Subauroral Crosstalk in POES/Metop TED proton channels

Wissing, Jan Maik; Yakovchuk, Olesya

doi:https://doi.org/10.5194/angeo-2023-33

Preprints

https://doi.org/10.5194/angeo-2023-33

Preprints

09 Nov 2023

| 09 Nov 2023

Status: this preprint was under review for the journal ANGEO but the revision was not accepted.

Subauroral Crosstalk in POES/Metop TED proton channels

Jan Maik Wissing and Olesya Yakovchuk

Abstract. Particle measurements are used for various purposes. The medium energy channels on the POES/Metop satellites, the MEPED channels, are widely used, for example in atmospheric ionization models, their low energy (eV and keV) counterpart, the TED detector, is not that popular. However the recent rise of the ionization/climate model altitudes will lead to increased interest in these low energy particle measurements.

This paper analyses MEPED and TED particle data from NOAA POES and Metop during 2001–2018 and shows that in particlular the TED proton channels (but also MEPED electron channels and MEPED proton channels P2 and P3) are contaminated by high background levels a L < 6. In some channels that may surpass typical auroral levels. We determined a Kp and channel dependent latitude boundary that might be used as (preliminary) cut of the contaminated area.

Received: 28 Oct 2023 – Discussion started: 09 Nov 2023

Download & links

Jan Maik Wissing and Olesya Yakovchuk

Status: closed

RC1:
'Comment on angeo-2023-33', Anonymous Referee #1, 28 Nov 2023

The paper investigates the quality of the POES/Metop TED proton channels, in particular the cross-talk-produced background in proton channels at sub-auroral latitudes. The paper aims at establishing a Kp and channel dependent latitudinal cut of the cross-talk dominated area that should be excluded from analyses. The topic is important since the interest in the low-energy data is increasing due to the connection with atmospheric ionisation related to climate models.
The presentation is clear apart from some minor language issues, typos, and a small issue with Fig. 2 (see minor comments). The conclusions reached by the authors (highlighting cross-talk contaminated proton fluxes) are substantial. The title and the abstract are pertinent and understandable. Appropriate credit is given to earlier work, as far as I am aware.
The only substantive comment on the analysis I have is that the authors are clumping all longitudes into one although the effects of contamination seem to be very different in the various (APEX 110 km) longitude and between north and south. I wonder if the data could be used in a bit broader latitude range if the longitude would be pieced in sectors. Also the clear north-south asymmetry might be considered. The contaminating electrons drift eastward and their mirroring altitudes are different in the north and south hemispheres, which is clearly visible in Fig. 1: in the north, regions between 40° and 120° longitude, for example, are almost free of contamination. Thus, the latitude cutoffs found by the authors might be overly conservative in some spatial regions. This could be pointed out as a topic of further study.
Minor issues:

- line 26: "high relativistic" -> "highly relativistic"

- line 26: "Inspite of these data are" -> "Despite of these data being"

- line 50: "by dividing the energgy range" -> "by dividing by the width of the energy range"

- line 51: "geometric correction factors" -> "geometric factors" (?)

- line 55: "high-energetic" -> "high-energy"

- line 56: "high-energetic" -> "high-energy"

- line 78: "Therefore is" -> "Therefore it"

- line 87: "prominantly" -> "prominently"

- line 92: "in mod. APEX" spell out the abbreviation!

- line 111: "dropc" -> "drops"

- line 122: "(Yando et al., 2011)" -> "Yando et al. (2011)"

- Fig. 2: the vertical lines are not very visible at first glance on top of the curves. Please consider making them longer, for example, to allow the reader to spot them immediately.

