Study on the Diurnal Dynamic Changes and Prediction Models of the Moisture Contents of Two Litters
Round 1
Reviewer 1 Report
Forests. Manuscript ID 681500
Study on the diurnal dynamic changes and prediction models of moisture contents of two litters
In this work a comparison between measured and modeled litter moisture content of two forestry species is showed. This kind of studies are important for improving fire danger and fire risk models. The issue of the paper is also relevant in a context of climate change where the increasing temperatures and changes in the precipitation regimes will impact fuel moisture dynamics and, consequentely, fire regimes. Nevertheless, the manuscript has significant points that need to be further managed before a publication in an international scientific journal. Following are my mayor concerns and comments.
I found some difficulties in reading the Introduction. The objective is clear, but not the relevance of your study. This is because, in my opinion, you need to better describe the state of the art. Issues as the role of fuel moisture sub-models in the fire danger models should be discussed, as well as the tipology (empirical, physical) of available fuel moisture models. This mainly because, in your work, I do not see particular novelty in the Methods section and, as a consequence, your Introduction should clearly justify the objective of your work for an international reader. Please give more arguments for this. E.g. there is no mention of of other fuel moisture models and of its performance, the fire regime of the area is not described, as well as the extension and importance of the two studied forest tipologies at national (China) and international scale.
Materials and Methods. Is there a particular motivation for chosing the selected models? With your data you can apply also other fuel moisture models (maybe obtaining better results..)
Six bags (3 per species), and only 7 measurements days means a quite short database. In addition, the measurements were conducted at local scale. It perhaps impacted the results. For instance, shadow and sun position of samples greatly influence the moisture content. This can be relevant for results esposed in an international journal. Please discusse it. Furthermore, the weather station height was 1.5 m, but the WMO rules indicate 2.0 m. It means that your data will have difficulties in a comparison with most of studies where the weather station height is 2.0 m.
Some minor comments
L11: “It is necessary to determine...” why it is necessary?
L13-16 I suggest to indicate before the objective, then what has been done
L16 “regression model were selected to establish the moisture content prediction model”: to establish the model??
L25 the direct estimation method is not described so far
L30 and following, The fire risk definition should be better stated. Most of the accepted definitions do not include the ignition factors estimation you cited
L33-36 not clear
L36: frequent occurrence? Maybe “increasing occurrence”
L38 more recent citations?
L56-65. This is a crucial point of the paper. In your description it seems that you need to choose one or another method. Please rewrite this sentence
L75 prediction?
L77: which fire risk system??
L83 the sentence about errors should be better discussed. Which kind of errors? Biases are related to the model objectives. Maybe is better to talk about uncertainties
L128 is there a particular motivation for using hm instead of the universal diffuse km
L132 and other. Units are space separated from numbers, 2.8 °C and not 2.8°C
L143 elevation I suppose
L143 canopy density unit?
L143 Litter, not liter
Author Response
In this work a comparison between measured and modeled litter moisture content of two forestry species is showed. This kind of studies are important for improving fire danger and fire risk models. The issue of the paper is also relevant in a context of climate change where the increasing temperatures and changes in the precipitation regimes will impact fuel moisture dynamics and, consequentely, fire regimes. Nevertheless, the manuscript has significant points that need to be further managed before a publication in an international scientific journal. Following are my mayor concerns and comments.
1. I found some difficulties in reading the Introduction. The objective is clear, but not the relevance of your study. This is because, in my opinion, you need to better describe the state of the art. Issues as the role of fuel moisture sub-models in the fire danger models should be discussed, as well as the tipology (empirical, physical) of available fuel moisture models. This mainly because, in your work, I do not see particular novelty in the Methods section and, as a consequence, your Introduction should clearly justify the objective of your work for an international reader. Please give more arguments for this. E.g. there is no mention of of other fuel moisture models and of its performance, the fire regime of the area is not described, as well as the extension and importance of the two studied forest tipologies at national (China) and international scale.
Answer:
The role of fuel moisture content sub-models in the fire danger models and the tipology (empirical, physical) of available fuel moisture models has been increased as required. The fuel moisture models and of its performance has been increased as required. The fire regime of the area (Heilongjiang Province, China) has been increased as required.
2.Materials and Methods. Is there a particular motivation for chosing the selected models? With your data you can apply also other fuel moisture models (maybe obtaining better results..)
Answer: After preliminary research, Nelson model and Simard Model can achieve better prediction results in this area, so we choose these two models.
Thank you very much for your comment, other equilibrium moisture content models will also be considered in later research.
3. Six bags (3 per species), and only 7 measurements days means a quite short database. In addition, the measurements were conducted at local scale. It perhaps impacted the results. For instance, shadow and sun position of samples greatly influence the moisture content. This can be relevant for results esposed in an international journal. Please discusse it. Furthermore, the weather station height was 1.5 m, but the WMO rules indicate 2.0 m. It means that your data will have difficulties in a comparison with most of studies where the weather station height is 2.0 m.
