the high-frequency receivers - was any prelaunch testing done to ensure electromagnetic compatibility of the spacecraft hardware and the radiometer? Does the radiometer noise measured on-orbit increase when the communications transmitter is on?
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC3 -
AC5: 'Reply on RC3', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC5-supplement.pdf
-
AC5: 'Reply on RC3', Patrick Eriksson, 08 Jul 2025
-
CC1: 'Comments on egusphere-2025-1769', Tim Hewison, 26 May 2025
L175: The scan rate is constant, giving an along-track distance between footprints of about 9.0 km
- I calculate 8.2km, assuming an altitude of 595km. The orbital velocity would be 7.6km/s, which would correspond to 9.0km, but the track velocity at the Earth’s surface and corresponding scan spacing would be proportionately less by 6371/(6371+595).
§3.2: I suggest to make it clear than the operational altitude of AWS has been a fairly constant 599km since 2024-12-01.
- https://celestrak.org/NORAD/elements/graph-orbit-data.php?CATNR=60543
L315: Please highlight and justify the departure from the usual convention is to report bandwidths between 3dB points (not 6dB).
L344: The minimum and maximum FWHM of the nadir response for some selected frequencies are reported in Table 4.
- Is it reasonable to provide a mean value?
L349: Over what period were the standard deviations evaluated?
- The longer the period, the greater the fraction of 1/f noise will be included in this estimate of NEDT.
- It is also noteworthy that these OG values are generally higher than the In Orbit values, despite being evaluated at a lower antenna temperature (if true), where lower values would normally be expected.
L361 + L509: How was the In-Orbit NEDT evaluated? Is this based on Deep Space on OBCT views?
L364: How is the short-term stability defined?
L367: How were the inter-pixel error and orbital stability quantified?
L419: What are the 5 atmospheric scenarios used to define the SRF sensitivity?
L420: How is the SRF sensitivity actually quantified in Table 4? Is this the mean difference between the brightness temperatures simulated by ARTS with the actual SRF and with the boxcar approximation? The difference will be scene dependent – can you quantify its variance?
(Out of interest, I have conducted a similar analysis, based on the RTTOV coefficients for actual SRF and boxcar approximation to the specification, using a global sample from ERA5 – and found much mean differences of a similar magnitude to those reported here in clear sky – but much larger differences when cloud was added.)Citation: https://doi.org/10.5194/egusphere-2025-1769-CC1 -
AC6: 'Reply on CC1', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC6-supplement.pdf
-
RC4: 'Comment on egusphere-2025-1769', Anonymous Referee #4, 26 May 2025
Review of " The Arctic Weather Satellite radiometer" by Eriksson et al.
The manuscript describes the Arctic Weather Satellite (AWS) radiometer instrument, which is a prototype demonstrator for the envisioned EPS-Sterna constellation. A great amount of detail is spent describing the instrument design, subsystems, and pre-launch measurements of spectral and spatial response functions, with a focus on characteristics that are different from previous cross-track instruments such as ATMS and MHS. Since some of the channels (325 GHz) are new, there is an analysis of the expected impacts of these channels due to their increased sensitivity to cloud ice from the 183 GHz channels. Finally, some post-launch measurements and performance statistics are presented, indicating general good agreement with pre-launch expectations.
Summary of review:
This manuscript has a good balance of technical information about the instrument (which along would probably not be suitable for AMT) along with a brief but informative analysis of the expected information content in the novel channel set (which is quite suitable for AMT). Considering that the instrument has been on orbit for several months at the time this manuscript was submitted, I would expect more comprehensive on-orbit performance information than just the individual channel noise estimates. It would be useful to see these, along with comparisons to heritage instruments such as ATMS and MHS, to demonstrate the viability of the "new space", small satellite approach. Otherwise, this is an excellent and informative manuscript for users of AWS data.Specific comments:
Figure 13: It would be helpful to illustrate the locations of the AWS 3x and 4x channel bands in these plots.Line 215: It is mentioned that the instrument is designed to minimize geolocation error here, but no geolocation accuracy statistics are presented. Especially considering the novel feedhorn arrangement, this would be particularly useful to assess in this manuscript.
