Dear Referee, thank you for the positive review. We modified the manuscript accordingly to yours comments and answered your questions below. 

1 My main suggestion is to insert a geological section, even simplified, of the area under study so that you can better confirm your hypotheses about the processes in place.

Effectively, we didn't discuss the geology in this paper as we did it for the first paper \citep{cavalie2015}. But, we agree that it would be good to add few elements and to refer specifically to the figures of the previous paper to backup the hypothesis of subsidence compaction. We added this section in the introduction :

Cavalié et al. (2015) published a first study showing that between 2003 and 2011 (the acquisition period of Envisat) \textbf{the Var delta as well as the airport that is located at its mouth, is subsiding. The spatial extent of this subsidence is strictly limited to the quaternary alluvium deposits of the delta and Var riverbed (Figure 4 in cavalié et al., 2015). Actually, on both sides of the riverbed, the subsidence rate quickly drops to zero where the transition from alluvium to conglomerate occurs.  Moreover, the downward displacement rate increases toward the sea as the sediment layers get thicker and more recent (Figure 6 in cavalié et al., 2015). Indeed, it ranges from less than 1 mm/yr in the Var valley to a maximum rate of 10 mm/yr on the airport platform where sediments got brought in the 1970s to built the runways.

2  I would also suggest that you include some references to the theoretical bases concerning creep phenomena (consolidation processes?) in clayey soils, appropriately accompanied by bibliographical references.

Soils and rocks can exhibit creep behavior, which is the development of time-dependent strains at a state of constant effective stress (Bland, 1960; Findley et al., 1976; Jaeger and Cook, 1979). Creep behavior influences the long-term stability of grounds and movement of slopes. This time-dependent material behavior exhibits viscoelastic or viscoplastic characteristics that can be reproduced with different creep models of increasing complexity depending on the type of material and loading conditions (Jaeger and Cook, 1979). Several constitutive laws have been introduced in the past to study creep and this still is an active field of research in the rock physics labs and geophysical field studies. 

Creep is the tendency of solid material to deform permanently under certain load that depend on time and temperature. Typical creep process has three phases which are primary creep (creep rate decreasing over time), secondary creep (constant creep rate) and tertiary creep (increasing creep rate until failure) as shown in Fig. 5a. 

In this work, tertiary creep is not modelled. We used the Burgers model, composed of a Kelvin model and a Maxwell model (Jiand and Wang, 2022), which is well adapted to accurately describe the characteristics of the primary and secondary creep stages (Jaeger and Cook, 1979), a behavior representative of the surface displacement measured by InSAR on the airport platform. A number of research works have previously demonstrated that this creep model was successfully used to model deformation of soils and surface displacement of landslides (You et al., 2013; Liao et al., 2022)

Bellow are the new references that we added in the manuscript :

•  Bland, D. R., The Theory of Linear Viscoelasticity, New York: Pergamon Press, 1960.
•  Findley, W. N., J. S. Lai, and K. Onaran, Creep and Relaxation of Nonlinear Viscoelastic Materials, New York: North-Holland Publishing Company, 1976.
•  Jiang Z., Wang H., 2022, Study on Shear Creep Characteristics and Creep Model of Soil-Rock Mixture Considering the Influence of Water Content, Front. Phys., 21 June 2022, Sec. Interdisciplinary Physics Volume 10 - 2022, https://doi.org/10.3389/fphy.2022.819709
• Liao, M., Cui, D.; Bao, X., Qiao, Z., Zhao, C. Creep Characteristics of Soil in the Sliding Zone of Huangtupo Landslide. Appl. Sci. 2022, 12, 12439. https://doi.org/10.3390/app122312439

3. Title and other comments

We agree that starting with 28 years is not the best way for a title. As the duration of the subsidence is important, we change for : 

"Three decades of coastal subsidence on the slow-moving Nice-Côte d’Azur airport area (France) revealed by InSAR : Insights into the deformation mechanism"

We also modified the manuscript according for your comments :

•  L25: done
• L44: There are actually 2 brackets and global isostatic adjustment and tectonics are included in natural phenomena.
• L48: We added a recent reference 
• The comment focused only on the constant rate and not on the nature of the phenomena 
  
  
  
   

Dear Referee, thank you for the positive review. We modified the manuscript accordingly to yours comments and answered your questions below. 

1.  About the structure of the manuscript. The authors follow a structure of InSAR method – InSAR measurements – uncertainty – creep modeling, with methods, results, and discussion together. If possible, I suggest reorganizing them into Methods, Results, and Discussion

It seems important to us to separate the observations from the modelling. Actually, InSAR measurement are facts and show clearly the ongoing subsidence of the airport platform at least since 1992 (date of the first ERS SAR images) and probably since the late 1970s when the platform was built. The noise analysis allows also to understand the accuracy of the measurements. The modelling part is thus fed by the robust InSAR measurements and it seems logical to introduce InSAR and Modelling one by one.  

