the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Nov 2023
A global behavioural model of human fire use and management: WHAM! v1.0
Abstract. Fire is an integral ecosystem process and a major natural source of vegetation disturbance globally. Yet at the same time, humans use and manage fire in diverse ways and for a huge range of purposes. Therefore, it is perhaps unsurprising that a central finding of the first Fire Model Intercomparison Project was simplistic representation of humans is a substantial shortcoming in the fire modules of dynamic global vegetation models (DGVMs). In response to this challenge, we present a novel, global geospatial model that seeks to capture the diversity of human-fire interactions. Empirically-grounded with a global database of anthropogenic fire impacts, WHAM! (the Wildfire Human Agency Model) represents the underlying behavioural and land system drivers of human approaches to fire management and their impact on fire regimes. WHAM! is designed to be coupled with DGVMs (JULES-INFERNO in the current instance), such that human and biophysical drivers of fire on Earth, and their interactions, can be captured in process-based models for the first time. Initial outputs from WHAM! presented here are in line with previous evidence suggesting managed anthropogenic fire use is decreasing globally, and point to land use intensification as the underlying reason for this phenomenon.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1993 KB)
-
Supplement
(994 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1993 KB) - Metadata XML
-
Supplement
(994 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2023-2162', carolina ojeda leal, 29 Nov 2023
First of all, the acronym WHAM is an excellent catchphrase, congrats on that. This paper presents an updated version of DGVM to understand human-fire dynamics globally, which is a valuable source of future fire trends. I suggest adding the most recent paper of Fischer et al. (https://doi.org/10.1016/j.crm.2023.100573) and the paper of Smith et al. (https://doi.org/10.1093/biosci/biv182) to discuss the difference between the terms proposed by those authors: Global Adaptation Mapping Initiative (GAMI) fire-human adaptation database and firescapes.
Furthermore, I wonder what is the difference between the land-fire systems (LFS) proposed here to classify the anthropogenic fire regimes (AFRs) with the categories proposed by Blondel in his paper (https://doi.org/10.1007/s10745-006-9030-4) or traditional land use land cover obtained with GIS used in other articles (https://doi.org/10.5194/nhess-21-3663-2021).
Also, please add in line 525 another explanation of land use intensification in South America, it is not just clearing the land for cattle. For example, in Chile, we have increased over the decades the available land for pyrophitic species (like pine and eucalyptus) and urban sprawl.
Lastly, the paper was written in a highly technical language that was hard to read, and therefore, WHAM will be hard to replicate for scientists in other countries.
Citation: https://doi.org/10.5194/egusphere-2023-2162-CC1 -
CC2: 'Comment on egusphere-2023-2162', Sam Rabin, 08 Feb 2024
I read this manuscript with great interest, as I'm pondering how to include management fires in a DGVM myself. DAFI and WHAM! represent huge leaps forward in this area. I say this because I don't want my questions and comments to come off as critical for no reason! Rather, I have so many because I'm so interested in this subject and think WHAM! is really important.
Questions
- Lines 193-196: What exactly is the difference between tendency and extent? How were they separately parameterized?- Line 305: What "initial model outputs" are being referred to here?
- Lines 323-324: It might be worth mentioning here that fire *control* applies only to managed fires whereas fire *suppression* applies only to unmanaged fires. (This is clear from the very helpful Fig. 1, but it could use reiterating.)
- Lines 359-360: This says that "evaluation of model outputs here focuses on managed fire only," but Sects. 3.1.2 and 3.1.3 look at, respectively, unmanaged fire and its suppression.
- Lines 382-383: Isn't fire return interval typically considered just the inverse of annual burned fraction? In which case this wouldn't be an independent evaluation.
- Lines 485-487: Does this mean that "extensive livestock farming" occurs outside the LUH2 "pasture" area? If not, how does land cover affect managed fire used there?
- Fig. 11:
- Subplot A shows "crop" (presumably residue burning + field prep) whereas subplot B shows only crop residue burning. Why?
- What are the individual dots in subplot B? Each continent in each year? If so, what year range? and does subplot A also include individual years?
- Caption says that "increased HDI consistently leads to decreased pasture fire," but this isn't true for Africa—a pretty important continent when it comes to pasture burning! (Similar issue with lines 495-496.)
