Wavelength Calibration for the LIBS Spectra of the Zhurong Mars Rover

Round 1

Reviewer 1 Report

This paper analysis the drift between the onboard spectra and laboratory reference of the MarsSCoDe instrument onboard the Zhurong rover. It is reported that the CCD temperature is the source of the wavelength drift. Subsequently, the authors proposed a quadratic function to correct the errors reported.

 

This is an excellent paper with clear objectives, solutions and analysis. All important information and experimental data are clearly presented and properly analysed. Congratulations to the excellent work!

 

I have only a few minor suggestions as follows.

 

37: LIBS is remote sensing detection -> LIBS is a remote sensing based detector

 

38: do not need pretreatment -> do not need any pretreatment

 

73: Mounted on the left side of the … -> it would be nice to have an illustrative figure, here or in the following section, showing the LIBS, SWIR, MIC, and MCCT instruments onboard the Zhurong rover.

 

127: in Table S1 (see Supplementary Materials)

 

136-139: It’s up to the authors but I think a flow diagram with a few lines of explanation in the text would make this paragraph more elegant.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript present a wavelength calibration method for MarSCoDe LIBS between the drift of the MarSCoDe Ti plate LIBS spectrum relative to the reference spectra from lab and the corresponding LIBS spectrometer CCD temperature. the drift of a certain channel is calculated by the corresponding CCD temperature, and then the wavelength of the spectrum is calibrated by the drift. This is a very meaningful work, but there are still some problems that need to be modified, the manuscript is original and I support publication but only after revision according to my queries. The list of my questions is below: 

1. As mentioned in the manuscript, there is a quadratic function relationship between spectral drift and CCD temperature. I want to know whether CCD temperature is the major factor or the only factor for spectral drift. If it is the major factor, can the wavelength drift caused by other reasons be ignored, such as instrument jitter.

2. Due to the differences in environmental conditions such as pressure and temperature between Mars and the ground, whether this wavelength calibration algorithm is universal, the ground environment is relatively complex, and whether this calibration relationship based on CCD temperature can be applied in the laboratory environment or some detection equipment based on LIBS technology.

3. In the manuscript, the calibration peak should be marked in Figure 1, mainly including element and wavelength information.

4. In the wavelength calibration process, whether the characteristic lines of the same element or the combination of characteristic lines of different elements are used. If it is the characteristic lines of the same element, will the differences between different elements be considered ? Wavelength drift of different elements may be different. If it is the combination of characteristic lines of different elements, Whether some trace element signals will be considered. because the spectral line intensity of some trace elements is weak.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript entitled "Wavelength calibration for the LIBS spectra of the Zhurong Mars rover" (ID: 2149412) is a practical investigation for LIBS space application, and the wavelength calibration of Zhurong's spectrometer is the main point. According to my review, I would like to recommend MINOR REVISION to this submission, and the specific comments are listed as follows:

1) The language has to be carefully polished. Some paragraphs are difficult to be understood.

2) The calibration process has to be specified with more details. For example, the signals with SNR > 3 were selected as reference peaks for the calibration. The 60 spectra were obtained at the same sampling point? or different points?

3) The Ti sample was used as the standard sample for the calibration. But the reason has to be specified because the Fe sample could also be applied as the standard. 

4) According to my review, authors pay much attention to the wavelength calibration. But the results showed that the wavelength drift was far less than the spectral resolution. So, the intensity calibration might be more important in the application.

5) In LIBS, the plasma properties would alter the line profile of spectra. So, the wavelength drift might be also caused by the spectra deformation resulted from the plasma?

6) If possible, it would be better to offer the details of spectrometer (pictures, 3D models, etc.). In the application, I would like to suggest authors to put the mineral picture in the manuscript to attract the readers.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

Interesting manuscript. No particular observations except for soem details and the presentation of some figures:

 

page 3 line 136: "Onboard data were processed in the following steps: 1) averaging the 60-shot LIBS 136 spectra..." by applying this procedure, could you provide in the text the standard deviation of the intensity if the average spectrum (maximum value is enough)? It could be useful for having an idea of the reproducibility of the data and for a precise quantification of the different materials.

Page 4 lines 173-174: "The emission peaks selected as calibration peaks are easily excited by 173 the laser, have less fluctuation in peak intensity, and are relatively uniformly distributed 174 over the entire spectrum (Figure 1, Table S2)..."

I agree with the authors that the fluctuation of the emission peaks should be minimized and ideally made null, but some of the spectral emission lines chosen for calibration appear really weak, for example the Ti peak at 578.75 nm on CH3 is barely distinguishable from the background and so also O II at 649.372 nm

For these weaker peaks, did the authors estimate the signal-to-background ratio? If so, could they include it in the text?

Figure 3: The fitting functions are not correctly written in the legend. For example for CH1 it is reported : drift = 0.0047x²+-0.2578x3.7412. It should be:

drift = 0.0047x² -0.2578x + 3.7412. Similar corrections for CH2, CH3.

Please correct

Figure 4: please enlarge spacings among the graphs to make more readable the captions of the x and y axes.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx