Spatial–Temporal Changes and Driving Factor Analysis of Net Ecosystem Productivity in Heilongjiang Province from 2010 to 2020

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper takes Heilongjiang Province as a case study, builds an improved CASA model and soil respiration model to estimate vegetation NEP from 2010 to 2020, and analyzes the spatiotemporal variations of NEP and its driving 16 factors.

 

1.       The abstract is too long. The result could highlight the key findings. The policy suggestion can be added in the abstract.

2.       As for the key words, the authors can consider to delete “spatiotemporal variation”, and added some other words, except the title.

3.       The introduction needs more detailed description about the novelty and significance of the study. Why did the authors select Heilongjiang as the study area.

4.       The conclusions can be further summarized according to the results. It’s better to propose some policy recommendations.

5.       As for the method, which land use types were considered in the assessment of NEP? What is the relationship between land cover types and NEP results?

6.       Figure 1 shows the land cover types. Which year is the land cover types?

7.       Figure 4, the legend shows “Forest/50、Cropland/50”. What is their meaning?

8.       Figure 9, the meaning of X1-X8 should be explained.

Comments on the Quality of English Language

please check the whole muanscript and polish the language.

Author Response

Comments1: The abstract is too long. The result could highlight the key findings. The policy suggestion can be added in the abstract.

Response1: Thank you for pointing this out. We agree with this comment. We have reorganized the abstract, removing some redundant information to ensure that it is more concise and still covers the main purpose, methods, results, and conclusions of the study. In addition, I will use concise language to directly point out the main results of the research, avoiding vague or general expressions; In addition, considering the practical value of this study, I have added a brief policy recommendation at the end of the abstract.

 

Comments2: As for the keywords, the authors can consider to delete "spatiotemporal variation”, and added some other words, except the title.

Response2: Thank you for pointing this out. We agree with this comment. I have carefully considered your suggestion and have decided to make corresponding adjustments to the keywords. I will remove "spatiotemporal variation" from the keyword list and add the keyword "CASA model" to more comprehensively cover the theme and focus of the paper.

 

Comments3: The introduction needs more detailed description about the novelty and significance of the study. Why did the authors select Heilongjiang as the study area.

Response3: Thank you for pointing this out. We agree with this comment.In response to your suggestion, first of all, I will conduct a more in-depth analysis of the current research status in the field, especially the existing achievements and shortcomings directly related to this study. By comparing the limitations of existing research with the innovative points of this study, clarify the starting point and necessity of this research. Secondly, I further emphasize the importance and significance of this study. As for why Heilongjiang was chosen as the research area, it is mainly based on the following considerations: ecological importance: Heilongjiang is an important component of the "Northeast Forest Belt" in China's "Two Screens and Three Belts" ecological security strategy layout, with abundant forest resources and diverse ecosystem types. The carbon cycle status in this region is of great significance for maintaining national ecological security and promoting regional sustainable development. Regional representativeness: Heilongjiang is located in a high latitude region, and its climate conditions and ecological environment have unique representativeness. Studying the characteristics and driving mechanisms of NEP changes in this region can help reveal the universal patterns and unique phenomena of carbon cycling in high latitude ecosystems. Data accessibility: With the development of remote sensing technology and the improvement of meteorological observation networks, the Heilongjiang region has accumulated a large amount of high-quality remote sensing images, meteorological observations, and socio-economic data. These data provide a solid foundation for conducting in-depth NEP research.

 

Comments4: The conclusions can be further summarized according to the results. It’sbetter to propose some policy recommendations.

Response4: Based on your suggestion, first of all, I will conduct a more comprehensive review and analysis of the research results, and summarize the main conclusions in more concise and accurate language. Then, based on the conclusion, combined with the actual situation and existing research results, targeted policy recommendations are proposed.

 

Comments5: As for the method, which land use types were considered in the assessment of NEP? What is the relationship between land cover types and NEP results?

Response5: Thank you for your review and suggestions. When evaluating NEP, we considered six types of land use: forest land, grassland, construction land, farmland, wetlands, and water bodies. There is a significant correlation between land cover types and NEP results. Specifically, different land cover types exhibit different characteristics in carbon absorption and release due to differences in vegetation types, biomass, soil conditions, and other factors. For example, forest covered areas have high NEP values due to lush vegetation and strong photosynthesis, which can absorb a large amount of carbon dioxide; Although grasslands and farmland also have a certain carbon absorption capacity, they are greatly affected by human activities. Improper management may lead to an increase in carbon emissions and a decrease in NEP values; Wetlands play an important role in carbon storage and conversion due to their unique ecological functions, and their NEP values are influenced by various factors such as hydrological conditions and vegetation types. In addition, our research also found that changes in land cover types have a significant impact on NEP results. Through ecological restoration and afforestation measures, forest area can be increased, carbon sequestration capacity can be enhanced, and NEP value can be improved.

