Review Reports - Phenological Changes of Mongolian Oak Depending on the Micro-Climate Changes Due to Urbanization

Round 1

Reviewer 1 Report

Comment for manuscript submitted to Remote Sensing

Title: Phenological changes of plants depending to the micro-climate changes due to urbanization

Authors: A Reum Kim, Chi Hong Lim, Bong Soon Lim, Jaewon Seol, and Chang Seok Lee

In this manuscript, the authors attempted to combine phenological camera data with MODIS satellite images to explore the beginning dates of greening of Quercus Mongolica forest under different land use intensities, and to monitor the velocity of sap flow, the physiological changes, to interpret the changes for vegetation phenology. The aim of the manuscript was clear, and the methods applied were reasonable. But the conclusion was not so robust and and the expressions were not logical enough. I have the following concerns need to be addressed by the authors.

My major concerns are as the following:

1. In the introduction section, you told that urban area was an important area for phenology, however, did you take into consideration other effects of human activities, such as irrigation on vegetation phenology in urban areas？

2. Through the map of study area, we can find the complicated topography of the research area. Could you please make it clear how topography affected your research results?

3. The selection basis for interest area (ROIs in the manuscript) was not sufficient.

4. Introduction to "urban heat island effect" in the manuscript was detailed, but analysis on temperature changes was not conducted so thoroughly to support the UHI effects on different research sites.

I have also some minor points as follows:

(1) Line 14-15: please point out the object vegetation type UHI may affect.

(2) Line 20: for the first time appearing in the abstract section, what is AGDD?

(3) Line 20-21: you said that the leaf unfolding date was earlier than expected. By the way, what was your expectation?

(4) Line 51-57: in this paragraph you told that urban area was an important area for phenology, however, did you take into consideration other effects of human activities, such as irrigation on vegetation phenology in urban areas

(5) Some compilation errors appeared in line 38,174, 180, 182, and 208 et al. Please revise completely.

(6) Regarding the calculation of urbanization rate, please add the reference here if the methods was put forward by others, or tell in more details if this was a method you presented first.

(7) I noticed that there were two sites representing the city center and the suburb respectively. Why was the urbanization level of the sites in the suburbs higher?

(8) In the first paragraph of 2.2, you told that the digital camera recorded three times a day, but you only chose the noon time as the experimental data. Please tell the reason.

(10) The length of observation period of camera in this study was only five months (from 2016/2/19 to 2016/7/18), but why not for a whole growing season?

(11) There were many problems with the figures and tables in the manuscript. The most serious problem is there were so many tables and figure. Please check to remove duplicate figures and tables. Other problems related to figures and tables:
(a) Figure 3 and Figure 4: the legend was not very clear.
(b) Figure 4: as digital camera observation period was five months from 2016/2/19 to 2016/7/18, the starting and ending dates for the horizontal axis were not correct. Please check it.
(c) Table 6: the data for the last column were obviously wrong.
(d) Figure 5 and figure 6: the horizontal coordinates for these two figures ((a) and (b)) were inconsistent.
(f) Figure 8: was the unit for AGDD right? Similar errors in table7.
(g) The vertical axis data of figure7 were not consistent with the observation value given in table4?
(h) Similarly, figure8 does not support the content described in the first paragraph of 3.2. In addition, since your AGDD was obtained through two sets of methods, why only one was reflected in figure8?
(i) Take consideration of combining Table 9 and Table 10. In addition, these data did not fully support what you said in conclusion 3.3.
(j) A large number of tables and figures in the manuscript could be simplified and merged, and the placements of some of them were not very suitable, such as Table8, Figure10, etc.

(12) The manuscript did not achieve its original goals. Please strengthen the discussion and conclusion section. Some parts of the discussion section will be more suitable to appear in the Introduction section.

Based on above, I would like to recommend a major revision decision.

