The Development of Dark Hyperspectral Absolute Calibration Model Using Extended Pseudo Invariant Calibration Sites at a Global Scale: Dark EPICS-Global

Round 1

Reviewer 1 Report

Dear authors,

In the manuscript “The Development of Dark Hyperspectral Absolute Calibration Model Using Extended Pseudo Invariant Calibration Sites (EPICS) at a Global Scale: Dark EPICS-Global” the author developed the DAHAC model using Landsat-8 (L8) Operational Land Imager (OLI), Sentinel-2A (S2A) Multispectral Instrument (MSI), and Earth Observing One (EO-1) Hyperion. And compare the results of DAHAC model with satellite measurements that shows the accuracy within 0.01 unit reflectance across overall spectral bands. This manuscript is well organized, and the drawn conclusions are coherent with the obtained results. I believe your manuscript may be useful for wide audience. However, there are some issues which need addressing. They are listed as

·       During cross-validation; is "Spectral Band Adjustment Factor" taken care of or not? Please, check

·       Figure 2, Convert Latitude and Longitude to Degree Decimal (DD) and escape so many zeros.

·       Figure 3, is so clumsy (most of the variables are overlapped), kindly make it attractive or it is better to can remove it.

·       Figure 5, the average TOA reflectance of all images with standard deviation could give more information than this figure (most of variables are overlapped).

·       Figure 6, is so clumsy (most of the variables are overlapped), kindly make it attractive or it is better to can remove it.

·       Table-3, kindly use 2 decimal digit values in table.

·       Lines 24 – 24: Please, arrange the keywords alphabetically.

·       Figure 25, Add legend.

·       Please, check all situations when en-dashes should be used (e.g., for ranges, between years etc.).

·       Grammar and English language must be improved in the entire MS.

·         Grammar and English language must be improved in the entire MS.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

I would like to thank the authors for such an interesting and thorough work.

I read the manuscript carefully and I have one comment about the low uncertainty of the DAHAC model with Monte Carlo simulation.

-        Maybe the initial distribution function (mean and sd) for all the coefficients could be specified

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. For section 2.4.1, could you please describe more details about the BRDF model coefficient inversion and validation? Did you apply cross-validation approach? It is better to use different portions to train and test the BRDF model coefficients. 

2. Similar as item 1, for section 3.1, did you use the same DaHD profiles to invert for DAHAC coefficients and then validate the DAHAC model?

3. As studied in paper "Impact of wavelength shift in relative spectral response at high angles of incidence in landsat-8 operational land imager and future landsat design concepts", the RSR will shift towards shorter wavelength as VZN increases, this could lead to changes in sensor received radiance. In section 3.1, when convolve hyperspectral TOA reflectance with RSR of L8, did you consider the RSR shift effect when VZN is large? If not, it's better to mention it in the conclusion section (section 6, line 677-678).

4. There is no legend in Figure 25. 

5. For Table 3, 426nm is not key wavelength band. Could you please update the table content using a more frequently used spectral band? For example, center wavelength of any bands of Landsat 8 OLI. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

accept to publish in current form