Classification of Crop Area Using PALSAR, Sentinel-1, and Planet Data for the NISAR Mission

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The crop classification using multi-frequency SAR image and optical data have been assessed for the upcoming NISAR mission. The coefficient of variation has been used for the classification. The manuscript does not have new finds for the crop classification using multi-frequency data. It is suggested to make the assessment of feature importance using Random Forest classifier.

1 It is suggested to increase the number of training and validation samples for the experiment.

2 It is suggested to use sigma-naught, but not gamma-naught for the SAR data. In addition, it is suggested to make speckle filter for SAR image before classification.

3 There are some uncertainties for the CV. It is suggested to use some classification methods for the classification, such as RF, which can assess the importance of features used.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this manuscript the algorithm for classifying crop areas from multi-frequency SAR and optical data is evaluated for the NISAR mission using ALOS-2, Sentinel-1A and Planet data.

Here are my suggestions.

(1)    ALOS 2 and Sentinel-1A respectively work on L- and C- band, but NISAR works on L- and S-band. How is the ALOS-2 and Sentinel-1A related to NISAR?

(2)    In section 4, more details about the simulated NISAR data is suggested. How did you obtain simulated NISAR using ALOS 2 and Sentinel-1A data? How did you get S-band data? Detailed introduction about the simulation method is suggested.

(3)    It is not clear how the methodology helps NISAR data classification and it is suggested to be further clarified.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Very interesting article, well focused on new SAR mission pre-launch evaluation, very clear with well addressing problems and issues. It should be published in present form giving common ground for scientific comunity for further study on different test sites following this methodology.

Author Response

Dear Reviewer, thank you for your appreciation of our paper and the work you have done on it.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors provide a reasonable explanation for the issues raised in last version. However, NISAR contains L-band and S-band, and the paper mainly uses the L-band ALOS-2 data to simulate the L-band data of NISAR, and does not analyze the S-band. This point can be further clarified in the paper.

In fact, there may be a lot of differences in the scattering characteristics of crops in the L/S/C bands, and the C-band scattering characteristics analyzed based on the Sentinel-1A data may not necessarily be the same as the S-band scattering characteristics.

Overall, S-band SAR data are relatively scarce at present, and the paper is suggested to do some discussions on how to utilize L/S-band data from NISAR to further improve crop classification.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf