Greening the Urban Landscape: Assessing the Impact of Tree-Planting Initiatives and Climate Influences on Miami-Dade County’s Greenness

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I believe that this study, which identifies the climatic impacts of spatial and temporal variations in greenness, is meaningful in terms of the planting initiative. But I have some questions.

1.     Landsat images were used to detect variations in greenness. However, the climate impacts used data measured at the weather station. Is it possible to directly compare it? Comparing climatic impacts of greenness using the results of temporal changes in temperature and precipitation seems to be ambiguous results. Also, the results do not seem to suggest a direct comparison. Is it possible to utilize meteorological data with spatial distribution (e.g. modeling data, etc.)?

2.     Discussion is the part where the results derived from this study are discussed. It is written as if it were a conclusion. Also, there are no conclusions from this study.

 

3.     There are a lot of typos overall.

Comments on the Quality of English Language

I think that the quality of English is good.

Author Response

I believe that this study, which identifies the climatic impacts of spatial and temporal variations in greenness, is meaningful in terms of the planting initiative. But I have some questions.

1. Landsat images were used to detect variations in greenness. However, the climate impacts used data measured at the weather station. Is it possible to directly compare it? Comparing climatic impacts of greenness using the results of temporal changes in temperature and precipitation seems to be ambiguous results. Also, the results do not seem to suggest a direct comparison. Is it possible to utilize meteorological data with spatial distribution (e.g. modeling data, etc.)?

As highlighted in the discussion section of the manuscript, we acknowledge the inherent challenge in directly comparing climatic data with greenness data due to the discrepancy in scale. Landsat images provide a valuable tool for assessing temporal changes in greenness at a spatial scale, while weather station data offer localized measurements of climatic impacts. The discrepancy in scale is indeed an important consideration, and we have addressed this limitation in the manuscript.

2. Discussion is the part where the results derived from this study are discussed. It is written as if it were a conclusion. Also, there are no conclusions from this study.

Thank you for your insightful feedback. In response, we have developed the discussion section and added a proper conclusion.  Under discussion, we now summarize the results (as recommended by reviewer 3).

3. There are a lot of typos overall.

We have carefully edited the manuscript for typos.

 

Reviewer 2 Report

Comments and Suggestions for Authors

The conducted case study is an example utilizing well-known sensors and methods for monitoring and spatial analysis of greening changes in Miami-Dade County. The topic is interesting, but the approach to scientific evidence is not entirely correct and understandable. I have a few comments, and I ask the authors to address them. During the article revision, please highlight changes in the text in yellow.

Introduction: The description in this section does not clearly indicate what the authors want to propose as new to the global scientific community. The description of the research objectives is also unclear in this section. The authors suggest that there are two research objectives, but I identified more (71-74; 93-95; 112-113; 123-125; 144-146). Please clarify which objective is primary and which are specific objectives that indirectly result from achieving the main objective. Perhaps there are too many repetitions causing confusion. The authors attempt to determine the significance of model assessment variables; therefore, to enhance the scientific value of the article, it is advisable to state/formulate the research hypothesis at the beginning.

Methods: Please explain how Census blocks were determined. Were they equal-sized fields or of different sizes determined by property boundaries/road networks?

Apart from that the reasoning is logical, and the methods are appropriate. The authors correctly interpreted the research results. After introducing the necessary explanations, I recommend the article for publication.

Author Response

The conducted case study is an example utilizing well-known sensors and methods for monitoring and spatial analysis of greening changes in Miami-Dade County. The topic is interesting, but the approach to scientific evidence is not entirely correct and understandable. I have a few comments, and I ask the authors to address them. During the article revision, please highlight changes in the text in yellow.

Introduction: The description in this section does not clearly indicate what the authors want to propose as new to the global scientific community. The description of the research objectives is also unclear in this section. The authors suggest that there are two research objectives, but I identified more (71-74; 93-95; 112-113; 123-125; 144-146). Please clarify which objective is primary and which are specific objectives that indirectly result from achieving the main objective. Perhaps there are too many repetitions causing confusion. The authors attempt to determine the significance of model assessment variables; therefore, to enhance the scientific value of the article, it is advisable to state/formulate the research hypothesis at the beginning.

We have rewritten certain sentences of the introduction to make the primary and secondary objectives more explicit:

“The primary objective of this study is to assess the effectiveness of the Miami-Dade tree planting initiatives in increasing urban canopy by conducting an analysis of the changes in greenness across Miami-Dade County. Secondarily, the paper seeks to explore how environmental factors such as rainfall and temperature may have contributed to these changes.”