Answer:
In this study, cross validation is chosen to avoid the problem of short database. Although the slope, aspect and canopy density have effect on the dynamic of moisture content of litter, if these factors are taken into consideration to study the diurnal dynamic process of moisture content and prediction models, many uncertain factors will be added. Therefore, this study select only one forest stand for each type of litters. And has been increased in the manuscript. After comparison, there is almost no difference between the meteorological data at 1.5 m and 2.0 m in the forest, so it has litter impact on the later results and discussion. Related description have been deleted. In addition, in some moisture content research literature, the height of weather station is also set at 1.5 m. (Sun P, Yu HZ, Jin S (2015) Predicting hourly litter moisture content of larch stands in daxinganling region, china using three vapour-exchange methods. International Journal of Wildland Fire,24(1), 114-119. doi: 10.1071/WF14098)
Some minor comments
4. L11: “It is necessary to determine...” why it is necessary?
Answer:
The error in the model of litter moisture content prediction, which are mainly due to the neglect of the diurnal variation process. Consequently, it is necessary to determine the diurnal variation of litter moisture content and establish a high-precision prediction model.
It has been increased in the abstract.
5.L13-16 I suggest to indicate before the objective, then what has been done
Answer:
It has been modified as required.
6.L16 “regression model were selected to establish the moisture content prediction model”: to establish the model??
Answer:
It has been modified as required.
7. L25 the direct estimation method is not described so far
Answer:
It has been increased in the introduction. And some described in the section of model description.
8. L30 and following, The fire risk definition should be better stated. Most of the accepted definitions do not include the ignition factors estimation you cited
Answer:
It has been modified as required.
9. L33-36 not clear
Answer:
It has been modified as required.
10. L36: frequent occurrence? Maybe “increasing occurrence”
Answer:
It has been modified as required.
11. L38 more recent citations?
Answer:
It has been increased recent citations.
12.L56-65. This is a crucial point of the paper. In your description it seems that you need to choose one or another method. Please rewrite this sentence
Answer:
It has been rewrited as required.
13. L75 prediction?
Answer:
Prediction has been replaced by value.
14. L77: which fire risk system??
Answer:
The Canadian Fire Risk Forecasting System, the noon temperature, humidity, wind speed and the first 24 h of rainfall are used as predictors to calculate the litter moisture content. The United States, meteorological elements, such as the daily maximum and minimum temperature, are used.
15. L83 the sentence about errors should be better discussed. Which kind of errors? Biases are related to the model objectives. Maybe is better to talk about uncertainties
Answer:
It has been increased as required.
16. Two main reasons for the errors are as follows: one is the error is caused by oversimplification of the diurnal variation of the moisture content of litter (Nelson, 2000; Matthews, 2006), and the other is the errors caused by extrapolation and poor applicability (Wotton et al. 2005; Wotton and Beverly, 2007).
Answer:
For specific type of litter, it seems the main reason that do not take the influence of diurnal variation of meteorological elements on the change of litter moisture content into considered.
17.L128 is there a particular motivation for using hm instead of the universal diffuse km
Answer:
It has been modified as required. All hm changed to km.
18.L132 and other. Units are space separated from numbers, 2.8 °C and not 2.8°C
Answer:
It has been modified as required.
19. L143 elevation I suppose
Answer:
It has been modified as required.
20. L143 canopy density unit?
Answer:
Canopy density is dimensionless.
21. L143 Litter, not liter
Answer:
It has been modified as required.
Reviewer 2 Report
Thanks to the authors for presenting this research into fuel moisture of leaf litter in a Mongolian ecosystem. This research does advance fire research in a geographical area that is under-represented in the literature. The main issue is that the authors need to bring more clarity in their description of the statistical methods. I realize that the differential equations (Byram etc.) can be run with low sample sizes, but this might also affect the ability to generalize from them. Also, the writing is hard to follow in places and needs some editing for clarity.
Line 10: Moisture is a major factor but not the sole detriment. Rephrase.
Line 30-36: This sentence is hard to follow. Rewrite.
Line 40: replace terms in quotation marks with suppression.
Line 43: Accuracy of what?
48: Most important object? Replace “object” with variable. Also is forest fuels always the most important variable for fire forecasting? Modify or provide evidence this is the case.
49: replace ignition source with fuel.
53-54: These lines get to the heart of the matter. Emphasize.
Line 72: unclear rewrite.
Line 84: Citation for evidence of the production errors?
112:113 Adding more detail about these fire ecosystems would help the reader put the moisture content trends in context.
Line 140: What constitutes a plot and replicate samples from those plots needs to stated directly and clearly. Is the total number of plots 2?
Line 154-155. Again the relationship between samples and the total n is not clear here. So a total of 6 samples with three from each plot? Where does the 168 set of data come in?