- Lines 481-482: What do you mean by "land use intensification"? Same for lines 525-531. LU90 doesn't seem able to distinguish between extensification (which is usually defined as increasing area of a land use type) vs. intensification (increasing management inputs/technology on existing cropland/pasture/etc.).
- ... Actually, I think I may have misinterpreted LU90. Typically when I read "land use," it's in the context of land use *area*, but I think that's included in LC90 instead. (Typically when I read "land cover" it refers to the areas of different types of vegetation, regardless of whether e.g. a hectare of grassland is "natural" or grazed by livestock.) I think a more complete explanation of what you mean by "land use" vs. "land cover" is warranted; note that they seem to be conflated in Table 3 Column 2. I also suggest changing the name of the LU90 experiment to something like SE90 (socioeconomic conditions 1990) instead.
- Still, the LU90 counterfactual doesn't seem able to distinguish between what I understand as "land use intensification" vs., say, increasing regulatory control of burning.
- How were the free parameters for vegetation constraint and AFR effect parameterized?
- Surprising trends in Africa:
- Lines 411-412: How does the modeled increase in Africa pasture burning square with Andela et al. (2014)?
- Fig. 9: Same question for crop fires—or are they even really considered in Andela et al. (2014)?Comments
- Lines 95-96 say that "the parameterisation of WHAM! presented in the main text takes relevant biophysical input variables from JULES model outputs," and Table 3 includes what seem to be JULES outputs from Best et al. (2011) and Clark et al. (2011) (although citations for those are missing from the reference list, so I can't be sure). But then that seems to be contradicted in some places. I think you're saying that the presented results are with WHAM! "offline" (or as you say, standalone), whereas in the future you hope to run JULES and INFERNO together at the same time. I might be getting confused because I consider what you have here to be a "one-way" coupling, whereas you hope to move to a "two-way" coupling. See Robinson et al. (2018). Modelling feedbacks between human and natural processes in the land system. Earth System Dynamics, 9(2), 895–914. doi: 10.5194/esd-9-895-2018. Anyway, here's where I got confused:
- Lines 334-335 (Table 6 caption): "These initial values will be updated during calibration of the planned coupled model with JULES-INFERNO."
- Lines 359-360 say that the coupling hasn't happened yet.
- Lines 448-449: "The relationship of suppression intensity to numbers of unmanaged fires will be determined through the planned coupling with JULES-INFENRO." (Note "INFENRO" typo, and that this sentence is confusing anyway—not sure what it's trying to say.)
- Table 3: All sources are missing from reference list except for Hurtt et al. (2020).
- Lines 401-407: It seems good that WHAM!'s burned area is almost always less than the GFED5-minus-GFED4 difference, since not all of that difference will be due to managed fires. (Although some managed fires will have been captured in GFED4 already.)
- Fig. 4: Interpretation of this figure seems limited to evaluating regional trends, so it might make more sense to just show the 1990-2014 difference rather than separate 1990 and 2014 maps. However, since some inputs to WHAM! (namely potential evapotranspiration and ecosystem NPP) can exhibit strong interannual variation, it might be more appropriate to map either (a) the difference between means of 5-year periods or (b) trends determined from a linear regression in each gridcell.
- Lines 439-441:
- I don't think Fig. 6b is informative as to the clustering of unmanaged fires in the WUI; the spatial resolution is just too coarse.
- How would a DGVM help determine industrialized firefighting effort? Isn't that what WHAM! is supposed to do?
- It would be helpful to briefly mention what is needed from a DGVM (could use JULES-INFERNO as an example) to translate from number of fires to burned area.
- Fig. 7: Same comments as for Fig. 4.
- Lines 460-461 mention global trends and refer to Fig. 9, but that figure has no "Global" panel.
- Fig. 10b: The caption should probably read "difference" instead of "change," the latter of which implies a change over time—especially when juxtaposed with 10a which does show change over time.
- Lines 496-498: Asia and South America don't seem to have any dots around HDI 0.6.
- I think it would be easier on the reader to have "performance evaluation" type results (lines 401-407/Fig. 3b, Sect. 3.2) in a single section. This should probably be the first section of the Results, before scientific evaluation begins. So, e.g., Sect. 3.1.1 would be comparison against GFED4-5 difference, Sect. 3.1.2 would be comparison against GFED5 crop burned area.
