Calibration of the DSCOVR EPIC visible and NIR channels using MODIS Terra and Aqua data and EPIC lunar observations

Geogdzhayev, Igor V.; Marshak, Alexander

doi:https://doi.org/10.5194/amt-11-359-2018

Articles | Volume 11, issue 1

https://doi.org/10.5194/amt-11-359-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-11-359-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 11, issue 1

Research article

|

17 Jan 2018

Research article |

| 17 Jan 2018

Calibration of the DSCOVR EPIC visible and NIR channels using MODIS Terra and Aqua data and EPIC lunar observations

Igor V. Geogdzhayev and Alexander Marshak

Download

Final revised paper (published on 17 Jan 2018)
Preprint (discussion started on 09 Aug 2017)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of AMT-2017-222', Anonymous Referee #2, 02 Sep 2017
- AC1: 'Response to Anonymous Referee #2 comments', Igor Geogdzhayev, 26 Oct 2017
RC2: 'Calibration of the DSCOVR EPIC visible and NIR channels using MODIS and EPIC lunar observations', Anonymous Referee #1, 28 Sep 2017
- AC2: 'Response to Anonymous Referee #1 comments', Igor Geogdzhayev, 26 Oct 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Igor Geogdzhayev on behalf of the Authors (26 Oct 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 Oct 2017) by Brian Kahn

RR by Anonymous Referee #1 (11 Nov 2017)

Suggestions for revision or reasons for rejection

Summary
This study calibrates the EPIC imager channels with corresponding MODIS band calibration using two different methods. They apply SBAF and show that the regression offsets are closer to a zero offset in version 2 EPIC, which has accounted for stray light. The EPIC absorbing channels were calibrated using lunar targets after adjusting for a small spectral shift. The absorbing channel calibration using this method was compared to ROLO and was found to be within 10%. This paper is ready for review after the following issues have been addressed.

I appreciate the authors adding a section of how the EPIC observations can be used scientifically.
I have few more minor comments. After these have been addressed I approve the paper for publication. I do not need to review the revised paper.

Specific Comments

Page 2 line 8. glints-> glint

Page 3 line 12. EPIC has 12-bit radiometric resolution. It samples the Earth 10-20 times a day. It combines 4 pixels (8^2 km) into 1 pixel (18^2 km). It seems odd, to lower the spatial resolution, rather, than lower the radiometric bit rate from 12 to 10 bits, or to increase the number of Earth images per day. Many geostationary instruments have only 10-bit resolution. Can the authors explain why keeping a 12-bit resolution is so important for the EPIC retrievals?

Page 3 line 21. For a sensor at L1 it seems quite odd to make the spectral response functions extremely narrow, especially if you are trying to capture as much light as possible in order to increase the signal to noise ratio. The MODIS sensor is capable of retrieving clouds, aerosols, land use properties, with broad spectral bands. What retrievals need such narrow spectral response functions?

Page 3 line 27. Is there a reference that describes the Version 2 improvements?

Page 4 line 7. What is the dwell time to make an observation? How much does the Earth rotate underneath during the measurement time?

Figure 4. Are these all the EPIC and MODIS matches from June 2015 and March 2017? Please state this in the figure caption.

Page 4 line 28. The pixel homogeneity was the greatest limiting factor. Less than 5000 matches were found, during a time period of 1 ¾ years. The pixel homogeneity of both MODIS and EPIC must be small. This is one way to avoid having to deal with navigation differences.

Has the version 2 navigation improved over version 1? What makes the navigating this sensor so difficult?

In the acknowledgements, you state that Karen Blank helped out with the EPIC image geolocation. Is this an improvement to the version 2 navigation that she implemented? To make the EPIC data more useful, how was the EPIC data for this study navigated improved upon, which was found in the version 2 data?

Page 5 line 4. It is fortunate that the EPIC sensor is not degrading over time. If it were degrading the first approach could not be used to monitor the degradation. Also, if one picks a reflectance threshold using the same channel that needs to be calibrated, and it is degrading, the true sensor degradation would be underestimated. Would it not be better approach to use a frequency threshold, as you suggested as 10% or 15%?

Page 5 line 12. But the first approach is crucial since it verifies that the sensor response is linear. However, the lack of sampling over the dynamic range, the linearity of the sensor cannot be verified, using approach one in its current form. But, I do agree that approach one does show that the stray light corrections are causing the linear regressions offset to be closer to zero.

Figure 5. The 780nm M/E ratio in the caption of 1.141 does not match the M/E in the plot of ~1.4.

Page 7 line 10. SCHIAMACHY -> SCIAMACHY

Page 9 line 13. Did the full moon and new moon data provide the same channel ratios? Which ratios are provided on line 15? Can the lunar phase symbols be used rather open circles to assure the readers that the ratio is the same for new and full moon conditions?

Hide

ED: Publish subject to minor revisions (review by editor) (17 Nov 2017) by Brian Kahn

AR by Igor Geogdzhayev on behalf of the Authors (21 Nov 2017) Author's response Manuscript

ED: Publish as is (27 Nov 2017) by Brian Kahn

Short summary

The unique Earth view of the Deep Space Climate Observatory (DSCOVR) Earth Polychromatic Imaging Camera (EPIC) orbiting at the point of equal attraction from the Earth and the Sun can significantly augment the low-orbit remote sensing of aerosols, clouds and gases. We derive the relationship between the digital counts and the reflected sunlight intensity for some EPIC channels using collocated Earth views from EPIC and Moderate Resolution Imaging Spectroradiometer (MODIS) and EPIC moon views.