Figure 7 caption needs more detail to explain what is happening in the graph. Confidence intervals?
Line 366. How much of this difference might be attributable to chance or measurement error vs that attributable to the model itself?
Line 403: Does the data presented here suggest this methods is ready for field use without further testing?
Line 407. Conclusion. Start conclusion at end of line 416. A detailed account of the moisture trends is not needed for conclusion.
Line 468. Check references. Fire study in a cancer journal?
European Journal of Surgical Oncology
Author Response
1.Thanks to the authors for presenting this research into fuel moisture of leaf litter in a Mongolian ecosystem. This research does advance fire research in a geographical area that is under-represented in the literature. The main issue is that the authors need to bring more clarity in their description of the statistical methods. I realize that the differential equations (Byram etc.) can be run with low sample sizes, but this might also affect the ability to generalize from them. Also, the writing is hard to follow in places and needs some editing for clarity.
Answer:
In this study, cross validation is chosen to avoid the problem of short database. And the writing has been improved.
2.Line 10: Moisture is a major factor but not the sole detriment. Rephrase.
Answer:
It has been rewrite as required.
The possibility of forest fire and fire behavior are mainly affected by moisture content of litter, which is very important for forest fire risk forecasting.
3. Line 30-36: This sentence is hard to follow. Rewrite.
Answer:
It has been rewrite as required.
4. Line 40: replace terms in quotation marks with suppression.
Answer:
It has been modified as required.
5. Line 43: Accuracy of what?
Answer:
The accuracy of fire occurrence and potential damages predictions. It has been increased.
6. 48: Most important object? Replace “object” with variable. Also is forest fuels always the most important variable for fire forecasting? Modify or provide evidence this is the case.
Answer:
It has been modified as required and increased the evidence.
7. 49: replace ignition source with fuel.
Answer:
It has been modified as required.
8. 53-54: These lines get to the heart of the matter. Emphasize.
Answer:
It has been rewrite as required.
9. Line 72: unclear rewrite.
Answer:
It has been modified as required
10. Line 84: Citation for evidence of the production errors?
Answer:
It has been increased as required
11. 112:113 Adding more detail about these fire ecosystems would help the reader put the moisture content trends in context.
Answer:
The fire regime of the area (Heilongjiang Province, China) has been increased as required.
12. Line 140: What constitutes a plot and replicate samples from those plots needs to stated directly and clearly. Is the total number of plots 2?
Answer:
For each litter type, select a plot, and randomly set three samples points in each plot for moisture content monitoring, to ensure three repetitions. It has been modified as required.
13. Line 154-155. Again the relationship between samples and the total n is not clear here. So a total of 6 samples with three from each plot? Where does the 168 set of data come in?
Answer:
For each litter type, select a plot. For each plot, three plastic basket for litter was arranged. The samples were weighed once every 1 h for a total of 7 days. After the experiment, the basket was dried at 105℃ for 24 h, and the dry weight was recorded. Then, the surface litter bed moisture content was obtained according to the formula for calculating the moisture content (the ratio of the wet weight to the dry weight of the litter bed). A total of 24168 sets of data were obtained for each basket and a total of 168=504 sets of data were obtained for each plot. The arithmetic mean of the moisture content of the three baskets of each plot was calculated as the moisture content value of the plot. and randomly set three samples points in each plot for moisture content monitoring, to ensure three repetitions for 7 days.
It has been modified as required.
14. Figure 7 caption needs more detail to explain what is happening in the graph. Confidence intervals?
Answer:
The detail description has been increased.
When the moisture content range of litter being different in Fig.7, it shows the difference between the predicted and measured values of different methods.
15. Line 366. How much of this difference might be attributable to chance or measurement error vs that attributable to the model itself?
Answer:
The discussion of this difference has been increased.
The packing ratio of litter bed and the physical properties of litter have significant impact on the dynamic change of moisture content (Zhang, 2019). Therefore, the different water holding capacity of the two types of litter bed may be caused by the different bed structure and physical properties of litter.
16. Line 403: Does the data presented here suggest this methods is ready for field use without further testing?
Answer:
For the litter in the field in this area, they can be directly used without testing. Due to the spatial heterogeneity of moisture content, it is necessary to extrapolate the litter in other areas.
17. Line 407. Conclusion. Start conclusion at end of line 416. A detailed account of the moisture trends is not needed for conclusion.
Answer:
It has been modified as required.
18. Line 468. Check references. Fire study in a cancer journal? European Journal of Surgical Oncology
Answer:
It has been modified as required.
Round 2
Reviewer 1 Report
Dear Authors, thank you for accepting my comments. The manuscript is now improved.
Best regards
Author Response
Comment:
Dear Authors, thank you for accepting my comments. The manuscript is now improved.
Best regards
Answer:
Thank you very much for your review of my manuscript. All your valuable Comments have been revised as required.
Thank you.