Lines 365-368: A lot of information is given here about parameters that are important for real-world radiometric accuracy and at the end of the paragraph, it is stated that the accuracy is better than 1 K for all channels. However, how exactly was this determined (e.g., how was non-linearity assessed, was the on-orbit thermal cycle modeled during the calibration testing, what on-orbit maneuvers were used to calculate spillover)?
Line 434: I believe this should be "The impact of measured SRFs is hard to *assess*."
Table 4: Are the on-orbit NEDT values also scaled to a 300 K scene? I would assume so but it is not explicitly stated. Also, are they derived from the variance of counts in the warm calibration sector, cold calibration sector, or both (and what receiver temperature was assumed)? Also, since it is mentioned in the Discussion section, it would be nice to have the striping index added to this table, since that is a standard performance parameter for microwave radiometers.
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC4 -
AC7: 'Reply on RC4', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC7-supplement.pdf
-
AC7: 'Reply on RC4', Patrick Eriksson, 08 Jul 2025
Status: closed
-
RC1: 'Comment on egusphere-2025-1769', Anonymous Referee #1, 02 May 2025
The manuscript entitled "The Arctic Weather Satellite Radiometer" provides a concise yet comprehensive overview of the AWS mission concept, its objectives, future developments (Sterna), and the functioning of the microwave radiometer. I particularly appreciated the completeness of the work—covering the description of the radiometer, the scan geometry, the antenna response, the data formats and the dissemination chain.
The manuscript also explains how the achievement of the SRF and NEDT targets was assessed, while highlighting and discussing the issues encountered and their causes (NEDT targets unmet for AWS18 and AWS42). I also appreciated the analysis of the Jacobians under all-sky conditions and the application to a real case (Cyclone Dikeledi), as these provide preliminary insight into the response at 325 GHz. While the analyses are simplified, I understand that a concise presentation is appropriate in this context.
In my opinion, the paper is well organized, of high quality, and should be published, subject only to minor/technical corrections.
Minor/technical corrections and suggestions:
- Line 321: The hyperlink/URL does not appear to be functional.
- Lines 80 and 538: The text refers to the "new space" philosophy/principles. It would be helpful to include a reference or add one or two sentences briefly explaining this concept for readers who may not be familiar with the term.
- Figures 8, 12, 13, 14, 15, and 16: Axis values are repeated and not easily interpretable. It would be beneficial to clarify what each axis represents and to indicate the units of measurement, where applicable. Additionally, the colorbar scale in Figure 15 should be corrected.
Minor textual corrections:
- Line 113: "as this set of channels offer a better basis" → "as this set of channels offers a better basis"
- Line 151: "is rectangular and allow for a" → "is rectangular and allows for a"
- Line 259: "The contribution of the far sidelobes are corrected" → "The contribution of the far sidelobes is corrected"
- Line 260: "A set of pitch and roll manoeuvres were performed" → "A set of pitch and roll manoeuvres was performed"
- Line 375: "the impact of clouds and precipitation (i.e. hydrometeors) are negligible" → "the impact of clouds and precipitation (i.e. hydrometeors) is negligible"
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC1 -
AC2: 'Reply on RC1', Patrick Eriksson, 08 May 2025
We thank you for the kind words about the manuscript.
Unfortunately, the NWP-SAF page with AWS SRFs has recently been moved. The new link is:
https://nwp-saf.eumetsat.int/downloads/rtcoef_info/mw_srf/rtcoef_aws_1_aws_srf.html
As explained in an author comment, something went wrong when converting our PDF manuscript to the preprint version. Thanks to your comment, we discovered this. The issues you brought up should now have been fixed. Please note that different colour scales are needed in Fig. 15 to clearly illustrate the features discussed in the text.
