Remote Sensing of Coastal Vegetation Phenology in a Cold Temperate Intertidal System: Implications for Classification of Coastal Habitats

Round 1

Reviewer 1 Report

Summary

The phenology of vegetation plays various essential ecological and biogeochemical roles that benefit both terrestrial and marine ecosystems. This study using remote sensing technique to study the effect of phenology of vegetation in the intertidal ecosystems in large-scale and time-serious. Therefore, the work of this paper is of great significance.

This study has carried out very detailed work. The structure is logical, the figures are of good quality and the historical background has given credit as is relatively appropriate, the discussion is impressed. One note that the process for remote sensing data to obtain the surface reflectance should need been more clarified, including atmosphere correction, cloud removal and so on, which is the convince foundation of your results. Overall, I think this manuscript can be considered if the author could adequately address the comments below.

 

Specific/Detailed Comments

1.     Figure 1: demonstrate the remote sensing image source and obtain date in the caption.

2.     Figure 2: it shows the four zones have obvious different spectral characteristics (seen from the color), please add a subfigure of the multiple spectral characteristics from the satellite sensors for each zone to make it clear.

3.     Figures 5 and 14: nice presentation.

4.     Atmosphere correction for remote sensing data is necessary, which directly influence your results, the atmosphere correction for remote sensing data should need more clarified.

5.     The reason for the differences between Figures 10 and 11 should be more clarified.

6.     Tide effect and relationship to phenology evolution of vegetation the could be discussed if possible.

Author Response

Thank you so much for your comments and suggestions which has improved the quality and readability of the article.

 

Comment 1: One note that the process for remote sensing data to obtain the surface reflectance should need been more clarified, including atmosphere correction, cloud removal, and so on, which is the convince foundation of your results.

Response 1: As stated in the manuscript, the satellite images used in the present study are atmospherically corrected using ACOLITE, which performs well for the sensors used in the present study. We have used the dark spectrum fitting method used in ACOLITE consistently for all multispectral sensors.  We added the following statement :

Briefly, after cloud masking and gas transmission correction of top-of-atmosphere refelctance, ACOLITE finds multiple dark pixels in subscenes to define a “dark spectrum” based on the lowest reflectance values in any spectral bands. The dark spectrum is then used to estimate the atmospheric path reflectance according to the best fitting aerosol model. Sky and sun glint components are also estimated over water surfaces, but not for land surfaces

Note that, as stated in the manuscript, we have used only cloud-free satellite images in the present study.

 

Comment 2: Figure 1: demonstrate the remote sensing image source and obtain the date in the caption.

Response 2: The caption of Fig. 1 already contains the remote sensing image source and date of acquisition. It is a True colour image from PlanetScope sensor captured over Ile-Verte on September 3rd, 2019.

“True-colour image of Ile-Verte from PlanetScope captured on September 3rd, 2019.”

 

Comment 3: Figure 2: it shows the four zones have obvious different spectral characteristics (seen from the color), please add a subfigure of the multiple spectral characteristics from the satellite sensors for each zone to make it clear.

Response 3: The spectra are presented in the following figures 5 and 7. In order to avoid repetition, we have chosen not to present the spectra.

 

Comment 4: Atmosphere correction for remote sensing data is necessary, which directly influences your results, the atmosphere correction for remote sensing data should need more clarified.

Response 4: As stated above, atmospheric correction is important, and we have used ACOLITE with the dark spectrum fitting method which performs well in coastal waters. More information is now provided.

Moreover, the reason for selecting the ACOLITE model is also discussed in the manuscript as shown below.

“Many atmospheric correction (AC) algorithms can be applied to multispectral images, and it is a crucial step in the processing of remote sensing data for aquatic and coastal applications. Ideally, AC aims to separate the top-of-atmosphere observation by the satellite sensor into the signal from the atmosphere and the signal from the surface to retrieve surface reflectance [89]. By using PlanetScope and RapidEye sensors, we are limited by the atmospheric correction algorithm due to a lack of spectral bands, for example in the SWIR. Even though we can apply different atmospheric corrections, the algorithm applied to the image needs to be the same to compare the sensor together. Here we adopted ACOLITE as it allows the application of the dark spectrum fitting (DSF) atmospheric correction method to all imagery evaluated in this study. Furthermore, ACOLITE has been developed for the coastal environment and is currently widely used for aquatic-based applications, such as coastal waters monitoring [90–93]. The sensitivity of satellite-derived NDVI phenology to AC could have been quantified, which was out of the scope of this study.”

 

Comment 5: The reason for the differences between Figures 10 and 11 should be more clarified.

Response 5: The differences between Fig. 10 and Fig. 11 are stated in the third paragraph of section 3.2. However, to make its clearer to the readers, we have added the following sentences to the revised manuscript.

“Even though XGBoost^Season can classify the intertidal zone with a significantly good accuracy, it is not practical to train separate models for every month of the growing season. Therefore, we have trained a single XGBoost model (XGBoost^noSeason) (Table 6) using the same set of in situ spectra, without accounting for the seasonality.”

 

Comment 6: Tide effect and relationship to phenology evolution of vegetation could be discussed if possible.

Response 6: The spectra of underwater vegetation depend on the depth of water column, the inherent optical properties of the water column and the solar geometry. Therefore, it becomes very difficult to classify underwater vegetation. Therefore, as stated in the manuscript, in the present study, to avoid the effect of the water column on the intertidal vegetation spectra, we have only used the satellite images of the site that were captured in the low tide conditions.

 

Reviewer 2 Report

A great approach to evaluating remotely sensed imagery for mapping vegetation health and coverage in this regions. There are several figures that could be re-worked or re-envisioned for better presentation of the data, and it might help the flow of the manuscript if results are restructured. Instead of presenting each type of analysis in its own section, it might be helpful to present separate conclusions that were supported by the various analyses. Since many of the analyses showed the same results for individual species of vegetation, this might simplify the presentation. Overall the manuscript would be greatly improved with a thorough review of grammatical structure and making sure the figures are included in the appropriate sections in which they are discussed. Great work on this study and I am pleased that this will contribute to the overall foundational knowledge of how best to apply remote sensing to enhance monitoring of the natural environment.

Comments for author File: Comments.pdf

Author Response

Thank you so much for your comments and suggestions which has improved the quality and readability of the article.

 

All grammatical errors marked by the reviewer are now corrected in the revised manuscript.

 

Comment 1: Line 36: Consider rephrasing so not so awkward? "Management activity relies on the provision of spatial and explicit information...making it more important than ever..." or something like that?

Response 1: The sentence is modified in the revised manuscript.

“Management activity relies on the provision of spatial and explicit information on coastal vegetation distribution making it more important than ever”

 

Comment 2: Line 53: You're discussing mapping areas using remote sensing images, no? Maybe clarify the wording a bit?

Response 2: The sentence has been modified: “Understanding vegetation spectral reflectance variability is essential and can be used to map different vegetation species using remote sensing images [ 22, 23].”

 

Comment 3: Line 57: I wouldn't call it a "moment". Perhaps "the season when the difference between spectral bands is at its greatest" or something like that.

Response 3: The sentence is updated:

“It can help, for example, to identify the season when the difference between the band is at its greatest.”

 

Comment 4: Line 63: Are these the four dominant vegetation types? Or the four most prevalent types? If so, I would add that clarification.

Response 4: The four species studied in this research are the most prevalent types in the study area. The sentence is updated:

“The prevalent vegetation types included eelgrass (Zostera Marina), macroalgae (Ascophyllum nodosum, Fucus vesiculosus), saltmarsh cordgrass (Spartina alterniflora) and Creeping saltbush (Atriplex prostrata).”

 

Comment 5: Line 68: quantify the phenological changes of...?

Response 5: The sentence is updated:

“The second objective was to assess the potential of combining multispectral sensors onboard satellite constellations offering high spatial and temporal resolution (i.e., Sentinel-2 multispectral instrument (MSI), Landsat-8 operational land imager (OLI), RapidEye (RE) and PlanetScope (PS)) to quantify the phenological change of intertidal vegetation.”

 

Comment 6: Line 85: Might you be able to briefly define "icefoot"?

Response 6: The sentence is updated:

“In winter, temperatures are well below the freezing point and most of the bay is covered by landfast ice and ice formed along and attached to the shore, also known as icefoot [29], with a mean winter temperature of -10.8°C.”