Citation: https://doi.org/10.5194/angeo-2023-33-RC1
- CC1:
  'Reply on RC1', Jan Maik Wissing, 04 Dec 2023
  
  We thank reviewer #1 for the helpful comments and want to address the main issue regarding longitudinal differences and the north-south asymmetry in the crosstalk contribution. Reviewer #1 argues that the latitudinal cutoffs might be conservative in some (longitude and hemisphere specific) regions where the crosstalk contribution is low.
  We completely agree to that. The crosstalk contribution varies with longitude and shows hemispheric differences. Fig. 1 (left panel) gives an estimation on that. A minimum crosstalk contribution can be seen at about 70°E in the northern hemisphere. If we consider this as uncontaminated flux, the latitudinal cutoff removes about 4 degrees with real data which rapidly decreases to background levels.
  
  We updated Figure 3 and added another longitude, namely 160°E, where North and South have relatively low crosstalk contribution. The lower graph of the same color always represents the 160°E fluxes (or the according background count rates). Note that the color code also changed in order to increase the readability. The figure is a little crowded but we can see that all the TED proton channels show the same crosstalk maximum at L=4, even at 160°E. The dashed lines that mark the contaminated area also hold for both longitudes and crosstalk seems to be the main source of the measured TED proton particle flux within this area at both longitudes as well.
  
  Thus the presented approach of "global" cutoff latitudes is conservative and leads to an underestimation of the precipitating flux (area). But given that crosstalk seems to be the major subauroral contribution except for a very small longitudinal range (at about 60-80°E) in the northern hemisphere, a sectorial approach will probably not lead to significant improvement of the cutoff latitudes.
  As topic for further studies the authors would suggest a recalculation of all affected channels taking into account the TED background counts which might preserve the equatorial boundary in a more natural form than this approach does.
  
  Citation: https://doi.org/10.5194/angeo-2023-33-CC1
  - RC2: 'Reply on CC1', Anonymous Referee #1, 04 Dec 2023
    
    I see the point made by the authors and agree with their assessment. After they implemented the minor points I made, I can recommend acceptance.
    
    Citation: https://doi.org/10.5194/angeo-2023-33-RC2
    
    AC2: 'Reply on RC2', Olesya Yakovchuk, 26 Jan 2024
    
    All our answers can be found in the attachment of another reply.
    
    Citation: https://doi.org/10.5194/angeo-2023-33-AC2
RC3:
'Comment on angeo-2023-33', Anonymous Referee #2, 14 Dec 2023

The main objective of the paper is to investigate potential cross-talk in POES/Metop TED proton channels at sub-auroral latitudes and provide a preliminary Kp-dependent cutoff latitude where the observations can be used safely. It also presents evidence of cross-talk in the MEPED electron and proton detectors. I do, however, have major concerns about the validity of some of the results. Moreover, the following discussion are somewhat superficial and inaccurate which makes the conclusions potentially erroneous.

Major comments:
Missing discussion on the physics of cross-contamination:
Both introduction and discussion focus on relativistic electrons in the radiation belts being the source of cross-talk in the TED and MEPED proton and electron detectors. Besides stating that TED is a cylindrical electrostatic analyzer and MEPED has a passive shielding, there is no technical specification nor discussion on how the potential cross-talk from relativistic electrons could happen. I recommend a short overview of relevant technical specifications on which type of particles and and associated energies the detectors are designed to shield. This will provide a fundament for a much more relevant discussion on how contamination could still occur.

The particle data and the associated discussion:
Line124: “The MEPED electron channels mep0e1 to mep0e3 are sensitive to high energetic electron by design (compare Yando et al., 2011, for modeled sensitivity).” This is true, but it is very unlikely to be a problem. MEPED is integral channels which is designed to also measure relativistic electrons. The same relativistic electrons are counted in >30, >100 and >300 keV. Making this into differential channels effectively removes the contribution of relativistic electrons. Do the authors believe that the relativistic electrons penetrate the detector house and become non-relativistic electrons before hitting the detector? How is this only relevant for low Kp?
Are the MEPED electron data used in this study corrected for the well-established low-energy proton (210–2700 keV) contamination in the MEPED electron channels? Cross-contamination by proton was documented already by Evans and Greer (2004) and confirmed by Yando et al. (2011). It is strongly questionable to consider other sources of contamination before ruling out the well-known problems. I appreciate that the authors use the Omni detector to effectively exclude solar proton events, but it does not account for cross-talk of low energy protons. I highly recommend that the MEPED electron data is either corrected for low-energy proton contamination or excluded in the current study.