- It would be really helpful to have figures in the Supplement illustrating various parts of the model. E.g.:
- Geographic distribution of AFRs, AFTs, and LUSs.
- It would be great to have a section on characteristics of DGVMs that are amenable to use of WHAM! E.g., "written in Python/able to call out to Python" or "calculate fire annually".
- Higher-resolution figures would be nice, as the text in e.g. Fig. 3 looks funky.Typos etc.:
- Line 364: "Teckentrup et al, (2018) pearson's" should be "Teckentrup et al. (2018), Pearson's"
- Line 429 (Fig. 5 caption): Colon is red for some reason.
- Line 566: Comma should be deleted.
- Supplement A, Sect. 1: "...firstly to WHAM! more readily transferrable..."Citation: https://doi.org/10.5194/egusphere-2023-2162-CC2 -
RC1: 'Comment on egusphere-2023-2162', carolina ojeda leal, 22 Feb 2024
First of all, the acronym WHAM is an excellent catchphrase, congrats on that. This paper presents an updated version of DGVM to understand human-fire dynamics globally, which is a valuable source of future fire trends. I suggest adding the most recent paper of Fischer et al. (https://doi.org/10.1016/j.crm.2023.100573) and the paper of Smith et al. (https://doi.org/10.1093/biosci/biv182) to discuss the difference between the terms proposed by those authors: Global Adaptation Mapping Initiative (GAMI) fire-human adaptation database and firescapes.
Furthermore, I wonder what is the difference between the land-fire systems (LFS) proposed here to classify the anthropogenic fire regimes (AFRs) with the categories proposed by Blondel in his paper (https://doi.org/10.1007/s10745-006-9030-4) or traditional land use land cover obtained with GIS used in other articles (https://doi.org/10.5194/nhess-21-3663-2021).
Also, please add in line 525 another explanation of land use intensification in South America, it is not just clearing the land for cattle. For example, in Chile, we have increased over the decades the available land for pyrophitic species (like pine and eucalyptus) and urban sprawl.
Lastly, the paper was written in a highly technical language that was hard to read, and therefore, WHAM will be hard to replicate for scientists in other countries.
Citation: https://doi.org/10.5194/egusphere-2023-2162-RC1 -
RC2: 'Comment on egusphere-2023-2162', Sam Rabin, 23 Feb 2024
I read this manuscript with great interest, as I'm pondering how to include management fires in a DGVM myself. DAFI and WHAM! represent huge leaps forward in this area. I say this because I don't want my questions and comments to come off as critical for no reason! Rather, I have so many because I'm so interested in this subject and think WHAM! is really important.
Questions
- Lines 193-196: What exactly is the difference between tendency and extent? How were they separately parameterized?- Line 305: What "initial model outputs" are being referred to here?
- Lines 323-324: It might be worth mentioning here that fire *control* applies only to managed fires whereas fire *suppression* applies only to unmanaged fires. (This is clear from the very helpful Fig. 1, but it could use reiterating.)
- Lines 359-360: This says that "evaluation of model outputs here focuses on managed fire only," but Sects. 3.1.2 and 3.1.3 look at, respectively, unmanaged fire and its suppression.
- Lines 382-383: Isn't fire return interval typically considered just the inverse of annual burned fraction? In which case this wouldn't be an independent evaluation.
- Lines 485-487: Does this mean that "extensive livestock farming" occurs outside the LUH2 "pasture" area? If not, how does land cover affect managed fire used there?
- Fig. 11:
- Subplot A shows "crop" (presumably residue burning + field prep) whereas subplot B shows only crop residue burning. Why?
- What are the individual dots in subplot B? Each continent in each year? If so, what year range? and does subplot A also include individual years?
- Caption says that "increased HDI consistently leads to decreased pasture fire," but this isn't true for Africa—a pretty important continent when it comes to pasture burning! (Similar issue with lines 495-496.)
- Lines 481-482: What do you mean by "land use intensification"? Same for lines 525-531. LU90 doesn't seem able to distinguish between extensification (which is usually defined as increasing area of a land use type) vs. intensification (increasing management inputs/technology on existing cropland/pasture/etc.).