“This study aims to achieve its objectives through the examination of the temporal evolution of greenness in Miami-Dade County as measured by Landsat data.”

Methods: Please explain how Census blocks were determined. Were they equal-sized fields or of different sizes determined by property boundaries/road networks?

We now clarify under methods that the study uses the determination of Census blocks made by the U.S. Census and provide the following description:

“Census blocks, delineated by the U.S. Census Bureau every ten years, are statistical areas defined by visible and nonvisible boundaries, including roads, streams, and administrative divisions. They serve as the foundational building blocks for all geographic boundaries tabulated by the Census Bureau, ranging from city-like blocks in urban areas to large and irregular blocks in suburban and rural regions, providing a comprehensive, wall-to-wall coverage across the United States and territories for collecting basic demographic data.”

 

Apart from that the reasoning is logical, and the methods are appropriate, the authors correctly interpreted the research results. After introducing the necessary explanations, I recommend the article for publication.

 

We thank the reviewer for their recommendation.

Reviewer 3 Report

Comments and Suggestions for Authors

In the context of the importance of urban greening, this paper points out that trees and vegetation play a very important role in urban development, which is closely related to human physical and mental health while reducing the urban heat island effect. In this paper, the change in greenness between 2006 and 2019 in Miami-Dade County was analyzed experimentally, and the seasonal differences in greenness change were determined by the Mann-Kendall test. The possible effects of environmental factors such as rainfall and temperature on greenness changes were analyzed, and the impact of Miami-Dade County's tree planting program on urban greening was assessed.

 

However, the manuscript also contains several problems that require significant revision.

 

Major:

Overall, the study only used NDVI data to analyze vegetation changes in Miami-Dade County and some climate factors. The spatial resolution of Landsat is 30m. Whether it can meet the evaluation of "Tree Planting Initiatives". NDVI data represents the density information of vegetation. How to identify the vegetation types required for this study should be explained in the article. In addition, the analysis of NDVI and climate factors is in the results section, and the analysis of "Tree Planting Initiatives" is in the discussion section. The two parts are relatively separated, and the author needs to further strengthen the structure.

 

Minor:

1. Firstly, the abstract part is too wordy. The summary of suggestions should be concise and clear, highlighting the key points.

2. The structure of the introduction needs to be adjusted. The article mainly studies urban green spaces and should be placed in the first paragraph of the introduction. The significance of urban green space research and the importance of "Tree Planting Initiatives" should be in the second or third paragraph of the introduction.

3. In order to describe the content of the article more clearly, it is recommended to add a technical roadmap.

4. Landsat's revisit period is 16 days, and the seasonal data of NDVI chooses the three-month average or daily data. It is recommended to visualize four seasonal NDVI composite images.

5. The structure of the discussion section also needs to be adjusted. It is recommended to add subtitles. The discussion section also needs to highlight in-depth analysis before and after greening.

6. In analyzing the relationship between climate and greenness change, there is a lack of in-depth analysis of the spatial scale mismatch.

7. In the analysis section, although the study mentions possible factors, it is not possible to identify the specific drivers of greenness change.

8. The results section did not show the reasons for the statistically significant changes and the likelihood of their existence.

9. Analysis of specific drivers of greenness change lacks tree planting and other human factors.

10. In terms of research methods, the generalized linear model and the Mann-Kendall test were selected, and the suitability of these two methods for this experiment was not expressed in detail

11.  The difference in the acquisition time of Google Earth and Landsat images in the study data makes it impossible to fully align the start and end dates of the two, which can lead to inaccurate results. How can this error be reduced?

Author Response

In the context of the importance of urban greening, this paper points out that trees and vegetation play a very important role in urban development, which is closely related to human physical and mental health while reducing the urban heat island effect. In this paper, the change in greenness between 2006 and 2019 in Miami-Dade County was analyzed experimentally, and the seasonal differences in greenness change were determined by the Mann-Kendall test. The possible effects of environmental factors such as rainfall and temperature on greenness changes were analyzed, and the impact of Miami-Dade County's tree planting program on urban greening was assessed.

However, the manuscript also contains several problems that require significant revision.

Major:

Overall, the study only used NDVI data to analyze vegetation changes in Miami-Dade County and some climate factors. The spatial resolution of Landsat is 30m. Whether it can meet the evaluation of "Tree Planting Initiatives". NDVI data represents the density information of vegetation. How to identify the vegetation types required for this study should be explained in the article. In addition, the analysis of NDVI and climate factors is in the results section, and the analysis of "Tree Planting Initiatives" is in the discussion section. The two parts are relatively separated, and the author needs to further strengthen the structure.

