the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Aug 2023
National forest carbon harvesting and allocation dataset for the period 2003 to 2018
Abstract. Forest harvesting is one of the anthropogenic activities that most significantly affect the carbon budget of forests. However, the absence of explicit spatial information on harvested carbon poses a huge challenge in assessing forest harvesting impacts, as well as the forest carbon budget. This study utilized provincial-level statistical data on wood harvest, the tree cover loss (TCL) dataset, and a satellite-based vegetation index to develop a Long-term harvEst and Allocation of Forest Biomass (LEAF) dataset. The aim was to provide the spatial location of forest harvesting with a spatial resolution of 30 m and quantify the post-harvest carbon dynamics. The validations against the surveyed forest harvesting at 133 cities and counties indicated a good performance of the LEAF dataset in capturing the spatial variation of harvested carbon, with a coefficient of determination (R2) of 0.83 between the identified and surveyed harvested carbon. The linear regression slope was up to 0.99. Averaged from 2003 to 2018, forest harvesting removed 68.34 Mt C yr-1, of which more than 80% was from selective logging. Of the harvested carbon, 22%, 45%, 4%, and 29% entered the wood fuel, wood products, paper products, and residual pools, respectively. Direct combustion of wood fuel was the primary source of carbon emissions after wood harvest. However, carbon can be stored in wood products for a long time, and by 2100, almost 90% of the harvested carbon during the study period will still be retained. This dataset is expected to provide a foundation and reference for estimating the forestry and national carbon budgets. The 30 m × 30 m harvested carbon dataset from forests in China can be downloaded at https://doi.org/10.6084/m9.figshare.23641164.v2 (Wang et al., 2023).
- Preprint
(1709 KB) - Metadata XML
-
Supplement
(990 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on essd-2023-309', Anonymous Referee #1, 05 Oct 2023
This study generated nation-wide harvested carbon data through synthesizing the remote sensing data, statistical data and empirical equations/parameters, and further estimated the carbon stock in harvested wood. The dataset can help provide effective and accurate data and parameters for estimating carbon flux due to forest disturbance and harvesting. The major problems in this manuscript include: (1) The methods are not detailed and sound enough, which result in the less confidence for this work; (2) lack of a quantification of uncertainty; (3) some of the results seem unreasonable. Other specific comments/suggestions are listed below.
1. There are a lot of English writing issues. I won’t point them here. The authors should carefully revise the English writing sentence by sentence.
2. Some terminologies are not conformed throughout the manuscript. Please check through and make them consistent.
3. Line 107-140: This approach for calculating the selective harvesting is not valid enough since NDVI values varied greatly among years due to climate change and other disturbance events. The CCDC, VCT, LandTrendr and other similar algorithms are more professional and proved methods to detect partial forest disturbance. You can consider using one of those algorithms to replace your work.
4. Line 152-155: In fact, paper belongs to the wood product pool (pool 3). “wood fuel” can be renamed to “fuelwood”.
5. Line 153-159: To more accurate tracking the carbon fates in HWPs, varied time-scale HWP pools should be further divided. For example, divide HWP into 2 (such as paper), 5 (e.g., decorative uses), 20 (e.g., furniture), 50 (e.g., buildings) half-life HWP pools. In addition, a landfill carbon pool is needed to separate since the decay of the landfill is significantly different from the regular HWPs. The carbon decay from landfill should be separately simulated.
6. Line 158: In my memory, the half-life of paper in IPCC report is 2 years (4 years’ lifespan)? Please double check and confirm
7. Line 158: it is not reasonable and accurate to assign 100 years’ span (50 years for half-life) for all HWPs. As I suggest above, the HWPs should be separated more pools to accurately track the carbon dynamic.
8. Line 167-222: Most of the parameter values in this study are from the IPCC default values. Actually, more specific parameter values are available for China. Please retrieve more published papers from Chinese scholars related to forest harvesting or disturbance or harvested wood.
9. Line 185: the SWDS overlaps with the landfill carbon. I suggest discarding the concept of SWDS and using “Landfill” to replace.
10. Method: how to allocate the harvested tree organs (leaf/root/stem) to the four harvested carbon pools? All the harvested biomass is allocated into HWPs? The fractions for these organs vary greatly among tree species and ages. The allocation methods are needed to elaborate in the method.