The remaining language points will be fixed in the revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2025-1769-AC2 -
AC3: 'Reply on RC1', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC3-supplement.pdf
-
AC1: 'PDF of preprint corrected May 7!', Patrick Eriksson, 08 May 2025
A new version of the PDF was created on May 7. If you downloaded the preprint earlier than this date, please download it again. Your browser's cache may need to be cleaned to get the new version. You likely have the old, incorrect version if you notice labels and text missing in figures.
Background: An error occurred when the first version of the preprint was generated based on your PDF. Labels and some text of the figures were missing. We thank referee #1 for pointing this out. After contacting Copernicus, they have now generated a new version with the conversion errors corrected.
Citation: https://doi.org/10.5194/egusphere-2025-1769-AC1 -
RC2: 'Comment on egusphere-2025-1769', Anonymous Referee #2, 16 May 2025
General Comment:
This paper will be an important contribution to the AWS mission and payload that will tie together a collection of AWS papers and reports. I found the order of the sections to be awkward, and suggest following the order closer to the ESA SMOS paper referenced below. The new order brings the “why” up front where it’s now presently in the middle of the paper. Besides minor edits for the new flow, I see this mostly as cut-and-past of the present sections.
Present Contents order:
- Intro. 2. Radiometer 3. Mission 4. Pre-launch 5. Sim. 6. L1b quicklook 7. Conclusion
New Contents order:
- Intro 2. Science Obj. (previously the sim. Section) 3. Mission 4. Radiometer 5. Pre-launch 6. L1b quicklook 7. Conclusion
H. Kerr, P. Waldteufel, J. . -P. Wigneron, J. Martinuzzi, J. Font and M. Berger, "Soil moisture retrieval from space: the Soil Moisture and Ocean Salinity (SMOS) mission," in IEEE Transactions on Geoscience and Remote Sensing, vol. 39, no. 8, pp. 1729-1735, Aug. 2001, doi: 10.1109/36.942551
Specific Comment:
- The paper’s title should include the word Mission to go with Radiometer
- Merge radiometer background (Sect. 2.1) with intro section (Sect. 1.0). It wasn’t until 2.1 that I learned why Artic is in the name.
- With the transition of the Simulation section to Science Objectives, I’d move Table 1 to the new Sci. Obj. section to capture the targeted design specifications that meet the mission sci. obj. I see a statement on Line 368 of radiometric accuracy, but no target in Table 1.
- Per the ATBD, L1b is brightness temp., but the paper seems to consistently call it antenna temperature (traditionally ant. temp. is L1a, but AWS doesn’t seem to offer it).
- Sect. 4.0 Pre-launch Characterization: I didn’t see any discussion on non-linearity. The only thing I could find in the ATBD was a detector nonlinearity correction without mention on how it was derived. Also, it doesn’t look like AWS radiometer went through TVac calibration where an instrument-level non-linearity correct is derived? While I’m an advocate of the New Space approach, the non-linearity is very hard to derive on-orbit with its dependency on both scene brightness temperature and instrument kinetic temperature. What did the team do convince themselves that the non-linearity was marginal? While the detector NL can be the primary driver of system NL, how close the RF amplifiers are to the 1dB compression can be critical, too.
- Sect. 4.5: Can you include references or more details on this section? There seems to be all results with no information on how it was calculated.
- NEDT consistency (Sect. 2.3, 4.4, & 6.2): It doesn’t seem that the comparison of the various stated NEDTs are consistent. Just looking for more details and not re-analysis. 6 is the intrinsic NEDT equation, but there are other sources of NEDT contribution:
- S. Hersman and G. A. Poe, "Sensitivity of the Total Power Radiometer with Periodic Absolute Calibration," in IEEE Transactions on Microwave Theory and Techniques, vol. 29, no. 1, pp. 32-40, Jan. 1981, doi: 10.1109/TMTT.1981.1130283.
- Were they all convert to a 300K scene? What technique? Using rec. temp. or extrapolating NEDT at two other scene temps? Note that various estimates of NEDT are impacted by the calibration period (i.e., flicker noise).
