Effects of Forest Fire Prevention Policies on Probability and Drivers of Forest Fires in the Boreal Forests of China during Different Periods
Round 1
Reviewer 1 Report
The paper entitled “Effects of forest fire prevention policies on probability and drivers of forest fires in the boreal forests of China during different periods” used historical fire data from 1981 to 2020 and took the regulations that has significantly changed China's forest fire prevention. The paper was written in a good shape; however, it needs some minor revisions:
1. Abstract and introduction: The aim of the study is not clear, please write about objectives in abstract and introduction 2. Figure 1: Legend of maps have a low quality which is hard to read easily. Please change it. 3. Section 2.2.2: give more details such as the values of average temperature (°C), maximum temperature (°C), daily temperature range (°C), average relative humidity (%), average surface air temperature (°C), maximum surface air temperature (°C), minimum relative humidity (%), average wind speed (m/s), sunshine hours, and accumulated precipitation (mm/24h). in a table. 4. LN 145: How much authors sure that forest fires cause by climate no by human and mistakes made by man? 5 Section 2.2.3: Do you have information about the vegetation species for 1985? How it changes during the time? 6. Section 2.3.3. Why didn’t the authors use RMSE or MAE to evaluate the models? 7. Discussion: What are the limitations in you study? Write about limitationsAuthor Response
请参阅附件
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper asks relevant questions to recognize the impact of prevention policies on the risk of forest fire in a relatively unknown region of China. Through these different aspects, it is therefore of certain interest. However, in its current version it has many formal and substantive weaknesses. As not specialist of modeling methodological approach, the reviewer focuses on his area of expertise on fire history and spatial analysis.
From a formal point of view, there is a lot to improve. There are several points that show that the authors did not pay enough attention to the finalization of the text.
- e.g.: why do § 4.1 (l.467) and 4.2 (l.495) have the same title: “Changes in forest fire risk zones and fire causes in different periods”. It is the same for § 4.3 (l.552) and 4.4 (l.609): “Implications for forest fire modeling and management”.
- In the text, especially the introduction, the authors use different surface units (ha, hm², km²). This makes it difficult to understand the proportions between burned areas and types of vegetation, for example.
- Figures 2 and 3 which summarize the results should be improved. In this form, they do not make it possible to compare the changes between the 3 periods because the scales of the graphs are different. What does also mean the caption “Time” on the right axis?
Tables 1 and 2 also have difficulty in understanding even if the abbreviations used are explained in the appendix.
These are some examples of possible improvements, but the header deserves a deep rewrite.
In terms of content there are also important improvements to be made.
The first question concerns the nature of the data used. The authors admit that there are limitations regarding the data issued from official government sources and that it is crucial to check their reliability (l.610-611). All the reasoning is however developed in the paper without these precautions. Are the quality, precision or sincerity of the statements of the administration comparable throughout the period studied? For example, we can wonder about the way in which the causes are assigned to fires. This is all the more important as one of the objectives is to assess the impact of fire prevention policies by the authorities. More practically the data on burned areas are not discussed. How are they estimated? Again, is there consistency in the observations over the 40 years of the study? Can we have accuracy to the hundredth of hm² as the authors use it? The map of fig.1c represents the fires by points, which does not allow appreciating their importance on the surface.
Capital notions of the paper are sometimes imprecise or insufficiently defined, sometimes misunderstood. For example in l.178 “One of the main causes of fire is the type of vegetation”: this sentence is inadequate. The type of vegetation can be a predisposing factor for the ignition and spread of forest fires, but is not a cause. The notions of “fire probability” and “fire risk” essential in the paper are never explained.
There are also many claims with little support from the results.
In the discussion, there is very often confusion between occurrence factor or cause of fire and simple correlation (l.518-520, l.527-531).
In the conclusion (l.649-650), the authors write: “The new and old Forest Fire Prevention Regulations have changed the drivers dominated by human factors, and the causes of fires”. This assertion is not demonstrated in the paper. There are certainly quantitative more or less important changes between the periods, but the causal link with the Regulations is not explained. This may possibly be implied for the reduction of human-made fires. But how can regulations explain the absolute and not only relative increase in lightning fires? This is one of the major weaknesses of the paper: the deductive approach implemented is not sufficiently supported by the data analysis.
l.444-447 the authors write: “It is important to note that the border regions with Russia and Mongolia, located in the north and southwest of the four periods, have medium and high fire risk, respectively. This suggests that a foreign fire poses a significant threat to the Daxing'an Mountains in Inner Mongolia”. This is not well demonstrated in Figure 7.
How can you explain: “the distance from the fire point to the watchtower is a driver for fire occurrence” or “most fire points occur at a significant distance from the watchtower” (l. 544-547)… Isn't there a bias?
These are some of the remarks that lead the reviewer to propose a major revision before the publication of this paper which has real potential.
Author Response
请参阅附件
Author Response File: Author Response.docx
Reviewer 3 Report
The manuscript researched the effects of forest fire prevention policies on fire probability and further revealed the drivers of forest fires in the boreal forests of China during four different periods. The study will be helpful for government to evaluate the fire prevention policies during different periods, and for policy-makers to optimize the fire management strategies to reduce potential fire risks.
The method used in this manuscript is appropriate, the data are abundant and reasonable, and the conclusion is credible. I like this manuscript very much. I suggest receiving the manuscript after minor revision.
1. P9 Line 366 (Fig.2): Merge a, b, and c into one figure with the same ranges of fire time and burned area. Additional, Figures b and c can be drawn in subfigure with a small range in Y-axis.
2. P9 Line 369 (Fig.3): The cause of fires should be classified into several categories, such as natural factors (lightning fire, ), human factors (smoking, paper money burning, ……), invasive factors (foreign country, foreign province, foreign county, ……), ……. Put the same subclasses within the same category together for easy analysis.
3. P14 Line 467 and P15 Line 495 (subsection 4.1 and 4.2): The titles of these two subsections are the same, please check them.
4. P16 Line 552 and P17 Line 609 (subsection 4.3 and 4.4): The titles of these two subsections are the same, please check them.
5. Further clarify the contributions of different factors during different periods to forest fire times and burned areas, as so to pay more attention to it(them) and prevent the occurrence of forest fires.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Thanks to the authors for this in-depth review of their manuscript.
All the reviewer's remarks have been taken into account in the new version or, at least, explained. The improvement is remarkable, even exemplary. Cartography and figures have been significantly improved.
This version can now be published.