 

Comments6: Figure 1 shows the land cover types. Which year is the land cover types?

Response6: In 2015. In revised draft 2.2 The Data Source and Pretreatment section has been supplemented with specific years.

 

Comments7: Figure 4, the legend shows “Forest/50 、Cropland/50” . What is their meaning?

Response7: Thank you for pointing out this out. Due to the significant difference in values between forests and cultivated land compared to grasslands and wetlands, only by dividing the values of forests and cultivated land by a certain value can the above four land cover types be displayed and compared on the same graph.

 

Comments8:Figure 9, the meaning of X1-X8 should be explained.

Response8:The specific meaning of X1-X8 is explained in detail in Article 2.3.4. Geodetectors.

 

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the invitation. This paper investigates the NEP in Heilongjiang Province, Northeast China. The topic is of practical significance and represents an interesting study. I have the following questions and suggestions regarding the content and results of this paper:

1.In line 167, the calculation of heterotrophic respiration is based on the empirical formula of Pei, Z.Y. et al. However, Pei, Z.Y. et al.'s experiments were conducted on the Qinghai-Tibet Plateau. Considering the differences in climate, vegetation types, and topography, applying this empirical formula to Heilongjiang Province requires further validation and discussion.

2.In lines 140-141, the spatial resolution of ERA5-Land monthly data is 0.1 degrees, and the paper resamples it to 500 meters. How is data accuracy addressed? Additionally, how is the issue of edge jaggedness resolved after resampling the 0.1-degree image data?

3.Have the NEP results obtained in this paper been validated or compared with the results of other studies in terms of spatial and temporal scales? How can the reliability of the results be ensured?

4.The CASA model improved by Zhu Wenquan et al. uses the vegetation index, but why is the vegetation index not included in the analysis when discussing the driving factors?

5.There is already a wealth of research on the calculation of NEP at the provincial scale, including studies on NEP in Northeast China. What are the innovative aspects of this paper compared to previous research?

6.In Figure 4, what is the reason for the large interannual fluctuation of grassland NEP?

7.The quality of the statistical figures needs to be improved.

Comments on the Quality of English Language

1.Attention to the Use of Technical Terms and Vocabulary: Review and ensure the accurate and consistent use of technical terms and specialized vocabulary throughout the manuscript.

2.Attention to the Use of Long Sentences: Break down lengthy sentences into shorter, more concise ones to enhance readability and clarity.

3.Improvement in Grammar: Correct grammatical errors, focusing on subject-verb agreement, tense usage, and punctuation to improve the overall quality of the writing.

 

Author Response

Comments1:In line 167, the calculation of heterotrophic respiration is based on the empirical formula of Pei, Z.Y. et al. However, Pei, Z.Y. et al.'s experiments were conducted on the Qinghai-  Tibet Plateau. Considering the differences in climate, vegetation types, and topography,applying this empirical formula to Heilongjiang Province requires further validation and discussion.

Response1:Thank you for pointing this out.Provincial Rationality of Soil Heterotrophic Respiration (Rh), in this paper in 3.5. Analysis of NEP driving factors Through comparison with other research results, it is found that the Rh estimation in this paper is basically consistent with the estimation results of soil microbial respiration in Gannan Prefecture and the Yellow River Basin using Pei Yongzhi's formula by Chen Di and Faye Wong, indicating that the estimation of soil microbial respiration in Heilongjiang Province using Pei Yongzhi's formula is reliable.

 

Comments2:In lines 140-141, the spatial resolution of ERA5-Land monthly data is 0.1 degrees, and the paper resamples it to 500 meters. How is data accuracy addressed? Additionally, how is the issue of edge jaggedness resolved after resampling the 0.1-degree image data?

Response2:Thank you for pointing this out.In response to your question about the spatial resolution and data accuracy of ERA5-Land monthly data, firstly, we use a bilinear interpolation method in the resampling process, which performs well in balancing computational efficiency and data accuracy. At the same time, we evaluated the accuracy of the resampled data by comparing it with other high-resolution datasets and ground observation data. In addition, for edge aliasing after resampling, we noticed the edge aliasing problem that may occur after resampling, and took various measures to reduce its impact. First, we chose an interpolation algorithm that suited the needs of this study. Secondly, a Gaussian filter was applied for post-processing after resampling to further smooth the edges. In addition, we have adjusted the interpolation parameters to optimize the processing effect. With these measures, we have succeeded in reducing edge aliasing and improving the overall quality of the data.

 

Comments3: Have the NEP results obtained in this paper been validated or compared with the results of other studies in terms of spatial and temporal scales? How can the reliability of the results be ensured?

Response3: Thank you for pointing this out.To ensure the reliability and accuracy of NEP results, we employ a variety of verification methods. First, we use the improved CASA model to estimate NEP, which has been widely used in similar studies and compared with other model estimation results, and the results show that our estimation results have high reliability. Second, we found that our NPP estimates had a high correlation coefficient with the MOD17A3HGF NPP product dataset by comparing them with the MOD17A3HGF NPP product dataset, which further validated our estimates. In addition, we compared estimates of soil microbial respiration with other studies and found that our estimates were in good agreement with those of other studies. These validation methods ensure the high reliability and accuracy of our NEP results.

 

Comments4: The CASA model improved by Zhu Wenquan et al. uses the vegetation index, but why is the vegetation index not included in the analysis when discussing the driving factors?

Response4: Thank you for pointing this out.We acknowledge that the vegetation index plays a crucial role in ecological research, particularly in estimating plant net primary productivity (NPP) using the CASA model. The vegetation index can reflect the information of surface vegetation coverage, growth status and seasonal changes, and is a bridge connecting vegetation growth with environmental factors such as climate and soil. The improved CASA model by Zhu et al. effectively improved the precision and accuracy of NPP estimation by introducing vegetation index as one of the key input parameters. However, when discussing the drivers of NEP, we do not directly analyze the vegetation index, which is mainly based on the following considerations: First, given the complexity of the definition of NEP, NEP is not only affected by plant production activities, but also regulated by soil respiration processes. Therefore, when analyzing the drivers of NEP, we need to comprehensively consider a variety of factors that affect NPP and Rh, including but not limited to climate, soil, vegetation type, human activities, etc. Although the vegetation index is closely related to NPP, it does not directly reflect the change of Rh, so the analysis of vegetation index alone may be limited to fully understand the driving mechanism of NEP. Second, when discussing the drivers of NEP, our research focus may shift more to those factors that can affect both NPP and Rh, such as climate change (temperature, precipitation, etc.), topographic conditions (altitude, altitude, etc.). The changes of these factors can directly affect the growth of vegetation and the process of soil respiration, which can have a significant impact on NEP. In addition, although we did not directly analyze the driving effect of vegetation index on NEP, the changes in vegetation index have actually been indirectly reflected in the changes in NEP by affecting NPP. In the CASA model, vegetation index is one of the important parameters for estimating NPP, and its change will directly lead to the change of NPP estimation results, which in turn will affect the estimation results of NEP. Therefore, although we do not directly include the vegetation index as the object of analysis when discussing the drivers of NEP, this does not mean that we ignore the importance of vegetation index in the estimation of NEP. On the contrary, we fully consider the impact of vegetation index on NPP by using the improved CASA model of Zhu Wenquan et al., and further explore other key factors affecting NEP on this basis.

 

Comments5: There is already awealth of research on the calculation of NEP at the provincial scale,including studies on NEP in Northeast China. What are the innovative aspects of this paper compared to previous research?

Response5: Thank you for pointing this out.Thank you very much for your in-depth review of this paper and your valuable comments. In response to your suggestion about the innovation of estimating NEP in Heilongjiang Province at the provincial scale compared with previous studies, we believe that it mainly includes the following aspects: first, model optimization and comprehensive application. In this paper, the improved CASA model of Zhu et al. was used to estimate the net primary productivity (NPP) of plants, and the soil respiration model was combined to achieve an accurate estimation of NEP. The comprehensive application of this model can more comprehensively reflect the carbon cycle process of the ecosystem than the single model, and improve the accuracy and reliability of NEP estimation. Second, high-resolution data analysis. In this study, high-resolution data sources such as MODIS and meteorological data were utilized. This high-resolution data analysis enables us to capture the spatiotemporal variation characteristics of NEP in more detail, which provides more abundant information for understanding the dynamics of the regional carbon cycle. Thirdly, the comprehensive application of multivariate statistical analysis methods: in addition to the traditional trend analysis methods (such as Theil-Sen and Mann-Kendall), multivariate statistical analysis methods such as Hurst index and geographic detector are also introduced. The application of these methods not only reveals the temporal and spatial trends of NEP, but also deeply explores the driving mechanism behind it, providing a scientific basis for formulating regional carbon management policies. Finally, an in-depth study of regional specificity. Heilongjiang Province is an important ecological region in China, and the change of NEP has an important impact on the climate and ecological environment of the region and even the whole country. This paper focuses on this specific region of Heilongjiang Province, and provides a new perspective and reference for the study of regional carbon cycle by deeply analyzing the temporal and spatial variation characteristics and driving factors of NEP.

 

Comments6: In Figure 4, what is the reason for the large interannual fluctuation of grassland NEP? 

Response6: Thank you for pointing this out.In Figure 4, the NEP of grassland shows large interannual fluctuations, mainly due to the complexity of grassland ecosystems and their influence by multiple environmental factors. Grassland ecosystems are very sensitive to climate change, including meteorological factors such as temperature, precipitation, and light, as well as human activities such as land use change and grazing. The interannual variation of these factors may lead to changes in the productivity, soil condition, and vegetation cover of grassland ecosystems, which in turn may affect the NEP of grassland. In addition, the process of restoration and reconstruction of grassland ecosystems may also lead to fluctuations in NEP. In order to explain this fluctuation more accurately, we will further explore the dynamics of grassland ecosystems and their influencing factors in future studies.

 

Comments7: The quality of the statistical figures needs to be improved.

Response7: Thanks to the judges for their suggestions. To improve the quality of my paper, I make improvements to the statistical charts to ensure that they convey the key information of the research clearly and accurately. I check the accuracy of the charts to make sure the data is error-free, and consider using different colors, fonts, and labels to improve the legibility of the charts.

Reviewer 3 Report

Comments and Suggestions for Authors

The paper presents an application of remote sensing for detecting key factors for NEP in Heilongjiang Province, China. However, details of the research problem, methodology, and discussion need to be better explained for the paper to be considered for publication. Below, there are some specific comments that should be considered in a resubmission.

1. Avoid acronyms in the title, for example NEP.

2. Generally, the abstract needs to be restructured, as the way it is currently written, it lacks connections between sentences, making it difficult to understand how each result was obtained and for what purpose. Additionally: i) The first time an acronym appears, it should be spelled out in full in the text, e.g., line 13; ii) The objective of the paper should be made clear; iii) the methodology should be better detailed to clarify which data were used and for what purpose; iv) On line 25, and beyond, future scenarios are mentioned. If future modelling was conducted, it should be made clear in the methodology, even in the abstract.

3. The knowledge gaps are clear in the introduction, but the authors focus only on how they will apply a method, without clearly expressing the objective, or the research question, of the paper to address the delineated knowledge gap. This needs to be clearer in the introduction, for example, what is the research question? What are the objectives?

4. To better understand the database, it would be appropriate to add this information in a table, indicating, at least, the type of data, spatial and temporal resolution, and source (reference).

5. The pre-processing mentioned on line 124 should be detailed. All data that underwent pre-processing should be indicated.

6. Several acronyms in the text are not accompanied by their meanings, ie., IGBP and GDP. This should be corrected in a new submission.

7. It is not clear in the methodology how each piece of data was used in the presented equations. For instance, the authors present an NPP equation but indicate that they used a MODIS product that already provides this data. So, what exact calculation was carried out here? How was APAR calculated? For what purpose was the NDVI used? The same applies to the other equations; the authors should always indicate how the collected data were used to generate their results.

8. Another point that needs to be clear in the methodology is the choice of the period for temporal analysis. Why were older and more recent MODIS data not used? Is there any event in the Province during this period that justifies limiting the analysis?

9. Additionally, it is suggested that all algorithms used to be offered as supplementary material or on a developer platform.

10. The methodology text does not make it clear how the model was validated for the study area. This is only addressed in the results. Without in-situ data, it is difficult to verify the actual validity of the work, making this a significant limitation. The authors should clearly address this limitation in both the methodology and the discussion of the article. Additionally, it should be clear in the methodology which data were used for validation. Although there is a "Limitations and Future Work" section, the article's writing is limited to addressing general knowledge about the limitations of remote sensing or field research, without addressing the article's own limitations. Furthermore, the authors do not explain why they did not combine remote sensing data with field data.

11. Check the caption of Figure 7b.

12. I believe the term “technique” on line 414 is not well implemented, as CCFP and CCUS appear to be government programmes and not techniques.

13. The map in Figure 1 should be modified, including with better resolution so that the regions indicated in the discussion can be easily identified.

14. The conclusions are limited to summarising the findings without addressing possible research questions and how the article can contribute to future research in this study area.

Author Response

Comments 1: Avoid acronyms in the title, for example NEP.

Response 1: Thank you for pointing this out.We agree with this comment.We have carefully considered your suggestion on using acronyms in the title and made corresponding modifications. In the revised title, We have removed acronyms and replaced them with complete vocabulary to ensure that the title is clearer and easier to understand, while also helping to improve the readability and professionalism of the paper.

 

Comments 2:Generally, the abstract needs to be restructured, as the way it is currently written, it lacks connections between  sentences,  making  it  difficult  to  understand  how  each  result  was obtained and for what purpose. Additionally: i) The first time an acronym appears, it should be spelled out in full in the text, e.g., line 13; ii) The objective of the paper should be made clear; iii) the methodology should be better detailed to clarify which data were used and for what purpose; iv) Online 25, and beyond, future scenarios are mentioned. If future modelling was conducted, it should be made clear in the methodology, even in the abstract.

Response 2: Thank you for pointing this out.We agree with these comments.Regarding the restructuring of the abstract, I have reorganized and rewritten the abstract, paying special attention to the coherence and logic between the sentences. I try to make the association between each result clearer by adding connectives and phrases so that the reader can better understand how each study result was obtained, and what they were studied for. Regarding the treatment of the first use of the abbreviation, in the abstract I have spelled the abbreviation "NEP" for the first time, i.e., "net ecosystem productivity", and used the abbreviated form in subsequent texts. This ensures that the reader can understand exactly what it means. As for the clarification of the objectives of the paper, at the beginning of the abstract, I have further clarified the research objectives of the paper, that is, to estimate the net ecosystem productivity (NEP) of vegetation in Heilongjiang Province from 2010 to 2020 by using the improved CASA model and soil respiration model, combined with MODIS and meteorological data, and to analyze the characteristics and driving factors of its temporal and spatial variation. This can more clearly convey the research purpose and value of the paper.With regard to the elaboration of the methodological part, in the abstract, I have supplemented and refined the methodological part, clearly stating the data used (MODIS, meteorological data) and models (improved CASA model and soil respiration model), as well as the specific role and purpose of these data and research methods in the study. At the same time, I also mentioned the analysis of spatiotemporal variation characteristics and their driving factors using Theil-Sen, Mann-Kendall, Hurst exponential and geographic detectors. Regarding the description of the future scenario, in the abstract I have made a clear statement of what the future modeling is about, and indicated its specific description in the methodology section. At the same time, I also mentioned the overall trend and regional distribution of future changes in the abstract, so that readers can better understand the background and purpose of this part of the study. In fact, we did not do specific future modeling, but instead predicted and analyzed future trends based on existing data. This is also explained in detail in the methodology section.

 

Comments 3: The knowledge gaps are clear in the introduction, but the authors focus only on how they will apply a method, without clearly expressing the objective, or the research question, of the paper to address the delineated knowledge gap. This needs to be clearer in the introduction, for example, what is the research question? What are the objectives?

Response 3: Thank you for pointing this out.You point out that the introduction section does not clearly articulate the research objectives or research questions of the paper to fill in the gaps in knowledge. I fully understand your point of view and have made the following changes in the introduction: After introducing the knowledge gap, I clearly put forward the research question of this paper: "However, the in-depth research on the spatial and temporal distribution characteristics of vegetation NEP and its driving mechanism in Heilongjiang Province is still insufficient. Therefore, this paper aims to explore the following questions: What are the temporal and spatial variation characteristics of vegetation NEP in Heilongjiang Province from 2010 to 2020? What are the driving factors?

Immediately following the research question, I clearly stated the research objectives of this paper: "The goal of this paper is to estimate the NEP of vegetation in Heilongjiang Province based on multi-source data such as MODIS, meteorology, topography, and socio-economic data, using the improved CASA model and soil respiration model, and to deeply analyze its temporal and spatial variation characteristics and driving factors, in order to provide a scientific basis for ecosystem carbon management in Heilongjiang Province and beyond." "Through such revisions, I hope that the introduction can more clearly convey the research questions and research objectives of this paper, so that readers can better understand the research content and research significance of this paper. Thank you again for your review and valuable comments.

 

Comments 4: To better understand the database, it would be appropriate to add this information in a table, indicating, at least, the type of data, spatial and temporal resolution, and source (reference).

Response 4: Thank you for pointing this out.About adding database information to a table,

I think the type of data used in the article, the spatial and temporal resolution, and the data source are as detailed in 2.2. The Data Source and Pretreatment section has been described in detail to help readers better understand and grasp the situation of the database, and if this part is added to the table, it will make the content of this section duplicative and redundant.

 

Comments 5: The pre-processing mentioned online 124 should be detailed. All data that underwent pre- processing should be indicated.

Response 5: Thank you for pointing this out.Regarding the elaboration of the preprocessing part, I explained it in 2.2. In the first two paragraphs, Data Source and Pretreatment has clearly listed all the preprocessed data and described the specific steps and methods of pretreatment in detail to ensure that readers can fully understand the entire process of data pretreatment.

 

Comments 6: Several acronyms in the text are not accompanied by their meanings, ie., IGBP and GDP. This should be corrected in a new submission.

Response 6: Thank you for pointing this out.In the new submission, I give the full names and explanations of these abbreviations when they first appear in the article to ensure that readers can understand the content of the article accurately.

 

Comments 7: It is not clear in the methodology how each piece of data was used in the presented equations. For instance, the authors present an NPP equation but indicate that they used a MODIS product that already provides this data. So, what exact calculation was carried out here? How was APAR calculated? For what purpose was the NDVI used? The same applies to the other equations; the authors should always indicate how the collected data were used to generate their results.

Response 7: Thank you for pointing this out.Based on the comments of the journal editors, we have removed the formulas of the article, leaving a descriptive and simple part, and the specific calculation process is presented in the form of references. But the article details all the data needed and the role it plays in the research.

 

Comments 8: Another point that needs to be clear in the methodology is the choice of the period for temporal analysis. Why were older and more recent MODIS data not used? Is there any event in the Province during this period that justifies limiting the analysis?

Response 8: Thank you for pointing this out.Regarding the choice of time analysis period in the methodology, we do need to provide a more detailed explanation. When choosing the period from 2010 to 2020 as the time analysis period, we mainly considered the following points: first, the MODIS data in this time period is of high quality and good continuity, which can ensure the accuracy and reliability of our analysis; Secondly, the goal of our study was to evaluate the change characteristics of vegetation net ecosystem productivity (NEP) in Heilongjiang Province in the past decade, so the selection of this time period can better reflect the recent change trend. Finally, we did not find any significant events in this time period that significantly interfered with the NEP assessment, so we considered this time period appropriate. As to why older or more recent data were not used, this was mainly because we wanted to ensure uniformity and consistency of the data in order to more accurately reveal the characteristics of changes in the study area. At the same time, we also take into account factors such as the availability of data and the cost of processing. In summary, we chose 2010-2020 as the time analysis period based on a combination of data quality, research objectives, and practical feasibility.

 

Comments 9: Additionally, it is suggested that all algorithms used to be offered as supplementary material or on a developer platform.

Response 9: Thank you for pointing this out.We fully agree to make all algorithms used available as supplementary materials or as resources on the Developer Platform. In order to enhance the reproducibility and transparency of the paper, we plan to compile the detailed implementation, parameter settings, and necessary code snippets of all algorithms into supplementary materials and submit them with the paper. In addition, we are also considering uploading these algorithms to developer platforms, such as GitHub, so that readers can more easily access, use, and further develop.

 

Comments 10: The methodology text does not make it clear how the model was validated for the study area. This is only addressed in the results. Without in-situ data, it is difficult to verify the actual validity of the work, making this a significant limitation. The authors should clearly address this limitation in both the methodology and the discussion of the article. Additionally, it should be clear in the methodology which data were used for validation. Although there is a "Limitations and Future Work" section, the article's writing is limited to addressing general knowledge about the limitations of remote sensing or field research, without addressing the article's own limitations. Furthermore, the authors do not explain why they did not combine remote sensing data with field data.

Response 10: Thank you for pointing this out.As for model validation, the area involved in this study is large, and it is difficult to obtain the measured data of NPP at the same scale, so the two parameters of NPP and Rh are verified by comparing with other remote sensing data products and other model estimation results, respectively, so as to judge the reliability and scientificity of the NEP estimation results. However, due to the lack of actual ground observational data (in-situ data) in the study area, we really cannot directly verify the accuracy of the model, which is an important limiting factor. We made this clear in the discussion section and explained why we chose this approach for validation, noting that this is a limitation of the study. In terms of data usage, the data we used to validate the model mainly include MODIS remote sensing data and meteorological data, which have been quality controlled and screened during processing and analysis. We will clearly list these data in the methodology section and explain how they were used for model validation by citing references. As to why we did not combine remote sensing data and ground data, this is mainly because ground observation data is relatively scarce and unevenly distributed in the study area, which makes it difficult to combine the two types of data. We'll explain this further in the discussion section and explore how future research can be combined with a wider variety of data. Finally, we discuss the limitations of this study in more detail in the "Limitations and Future Work" section, including the limitations of model validation, and how the study can be improved and expanded in the future.