- The OG NEDT: did it use an ambient target in the Earth view that was assumed to be 300K? Since it was in operational mode (i.e., flight calibration period), this would be a good estimate of the NEDT.
- The IO NEDT: I’d assume this estimate comes from the calibration sectors? Were these extrapolated to 300K? Note, this technique doesn’t typically have all of the other NEDT contributors.
- Sect. 2.6 Scan Sequence: The ATBD mentioned two potential cold sky sectors (ATBD Sect. 3.6.1) and that “the cold sky measurement depends on the orbit and occur before or after the earth scene.” What was the final result? I’m interested for a couple of reasons: a) can you use both b) SSO satellites usually have a sun shield as the solar beta angle is fairly constant off to one side of the spacecraft throughout the year. I’m wondering how that impacted calibration with potential illumination of the rotating reflector.
- Sect. 3.1 Platform:
- What, if any, thermal control of the radiometer is there?
- Can you add something on geolocation target accuracy and/or point to your sensitivity study in Table 2 of AWS-OMN-RP-0002 Issue C? This report seems to have more info. than the referenced AWS-SMHI-RP-0002 Issue A?
- What on-orbit verification (e.g., Coastline Inflection Point technique) will be used to tune the geolocation parameters?
- What pre-launch measurements were made to confirm pointing knowledge? The antenna pattern is a start, but doesn’t include the alignment/transform between the instrument and the spacecraft LVLH control. Is it just the close placement of the star trackers to the payload and use a post-launch empirical correction based on CIP?
- Sect. 3.4 L1b:
- Consider adding a less detailed version of the ATBD Fig. 6 “Overview of the AWS instr. signal proc. chain” be included that allows the activities in the Pre-launch section be tied to the calibration algorithm?
- Regarding Line 260 starting with “The final L1b…”: Is this 2.5 times the channel’s FWHM projection on the surface?
- Sect 4.1 SRF: on line 307 “should be minimal” Is there any reference for this statement? The fact that the Front-end and back-ends were measured separately and combined can have SRF inaccuracies if the VSWR (i.e., match) of the back-end isn’t sufficient. The same VSWR issue can arise between the antenna system and Front-end. You can point to your SRF impact section to support that any residual inaccuracies from combining the separate SRFs would be minimal as a majority of the inaccuracies are captured by the SRF incorporated into RTTOV.
- Sect. 5.3 Impact of SRF:
- Please add more info. on the “five distinct atmospheres” or a reference.
- Regarding AWS14’s passband crossing over the absorption line, what spectral sampling did you use? Did you have a data point right on the absorption line? How did the RTTOV folks handle this?
- I’m not clear on the take away or point of the last paragraph. Because AWS measured the SRF and made them available to the community (e.g., RTTOV), you’re mitigating the impact of not having stringent bandpass requirements (definitely in line with “New Space” approach of replacing knowledge of passband instead of strict controlling of it). I think you should emphasize that you’re saying that any residual SRF uncertainties are marginal. Using the ATMS has an example isn’t the same in my opinion because they had the strict requirements in place, so the SRF was fairly close to the boxcar.
- Sect. 5.4 Impact of SRF: Is the impact in Table 4 the difference between zeroing out the 03 or using the climatology mean O3 profile? That is, are the numbers in Table 4 the residual error of using the mean O3 profile?
Technical Corrections:
None at this time.
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC2 -
AC4: 'Reply on RC2', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC4-supplement.pdf
-
RC3: 'Comment on egusphere-2025-1769', Anonymous Referee #3, 20 May 2025
This paper describes on the AWS microwave sounding instrument and the associated spaceflight mission to demonstrate its capabilities and performance. It is generally well-written and provides many engineering details on the design, construction, and testing of the instrument, and gives some insights into the decisions behind design choices and descopes relative to currently operational microwave sounders in the context of cost savings and suitability for future constellation implementation.