11. Method: I did not see how the pixel level AGB is obtained or calculated (only mentioned the provincial AGB data). This is important to ensure the data quality.
12. Method: If I understand right, the inventory AGB is used to calculate the harvested carbon; however, this static AGB data fails to track the forest regrowth after harvesting. This will result in some uncertainties, so a short discussion is needed to mention this limitation.
13. Method: there are too many parameters and their values. A or multiple Table(s) should be used to list all parameters, values and sources.
14. Line 218: the GWPs for CH4 in the new IPCC report should be used rather than the IPCC 2007.
15. Line 281: The harvested carbon here includes which carbon pools? All four carbon pools? Or just harvested wood biomass?
16. Line 304: 80% harvested carbon is from selective harvesting. This percentage is too high. In my mind, the main forest harvesting mode is clearcut, especially for the commercial forests. Any additional proofs to provide?
17. Figure 8: The annual variations in harvested carbon pools look unreasonable, for example, why the total carbon stock (all carbon pools add together) is still very high at 2100? Based on the decay equations and half-life span (50 half-life for HWPs), most of the harvested carbon should be released at 2100 (about 80 years from 2019-2100). Need to double check the calculation equations and their parameters, especially for the carbon decay equations.
18. Results: Uncertainty ranges should be quantified when present the research results (harvested carbon and decayed carbon) since there are a lot uncertainties associated to parameters and their values. Namely, each result should provide a variation range (±standard deviation). Single parameter values are provided in the equations, but actually these values could vary in a range.
19. Discussion: the estimated HWPs amounts in this study should be compared with previous studies. As I know, there are several studies published by the Chinese scholars also provided these data.
20. The uploaded data link on figshare does not work, please check the problem.Citation: https://doi.org/10.5194/essd-2023-309-RC1 -
RC2: 'Comment on essd-2023-309', Anonymous Referee #2, 18 Nov 2023
With intensified climate change, international organizations and national governments have highlighted the significant role of forests in mitigating climate change. This research attempts to establish a long-term harvest and allocation of forest biomass dataset, which provides guidelines for China’s carbon budget assessment. The analysis of forest carbon allocation would attract more attention in future research. However, I still have some questions, which I’m hoping authors can respond to:
- The dataset is the central component of this study, but its name doesn't appear to accurately describe its features. What are the benefits of this dataset, furthermore? The dataset's significance is not properly represented.
- It is necessary to introduce the design of the dataset establishment in further detail at the beginning of the approach. Otherwise, readers cannot catch your key points. I have other issues: what’s the difference between forest loss and selective logging? It is best to describe new concepts as they emerge in your research for the first time and demonstrate how they relate to earlier concepts.
- For the same reason, what does the term “potential selective logging areas” mean, and why?
- The research contains many acronyms. It is preferable to create a table so that they can be seen more clearly.
- Line 110: Why do you assume that? Is there a foundation?
- It is important to consider the basis upon which your approach or theory is based. This is the weakness of this study.
- Generally, linear regression is used to represent the average level. However, the scale of dataset established in this study is 30 m * 30m, therefore the test based on linear regression cannot be a reliable way to verify the accuracy of LEAF dataset.
- The data scales before and following the section on "carbon flows between HWPs" appear to be inconsistent, and the part appears to have little link with the preceding sections. It is not creative to just calculate the current state of carbon flows.
- Also, if I understand correctly, the core of the article's dataset construction is “the TCL dataset produced by Hansen et al. (2013)”, and then used Chinese coefficients to calculate forest carbon. How significant it is? Additionally, the coefficients are applied at the provincial level, which is significantly less precise than carbon sink observations from remote sensing.
Citation: https://doi.org/10.5194/essd-2023-309-RC2 -
RC3: 'Comment on essd-2023-309', Anonymous Referee #3, 08 Dec 2023
Wood harvesting is a significant land use activity that reduces forest biomass and affects carbon budget. This paper developed a spatial dataset of harvested carbon pools in China by downscaling province-level wood harvest statistics using satellite data. This dataset also includes the allocation of wood products over time. The idea of merging two data sources is very constructive, and I believe such a dataset of wood harvests is critical for understanding the scale and impact of forestry. While this dataset could have a very strong scientific contribution, I have several concerns, mainly regarding the methodology, which require clarification and careful examination by the authors.