SSA, 0 degree detector and the associated discussion:
The discussion section ends with suggesting that for the MEPED energy channels the 0 degree detector might observed the “out of nominal field-of-view” electrons as pointed out by Selesnick et al. (2020). This is a good point, and raises the question why the authors limit their study to the 0 degree detectors? If cross-talk contamination is real, it should be evident in both the 0/30 TED and 0/90 MEPED channels.
Moreover, the paper states that the potential contamination occurs predominantly at longitudes corresponding to South Atlantic Anomaly (SSA). Figure 2 also illustrates that this is mainly an SH issue. At the longitudes associated to the SSA the detectors' pitch angle coverage are altered as previously documented by Rodger et al. (2010). This will naturally elevate the particle count rate simply due to seeing a different part of the particle distributions. Potential cross-contamination should therefore be assessed in terms of longitudes.
Rodger, C. J., B. R. Carson, S. A. Cummer, R. J. Gamble, M. A. Clilverd, J. C. Green, J.-A. Sauvaud, M. Parrot, and J.-J. Berthelier (2010), Contrasting the efficiency of radiation belt losses caused by ducted and nonducted whistler-mode waves from ground-based transmitters, J. Geophys. Res., 115, A12208, doi:10.1029/2010JA015880.

Result, table 1: What are the criteria for “enhanced subauroral peak” in the different flux measurements?

Minor comments:
Introduction, line21-24:
The authors highlight the importance of the study based on its potential impact on lower thermospheric and mesospheric chemistry. However, this is not relevant for the proton energies that is the main target examined in this study. The typical energies 1-10 keV measured by the TED proton detector are depositing their energy at altitudes corresponding to the F-region. They are not relevant for the production of NOx in the lower thermosphere nor HOx in the mesosphere. A proton needs an initial energy of ~1 MeV to reach the lower thermosphere. Please, rephrase the introduction to avoid potential misunderstandings regarding its relevance and application.

Data:
In terms of the MEPED electron data it is unclear which data the authors have used. Line 51: “All particle count rates have been converted into differential flux by dividing the energy range and applying satellite/channel specific geometric correction factors (Evans and Greer, 2004).” To my knowledge data using these geometric factors are available only up to around 2014 as Green (2013) updated the geometric factors. How did the authors merge the data before or after 2014? Please, elaborate.

Discussion:
Line 106: It would be useful with a sentence to elaborate on the background correction by Green (2013).
Line 110-111: “Energetic electron crosstalk also explains the Kp dependence as Turner et al. (2012) point out that the radiation belt electron flux dropc by orders of magnitude during geomagnetic storms.” Although magnetopause shadowing can account for relativistic electron drop-outs, Turner et al. (2012) also shows how the radiation belt is rather quickly replenished and increased compared to pre-storm levels. Hence, I find the statement in 110-11 questionable and not supported by the reference.
Figure 3: Acronym CPS not defined in the paper. Which energies do “lower and higher channels” refer to?
Conclusion, point 1: 60 N/S cannot be both southward of the SAA.
Abstract, line 7: a L<6 -> at L<6
Result, line 95: extend -> extent
Discussion, line 111: dropc -> drop

Citation: https://doi.org/10.5194/angeo-2023-33-RC3
- AC1: 'Reply on RC3', Olesya Yakovchuk, 26 Jan 2024
  
  Please, see attachment.
  