As recommended by the reviewer, we have placed the analysis of tree planting impact (i.e., NDVI changes) under Results and its discussion under Discussion. We do not present any other analyses of tree planting because data on actual trees planted were not available from Miami-Dade County.

Why use NDVI in this study?  As noted in the introduction, there is a growing literature on the impact of greenness on health. Except for very few studies that include an experimental manipulation of planting trees, other studies typically use NDVI to associate to health outcomes.  Our intention was to determine if the tree planting initiative in Miami-Dade County was increasing greenness as measured by NDVI which our team has used in our many studies of greenness and health [1-3].  If tree planting was effective in increasing greenness, it would have a beneficial impact on the health of Miami-Dade residents.  Yet, we identified in our own studies of NDVI and health, that the number of blocks that were increasing from low tertile of greenness to the high tertile of greenness is very small. Therefore, we were interested in determining in this study/manuscript if in fact tree planting was resulting in the expected increased in greenness.

In response to your comment regarding the diversity of vegetation in Miami-Dade County, we acknowledge the importance of considering vegetation diversity in environmental research. For this reason, it was beyond the scope of this work to include an analysis of vegetation type as it did not directly address our aims.  We now note this under limitations.  From a theoretical perspective, the mechanisms through which health is impacted are likely to differ for NDVI/vegetation density vs. diversity of plant types, and hence, our emphasis on NDVI which has been used in the majority of studies investigating the relationship of greenness and health.

Minor:

1. Firstly, the abstract part is too wordy. The summary of suggestions should be concise and clear, highlighting the key points.

Following the reviewer’s suggestions, the abstract has been streamlined:

“In urban settings, trees and greenery play a vital role in environmental well-being and community vitality. This study explores the impact of Miami-Dade County's tree planting initiative on urban greenness and considers the influence of climate dynamics. Using Landsat data from 2006 to 2019, we find stable overall greenness, with 5.64% of the Census blocks exhibiting significant changes. Seasonal analysis reveals winter as prominent, with 61.47% of Census blocks showing increased greenness. Temperature and precipitation, especially post-2010, correlate with greenness changes. Despite a reported increase in tree cover from 14% to 20%, our findings show only 5-6% of Census blocks with statistically significant changes, highlighting the complexity of achieving substantial improvements in green canopy coverage. The study raises questions about the efficacy of large-scale tree planting initiatives in densely urbanized areas when human factors are not well understood. Implications for urban planning stress the importance of preserving green spaces and informed decision-making for enhancing vegetation cover in Miami-Dade County, emphasizing the need to consider local conditions, seasonal variations, policies and human factors in urban greening efforts.”

2. The structure of the introduction needs to be adjusted. The article mainly studies urban green spaces and should be placed in the first paragraph of the introduction. The significance of urban green space research and the importance of "Tree Planting Initiatives" should be in the second or third paragraph of the introduction.

We have edited portions of the first paragraph of the introduction in accordance with the suggestions by reviewer #2 that we believe addresses this concern. We also have included this sentence at the end of the second paragraph to highlight the importance of tree planting initiatives:

“In light of these compelling findings, fostering higher levels of greenness through tree planting initiatives emerges as a crucial strategy not only for environmental conservation but also for promoting public health and well-being.”

3. In order to describe the content of the article more clearly, it is recommended to add a technical roadmap.

A flow diagram of the methods concerning the Landsat imagery was created and was included as Supplementary Figure S2.

4. Landsat's revisit period is 16 days, and the seasonal data of NDVI chooses the three-month average or daily data. It is recommended to visualize four seasonal NDVI composite images.

The requested graphic was created as a supplementary figure. See Supplementary Figure S1.

5. The structure of the discussion section also needs to be adjusted. It is recommended to add subtitles. The discussion section also needs to highlight in-depth analysis before and after greening.

Sub headers in the Discussion section have been created:

4.1 Overview of Findings

4.2 Prior Greenness Studies in Miami-Dade County

4.3 Extent of Changes in Greenness

4.4 Million Trees Miami Initiative Evaluation

4.5 Limitations

4.6 Conclusion and Future Research

 

The initial paragraph of the Discussion section was also reworked to highlight the in-depth analysis of greenness changes in Miami- Dade County thereby providing a brief summary of the results.