 

Comment 7: figure 1 caption: Are these 13 "harvest sites" the same as the "observation points" in Figure 2? Maybe harvesting and observing were done at the same sites, but it's a bit confusing for the reader to know if we're talking about the same sites.

Response 7: Yes the harvest sites and observation points are the same. The caption of figure 2 is now updated to be consistent with the term use:

“Figure 2. Drone image collected on 1st September 2019 presenting the delimitation of the four zones of the study site and the transect of 13 harvest points.”

 

Comment 8: Line 91: I feel like this paragraph where you discuss and define the four zones for sampling/measurement should be down below in section 2.2 where you discuss the setup of those zones.

Response 8: The paragraph is moved to section 2.2 In situ measurement.

 

Comment 9: Table 1: Why was only one station sampled for zone CS? Might be helpful to explain.

Response 9: Only one station was sampled for CS.  At the beginning of the season (in May) when the stations were identified and located, it was difficult to identify the CS. Furthermore, we wanted to concentrate our research on the EG, SC, and MA. Moreover, being dependent on the tide, we had to manage the time of data collection.

A note is added to the table: “ Only one station was sampled for CS due to time constrain (i.e. low tide sampling) and difficulty to identify CS at the beginning of the season.”

 

Comment 10: Line 110: Was this change in sampling schedule due to a decrease in changes in vegetation? Or some other reason?

Response 10: It was due to practical reasons (another field campaign in August; courses at the university, etc) and because the vegetation was more stable in summer.  The sentence is updated:

“All stations were sampled during emersion during low tide, approximately every two weeks during the maximum growth period, for this study defined as mid-May to early August, and then every three to four weeks for the rest of the season until late October, due to a decrease of change in the vegetation in summer, or for practical reasons (instrument availability and other academic activities).”

 

Comment 11: line 114: Can you add some detail to how this allometry was determined?

Response 11: The paragraph have been updated:

“Plant allometry was determined for each station along with radiometric measurements and vertical photographys for vegetation percent cover every time. More specifically, the plant allometry for eelgrass (EG), the creeping saltbush (CS), and the saltmarsh cordgrass (SC) was determined every sampling date. For each havest site, the number of plant were calculated inside the quadrat of ∼2500 cm². Furthermore, the number of leaves by plant and their length for EG and SC were measured on five plants chosen randomly inside the station. For the CS, the length of the stem was measured using ramdomly five plants. The length of the leaves and stem were realised with a measuring tape directly on the field. Solely at the end of the season, the vegetation inside the quadrat was collected for biomass determination. The Leaf Area Index (LAI) was calculated using a similar method by [ 37 ] and applied for the EG and SC. Three shoots were collected onin the field and photographed with their leaves flattened out on a white board next to a ruler. Leaf surface area (SLeaf) was determined using the Wand Tool of the ImageJ software [ 38 ]. Leaf was measured twice for the photosynthetic leaf surface (using only the green parts of the leaves) and for total leaf surface (including the necrotic tissues). The LAI was then calculated based on the following by Watson [37] :”

 

Comment 12: Line 200: should you specify all months of what? It's obviously not all months of the year, but if you are going for all months in the growing season, maybe you could state that?

Response 12: The sentence is updated:

“To cover all months of the growing season, an image from August 2020 was used as no clear sky images were available in August 2019.”

 

Comment 13: Line 226: Throughout this paragraph, when you mention the decrease in length of the leaves or stems after the August peak in the growing season, are you talking about a decrease in the absolute length of each of these parts or a decrease in the rate of growth? Might be helpful to specify

Response 13: We refer to the absolute length of the leave’s. The paragraph is updated:

“Over the season, the number of leaves per shot, the leaves and stem maximum length for  Zostera marina (EG), Spartina alterniflora (SC) and Atripex prostrata (CS) were measured (Fig 4a). For the SC and the CS, the maximum length increased from May to August with lengths reaching about 35 cm (except for station S9 which remains shorter than 20 cm), which is 30 times longer than at the beginning of the season in mid-May. A sharp decrease in the absolute length of the stem and leaves occurred after the peak reached in August, with leaves length values ~10 cm in October. In contrast, EG leaf length at the beginning of the season was much longer (10 to 25 cm already) than the other species. It must be noted that the length of the leaf was inversely proportional to their elevation relative to the sea level, with longer leaves in deeper waters (S1-S3). More than twice as shorter leaves were observed at station S4 located at the fringe of the meadow in sightly shallower waters (Fig 2). The eelgrass leaves maximum length increased until September, reaching lengths >20 cm with a maximum of 80 cm, and their absolute length decreased at all stations thereafter.”