- ... Actually, I think I may have misinterpreted LU90. Typically when I read "land use," it's in the context of land use *area*, but I think that's included in LC90 instead. (Typically when I read "land cover" it refers to the areas of different types of vegetation, regardless of whether e.g. a hectare of grassland is "natural" or grazed by livestock.) I think a more complete explanation of what you mean by "land use" vs. "land cover" is warranted; note that they seem to be conflated in Table 3 Column 2. I also suggest changing the name of the LU90 experiment to something like SE90 (socioeconomic conditions 1990) instead.
- Still, the LU90 counterfactual doesn't seem able to distinguish between what I understand as "land use intensification" vs., say, increasing regulatory control of burning.
- How were the free parameters for vegetation constraint and AFR effect parameterized?
- Surprising trends in Africa:
- Lines 411-412: How does the modeled increase in Africa pasture burning square with Andela et al. (2014)?
- Fig. 9: Same question for crop fires—or are they even really considered in Andela et al. (2014)?Comments
- Lines 95-96 say that "the parameterisation of WHAM! presented in the main text takes relevant biophysical input variables from JULES model outputs," and Table 3 includes what seem to be JULES outputs from Best et al. (2011) and Clark et al. (2011) (although citations for those are missing from the reference list, so I can't be sure). But then that seems to be contradicted in some places. I think you're saying that the presented results are with WHAM! "offline" (or as you say, standalone), whereas in the future you hope to run JULES and INFERNO together at the same time. I might be getting confused because I consider what you have here to be a "one-way" coupling, whereas you hope to move to a "two-way" coupling. See Robinson et al. (2018). Modelling feedbacks between human and natural processes in the land system. Earth System Dynamics, 9(2), 895–914. doi: 10.5194/esd-9-895-2018. Anyway, here's where I got confused:
- Lines 334-335 (Table 6 caption): "These initial values will be updated during calibration of the planned coupled model with JULES-INFERNO."
- Lines 359-360 say that the coupling hasn't happened yet.
- Lines 448-449: "The relationship of suppression intensity to numbers of unmanaged fires will be determined through the planned coupling with JULES-INFENRO." (Note "INFENRO" typo, and that this sentence is confusing anyway—not sure what it's trying to say.)
- Table 3: All sources are missing from reference list except for Hurtt et al. (2020).
- Lines 401-407: It seems good that WHAM!'s burned area is almost always less than the GFED5-minus-GFED4 difference, since not all of that difference will be due to managed fires. (Although some managed fires will have been captured in GFED4 already.)
- Fig. 4: Interpretation of this figure seems limited to evaluating regional trends, so it might make more sense to just show the 1990-2014 difference rather than separate 1990 and 2014 maps. However, since some inputs to WHAM! (namely potential evapotranspiration and ecosystem NPP) can exhibit strong interannual variation, it might be more appropriate to map either (a) the difference between means of 5-year periods or (b) trends determined from a linear regression in each gridcell.
- Lines 439-441:
- I don't think Fig. 6b is informative as to the clustering of unmanaged fires in the WUI; the spatial resolution is just too coarse.
- How would a DGVM help determine industrialized firefighting effort? Isn't that what WHAM! is supposed to do?
- It would be helpful to briefly mention what is needed from a DGVM (could use JULES-INFERNO as an example) to translate from number of fires to burned area.
- Fig. 7: Same comments as for Fig. 4.
- Lines 460-461 mention global trends and refer to Fig. 9, but that figure has no "Global" panel.
- Fig. 10b: The caption should probably read "difference" instead of "change," the latter of which implies a change over time—especially when juxtaposed with 10a which does show change over time.
- Lines 496-498: Asia and South America don't seem to have any dots around HDI 0.6.
- I think it would be easier on the reader to have "performance evaluation" type results (lines 401-407/Fig. 3b, Sect. 3.2) in a single section. This should probably be the first section of the Results, before scientific evaluation begins. So, e.g., Sect. 3.1.1 would be comparison against GFED4-5 difference, Sect. 3.1.2 would be comparison against GFED5 crop burned area.
- It would be really helpful to have figures in the Supplement illustrating various parts of the model. E.g.:
- Geographic distribution of AFRs, AFTs, and LUSs.