  Citation: https://doi.org/10.5194/angeo-2023-33-AC1

Status: closed

RC1:
'Comment on angeo-2023-33', Anonymous Referee #1, 28 Nov 2023

The paper investigates the quality of the POES/Metop TED proton channels, in particular the cross-talk-produced background in proton channels at sub-auroral latitudes. The paper aims at establishing a Kp and channel dependent latitudinal cut of the cross-talk dominated area that should be excluded from analyses. The topic is important since the interest in the low-energy data is increasing due to the connection with atmospheric ionisation related to climate models.
The presentation is clear apart from some minor language issues, typos, and a small issue with Fig. 2 (see minor comments). The conclusions reached by the authors (highlighting cross-talk contaminated proton fluxes) are substantial. The title and the abstract are pertinent and understandable. Appropriate credit is given to earlier work, as far as I am aware.
The only substantive comment on the analysis I have is that the authors are clumping all longitudes into one although the effects of contamination seem to be very different in the various (APEX 110 km) longitude and between north and south. I wonder if the data could be used in a bit broader latitude range if the longitude would be pieced in sectors. Also the clear north-south asymmetry might be considered. The contaminating electrons drift eastward and their mirroring altitudes are different in the north and south hemispheres, which is clearly visible in Fig. 1: in the north, regions between 40° and 120° longitude, for example, are almost free of contamination. Thus, the latitude cutoffs found by the authors might be overly conservative in some spatial regions. This could be pointed out as a topic of further study.
Minor issues:

- line 26: "high relativistic" -> "highly relativistic"

- line 26: "Inspite of these data are" -> "Despite of these data being"

- line 50: "by dividing the energgy range" -> "by dividing by the width of the energy range"

- line 51: "geometric correction factors" -> "geometric factors" (?)

- line 55: "high-energetic" -> "high-energy"

- line 56: "high-energetic" -> "high-energy"

- line 78: "Therefore is" -> "Therefore it"

- line 87: "prominantly" -> "prominently"

- line 92: "in mod. APEX" spell out the abbreviation!

- line 111: "dropc" -> "drops"

- line 122: "(Yando et al., 2011)" -> "Yando et al. (2011)"

- Fig. 2: the vertical lines are not very visible at first glance on top of the curves. Please consider making them longer, for example, to allow the reader to spot them immediately.

Citation: https://doi.org/10.5194/angeo-2023-33-RC1
- CC1:
  'Reply on RC1', Jan Maik Wissing, 04 Dec 2023
  
  We thank reviewer #1 for the helpful comments and want to address the main issue regarding longitudinal differences and the north-south asymmetry in the crosstalk contribution. Reviewer #1 argues that the latitudinal cutoffs might be conservative in some (longitude and hemisphere specific) regions where the crosstalk contribution is low.
  We completely agree to that. The crosstalk contribution varies with longitude and shows hemispheric differences. Fig. 1 (left panel) gives an estimation on that. A minimum crosstalk contribution can be seen at about 70°E in the northern hemisphere. If we consider this as uncontaminated flux, the latitudinal cutoff removes about 4 degrees with real data which rapidly decreases to background levels.
  
  We updated Figure 3 and added another longitude, namely 160°E, where North and South have relatively low crosstalk contribution. The lower graph of the same color always represents the 160°E fluxes (or the according background count rates). Note that the color code also changed in order to increase the readability. The figure is a little crowded but we can see that all the TED proton channels show the same crosstalk maximum at L=4, even at 160°E. The dashed lines that mark the contaminated area also hold for both longitudes and crosstalk seems to be the main source of the measured TED proton particle flux within this area at both longitudes as well.
  
  Thus the presented approach of "global" cutoff latitudes is conservative and leads to an underestimation of the precipitating flux (area). But given that crosstalk seems to be the major subauroral contribution except for a very small longitudinal range (at about 60-80°E) in the northern hemisphere, a sectorial approach will probably not lead to significant improvement of the cutoff latitudes.
  As topic for further studies the authors would suggest a recalculation of all affected channels taking into account the TED background counts which might preserve the equatorial boundary in a more natural form than this approach does.
  