 

Comment 14: Figure 4: Perhaps using more distinct colors might help visually ID the profiles in this graph?

Response 14: The same color palette was used thought this paper. Indeed, the color blue is associated with EG, red with SC, and purple with CS. We prefer not changing the color palette for this figure for consistency with the other plots.

 

Comment 15: Line 248: Was this also due to tissue necrosis?

Response 15: No as there was almost no necrosis in the SC leaves. The decrease in LAI is mainly due to the dead of pigment in the tissue caused by the temperature change. 

A sentence is added to clarify: “ The seasonal evolution of EG LAI is explained with tissues growth until August and thereafter a decrease due to the appearance of necrosis on the leaves which reduced the LAI based on the green part of the leaves. As for the SC, the LAI increased from May to July (S11 ad S12) or August (S9 and S10) reaching values > 1 m2 m−2 (maximum close to 3 m2 m−2) and then decrease to values of 0.5 to 1 m2 m−2 in October due to the  pigments degradation in the tissue caused by the temperature change.

 

Comment 16: Line 256: Do you mean the other vegetation species?

Response 16: Yes, the correction is applied: “Despite a marked seasonality in terms of leaf length and LAI (Fig 4), the standard deviation of raw Rsurf is much lower for Spartina alterniflora (SC) compared to the other vegetation species”

 

Comment 17: Figure 5: Great findings here! But I feel the gray portion of "a" is a bit confusing since it's not explained, in legend or caption. Also, it might be helpful to reflect the differences between these two graphs in the y-axis label (e.g. "Reflectance-Raw" and "Reflectance-MSC corrected" or something like that)

Response 17: Thanks! The gray portion is actually the superposition of the standard deviation (shade area). The y-axis have been modified.

 

Comment 18: Line 277: I'm not sure this makes sense as written. Looking at the Zostera (EG) profile, there's a very low reflectance peak in the beginning of the season (early May) before jumping up to the highest reflectance for this season and then remaining mid to low peaks. Is this what you're referring to with the May 23rd trace that was different from the others? If so, I think this should be clarified.

Response 18: Yes. Very low reflectance peak at the beginning of the season (early May) before jumping up to the highest reflectance for this season at the end of May. Afterward, the spectra remain with mid to low peaks.

The sentence is updated: “The EG show a very low reflectance peak 290 at the beginning of the season (early May) before jumping up to the highest reflectance for this season at the end of May. Afterward, the spectra remain with mid to low peaks and higher reflectance at the beginning of the season while low values were observed in the middle and at the end of the season.”

 

Comment 19: Line 299: Did it change over the season, or changed at the beginning and end of the season?

Response 19: The sentence is updated: “The seasonal shift in the reflectance peak can be explained by the composition of the dominant pigment changing at the beginning and end the season.”

 

Comment 20: Line 305: This wording is a bit contrary between these two sentences. The first sentence you state MA and SC spectra don't change over the season due to low SAM values, but then you say there's a clear seasonal signal. Perhaps re-write for clarity?

Response 20: The wording is updated: “ Even though the SAM values are low for SC, we can observe a difference in the spectral shape between the beginning/end and the middle of the season, but the differences remained small in terms of SAM.”

 

Comment 21: Line 314: Do you really think the EG would have shown high growth, then partial necrosis, then growth again in the same season?

Response 21: From field observations, we could see that although necrosis was present on the leaves, the EG plants continued to grow. Thus, it might have been possible to observe rapid growth, then the appearance of necrosis and then growth again. As mentioned, the exact reason for this anomaly could not be confirmed due to unavailability of supporting data. At any moment, the EG spectra stay remarkably similar without a clear seasonal evolution.

 

Comment 22: Line 328: Isn't this the same finding for all the vegetations species, as stated in sentence 3?