Major comments:
The data generation are separated into two parts: 1) Harvested carbon, and 2) Decay of harvested carbon pools.
Part 1.
- The way that estimates selective logging seems problematic (Equation 5). The quantity of selective logging is determined by the percentage of NDVI reduction from the previous year.
How do the authors separate the regrowth after harvest and selective logging within the same grid cell? NDVI_diff can be the net result of regrowth and loss due to logging. In the algorithm, the actual selective logging will be considered as disturbance (undetected) when NDVI_diff is small. There is a risk of underestimating selective logging in areas with higher forest regrowth, where NDVI_diff is smaller.
- The way that calculates carbon loss due to clear-cutting seems straightforward. However, the parameter Coef (L95-100, Table S1) requires some data quality checks and some discussion of potential data uncertainties.
For example, there is a huge difference in Coef between Qinghai and Tibet despite similar ecozones (even though forests are rare over there). Coef is equivalent to wood basic density x carbon fraction of dry wood x biomass expansion factor. Assuming wood basic density and carbon fraction at 0.5, biomass expansion factor will be around 2.8 to get a Coef at 0.73 in Qinghai, which is extremely high. This raises a question of whether this Coef method is valid in areas with very low AGB values. Also, there is a typo in Table S1: Qinhai-> Qinghai.
- Can the authors explain why FAO data is substantially higher (40-50% more) than province level data (Figure 5)? What are the potential systematic biases between the two statistics? I recalled that LUH2 also used FAO data, so why is there a huge discrepancy between FAO and LUH2. Please explain by further elaborating the steps of calculation/aggregation of data.
Part 2.
- The paper uses confusing terminologies and too many acronyms, which makes it hard to follow.
Harvested wood products were allocated into wood fuel, paper, and wood products (L150, L230). How can the term “wood products” be used to refer both to the total sum, and to a sub-category? Suggest use longlived wood products.
Also, are landfill and solid waste disposal sites (SWDS) referring to the same thing?
The definition of lumber that includes paper and wood fuel (2.3.2) is unconventional. Lumber is typically understood as wood used for building and furniture.
- Allocation method - it is not clear to me how the authors allocate the harvested wood into the four pools (Section 2.2, L150-155).
For example, the residual pool includes wood used as wood fuel burned for energy, independent from the existing wood fuel category. How did the authors separate statistical data for wood fuel into dedicated wood fuel and residual wood being burned?
How did the authors treat the harvesting slashes such as branches, killed understory vegetation, and roots in these pools? I suspect they are not accounted for and I am not sure whether the statistical data would provide this information.
How did the authors treat bark during the processing? Is it allocated to the wood fuel, residual, or not accounted for?
- The spatial distribution of the decay harvested carbon after harvesting (Figure 7). Did the authors assign the subsequential changes in the carbon pools to their original harvest locations (grid cell and province level)? Considering that commercial wood is often transported and processed elsewhere, how is the spatial flow accounted for?
- Figure 8b. The linear increase in wood harvests at the beginning reflects the actual wood harvests from statistics during 2003-2018. Since you don’t have harvesting data since 2018, the post-2018 time series are solely decay pools (not reflecting the real-world situations). These two periods represent different meanings and should perhaps be presented separately for clarity. It is also hard to compare 8a and 8b as the units are different.
- Lifespan are 5, 100 years for paper and longlived products (IPCC, 2019a) in L155, while k for paper products and wood products were used as 0.347 and 0.023 (L180), respectively, according to the IPCC default values (IPCC, 2014). Therefore, half-life will be ln(2)/0.347 and ln(2)/0.023, rather than 2.5 and 50 years. Which set of values is ultimately used?
Minor comments:
- Figure 3. It is hard to compare a and b as the scales in the legends are so different. It also suggests that 30m resolution is less capable in capturing harvesting compared to the 0.1 degree.
Citation: https://doi.org/10.5194/essd-2023-309-RC3 -
AC1: 'Response to Referees', Daju Wang, 15 Jan 2024
Dear Editor and Referees,
We are very grateful to you and referees for your constructive comments and suggestions on our manuscript “National forest carbon harvesting and allocation dataset for the period 2003 to 2018” (MS No.: essd-2023-309). We have carefully studied the comments, and revised our manuscript accordingly. Consequently, our manuscript has been considerably improved.