  Citation: https://doi.org/10.5194/angeo-2023-33-CC1
  - RC2: 'Reply on CC1', Anonymous Referee #1, 04 Dec 2023
    
    I see the point made by the authors and agree with their assessment. After they implemented the minor points I made, I can recommend acceptance.
    
    Citation: https://doi.org/10.5194/angeo-2023-33-RC2
    
    AC2: 'Reply on RC2', Olesya Yakovchuk, 26 Jan 2024
    
    All our answers can be found in the attachment of another reply.
    
    Citation: https://doi.org/10.5194/angeo-2023-33-AC2
RC3:
'Comment on angeo-2023-33', Anonymous Referee #2, 14 Dec 2023

The main objective of the paper is to investigate potential cross-talk in POES/Metop TED proton channels at sub-auroral latitudes and provide a preliminary Kp-dependent cutoff latitude where the observations can be used safely. It also presents evidence of cross-talk in the MEPED electron and proton detectors. I do, however, have major concerns about the validity of some of the results. Moreover, the following discussion are somewhat superficial and inaccurate which makes the conclusions potentially erroneous.

Major comments:
Missing discussion on the physics of cross-contamination:
Both introduction and discussion focus on relativistic electrons in the radiation belts being the source of cross-talk in the TED and MEPED proton and electron detectors. Besides stating that TED is a cylindrical electrostatic analyzer and MEPED has a passive shielding, there is no technical specification nor discussion on how the potential cross-talk from relativistic electrons could happen. I recommend a short overview of relevant technical specifications on which type of particles and and associated energies the detectors are designed to shield. This will provide a fundament for a much more relevant discussion on how contamination could still occur.

The particle data and the associated discussion:
Line124: “The MEPED electron channels mep0e1 to mep0e3 are sensitive to high energetic electron by design (compare Yando et al., 2011, for modeled sensitivity).” This is true, but it is very unlikely to be a problem. MEPED is integral channels which is designed to also measure relativistic electrons. The same relativistic electrons are counted in >30, >100 and >300 keV. Making this into differential channels effectively removes the contribution of relativistic electrons. Do the authors believe that the relativistic electrons penetrate the detector house and become non-relativistic electrons before hitting the detector? How is this only relevant for low Kp?
Are the MEPED electron data used in this study corrected for the well-established low-energy proton (210–2700 keV) contamination in the MEPED electron channels? Cross-contamination by proton was documented already by Evans and Greer (2004) and confirmed by Yando et al. (2011). It is strongly questionable to consider other sources of contamination before ruling out the well-known problems. I appreciate that the authors use the Omni detector to effectively exclude solar proton events, but it does not account for cross-talk of low energy protons. I highly recommend that the MEPED electron data is either corrected for low-energy proton contamination or excluded in the current study.

SSA, 0 degree detector and the associated discussion:
The discussion section ends with suggesting that for the MEPED energy channels the 0 degree detector might observed the “out of nominal field-of-view” electrons as pointed out by Selesnick et al. (2020). This is a good point, and raises the question why the authors limit their study to the 0 degree detectors? If cross-talk contamination is real, it should be evident in both the 0/30 TED and 0/90 MEPED channels.
Moreover, the paper states that the potential contamination occurs predominantly at longitudes corresponding to South Atlantic Anomaly (SSA). Figure 2 also illustrates that this is mainly an SH issue. At the longitudes associated to the SSA the detectors' pitch angle coverage are altered as previously documented by Rodger et al. (2010). This will naturally elevate the particle count rate simply due to seeing a different part of the particle distributions. Potential cross-contamination should therefore be assessed in terms of longitudes.
Rodger, C. J., B. R. Carson, S. A. Cummer, R. J. Gamble, M. A. Clilverd, J. C. Green, J.-A. Sauvaud, M. Parrot, and J.-J. Berthelier (2010), Contrasting the efficiency of radiation belt losses caused by ducted and nonducted whistler-mode waves from ground-based transmitters, J. Geophys. Res., 115, A12208, doi:10.1029/2010JA015880.

