Evaluation of MODIS Dark Target AOD Product with 3 and 10 km Resolution in Amazonia

Round 1

Reviewer 1 Report

This study evaluates the MODIS DT AOD products against the AERONET measurements in the Amazonia region, focusing on the discrepancies between 3km and 10km products. The objective of this study is clear, and the outcomes will benefit the MODIS AOD user and algorithm team. Overall, the manuscript is well-written, and the results are presented and discussed in an appropriate manner. This manuscript can be accepted after considering the comments and suggestions I pointed out in the attached file. 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This is a review for the manuscript entitled “AOD evaluation using MODIS Dark Target with 3 and 10 km resolution in Amazonia” submitted for RS. The MS focuses on evaluated the MODIS DT 3km and 10km AOD products over Amazonia. The experiments is simple but convincing. However, some key points of the manuscript are less conclusive and remain a bit unclear.

 

The importance of studying the MODIS DT AOD accuracy over Amazonia needs to be emphasized in the MS, especially in Introduction. As a reader, I may prefer to get some more useful conclusions from the evaluation rather than validation itself. For example, what kind of studies are these validation data more applicable to in order to take full advantage of them.

 

Table1 AERONET 15min observation is temporal resolution not SR (Spatial Resolution)

 

For Fig.3, in addition to the several metrics listed by the authors, I am probably more concerned with the spatial coverage of the two DT algorithm. The authors should give the number or fractions of effective retrievals of the two algorithms in the study area.

 

The metrics among the different sites in Fig.4 are better summarized in a bar chart or table. RMB represents a percentage or a multiplier relationship? The average difference in RMB between MODIS DT products was 20%. How was it calculated? Please give a more visual diagram and figures.

 

Did the comparison between the MODIS DT 3 km and 10 km AOD products in Figure 6d only use the matched points over the stations? “The differences between the 3 km and 10 km products can be attributed to different ways of estimating the algorithms” How this conclusion can be seen from the scatter plot in Figure 6d?

 

Different spatial scales may also cause differences in accuracy. In other words, the author selected 5 x 5 pixels, which corresponds to the range of 50 km and 15 km for 10 km AOD and 3 km AOD, respectively. It would be interesting if the authors discuss the differences in accuracy at different scales chosen, although this could lead to very different conclusions.

 

In addition to the validation of all matching points, I would further focus on the difference between the two products on different time scales. Does the monthly average or even the annual average represent the real situation of Amazon? It is impossible to judge only from the fire point, because it is not known whether the difference among AOD and fire counts is caused by precision or other reasons.

 

From the conclusions given by the authors, it is known that they prefer 10km product (before discussing scale effects) for general analysis in Amazonia. But for some purpose we had to use higher resolution data to study aerosol transport and sources, and even population exposed content. Therefore, it would be better to add a simple evaluation of MODIS 1km MAISAC products, so that a clearer choice can be made in future research.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed my comments accordingly. I recommend this manuscript be accepted for publication.