High-Resolution Image Products Acquired from Mid-Sized Uncrewed Aerial Systems for Land–Atmosphere Studies

Round 1

Reviewer 1 Report

The manuscript introduces a methodology to calibrate and evaluate reflectance and surface skin temperature obtained from a multispectral and thermal imager and a sensor flown on board mid-size Uncrewed Aerial Systems (UAS). The instrumentation and the UAS flight plans are described in the first section, followed by the post-processing steps necessary to correct the images, particularly for changing lighting conditions. The second part of the paper is dedicated to comparing measurements from the UAS to results from satellite images and tower-based sensors. The conclusion assesses how the instrumentation introduced in the paper can be used for multiple applications from reflectance, and surface skin temperatures to other indexes as normalized vegetative index.

This work addresses interesting issues in imagery products raised by increasing the size of the platform along with the flight time. However, some improvements are required to make this manuscript more thorough in the validation process and more precise for the calibration steps. Revisions are recommended to strengthen this technical paper and make it contribute better to the literature.

Comments

1)    introduction: From line 50 to the end of the introduction, the text is well-structured with the appropriate references. However, the first three paragraphs, from lines 32 to 49 lack citations, are oversimplifying the role of satellites and ground-based measurements, and provide a poor comparison with UAS. Please, rework these paragraphs to emphasize the complementarity between each type of measurement. If you decide to list to limitations of satellite and tower-based measurements, you should also cite the constraints of deploying a mid-size UAS (flight authorization, cost of operations compared to small UAS, etc.)

2) post-processing: the steps put in place to correct the images are interesting but are not detailed enough. A small case study to correct images during a flight should be considered to make explicit the use of the DLS sensor in the process and what the image looks like before and after correction.

3) data quality and validation: section 3.2.2. multispectral imagery from two satellites and the UAS are compared. What point are you trying to make in this section? Is it a validation of UAS imagery reflectance? What do we learn from this comparison? A more detailed description is needed.

Specific comments

Please use international units, add aeronautical units in parenthesis when necessary.

Line 137: The Altum imager and the DLS2 sensor were mounted directly on the aircraft body. Other groups have chosen to set up the sensors on gimbles for them to remain level and limit the correction to apply due to the aircraft's attitude. How did you consider the UAS attitude in the results? What was the impact of the turns? Was the effect of the roll negligible?

Figure 2 B and C: These figures present a very poor design that doesn’t support the flight plan description very well. Please, update them.  

Line 242: DN (digital numbers?) is not defined before.

Line 298: “more than sufficient”? too imprecise, please specify.

Line 306: Did you use the mean reflectance calculated at each altitude with the 3x3 neighborhood pixels? What is the tarp size in pixels at each altitude? Is the tarp minimum size at 1800 m MSL 3x3 pixels?

Figure 4: The apparent reflectance curves of the bright tarp at 1500 m MSL and 1300 m MSL are clearly above the over curves, but it doesn’t seem to be an altitude effect. What are your assumptions to explain these gaps? Cloud cover? Aircraft attitude?  

Line 339: The flight plan is described as a lawnmower pattern at 520 m AGL and 1360 m AGL and a square-based ladder pattern to climb between the two levels. In figures 6. A and B, it seems that the square is visible along the two altitude-leveled patterns. This creates some confusion in reading the bias values between the Altum thermal imagery and the IRT as the largest biases are observed along the square. A separate figure focusing on the ascent/descent profiles should be added to present the comparison of the sensors based on altitude. Otherwise, either explain better in the text or remove the square from the bias map. Are the data points in figures 6. B and D only at 520 m AGL and 1360 m AGL? Is the square profile also included?

Figure 6:  The legend in figure A and C should be improved, it is barely readable. Why there are blank data in figure C? Please provide a short explanation to ease the reader’s understanding.

Line 345: Can you develop more about what you learned from the biases between the two altitudes?

Line 409: This is the first time that LES models are cited. You might want to elaborate more on how the imagery from mid-size UAS can be beneficial for LES models.

Line 453: Are there current projects deploying the ArcticShark and the TigerShark to leverage this type of payload for scientific analysis?

Line 440: remove the second conclusion.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper provides an application of commercially available thermospectral and multispectral cameras for environmental monitoring in small unmanned aerial systems (sUAS, <55 lbs) Group 3 classification UAS. The work is interesting based on a systematic protocol aimed at understanding application of thermospectral and multispectral cameras.

Advantages:

Work is considered interesting and relevant to the journal's scope, but has multiple editing errors in the paper, making it unprintable as an article. The researchers understand both the rationale but do not organize a coherent paper in a correct manner, plus it does not have a discussion.

Disadvantages.

1.          The abstract lacks to better reflect the objective of the work and its main result. synthesize in a better way to reduce, the text to comply with the requirements of the journal.

2.          The introduction needs to reflect a greater theoretical support of the research in terms of other experiences with this type of application.

3. There is not a very broad base of theoretical foundation (bibliography) to support the research, there are very few bibliographical references (24 in total) for a scientific article.

4.          The formula (1) in number 180-182 is described in a table, this does not comply with the standards of the journal, it should be corrected.

5.          Table 2 should be separated from formula (2).

6.          Figure 2 is separated into A and B, but each one is not explained, figure B is not well visualized because of its small size.

7.          Figure 3 should use A and B and describe each, not left and right.

8.          Figure 4 The graphics should be improved and use more appropriate symbols that can be differentiated in a unique way, the orange and brown colors are confused.

9.          Figure 8 should be a map, missing north, scale and coordinates.

10.        The document has a very serious error in the point (4. Conclusion) and another point appears as (6. Conclusion), the text is repeated, this is unmistakable in a scientific article, the document was not revised and was sent without a minimum editing process of standards.

11.        This document has no discussion; the document does not discuss its findings or confront its results with other research. There is no evidence that the author discusses its results with previous studies or confronts its work with contributions or problems found in other national or international studies.

12.        The title Data File Structure, Data Availability, and Code Availiability is irrelevant, this information could be placed as an annex, it does not make a significant contribution within the document and uses a text space that should be another section such as the discussion.

13.        It has grammatical and punctuation errors, in addition to the basic structure of the article.

In conclusion, the document has serious editing errors in its structure (numbering and sections of the article) and the way it is presented is not admissible for publication; it should have been edited and revised according to the parameters of the journal. The authors did not take the time to review the document before submission and sent a document with basic errors and poorly structured.

It has grammatical and punctuation errors, in addition to the basic structure of the article.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper has done some work around medium-sized UAV-based high-resolution imaging. However, there are still some issues that need further refinement.

1.     The Introduction part should be further condensed and revised. The discussion on current status needs to be more systematically summarized, rather than directly and simply listed. The novelty and contribution of this paper should be further summary and condensation as an academic paper, which are not very clear in current state.

2.     For the aerial imaging, the overlap ration is one of the key metrics which affects the imaging efficiency. More analysis should be provided on the reason for the need to guarantee the minimum overlap ratio of 75%.

3.     More information on the means to ensure high resolution imaging performance should be presented.

4.     Polarization enables target detection and target classification in complex backgrounds, have you considered adding polarization information to aerial imaging?

5.     What are the differences in conditions such as carrier vibration during flight of a medium-sized UAV compared to a small UAV?

6.     The impact of wind resistance on the imaging system is more apparent with faster flight speeds, especially for more variable and stronger wind. How to ensure high spatial resolution in your work?

NO

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have made a significant effort to account for the reviewer’ comments and improve the manuscript. However, some details were left behind.

Abstract, Table 1, and lines 61, 86, 210, 218, 223, etc.: Please change the units to the ones from the international system. This shouldn’t need to be repeated.

Line 488: No, you are not the first ones to propose to integrate UAS imagery into atmospheric models. Please, reformulate.

Author Response

Please check the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors solved the problems of the article

Author Response

Please check the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

No

Author Response

Please check the attachment.

Author Response File: Author Response.pdf