I will start with a few "big picture" comments. There has been considerable recent interest and activity in the area of earth observing small satellites that typically require sacrifices in performance, capability, and/or reliability in order to reduce costs and facilitate their implementation in an effective and expeditious manner. This paper does not adequately mention or reference prior art or previous work in this area (small satellite microwave radiometers/missions, in particular) to provide context or background for this work. As an example, here is an article that summarizes some of the recent missions and aspirations in this field: https://www.science.org/doi/pdf/10.1126/science.adr3312 - there are ongoing and upcoming small satellite weather sensing projects from the public and private sectors that are described herein. To what extent does the present work draw inspiration from, build upon, and surpass this prior work? Just a few sentences or a small paragraph on this would substantially improve the contextual narrative and set the stage for the impressive work to be described in the paper.
Regarding the technical content of the paper, many details are given in the paper, but I think there are a few basic pieces of information that are missing that would be of keen interest to the readership. For example, what is the mass, volume, power consumption, and data rate of the instrument (and even better, simple comparisons to what is flying now or is planned to fly)? There is some of this kind of information presented for the satellite bus, but instrument parameters would be more meaningful.
The instrument does not include "traditional" channels near 24 and 31 GHz due to the size of the reflector that would be needed. This is a reasonable design trade, but those are very important channels for the retrieval of total precipitable water - it would be useful to simply discuss the impact of this omission and how it might be mitigated.
Another design choice appears to be to fly the satellite at a lower altitude than current operational sensors (600 km versus 817 km). This of course yields better spatial resolution for a given reflector size, but at the cost of substantial footprint broadening at larger instrument scan angles. The spatial resolution at the scan edges (55 degrees) could be much too coarse for effective operational use. Some discussion of this point would be helpful - what are the AWS scan-edge resolutions and how do these compare with present systems, for example? And how does this impact the planned approach of spatially combining/aligning the footprints for the various bands?
Another compromise is the choice (reasonably) of a constant scan velocity versus and non-constant scan velocity (whereby the scan is slower over the earth and faster away from the earth, so that the integration time for earth-viewing footprints is longer, thus noise is lower). The penalty paid for this is approximately sqrt(2) in noise amplification (assuming scan accelerations consistent with current operational sounders). Again, some discussion of the regret of this would be useful, especially in light of the profound impact of the radiance noise in numerical weather prediction applications.
The terms "inter-pixel error" and "orbital stability" are used without definition - what are these and how were they quantitatively assessed?
I believe the L-band satellite communications transmitter frequency (1.7 GHz) falls within the IF bands of the high-frequency receivers - was any prelaunch testing done to ensure electromagnetic compatibility of the spacecraft hardware and the radiometer? Does the radiometer noise measured on-orbit increase when the communications transmitter is on?
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC3 -
AC5: 'Reply on RC3', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC5-supplement.pdf
-
AC5: 'Reply on RC3', Patrick Eriksson, 08 Jul 2025
-
CC1: 'Comments on egusphere-2025-1769', Tim Hewison, 26 May 2025
L175: The scan rate is constant, giving an along-track distance between footprints of about 9.0 km
- I calculate 8.2km, assuming an altitude of 595km. The orbital velocity would be 7.6km/s, which would correspond to 9.0km, but the track velocity at the Earth’s surface and corresponding scan spacing would be proportionately less by 6371/(6371+595).
§3.2: I suggest to make it clear than the operational altitude of AWS has been a fairly constant 599km since 2024-12-01.
- https://celestrak.org/NORAD/elements/graph-orbit-data.php?CATNR=60543
L315: Please highlight and justify the departure from the usual convention is to report bandwidths between 3dB points (not 6dB).
L344: The minimum and maximum FWHM of the nadir response for some selected frequencies are reported in Table 4.
- Is it reasonable to provide a mean value?
L349: Over what period were the standard deviations evaluated?
- The longer the period, the greater the fraction of 1/f noise will be included in this estimate of NEDT.
- It is also noteworthy that these OG values are generally higher than the In Orbit values, despite being evaluated at a lower antenna temperature (if true), where lower values would normally be expected.
L361 + L509: How was the In-Orbit NEDT evaluated? Is this based on Deep Space on OBCT views?
L364: How is the short-term stability defined?
L367: How were the inter-pixel error and orbital stability quantified?
L419: What are the 5 atmospheric scenarios used to define the SRF sensitivity?
L420: How is the SRF sensitivity actually quantified in Table 4? Is this the mean difference between the brightness temperatures simulated by ARTS with the actual SRF and with the boxcar approximation? The difference will be scene dependent – can you quantify its variance?
(Out of interest, I have conducted a similar analysis, based on the RTTOV coefficients for actual SRF and boxcar approximation to the specification, using a global sample from ERA5 – and found much mean differences of a similar magnitude to those reported here in clear sky – but much larger differences when cloud was added.)Citation: https://doi.org/10.5194/egusphere-2025-1769-CC1 -
AC6: 'Reply on CC1', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC6-supplement.pdf
-
RC4: 'Comment on egusphere-2025-1769', Anonymous Referee #4, 26 May 2025
Review of " The Arctic Weather Satellite radiometer" by Eriksson et al.
The manuscript describes the Arctic Weather Satellite (AWS) radiometer instrument, which is a prototype demonstrator for the envisioned EPS-Sterna constellation. A great amount of detail is spent describing the instrument design, subsystems, and pre-launch measurements of spectral and spatial response functions, with a focus on characteristics that are different from previous cross-track instruments such as ATMS and MHS. Since some of the channels (325 GHz) are new, there is an analysis of the expected impacts of these channels due to their increased sensitivity to cloud ice from the 183 GHz channels. Finally, some post-launch measurements and performance statistics are presented, indicating general good agreement with pre-launch expectations.
Summary of review:
This manuscript has a good balance of technical information about the instrument (which along would probably not be suitable for AMT) along with a brief but informative analysis of the expected information content in the novel channel set (which is quite suitable for AMT). Considering that the instrument has been on orbit for several months at the time this manuscript was submitted, I would expect more comprehensive on-orbit performance information than just the individual channel noise estimates. It would be useful to see these, along with comparisons to heritage instruments such as ATMS and MHS, to demonstrate the viability of the "new space", small satellite approach. Otherwise, this is an excellent and informative manuscript for users of AWS data.Specific comments:
Figure 13: It would be helpful to illustrate the locations of the AWS 3x and 4x channel bands in these plots.Line 215: It is mentioned that the instrument is designed to minimize geolocation error here, but no geolocation accuracy statistics are presented. Especially considering the novel feedhorn arrangement, this would be particularly useful to assess in this manuscript.
Lines 365-368: A lot of information is given here about parameters that are important for real-world radiometric accuracy and at the end of the paragraph, it is stated that the accuracy is better than 1 K for all channels. However, how exactly was this determined (e.g., how was non-linearity assessed, was the on-orbit thermal cycle modeled during the calibration testing, what on-orbit maneuvers were used to calculate spillover)?
Line 434: I believe this should be "The impact of measured SRFs is hard to *assess*."
Table 4: Are the on-orbit NEDT values also scaled to a 300 K scene? I would assume so but it is not explicitly stated. Also, are they derived from the variance of counts in the warm calibration sector, cold calibration sector, or both (and what receiver temperature was assumed)? Also, since it is mentioned in the Discussion section, it would be nice to have the striping index added to this table, since that is a standard performance parameter for microwave radiometers.
Citation: https://doi.org/10.5194/egusphere-2025-1769-RC4 -
AC7: 'Reply on RC4', Patrick Eriksson, 08 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1769/egusphere-2025-1769-AC7-supplement.pdf
-
AC7: 'Reply on RC4', Patrick Eriksson, 08 Jul 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 641 | 387 | 33 | 1,061 | 23 | 56 |
- HTML: 641
- PDF: 387
- XML: 33
- Total: 1,061
- BibTeX: 23
- EndNote: 56
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1