Characterizing Spatiotemporal Patterns of Land Deformation in the Santa Ana Basin, Los Angeles, from InSAR Time Series and Independent Component Analysis
Round 1
Reviewer 1 Report
The authors used the independent component analysis (ICA) method to decompose the InSAR time series deformation over Santa Ana basin in Los Angeles, and observed the seasonal deformation which may be associated with groundwater extraction and recharge. This study may provide useful information for monitoring ground deformation due to the groundwater changes. Overall, I think the manuscript is interesting and should be published after the following revision:
- In the abstract, pls. give more information about “traditional parameter estimation”.
- The geological setting should be added when introducing the study area.
- The figures should be behind the text, such as Figure 2 and 3.
- Figure.4, it is the deformation rate, not the accumulative deformation from 20150514 to 20180428.
- The legend should be added in some figures, such as Figures.9, 10,
- In section 4.1, it is difficult to conclude that the basin changes seasonally based on the annual deformation rate in Figure.5.
- give an interpretation of score in Figure.7.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors perform a multi-temporal InSAR analysis over Sant Ana Basin (LA) to evaluate the correlation between ground displacement derived from InSAR and ground water level.
In the first part of the paper they describe the methodology used to get the ground displacement data.
In section 2.1 they propose a SBAS approach, therefore I would suggest to cite the main paper
A small-baseline approach for investigating deformations on full-resolution differential SAR interferograms.
In section 2.2 the authors propose a ICA decomposition methodology to extract the dominant ground displacement patterns. Before performing this analysis I would suggest removing from the unwrapped phase possible planar component (related to orbital errors, APS, etc) and a stratigraphic component of the APS as it can be temporally correlated.
In Eq (8) there is the symbol A^- which is not defined, probably a typo.
Pag 11, line 304, there is a reference to Figure 6e, it should be 7e. I can not see the similarity between the first independent component of ICA and a sinusoidal function.
Pag 12, how do the authors evaluate the RMSE between GNSS data and INSAR? Do they compare the closest images? As the GNSS have a higher temporal sampling I would suggest to average the GNSS data on a temporal window equal to the satellite revisit time and evaluate the GNSS data in correspondence of the satellite acquisition date and use this comparison to evaluate the RMSE.
Pag 13. I would suggest to spatially visualize the lag value, the correlation coeff could be used to define the transparency value of the point to hide the lag value of the points which do not correlate.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
I included extensive editing notes in the manuscript. However, they are not sufficient to reach a publication quality. So, the authors are advised to find someone to help them to further improve the English of the paper.
In addition, I included a technical comment in the manuscript that I will repeat here. It concerns the use of line-of-sight (LOS) observations in the paper and the oversimplified jump to vertical deformation (e.g., subsidence) as if there are no horizontal movements. The look angle for Sentinel (~42 degrees for descending) is smaller than the look angles of the earlier satellites (Envisat, ERS-1/2), which were ~22 degrees. So, the geometry for Sentinel is such, that the LOS sensitivity to vertical movements (~0.75 of 1) is quite smaller than that of the earlier satellites (~0.93 of 1). At the same time, for Sentinel, the sensitivity to the eastward horizontal movements (~0.66 of 1) is quite larger than that for Envisat and ERS (~0.37 of 1). That is, for Sentinel, the sensitivity to horizontal movements is not much smaller than that to vertical movements. So, I suggest that the authors do not so readily ignore the horizontal movements, because the Sentinel LOS significantly reflects those too. Better stay with LOS only, and if they want to compare with GNSS, project those onto the LOS. In summary, the authors should tone down the consideration of LOS as if it reflects only vertical movements.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
I suggest to publish the manuscript as current version.
Reviewer 2 Report
The authors properly replied to my review