 

Comments 11: Check the caption of Figure 7b.

Response 11: Thank you for pointing this out.In the new document, the unreasonable expression of the title of Figure 7b has been modified.

 

Comments 12: I believe the term “technique” on line 414 is not well implemented, as CCFP and CCUS appear to be government programmes and not techniques.

Response 12: Thank you for pointing this out.I agree that the word 'technique' may not be accurate here. It is true that CCFP and CCUS are more of a government-driven project or strategy than a purely technical approach. To express this more accurately, I would change the 'technique' in the original text to 'programme', i.e., "CCFP and CCUS are the two decisive pathways to achieve carbon sequestration, and as government-driven projects, they should be systematically applied to achieve China's 2060 carbon neutrality goal." Such a statement is more in line with the actual nature of CCFP and CCUS, and it also conveys the meaning of the original text more accurately.

 

Comments 13: The map in Figure 1 should be modified, including with better resolution so that the regions indicated in the discussion can be easily identified.

Response 13: Thank you for pointing this out.To improve the reader's reading experience, I would consider increasing the resolution of the map, readjusting the map, and re-uploading it to make sure the areas indicated in the discussion are clearly recognizable.

 

Comments 14: The conclusions are limited to summarising the findings without addressing possible research questions and how the article can contribute to future research in this study area.

Response 14:Thank you for pointing this out.In response to the reviewers' comments, I will add a discussion of possible research questions to the conclusion and emphasize how this paper can provide inspiration and direction for future research.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

1.Line 353 (Table 2): If this study estimates the monthly average Rh in Heilongjiang Province using the empirical model developed by Pei, Z.Y. et al., the results should also be compared with other similar studies in the same region, rather than comparing with estimates from other regions. Comparing results from different regions in the manuscript does not validate the reliability of this study's findings.

2.To estimate Rh at a regional scale, one can use observational data and machine learning algorithms, or empirical models based on empirical fitting, such as the model proposed in "Hashimoto, Shoji, Nuno Carvalhais, Akihiko Ito, Mirco Migliavacca, Kazuya Nishina, and Markus Reichstein. 'Global spatiotemporal distribution of soil respiration modeled using a global database.' Biogeosciences 12, no. 13 (2015): 4121-4132," among others.

3.Line 310 (Figure 8): Why do the q-values of radiation in 2010 differ significantly from those in other years?

4.All data should be accompanied by detailed sources and preprocessing procedures.

Comments on the Quality of English Language

1.Pay attention to the use of long sentences, as they may cause ambiguity; shorter sentences can be used when appropriate.

2.Ensure grammatical consistency and the uniform use of technical terms, such as "NEP mean" and "average NEP."

Author Response

Comments 1: Line 353 (Table 2): If this study estimates the monthly average Rh in Heilongjiang Province using the empirical model developed by Pei, Z.Y. et al., the results should also be compared with other similar studies in the same region, rather than comparing with estimates from other regions. Comparing results from different regions in the manuscript does not validate the reliability of this study's findings.

Response 1: Thank you for pointing this out. We agree with this comment.Based on your suggestion, we will proceed with 3.5 The third section of the Analysis of NEP Driving Factors includes the monthly average soil heterotrophic respiration in the three provinces of Northeast China, as well as the monthly average Rh in Heilongjiang Province over the years, which is consistent with the estimated results in this article. Further accurate verification has confirmed the reliability of using Pei Yongzhi's formula to estimate soil heterotrophic respiration in Heilongjiang Province.

 

Comments 2: To estimate Rh at a regional scale, one can use observational data and machine learning algorithms, or empirical models based on empirical fitting, such as the model proposed in "Hashimoto, Shoji, Nuno Carvalhais, Akihiko Ito, Mirco Migliavacca, Kazuya Nishina, and Markus Reichstein. 'Global spatiotemporal distribution of soil respiration modeled using a global database.' Biogeosciences 12, no. 13 (2015): 4121-4132," among others.

Response 2: Thank you for your review and suggestions. We are in manuscript 3.5 The first paragraph of the Analysis of NEP Driving Factors cited this literature to make the content of the article more scientific and comprehensive.

 

Comments 3: Line 310 (Figure 8): Why do the q-values of radiation in 2010 differ significantly from those in other years?