Our detailed responses to the comments are in the supplement. Please note that the comments from the referees are in bold followed by our responses in regular text. The revised and newly added sentences have been highlighted in red. Major revisions and new references are marked in the manuscript and using tracked changes.
Thank you for your consideration.
Sincerely,
Daju Wang, Wenping Yuan, on behalf of all co-authors
Email: yuanwp3@mail.sysu.edu.cn
Status: closed
-
RC1: 'Comment on essd-2023-309', Anonymous Referee #1, 05 Oct 2023
This study generated nation-wide harvested carbon data through synthesizing the remote sensing data, statistical data and empirical equations/parameters, and further estimated the carbon stock in harvested wood. The dataset can help provide effective and accurate data and parameters for estimating carbon flux due to forest disturbance and harvesting. The major problems in this manuscript include: (1) The methods are not detailed and sound enough, which result in the less confidence for this work; (2) lack of a quantification of uncertainty; (3) some of the results seem unreasonable. Other specific comments/suggestions are listed below.
1. There are a lot of English writing issues. I won’t point them here. The authors should carefully revise the English writing sentence by sentence.
2. Some terminologies are not conformed throughout the manuscript. Please check through and make them consistent.
3. Line 107-140: This approach for calculating the selective harvesting is not valid enough since NDVI values varied greatly among years due to climate change and other disturbance events. The CCDC, VCT, LandTrendr and other similar algorithms are more professional and proved methods to detect partial forest disturbance. You can consider using one of those algorithms to replace your work.
4. Line 152-155: In fact, paper belongs to the wood product pool (pool 3). “wood fuel” can be renamed to “fuelwood”.
5. Line 153-159: To more accurate tracking the carbon fates in HWPs, varied time-scale HWP pools should be further divided. For example, divide HWP into 2 (such as paper), 5 (e.g., decorative uses), 20 (e.g., furniture), 50 (e.g., buildings) half-life HWP pools. In addition, a landfill carbon pool is needed to separate since the decay of the landfill is significantly different from the regular HWPs. The carbon decay from landfill should be separately simulated.
6. Line 158: In my memory, the half-life of paper in IPCC report is 2 years (4 years’ lifespan)? Please double check and confirm
7. Line 158: it is not reasonable and accurate to assign 100 years’ span (50 years for half-life) for all HWPs. As I suggest above, the HWPs should be separated more pools to accurately track the carbon dynamic.
8. Line 167-222: Most of the parameter values in this study are from the IPCC default values. Actually, more specific parameter values are available for China. Please retrieve more published papers from Chinese scholars related to forest harvesting or disturbance or harvested wood.
9. Line 185: the SWDS overlaps with the landfill carbon. I suggest discarding the concept of SWDS and using “Landfill” to replace.
10. Method: how to allocate the harvested tree organs (leaf/root/stem) to the four harvested carbon pools? All the harvested biomass is allocated into HWPs? The fractions for these organs vary greatly among tree species and ages. The allocation methods are needed to elaborate in the method.
11. Method: I did not see how the pixel level AGB is obtained or calculated (only mentioned the provincial AGB data). This is important to ensure the data quality.
12. Method: If I understand right, the inventory AGB is used to calculate the harvested carbon; however, this static AGB data fails to track the forest regrowth after harvesting. This will result in some uncertainties, so a short discussion is needed to mention this limitation.
13. Method: there are too many parameters and their values. A or multiple Table(s) should be used to list all parameters, values and sources.
14. Line 218: the GWPs for CH4 in the new IPCC report should be used rather than the IPCC 2007.
15. Line 281: The harvested carbon here includes which carbon pools? All four carbon pools? Or just harvested wood biomass?
16. Line 304: 80% harvested carbon is from selective harvesting. This percentage is too high. In my mind, the main forest harvesting mode is clearcut, especially for the commercial forests. Any additional proofs to provide?
17. Figure 8: The annual variations in harvested carbon pools look unreasonable, for example, why the total carbon stock (all carbon pools add together) is still very high at 2100? Based on the decay equations and half-life span (50 half-life for HWPs), most of the harvested carbon should be released at 2100 (about 80 years from 2019-2100). Need to double check the calculation equations and their parameters, especially for the carbon decay equations.