Result, table 1: What are the criteria for “enhanced subauroral peak” in the different flux measurements?

Minor comments:
Introduction, line21-24:
The authors highlight the importance of the study based on its potential impact on lower thermospheric and mesospheric chemistry. However, this is not relevant for the proton energies that is the main target examined in this study. The typical energies 1-10 keV measured by the TED proton detector are depositing their energy at altitudes corresponding to the F-region. They are not relevant for the production of NOx in the lower thermosphere nor HOx in the mesosphere. A proton needs an initial energy of ~1 MeV to reach the lower thermosphere. Please, rephrase the introduction to avoid potential misunderstandings regarding its relevance and application.

Data:
In terms of the MEPED electron data it is unclear which data the authors have used. Line 51: “All particle count rates have been converted into differential flux by dividing the energy range and applying satellite/channel specific geometric correction factors (Evans and Greer, 2004).” To my knowledge data using these geometric factors are available only up to around 2014 as Green (2013) updated the geometric factors. How did the authors merge the data before or after 2014? Please, elaborate.

Discussion:
Line 106: It would be useful with a sentence to elaborate on the background correction by Green (2013).
Line 110-111: “Energetic electron crosstalk also explains the Kp dependence as Turner et al. (2012) point out that the radiation belt electron flux dropc by orders of magnitude during geomagnetic storms.” Although magnetopause shadowing can account for relativistic electron drop-outs, Turner et al. (2012) also shows how the radiation belt is rather quickly replenished and increased compared to pre-storm levels. Hence, I find the statement in 110-11 questionable and not supported by the reference.
Figure 3: Acronym CPS not defined in the paper. Which energies do “lower and higher channels” refer to?
Conclusion, point 1: 60 N/S cannot be both southward of the SAA.
Abstract, line 7: a L<6 -> at L<6
Result, line 95: extend -> extent
Discussion, line 111: dropc -> drop

Citation: https://doi.org/10.5194/angeo-2023-33-RC3
- AC1: 'Reply on RC3', Olesya Yakovchuk, 26 Jan 2024
  
  Please, see attachment.
  
  Citation: https://doi.org/10.5194/angeo-2023-33-AC1

Jan Maik Wissing and Olesya Yakovchuk

Viewed

Total article views: 218 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
140	57	21	218	13	15

HTML: 140
PDF: 57
XML: 21
Total: 218
BibTeX: 13
EndNote: 15

Views and downloads (calculated since 09 Nov 2023)

Month	HTML	PDF	XML	Total
Nov 2023	19	8	3	30
Dec 2023	50	18	7	75
Jan 2024	32	12	4	48
Feb 2024	10	4	4	18
Mar 2024	17	13	2	32
Apr 2024	12	2	1	15

Cumulative views and downloads (calculated since 09 Nov 2023)

Month	HTML	PDF	XML	Total
Nov 2023	19	8	3	30
Dec 2023	50	18	7	75
Jan 2024	32	12	4	48
Feb 2024	10	4	4	18
Mar 2024	17	13	2	32
Apr 2024	12	2	1	15

Viewed (geographical distribution)

Total article views: 205 (including HTML, PDF, and XML) Thereof 205 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 25 Apr 2024

Short summary

We investigated a subauroral particle flux maximum (located at 60 N/S with focus at 0–100 E) seen in the NOAA POES detector, which we attributed to crosstalk contamination of radiation belt electrons. Affected are the processed TED proton channels (which should already be corrected by NOAA) and lower MEPED channels. The particle flux in the contaminated area can be more than factor 2 higher than typical auroral flux. The region of intense particle crosstalk may be removed by a latitudinal cut.


Total:	0
HTML:	0
PDF:	0
XML:	0