Response 22: Yes, but its even truer for EG because at the spring (end of May) and in the early October, the EG is the only green vegetation.

 

Comment 23: Figure 7: It would be helpful to the reader if you kept the naming convention the same throughout text and figures. Either use the scientific names throughout, or continue using the two letter abbreviation.

Also, I think it might help to use a gradient between two or three colors (e.g.  yellow to red, or green to blue, etc.) to show seasonal progression (or lack thereof), as it is, it's a little hard to follow the colors and whether they're in "order" for seasonal change.

Response 23: The two-letter abbreviation is now added to the figure 7. As for the color, the gradient visdis  have been use on the figure. To avoid confusion due to addition of too many colors, we have restricted to the standard blue, green, yellow, red, and purple, throughout the manuscript.

 

Comment 24: Figure's 8 and 9 should be here since they are referenced in that last section, not in section 3.2 starting here. Might help with the flow.

Response 24: The figures have been moved

 

Comment 25: Line 340: I would avoid using the term "moment" and instead opt for "time period", "season", "dates", etc.

Response 25: Yes you are right. The sentence is updated: “ This sensor was selected because i) it provides the best response to seasonal variability and availability over the critical dates for the vegetation phenology; and ii) it has 13 spectral bands including key bands in the red-edge portion of the spectrum that is suitable for vegetation mapping.

 

Comment 26: Figure 8: I think I would label these plots as the two letter abbreviation for each type of vegetation/sediment.

Response 26: The two letter appreviation has been added to the figure

 

Comment 27: This needs to be up under Figure 7 and 8. I really think this information needs to be presented in a different way. It is not at all intuitive what's being said here and having to figure out which is the species being compared to (or rather which species is missing from the color traces) makes this difficult to follow. Perhaps a pairwise comparison for each month? At very least, each of these plots needs to be labeled with the species being compared to.

Response 27: The figures have been moved

 

Comment 28: Figure 11: Caption needs to stand alone, not refer to another caption elsewhere. Also, this figure needs to be up beneath Figure 10 to facilitate comparison.

Response 28: The caption has been modified: “Classification of coastal and intertidal vegetation using XGBoost^noSeason applied to Sentinel- 2 time series.”

 

Comment 29: Figure 12 should be here so we can look at what is being discussed.

Response 29: With the modification made to the document, the figure 12 should be located to the right location.

 

Comment 30: line 399: Awkward as written, needs to be reworded.

Response 30: The sentence is update:

“For both species, we found an important variability in terms of LAI at some station at a given NDVI value.”

 

Comment 31: Line 399: Why were only EG and SC represented here in this analysis?

Response 31: The LAI was only measured on the EG and SC.

 

Comment 32: Line 425: This doesn't make sense as phrased

Response 32: The sentence is updated: “The shape and range is similar for both normalised and raw spectra for SD and MA, but the normalisation reduces the range for CS, SC and EG”

 

Comment 33: Line 432: What do the values in parentheses represent?

Response 33: The values in parentheses present the values obtain by the raw spectra (the non MSC-Corrected spectra).

“We can see a clear seasonal evolution for the CS with NDVI as obtained from MSC-corrected (raw) reflectance ranging between 0.44 and 0.76(0.35 and 0.85).”

However, we have now removed these values in the brackets to avoid confusion.

 

Comment 34: Line 433: If you're going to represent corrected and raw ranges, then you should be consistent with providing a range instead of individual peak or low points

Response 34: We have removed the NDVI values from raw data to avoid confusion. Now only the NDVI values from corrected data are shown.

 

Comment 35: These figures need to be in this section where they are discussed.

Response 35: The figure has been moved.

 

Comment 36: Line 474: This doesn't make sense and needs to be re-written.

Response 36: The sentence is updated: “By combining the time series of the two sensors, we can better see a complementarity of the data, offering a better understanding and monitoring of the phenology of the vegetation.”

 

Comment 37: Figure 15: The regions of interest are too small here to be of much value

Response 37: Even though the regions of interest are small, there is significant variability in the NDVI values. In addition, Figure 15 and 16 show the variations that can be observed over a larger area, showing that an evolution is observable and the impact of the tidal level on the images.