18. Results: Uncertainty ranges should be quantified when present the research results (harvested carbon and decayed carbon) since there are a lot uncertainties associated to parameters and their values. Namely, each result should provide a variation range (±standard deviation). Single parameter values are provided in the equations, but actually these values could vary in a range.
19. Discussion: the estimated HWPs amounts in this study should be compared with previous studies. As I know, there are several studies published by the Chinese scholars also provided these data.
20. The uploaded data link on figshare does not work, please check the problem.Citation: https://doi.org/10.5194/essd-2023-309-RC1 -
RC2: 'Comment on essd-2023-309', Anonymous Referee #2, 18 Nov 2023
With intensified climate change, international organizations and national governments have highlighted the significant role of forests in mitigating climate change. This research attempts to establish a long-term harvest and allocation of forest biomass dataset, which provides guidelines for China’s carbon budget assessment. The analysis of forest carbon allocation would attract more attention in future research. However, I still have some questions, which I’m hoping authors can respond to:
- The dataset is the central component of this study, but its name doesn't appear to accurately describe its features. What are the benefits of this dataset, furthermore? The dataset's significance is not properly represented.
- It is necessary to introduce the design of the dataset establishment in further detail at the beginning of the approach. Otherwise, readers cannot catch your key points. I have other issues: what’s the difference between forest loss and selective logging? It is best to describe new concepts as they emerge in your research for the first time and demonstrate how they relate to earlier concepts.
- For the same reason, what does the term “potential selective logging areas” mean, and why?
- The research contains many acronyms. It is preferable to create a table so that they can be seen more clearly.
- Line 110: Why do you assume that? Is there a foundation?
- It is important to consider the basis upon which your approach or theory is based. This is the weakness of this study.
- Generally, linear regression is used to represent the average level. However, the scale of dataset established in this study is 30 m * 30m, therefore the test based on linear regression cannot be a reliable way to verify the accuracy of LEAF dataset.
- The data scales before and following the section on "carbon flows between HWPs" appear to be inconsistent, and the part appears to have little link with the preceding sections. It is not creative to just calculate the current state of carbon flows.
- Also, if I understand correctly, the core of the article's dataset construction is “the TCL dataset produced by Hansen et al. (2013)”, and then used Chinese coefficients to calculate forest carbon. How significant it is? Additionally, the coefficients are applied at the provincial level, which is significantly less precise than carbon sink observations from remote sensing.
Citation: https://doi.org/10.5194/essd-2023-309-RC2 -
RC3: 'Comment on essd-2023-309', Anonymous Referee #3, 08 Dec 2023
Wood harvesting is a significant land use activity that reduces forest biomass and affects carbon budget. This paper developed a spatial dataset of harvested carbon pools in China by downscaling province-level wood harvest statistics using satellite data. This dataset also includes the allocation of wood products over time. The idea of merging two data sources is very constructive, and I believe such a dataset of wood harvests is critical for understanding the scale and impact of forestry. While this dataset could have a very strong scientific contribution, I have several concerns, mainly regarding the methodology, which require clarification and careful examination by the authors.
Major comments:
The data generation are separated into two parts: 1) Harvested carbon, and 2) Decay of harvested carbon pools.
Part 1.
- The way that estimates selective logging seems problematic (Equation 5). The quantity of selective logging is determined by the percentage of NDVI reduction from the previous year.
How do the authors separate the regrowth after harvest and selective logging within the same grid cell? NDVI_diff can be the net result of regrowth and loss due to logging. In the algorithm, the actual selective logging will be considered as disturbance (undetected) when NDVI_diff is small. There is a risk of underestimating selective logging in areas with higher forest regrowth, where NDVI_diff is smaller.
- The way that calculates carbon loss due to clear-cutting seems straightforward. However, the parameter Coef (L95-100, Table S1) requires some data quality checks and some discussion of potential data uncertainties.