Understanding the deformation is more complex because it requires a detailed knowledge of the platform underground. Such work has been done in few papers cited in the article, but it can't give the full picture. Moreover, models are always a simplified view of reality, although modelling is the only way to have an idea about how the airport deformation will develop. Thus, it was important to add a modelling section (that we didn't do in Cavalié et al. (2015)). But, we think it's important to keep separate from the observations. 

Finally, our discussion/conclusion brings face to face observations and modelling to elaborate some recommendations about how to handle this particular situation where a critical infrastructure might be at risk. 

To summarize, I would not group InSAR and modelling methods under one section.

2.  The authors used the Heaviside step functions for reconstructing the time series. Why this function was chosen? Are there some other functions that can also be applied in this situation? Some information/discussion can be added.

InSAR measurements have been computed from three independant datasets (corresponding to 3  different satellites generations : ERS, Envisat and Sentinel-1) with no overlapping periods. So direct measurements cannot give the offsets from one time series to the following one (red dots on the figure 3d of the paper). To simulate the 2 "artificial" offsets (between ERS and Envisat, and between Envisat and Sentinel-1), the Heaviside step is the most natural way (and standard way) to do. But it is only to simulate the offset due to the lack of overlap between time series. Otherwise, we assume a logarithmic evolution of the subsidence.   

3. Similar to the above comment, why viscoelastic Burger’s model was used? Are there some other models appropriate to this study area? Some information/discussion can be added.

We reproduce here the response to a similar question of the reviewer 1:

Soils and rocks can exhibit creep behavior, which is the development of time-dependent strains at a state of constant effective stress (Bland, 1960; Findley et al., 1976; Jaeger and Cook, 1979). Creep behavior influences the long-term stability of grounds and movement of slopes. This time-dependent material behavior exhibits viscoelastic or viscoplastic characteristics that can be reproduced with different creep models of increasing complexity depending on the type of material and loading conditions (Jaeger and Cook, 1979). Several constitutive laws have been introduced in the past to study creep and this still is an active field of research in the rock physics labs and geophysical field studies. 

Creep is the tendency of solid material to deform permanently under certain load that depend on time and temperature. Typical creep process has three phases which are primary creep (creep rate decreasing over time), secondary creep (constant creep rate) and tertiary creep (increasing creep rate until failure) as shown in Fig. 5a. 

In this work, tertiary creep is not modelled. We used the Burgers model, composed of a Kelvin model and a Maxwell model (Jiand and Wang, 2022), which is well adapted to accurately describe the characteristics of the primary and secondary creep stages (Jaeger and Cook, 1979), a behavior representative of the surface displacement measured by InSAR on the airport platform. A number of research works have previously demonstrated that this creep model was successfully used to model deformation of soils and surface displacement of landslides (You et al., 2013; Liao et al., 2022)

3. We also modified the manuscript according to your comments :

•  L2: done
• L5: ok
• L7 : ok
• L12 : we changed the sentence.
• L18 : done
• L23: done
• L34: done
• L49: we replaced "Indeed" by "And"
• L56: done
• L57-58: we agree that it sounds strange and we removed this sentence
• Line 76: Yes you did understand it properly. The mass was dropped from the top of a crane. I doubled check the reference and it turns out that it is 23 m and not 22 m... I corrected the number in the manuscript. 
• L84: We changed "Since" for "Since then"
• Caption Figure 2: we change the sentence by adding "image pairs processed into"
• L108-109: Resolution in raw SAR images are different for range and azimuth axis and also depend on the satellite. For ERS and Envisat, SAR images have a better resolution in azimuth (by a factor 5) compared to range. On the contrary, Sentinel SAR images have a better resolution in range (by a factor 4) compared to azimuth. So, if we want a ~square ground pixel, we need to add extra multilooking in the better resolved direction.
  
   
• L119: ok
• L145: Spatially, pixels can be affected by different sources of noise. One type is due to the distribution of scatterers inside a resolution cell (or pixel) and of their temporal evolutions. In this case, the noise is really pixel dependent. For example, if vegetation grows between the 2 SAR acquisitions or if the ground get eroded, the pixel coherence will decrease and the measurement will be more noisy. On the contrary, pixels on urbanized areas usually show a low noise level as the scatterers (roads, buildings) inside a pixel do not change between the acquisitions. Therefore, it would not make sense to study the noise level of the interferograms in the hills nearby the airport as vegetation and erosion are two factors that will increase the noise content of those pixels compared to the pixels located on the airport platform. Moreover, wave delays in the atmosphere are also heterogeneous and thus impact pixels differently. Studying the noise of the area of interest is thus better if it is possible. 
• L160-161: we agree and completed the paragraph
• L190: This is the uncertainty we estimated previously (Figure 4).