“The objective of this study was to investigate changes in greenness (i.e., vegetation amount/density) across Miami-Dade County over a 13-year period (2006-2019) and to evaluate how these changes align with the objectives of the Million Trees Miami project (2011-ongoing). Only 6.76% of the studied Census blocks displayed statistically significant changes, predominantly reflecting positive trends. These results suggest a lack of impact by the tree planting initiative on the overall greenness of Miami-Dade County. Additionally, the study aimed to ascertain whether alterations in spatial and temporal greenery patterns were associated with shifts in climate patterns. Winter, summer, and fall showed statistically significant increases in greenness, with winter having the highest proportion of Census blocks with significant changes (61.47%).”

6. In analyzing the relationship between climate and greenness change, there is a lack of in-depth analysis of the spatial scale mismatch.

We appreciate the reviewer's careful consideration of our work and their insightful comments. The reviewer is correct about the disparity of scale between greenness and climate. We now address this under limitations:

“Our use of the Census block as the unit of analysis facilitated assessment of greenness variability at a very high resolution. In contrast, our climate data is derived from climate stations and therefore lacks a comparable spatial scale. While we can discuss broad-scale climate changes for the entire region (Figures 2, 3, 10, & 11), establishing statistical linkages proves challenging due to the  different spatial scales.”

7. In the analysis section, although the study mentions possible factors, it is not possible to identify the specific drivers of greenness change.

We appreciate your feedback regarding the identification of specific drivers of greenness change. We would like to clarify that our study did not explicitly aim to identify specific drivers; rather, our focus was on examining overall patterns and trends. It is important to also point out that we did review how minimum temperature and precipitation might relate to changes in greenness in our analysis. However, as discussed in the manuscript, we cannot directly say that these climatic variables were drivers of the observed changes in greenness.  

8. The results section did not show the reasons for the statistically significant changes and the likelihood of their existence.

We now address your comment in the Discussion section as follows:

“Many factors, including species selection, local conditions, and maintenance practices, in-fluence the growth and establishment of newly planted trees. These saplings need time to establish their root systems, develop foliage, and contribute significantly to the overall greenness of their location. As such, while the ‘Million Trees Miami’ project may hold the potential to significantly impact the urban canopy in the long term, the observed changes in greenness, found predominantly within the winter season only, across the study period are much more likely to be influenced by a combination of factors linked to changing dynamics of this season such as lower overnight minimum temperatures and changes in distribution of precipitation, and canopy preservation. If the changes in greenness were a result of increased tree canopies and tree counts, we would expect to see a much more consistent increase in greenness levels year-round, not just in winter.”

In this new paragraph in the Discussion, we highlight that the observed changes in greenness during the winter season are likely influenced by factors such as lower overnight minimum temperatures and changes in precipitation distribution.

9. Analysis of specific drivers of greenness change lacks tree planting and other human factors.

An in-depth analysis looking into how specific drivers of greenness change influenced the observed changes is beyond the scope of our paper. However, we now address the reviewer’s comment under in the Discussion:

“In addition, there are a number of factors that may have counteracted the possible effects of tree planting initiatives including significant urban development, existing tree removals, mature tree pruning, and other human-induced reductions in canopy.”

10. In terms of research methods, the generalized linear model and the Mann-Kendall test were selected, and the suitability of these two methods for this experiment was not expressed in detail

To address the reviewer’s concern, we have revised the Methods section to include a more detailed explanation of the suitability of the generalized linear model (GLM) and the Mann-Kendall test as follows:

“To identify significant changes in greenness throughout the study period, we analyzed the NDVI data over time using utilized Mann-Kendall tests on the Landsat data. This non-parametric test was used to detect monotonic trends in greenness over time.”

“These linear regression analyses are particularly suitable for our aim investigation as it helps to determine if linear relationships exists over time within a dataset. The dependent variable in the GLM were the climatic variables, and the independent variables was time (in years).”

11. The difference in the acquisition time of Google Earth and Landsat images in the study data makes it impossible to fully align the start and end dates of the two, which can lead to inaccurate results. How can this error be reduced?

The acquisition time disparity between Google Earth and Landsat images in our study data posed a challenge in aligning their start and end dates accurately, potentially introducing inaccuracies. To address this concern, we now use imagery that more closely aligns with the start and end dates of Landsat images for the Google Earth comparisons. Consequently, we have adjusted the relevant figures; accordingly, for instance, the end date of Figure 8 has been revised to 2019 instead of 2022. However, due to the unavailability of imagery from 2019 for the location depicted in Figure 9, we substituted it with an image from February 2020.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I think that this manuscript was revised well.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have addressed my concerns well. Now I recommend it could be accepted in Remote Sensing.