Response 3: Thank you for your review and suggestions. The radiation q value in 2010 showed significant differences compared to other years, possibly due to the higher solar radiation, suitable temperature, and precipitation distribution experienced in 2010, which were beneficial for plant photosynthesis and growth, thus directly affecting NEP. In addition, it is also possible that in 2010, the ecosystem in Heilongjiang Province was in a relatively stable or restored state, with high vegetation coverage and good soil conditions, resulting in a higher utilization efficiency of solar radiation by the ecosystem.

 

Comments 4: All data should be accompanied by detailed sources and preprocessing procedures.

Response 4: Thank you for your review and suggestions. In order to enhance the credibility of the paper, following your suggestion, we have revised section 2.2 For each set of data used in Data Source and Pretreatment, a detailed list of its sources is provided, including the name of the dataset, the publishing institution, the source (such as website links), and the time frame of data collection. In addition, we also provide a detailed description of the data preprocessing steps, including the use of software such as MRT and ArcGIS, data cleaning (such as outlier detection and processing), and data synthesis.

 

Comments on the Quality of English Language

Comments 1：Pay attention to the use of long sentences, as they may cause ambiguity; shorter sentences can be used when appropriate.

Response 1: Thank you very much for your valuable suggestion. We have carefully reviewed the sentence structure in the paper and have indeed found that some places use longer sentences, which may cause difficulties or ambiguity in readers' understanding. In order to improve the readability and clarity of the paper, we will follow your suggestion to split and reconstruct long sentences, using shorter sentences to express the same meaning. At the same time, I will also pay more attention to the logic and coherence of sentences, ensuring that each sentence can convey information clearly and accurately. The revised paper will be easier to understand and avoid possible ambiguity.

 

Comments 2：Ensure grammatical consistency and the uniform use of technical terms, such as "NEP mean" and "average NEP."

Response 2: Thank you very much for your detailed guidance. In order to enhance the professionalism and readability of the paper, we comprehensively checked the grammar of the paper to ensure consistency in sentence structure, tense, voice, etc., and to avoid grammar errors or inconsistencies. In addition, we have unified and standardized the technical terms used in the paper. Specifically, we have unified standard terms such as' NEP mean 'or' average NEP 'throughout the text.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The authors' revisions have improved the paper’s clarity regarding its research aims. However, methodological details require further elaboration. For instance, the CASA model (including its acronym's meaning) remains undefined (i.e., what is this model method? how was it implemented? what variables make up this method?), and the reasons for using each database, i.e., MOD13A1 (include the meaning of the acronym NDVI), MOD16A2, and MCD12Q1 are not clarified. Given the methodology's crucial role in scientific papers, this should be more explicit. The cover letter addressed the analysed period, but this information is absent from the main text. Additionally, the discussion of limitations needs to be sustained by relevant literature, exploring how other global remote sensing studies have handled similar constraints and why it was not employed in this paper. Overall, in its present form, it is difficult to verify and reproduce the methodology, and its limitations are inadequately addressed in the discussion.

Author Response

Response: Thank you for your review and suggestions. We are pleased to learn that the paper has become clearer in explaining the research objectives. At the same time, we fully understand your need for further elaboration on methodological details. We did not fully explain the CASA model and its related details in previous versions. We are revising version 2.3.1 The second paragraph of the NEP estimation model clearly explains the specific meaning, implementation method, and variables that make up the method, ensuring that readers can fully understand the application of the model. For the databases MOD13A1 (including its abbreviation NDVI), MOD16A2, and MCD12Q1 used, we have explained in detail the specific meanings and preprocessing processes of these databases in the revised version. As for the specific use of all data in the model, they have been marked in the form of references. The information regarding the analysis period mentioned in the attached letter has been supplemented in the revised version to ensure that readers can clearly understand the time frame of the study. Regarding the discussion on limitations, we understand your suggestion to support the analysis of local limitations by citing relevant literature and exploring how other global remote sensing studies have addressed similar limitations. However, in the current paper, we have conducted a comprehensive and in-depth analysis of the limitations of the research, and pointed out the direction of future work, including improving observation techniques, optimizing remote sensing data processing methods, improving model simulations, etc., to enhance the accuracy and completeness of the data. Regarding the issue of validation and reproducibility methodology you mentioned, we have provided as detailed a description of our research methodology and data processing process as possible in the paper to ensure that other researchers can understand and reproduce our work. Of course, we also understand that there may be some challenges encountered in practical operations, which is also the direction we need to continue improving in our future work. As for why other studies were not cited to address similar limitations, this is mainly because our research focuses on in-depth analysis within specific geographic regions (such as Heilongjiang Province) and research areas (such as net ecosystem productivity), while other studies may focus on different regions or areas, so their methods and results may not be fully applicable to our research.

Author Response File: Author Response.docx