 

Comment 38: Figure 16: As previously stated, the captions should stand alone.

Response 38: The caption has been modified:

“Time series of NDVI from PlanetScope sensor acquisition obtained at low tide and zoomed in the region of interest. Homogenous region of interest (ROIs) for the four vegetation types and sediment where the NDVI values were extracted are also shown.”

 

Comment 39: Line 617: This needs to be separated into two topics, the first, which species showed changes in phenology, and then second discuss which species showed changes in in situ reflectance spectra.

Response 39: The sentence has been reworked:

“We identified a significant seasonal change in phenology of saltmarsh cordgrass and creeping saltbush. Even though some seasonal change can be observed for some vegetation types, no significant changes were observed in the in situ reflectance spectra for eelgrass and macroalgae”

 

Comment 40: Line 627: What about the initial coverage of the quadrats? Would help to complete the thought.

Response 40: The coverage of the quadrats have been add to the sentence:

“By combining Sentinel-2 and Planet imagery, we showed that the seasonal evolution of eelgrass NDVI was more evident than with in situ measures, likely because of the initial coverage of the quadrats (2500 cm²).”

 

Reviewer 3 Report

The paper presents a study of vegetation coverage colours on a transient coastal marsh during ice-free season alternating flood and ebb tides. The study is located in the St-Laurent estuary. The raw materials are images enable from different satellites.

The paper is well written and the paper flow is excellent. The scientific protocol is clearly presented. Several suggestions are proposed. The main drawback of the paper is the missing “s” all over the pages; where singular and plural are, often miss. We propose an author reading for this point of grammar/clarification.

Science:

The position of the sun (section 3.1.2), the presence of a breeze could affect the vegetation colours. The two sides of a leaves could have different colours. A breeze could change the ocean surface colours. The paper could mentioned if a hypothesis concerns sun position and meteorological condition on wind.

The linear regression in Fig12.A could be another one, if two groups of points are considered: one < 0.5, the second > 0.5. This last one follows a more vertical line with a negative derivative.

Edition:

A sentence could present the word “phenology”

Maybe separate the 3 items after Fig2):, line 93

Standard deviation of what ? line 136

“Spectra are spectraly different”, could be improved, lined 332

Maybe define r in a sentence, line 398

Edit “without issues” line 559, “number of the band” line 562, “increased in number” line 568

Edit refs n° 8 [923]?, 15 geo?, editors n°27, n° 28, n°37, link n°29, “.,” n° 31, university n°53, n°51, pages n° 78 and n° 80, Capital letter n°32

Typos:

Photographys line 112, by Watson line 123

Sets line 138, are line 159, images line 183, homogenous line 258, Table 6;. line 359, Pearson line 405, resolutions lines 419 539, types line 454, stations line 462, limits line 466, saltmarsh line 483, check different line 492, it line 510, board of a wide line 516, demonstrates line 534, moments line 558, depends line 588, types line 613, correlates line 623, stations line 400, island line 353

Author Response

Thank you so much for your comments and suggestions which has improved the quality and readability of the article.

 

Comment 1: The paper could mentioned if a hypothesis concerns sun position and meteorological condition on wind.

Response 1: As stated by Qu (2018), “In most of the cases, the surface reflectance depends on the wavelength and incident/viewing angle according to the optical properties and structures of the surface.” As all the spectra all collected in the air, therefore, there is no effect of the slope of the water surface which is a function of surface wind speed and direction. Moreover, as stated in the manuscript, the effect of solar geometry and atmospheric conditions are corrected.

Qu (2018) is added as reference in the revised manuscript.

“Because the Rsurf were not collected on the same day and time of the day, under different sky conditions, sun-view geometry, and instrument settings (i.e., integration time) (Table 2), variations in magnitude due to illumination differences were present [40].”

40. Qu, Y. Sea Surface Albedo. In Comprehensive Remote Sensing; Liang, S., Ed.; Elsevier, 2018; pp. 163–185

 

Comment 2: The linear regression in Fig12.A could be another one, if two groups of points are considered: one < 0.5, the second > 0.5. This last one follows a more vertical line with a negative derivative.

Response 2: We test many linear regressions for the Fig12. The one shown in this paper present the best relation with between the LAI and the NDVI.