For example, there is a huge difference in Coef between Qinghai and Tibet despite similar ecozones (even though forests are rare over there). Coef is equivalent to wood basic density x carbon fraction of dry wood x biomass expansion factor. Assuming wood basic density and carbon fraction at 0.5, biomass expansion factor will be around 2.8 to get a Coef at 0.73 in Qinghai, which is extremely high. This raises a question of whether this Coef method is valid in areas with very low AGB values. Also, there is a typo in Table S1: Qinhai-> Qinghai.
- Can the authors explain why FAO data is substantially higher (40-50% more) than province level data (Figure 5)? What are the potential systematic biases between the two statistics? I recalled that LUH2 also used FAO data, so why is there a huge discrepancy between FAO and LUH2. Please explain by further elaborating the steps of calculation/aggregation of data.
Part 2.
- The paper uses confusing terminologies and too many acronyms, which makes it hard to follow.
Harvested wood products were allocated into wood fuel, paper, and wood products (L150, L230). How can the term “wood products” be used to refer both to the total sum, and to a sub-category? Suggest use longlived wood products.
Also, are landfill and solid waste disposal sites (SWDS) referring to the same thing?
The definition of lumber that includes paper and wood fuel (2.3.2) is unconventional. Lumber is typically understood as wood used for building and furniture.
- Allocation method - it is not clear to me how the authors allocate the harvested wood into the four pools (Section 2.2, L150-155).
For example, the residual pool includes wood used as wood fuel burned for energy, independent from the existing wood fuel category. How did the authors separate statistical data for wood fuel into dedicated wood fuel and residual wood being burned?
How did the authors treat the harvesting slashes such as branches, killed understory vegetation, and roots in these pools? I suspect they are not accounted for and I am not sure whether the statistical data would provide this information.
How did the authors treat bark during the processing? Is it allocated to the wood fuel, residual, or not accounted for?
- The spatial distribution of the decay harvested carbon after harvesting (Figure 7). Did the authors assign the subsequential changes in the carbon pools to their original harvest locations (grid cell and province level)? Considering that commercial wood is often transported and processed elsewhere, how is the spatial flow accounted for?
- Figure 8b. The linear increase in wood harvests at the beginning reflects the actual wood harvests from statistics during 2003-2018. Since you don’t have harvesting data since 2018, the post-2018 time series are solely decay pools (not reflecting the real-world situations). These two periods represent different meanings and should perhaps be presented separately for clarity. It is also hard to compare 8a and 8b as the units are different.
- Lifespan are 5, 100 years for paper and longlived products (IPCC, 2019a) in L155, while k for paper products and wood products were used as 0.347 and 0.023 (L180), respectively, according to the IPCC default values (IPCC, 2014). Therefore, half-life will be ln(2)/0.347 and ln(2)/0.023, rather than 2.5 and 50 years. Which set of values is ultimately used?
Minor comments:
- Figure 3. It is hard to compare a and b as the scales in the legends are so different. It also suggests that 30m resolution is less capable in capturing harvesting compared to the 0.1 degree.
Citation: https://doi.org/10.5194/essd-2023-309-RC3 -
AC1: 'Response to Referees', Daju Wang, 15 Jan 2024
Dear Editor and Referees,
We are very grateful to you and referees for your constructive comments and suggestions on our manuscript “National forest carbon harvesting and allocation dataset for the period 2003 to 2018” (MS No.: essd-2023-309). We have carefully studied the comments, and revised our manuscript accordingly. Consequently, our manuscript has been considerably improved.
Our detailed responses to the comments are in the supplement. Please note that the comments from the referees are in bold followed by our responses in regular text. The revised and newly added sentences have been highlighted in red. Major revisions and new references are marked in the manuscript and using tracked changes.
Thank you for your consideration.
Sincerely,
Daju Wang, Wenping Yuan, on behalf of all co-authors
Email: yuanwp3@mail.sysu.edu.cn
Data sets
National forest carbon harvesting and allocation dataset for the period 2003 to 2018 Daju Wang, Peiyang Ren, Xiaosheng Xia, Lei Fan, Zhangcai Qin, Xiuzhi Chen, Wenping Yuan https://doi.org/10.6084/m9.figshare.23641164.v2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 511 | 113 | 33 | 657 | 42 | 30 | 31 |
- HTML: 511
- PDF: 113
- XML: 33
- Total: 657
- Supplement: 42
- BibTeX: 30
- EndNote: 31
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1