Author Response

Response to reviewer’s comments

My major concerns are as the following:

In the introduction section, you told that urban area was an important area for phenology, however, did you take into consideration other effects of human activities, such as irrigation on vegetation phenology in urban areas？
☞ The Mongolian oak forest targeted in this study is located in urban areas but in mountainous areas escaped from direct human interference, so it is not affected by Irrigation.
Through the map of study area, we can find the complicated topography of the research area. Could you please make it clear how topography affected your research results?
☞ We added the following contents in our manuscript to explain such an effect due to a difference of topographic condition: According to Richardson et al. (2019), increases in latitude 1°N, longitude 1°E, and altitude 100 m result in delays of 2.6 ± 0.2 days, 0.6 ± 0.1 days, and 2.1 ± 0.2 days, respectively in leaf unfolding date. Based on this information, this study corrected the difference of leaf unfolding dates due to differences in latitude, longitude, and altitude among the study areas. Lines 205-218.
The selection basis for interest area (ROIs in the manuscript) was not sufficient.
☞ We explained the selection basis for ROI (the region of interest) in our manuscript. Lines 167-177.
Introduction to "urban heat island effect" in the manuscript was detailed, but analysis on temperature changes was not conducted so thoroughly to support the UHI effects on different research sites.
☞ We explained the effect based on AGDD and advanced green-up date in urban center sites. Furthermore, we explained the effect also based on the difference between the observed and expected dates of the green-up. Lines370-377.

I ha`ve also some minor points as follows:

(1) Line 14-15: please point out the object vegetation type UHI may affect.

☞ We revised our manuscript by accepting reviewer’s comment. Lines 14-15.

(2) Line 20: for the first time appearing in the abstract section, what is AGDD?

☞ We revised our manuscript by accepting reviewer’s comment. Line 20.

(3) Line 20-21: you said that the leaf unfolding date was earlier than expected. By the way, what was your expectation?

☞ We designated a reference site to clarify the differences due to land use intensity among sites. Mt. Jeombong site is the reference site. Mt. Jeombong site maintains a healthy and integrate stand of Q. mongolica as it is designated as a reserve by Korea Forest Service and thereby is escaped from artificial interferences. Expected date is leaf unfolding date calculated by assuming unurbanized state, that is, based on Mt. Jeombong site.

☞ As was mentioned before, the Mongolian oak forest targeted in this study is located in urban areas but in mountainous areas escaped from direct human interference, so it is not affected by Irrigation.

(5) Some compilation errors appeared in line 38,174, 180, 182, and 208 et al. Please revise completely.

☞ We revised our manuscript by accepting reviewer’s comments. Lines 40, 187, 196, 198, 229.

(6) Regarding the calculation of urbanization rate, please add the reference here if the methods was put forward by others, or tell in more details if this was a method you presented first.

☞ We addressed the method in our manuscript. Lines 116-119.

(7) I noticed that there were two sites representing the city center and the suburb respectively. Why was the urbanization level of the sites in the suburbs higher?

☞ We miscalculated by mistake. We revised the data. Lines 119-122, 129-130, 276-277, 291-293.

(8) In the first paragraph of 2.2, you told that the digital camera recorded three times a day, but you only chose the noon time as the experimental data. Please tell the reason.

☞ In good weather conditions, only the 12:30 images were used for the analysis to avoid direct sunlight and maintain consistency. Because the images taken at 09:00 and 14:30 are highly affected by sunlight, so the brightness of the images is often not constant. In the case of poor weather conditions such as rain or fog, images taken at different times were used.

(10) The length of observation period of camera in this study was only five months (from 2016/2/19 to 2016/7/18), but why not for a whole growing season?

☞ The response to climate change is known to be most evident in spring phenology, and we carried out this study focusing on spring phenology.

(a) Figure 3 and Figure 4: the legend was not very clear.

☞ We revised our manuscript by accepting reviewer’s comments. Lines 246-255.

(b) Figure 4: as digital camera observation period was five months from 2016/2/19 to 2016/7/18, the starting and ending dates for the horizontal axis were not correct. Please check it.

☞ 2016/2/19 - 2016/7/18 is the period that MODIS video was collected. We added digital camera observation period in the text (Table 1.) To improve resolution of the graph, only the corresponding period (Doy 60-240) was expressed.
(c) Table 6: the data for the last column were obviously wrong.

☞ We revised our manuscript by accepting reviewer’s comments.
(d) Figure 5 and figure 6: the horizontal coordinates for these two figures ((a) and (b)) were inconsistent.

☞ We revised our manuscript by accepting reviewer’s comments.
(f) Figure 8: was the unit for AGDD right? Similar errors in table7.

☞ We revised our manuscript by accepting reviewer’s comments.

(g) The vertical axis data of figure7 were not consistent with the observation value given in table4?

☞ We revised our manuscript by accepting reviewer’s comments.