 

Comment 3: A sentence could present the word “phenology”

Response 3: The definition of the word phenology is presented in the paper.

“In addition, understanding the seasonal evolution, i.e., the phenology [24], of the reflectance spectra of key vegetation type can provide additional clues for distinguishing them from space [25,26].”

 

Comment 4: Maybe separate the 3 items after Fig 2):, line 93

Response 4: The sentence has been updated:

“The sampling area was divided in four zones based on the frequency of immersion, which is a function of elevation, the ice in winter, the water temperature and salinity (Fig 2). The zones are: 1) the eelgrass bed of Zostera marina (EG) ; 2) the mud flat with microphytobenthos (SD: sediment) and macroalgae zone composed of Fucus and Ascophyllum mainly attached to scattered boulders (MA); 3) the low marsh Cordgrass zone dominated by Spartina alterniflora (SC) and 4) High marsh zone mainly dominated by Atriplex prostrata (CS, creeping saltbush) [30,31,33].”

 

Comment 5: Standard deviation of what ? line 136

Response 5: The sentence has been updated:

“For each spectrum recorded, the ASD collects five spectra and computes the average and standard deviation of the spectra.”

 

Comment 6: “Spectra are spectraly different”, could be improved, lined 332

Response 6: The sentence has been updated:

“The sediment spectra are different from all of the vegetation spectra (Fig 9e).”

 

Comment 7: Maybe define r in a sentence, line 398

Response 7: Eq. 6 is now added to the revised manuscript to define r.

 

Comment 8: Edit “without issues” line 559,

Response 8: The sentence has been updated:

“The spatial resolution of PlanetScope images provide many details, but not without complication.”

 

Comment 9: Edit “number of the band” line 562,

Response 9: The sentence has been updated:

“The low spectral resolution of the first Dove generation limits its use for the vegetation type classification”

 

Comment 10: Edit “increased in number” line 568

Response 10: The sentence has been updated:

“The next generation of PlanetScope sensors launch in January 2022 with a greater number of bands (eight bands) and radiometric signal may help resolving most of the limitations of the early sensor fleets.”

 

Comment 11: Edit refs n° 8 [923]?,

Response 11: Corrected

 

Comment 12: 15 geo?

Response 12: Corrected

 

Comment13: editors n°27, n° 28, n°37,

Response 13: Corrected

 

Comment 14: link n°29

Response 14: Corrected

 

Comment 15: link n° 31,

Response 15:  Corrected

 

Comment 16: university n°53,

Response 16: Corrected

 

Comment 17: university n°51,

Response 17: Corrected

 

Comment 18: pages n° 78 and n° 80,

Response 18: Corrected

 

Comment 19: Capital letter n°32

Response 19: Corrected

 

Comment 20: Photographys line 112

Response 20: Accepted

 

Comment 21: by Watson line 123

Response 21: Accepted

 

Comment 22: Sets line 138

Response 22: Accepted

 

Comment 23: are line 159

Response 23: Accepted

 

Comment 24: images line 183

Response 24: Accepted

 

Comment 25: homogenous line 258

Response 25: Accepted

 

Comment 26: Table 6;. line 359

Response 26: Accepted

 

Comment 27: Pearson line 405

Response 27: Accepted

 

Comment 28: resolutions lines 419 539

Response 28: Accepted

 

Comment 29: types line 454

Response 29: Accepted

 

Comment 30: stations line 462

Response 30: Accepted

 

Comment 31: limits line 466

Response 31: Accepted

 

Comment 32: saltmarsh line 483

Response 32: Accepted

 

Comment 33: check different line 492

Response 33: Accepted

 

Comment 34: it line 510

Response 34: Accepted

 

Comment 35: board of a wide line 516

Response 35: Accepted

 

Comment 36: demonstrates line 534

Response 36: Accepted

 

Comment 37: moments line 558

Response 37: The term moments was changed by “stages”

 

Comment 38: depends line 588

Response 38: Accepted

 

Comment 39: types line 613

Response 39: Accepted

 

Comment 40: correlates line 623

Response 40: Accepted

 

Comment 41: stations line 400

Response 41: Accepted

 

Comment 42: island line 353

Response 42: Accepted