(h) Similarly, figure8 does not support the content described in the first paragraph of 3.2. In addition, since your AGDD was obtained through two sets of methods, why only one was reflected in figure8?

☞ We revised our manuscript by accepting reviewer’s comments. And the figure is a graph showing the correlation between the green-up dates derived through AGDD and the urbanization rate. Lines 301-303.

(i) Take consideration of combining Table 9 and Table 10. In addition, these data did not fully support what you said in conclusion 3.3.

☞ We revised our manuscript by accepting reviewer’s comments.

(j) A large number of tables and figures in the manuscript could be simplified and merged, and the placements of some of them were not very suitable, such as Table8, Figure10, etc.

☞ We revised our manuscript by accepting reviewer’s comments.

Based on above, I would like to recommend a major revision decision.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments on “Phenological changes of plants depending to the micro-climate changes due to urbanization”

This study estimated leaf unfolding dates from multiple indicators, including digital camera, satellite vegetation index, and a novel tool sap flow. The authors found an earlier leaf unfolding date by increasing urbanization, which can be considered as an additional evidence for the influence of urban hear island on vegetation phenology. This manuscript needs substantial revisions and its publication in RS is depending on the improvement. Please carefully consider my following concerns.

1, The introduction section should be rewritten. Some information is missing. For example, what is the advantage to investigate leaf unfolding date by sap flow? Whether it is a physiological indicator of phenology and has a better ability to reflect phenology of evergreen plants?

2, Line 56, The main purpose to investigate urban vegetation phenology does not focus on the influence of phenological changes on climate because of the small areas of urban vegetation. The urban provides a scenario to study how vegetation phenology changes when temperature increases in the future.

3, Line 61, Some necessary literatures are missing. Please cite “Cao, R.Y., Shen, M.G., Zhou, J., et al., 2018. Modeling vegetation green-up dates across the Tibetan Plateau by including both seasonal and daily temperature and precipitation. Agricultural and Forest Meteorology, 249, 176-186.”

4, Line 75, what is “a study method”?

Line 85, revise “think”

Line 87, “purpose”-objectives

5, Line 123-134, The MODIS data has a spatial resolution of 500 m, whether it is appropriate for phenology detection at the study site?

6, Line 165, how you process the MODIS time-series data? How to remove the residual noise in the EVI time-series data?

7, Line 174, format error

8, Lines 191-193, how to correct the influence of geographic location on phenology? Please give more details.

9, Line 214, “Figure 4”-It should be figure 3

10, I suggest put the figures 3 and 5 together, and put the figures 4 and 6 together.

11, Line 239, what is the “latitude and altitude correction”

12, Line 249 “green-up start date”-green-up date

13, Figure 7, please give the significant level

14, Figure 9, what are the short lines in the bottom of each panel?

15, The discussion section must be improved. Results are repeated in this section, please remove these sentences.

Lines 324-326, format errors

Section 4.2, please discuss the additional information provided by sap flow, and what is the implication for vegetation phenological monitoring?

16, please rewrote the Conclusion section. The current conclusion is ambiguous.

17, the manuscript needs substantial language editing by native speakers.

Author Response

Response to reviewer’s comments

☞ We revised our manuscript by accepting reviewer’s valuable comment. Lines 78-98.

☞ We addressed such contents in Introduction section. Lines 56-61.

But we reinforced the part by accepting reviewer’s comments.

☞ We reinforced necessary literatures by accepting reviewer’s comments.

4, Line 75, what is “a study method”?

☞ It means a study method to confirm the phenological phases of plants through seasonal changes of the physiological response of the plant such as sap flow.

Line 85, revise “think”

☞ We revised this part of our manuscript by accepting reviewer’s comment.

Line 87, “purpose”-objectives

☞ We revised this part of our manuscript by accepting reviewer’s comment.

5, Line 123-134, The MODIS data has a spatial resolution of 500 m, whether it is appropriate for phenology detection at the study site?

☞ As the reviewer points out, MODIS images are useful for monitoring a wide range of areas due to their low resolution, but are not appropriate for monitoring phenological changes in plant species as the site based data source. A challenge of phenological research is to integrate remotely sensed observations obtained at different spatial and temporal scales to provide information that contains both a high temporal density and ﬁne spatial scale observations. In this study, we solved this problem by improving the level of spatial detail and the temporal density required for eﬃcient monitoring of vegetation phenology by applying two remote sensing techniques, MODIS and digital camera images.

6, Line 165, how you process the MODIS time-series data? How to remove the residual noise in the EVI time-series data?

☞ For MODIS data analysis, we followed the same procedures used for analysis of the digital camera data. The VI was obtained using the smooth curve fitting method to remove variation and to gather trends because interpretation error can occur due to data errors and variation depending on weather conditions. In order to derive phenological event dates from phenological time-series data, we used a sigmoid-based equation: In the sigmoid model, phenological transition dates were attained by obtaining local extrema in the rate of change of curvature K (hereafter, abbreviated as curvature K). Lines 190-193

7, Line 174, format error

☞ We revised this part of our manuscript by accepting reviewer’s comment. Line 187

8, Lines 191-193, how to correct the influence of geographic location on phenology? Please give more details.

☞ We revised our manuscript by adding the following contents: According to Richardson et al. (2019), increases in latitude 1°N, longitude 1°E, and altitude 100 m result in delays of 2.6 ± 0.2 days, 0.6 ± 0.1 days, and 2.1 ± 0.2 days, respectively in leaf unfolding date. Based on this information, this study corrected the difference of leaf unfolding dates due to differences in latitude, longitude, and altitude among the study areas. Lines 205-218.

9, Line 214, “Figure 4”-It should be figure 3

☞ We revised the part of manuscript. Line 235.

10, I suggest put the figures 3 and 5 together, and put the figures 4 and 6 together.

☞ We revised our manuscript by accepting reviewer’s comment. Lines 246-255.

11, Line 239, what is the “latitude and altitude correction”

☞ As was mentioned above, we corrected the leaf unfolding date by reflecting the differences in latitude, longitude, and elevation. Lines 205-218.

12, Line 249 “green-up start date”-green-up date

☞ We revised our manuscript by accepting reviewer’s comment.

13, Figure 7, please give the significant level

☞ We revised our manuscript by accepting reviewer’s comment. Line 287.

14, Figure 9, what are the short lines in the bottom of each panel?

☞ We revised the parts by removing them. Lines 314-315.

15, The discussion section must be improved. Results are repeated in this section, please remove these sentences.

☞ We revised our manuscript by accepting reviewer’s comment. Lines 344-359, 423-430.

Lines 324-326, format errors

☞ We revised our manuscript. Line 330.

Section 4.2, please discuss the additional information provided by sap flow, and what is the implication for vegetation phenological monitoring?

☞ We revised our manuscript by accepting reviewer’s comment. Lines 389-395, 405-422.

16, please rewrote the Conclusion section. The current conclusion is ambiguous.

☞ We revised our manuscript by accepting reviewer’s comment. Lines 488-518.

17, the manuscript needs substantial language editing by native speakers.

☞ We’ll get language editing by native speaker.

Author Response File: Author Response.pdf

Reviewer 3 Report

See the attached file with my comments

Comments for author File: Comments.pdf

Author Response

Response to reviewer’s comments

1.Line 16, 17 Region or nation should be mentioned

☞ We revised our manuscript by accepting reviewer’s comment. Lines 15-17.

2.Avoid using we, I, in an article writing, correct this where applicable in the text

☞ In Method and Discussion sections, we used subjective ‘we’ several times. However, as a result of the review, it is not particularly awkward, so we’d like to keep the current state. But I would like to leave the amendment to the Language Editing process.

3.Line 50 [23,24] Cite relevant and recent studies here See the suggested few below;

https://doi.org/10.1016/j.scitotenv.2020.144224

https://doi.org/10.1007/s12665-018-7252-6

https://doi.org/10.1073/pnas.1911117117

https://doi.org/10.1007/s12145-019-00391-2

☞ We revised our manuscript by accepting reviewer’s valuable comments.

4.Line 73 [16, 42] Cite these studies here;

https://doi.org/10.1016/j.inpa.2018.02.002

https://doi.org/10.1016/j.jenvman.2021.112112

☞ We cited references that reviewer recommended.

5.What are the tower types of these sites? (Mts. Mido, Umyeon, Cheonggye and Bulam)

☞ Because there were no towers in these areas, cameras were installed in trees similar to towers in height.

6.Table 2. Urbanization rate, For what year? year and source should be known

☞ We added them in the text. Lines 116- 119.

7.Line 160 we

☞ As was mentioned above, in Method and Discussion sections, we used subjective ‘we’ several times. However, as a result of the review, it is not particularly awkward, so we’d like to keep the current state. But I would like to leave the amendment to the Language Editing process.

8.Figure 2. Date here shows that the field work was done 7 years ago, there will be a lots of changes by now, any justification for this?

☞ It was old one. We exchanged the photo into new one. Lines 175-177.

9.Table 6. Difference in days is 0? correct this accordingly

☞ We revised that by accepting reviewer’s valuable comments. Lines 258- 260.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Most of my concerns were addressed, but the manuscript still need to be improved for the following points:

1. In section 2.2, please tell the number of channels, and the most important channels applied for the application in your work when introduced the digital camera. This means, besides introduction to the items including in table 3, you should add more important information, such information and the methods for ExG.

2. Section 2.5, as to the function of sap flow value in the work, there are still some serious problems.

(1) what was "the easonal trajectory of sap flow (appeared in line 230, revised version)? I guess should be "seasonal".

(2) "The seasonal trajectory of sap flow was interpreted using curvature K (formula 3) based on the daily sap velocity Transition date of the sap flow was compared with the phenological transition date obtained from digital camera and MODIS installed at the same site". How to calculate curvature K? Where is formula 3? But in above text, formula was for calculation of EVI of MODIS imagery.

These were just two obvious problems, but please check the whole text very carefully to avoid similar problems.

Author Response

Response to reviewer’s comments

Reviewer 1

Most of my concerns were addressed, but the manuscript still need to be improved for the following points:

In section 2.2, please tell the number of channels, and the most important channels applied for the application in your work when introduced the digital camera. This means, besides introduction to the items including in table 3, you should add more important information, such information and the methods for ExG.

☞ We revised our manuscript by accepting reviewer’s comment. Table 3 and Lines 171-177.

Section 2.5, as to the function of sap flow value in the work, there are still some serious problems.

(1) what was "the easonal trajectory of sap flow (appeared in line 230, revised version)? I guess should be "seasonal".

☞ We revised our manuscript by accepting reviewer’s comment. Line 215.

☞ We revised our manuscript by accepting reviewer’s comment. Line 216.

These were just two obvious problems, but please check the whole text very carefully to avoid similar problems.

☞ We checked our manuscript carefully.

Author Response File: Author Response.pdf

Reviewer 2 Report

The revised version has been improved. It can be accepted by RS after addressing the following comments:

1, for the conclusion and abstract, it would be better to summary some quantitative findings，e.g., how many days advanced by increasing temperature one degree for the species at your study site.

2, Introduction section, lines 96-116, please combine the two paragraphs into one paragraph;

3, my previous comment: how you process the MODIS time-series data? How to remove the residual noise in the EVI time-series data?

please note that in the MODIS phenology product (version 5), to remove the redidual noise, a median smooth filter was applied to the EVI time-series data before logistic curve fitting.

4, green-up start date - change to green-up date throughout the manuscript.

5, my previous comment: Figure 7, please give the significant level -please give the P value.

Author Response

Response to reviewer’s comments

Reviewer 2

The revised version has been improved. It can be accepted by RS after addressing the following comments:

1, for the conclusion and abstract, it would be better to summary some quantitative findings，e.g., how many days advanced by increasing temperature one degree for the species at your study site.

☞ We revised our manuscript by accepting reviewer’s comment. Lines 17-39, 375-382, 526-532.

2, Introduction section, lines 96-116, please combine the two paragraphs into one paragraph;

☞ We revised our manuscript by accepting reviewer’s comment. Lines 86-105.

3, my previous comment: how you process the MODIS time-series data? How to remove the residual noise in the EVI time-series data?

please note that in the MODIS phenology product (version 5), to remove the redidual noise, a median smooth filter was applied to the EVI time-series data before logistic curve fitting.

☞ In this study, we applied exponentially weighted moving average method to EVI time-series data for removing the residual noise. We addressed the procedure in lines 202-208.

4, green-up start date - change to green-up date throughout the manuscript.

☞ We revised our manuscript by accepting reviewer’s comment.

5, my previous comment: Figure 7, please give the significant level -please give the P value.

☞ We revised our manuscript by accepting reviewer’s comment. Lines 302-303.

Author Response File: Author Response.pdf