- It would be great to have a section on characteristics of DGVMs that are amenable to use of WHAM! E.g., "written in Python/able to call out to Python" or "calculate fire annually".
- Higher-resolution figures would be nice, as the text in e.g. Fig. 3 looks funky.Typos etc.:
- Line 364: "Teckentrup et al, (2018) pearson's" should be "Teckentrup et al. (2018), Pearson's"
- Line 429 (Fig. 5 caption): Colon is red for some reason.
- Line 566: Comma should be deleted.
- Supplement A, Sect. 1: "...firstly to WHAM! more readily transferrable..."ReplyCitation: https://doi.org/10.5194/egusphere-2023-2162-RC2 - AC1: 'Response to reviewers', Oliver Perkins, 22 Mar 2024
Interactive discussion
Status: closed
-
CC1: 'Comment on egusphere-2023-2162', carolina ojeda leal, 29 Nov 2023
First of all, the acronym WHAM is an excellent catchphrase, congrats on that. This paper presents an updated version of DGVM to understand human-fire dynamics globally, which is a valuable source of future fire trends. I suggest adding the most recent paper of Fischer et al. (https://doi.org/10.1016/j.crm.2023.100573) and the paper of Smith et al. (https://doi.org/10.1093/biosci/biv182) to discuss the difference between the terms proposed by those authors: Global Adaptation Mapping Initiative (GAMI) fire-human adaptation database and firescapes.
Furthermore, I wonder what is the difference between the land-fire systems (LFS) proposed here to classify the anthropogenic fire regimes (AFRs) with the categories proposed by Blondel in his paper (https://doi.org/10.1007/s10745-006-9030-4) or traditional land use land cover obtained with GIS used in other articles (https://doi.org/10.5194/nhess-21-3663-2021).
Also, please add in line 525 another explanation of land use intensification in South America, it is not just clearing the land for cattle. For example, in Chile, we have increased over the decades the available land for pyrophitic species (like pine and eucalyptus) and urban sprawl.
Lastly, the paper was written in a highly technical language that was hard to read, and therefore, WHAM will be hard to replicate for scientists in other countries.
Citation: https://doi.org/10.5194/egusphere-2023-2162-CC1 -
CC2: 'Comment on egusphere-2023-2162', Sam Rabin, 08 Feb 2024
I read this manuscript with great interest, as I'm pondering how to include management fires in a DGVM myself. DAFI and WHAM! represent huge leaps forward in this area. I say this because I don't want my questions and comments to come off as critical for no reason! Rather, I have so many because I'm so interested in this subject and think WHAM! is really important.
Questions
- Lines 193-196: What exactly is the difference between tendency and extent? How were they separately parameterized?- Line 305: What "initial model outputs" are being referred to here?
- Lines 323-324: It might be worth mentioning here that fire *control* applies only to managed fires whereas fire *suppression* applies only to unmanaged fires. (This is clear from the very helpful Fig. 1, but it could use reiterating.)
- Lines 359-360: This says that "evaluation of model outputs here focuses on managed fire only," but Sects. 3.1.2 and 3.1.3 look at, respectively, unmanaged fire and its suppression.
- Lines 382-383: Isn't fire return interval typically considered just the inverse of annual burned fraction? In which case this wouldn't be an independent evaluation.
- Lines 485-487: Does this mean that "extensive livestock farming" occurs outside the LUH2 "pasture" area? If not, how does land cover affect managed fire used there?
- Fig. 11:
- Subplot A shows "crop" (presumably residue burning + field prep) whereas subplot B shows only crop residue burning. Why?
- What are the individual dots in subplot B? Each continent in each year? If so, what year range? and does subplot A also include individual years?
- Caption says that "increased HDI consistently leads to decreased pasture fire," but this isn't true for Africa—a pretty important continent when it comes to pasture burning! (Similar issue with lines 495-496.)
- Lines 481-482: What do you mean by "land use intensification"? Same for lines 525-531. LU90 doesn't seem able to distinguish between extensification (which is usually defined as increasing area of a land use type) vs. intensification (increasing management inputs/technology on existing cropland/pasture/etc.).
- ... Actually, I think I may have misinterpreted LU90. Typically when I read "land use," it's in the context of land use *area*, but I think that's included in LC90 instead. (Typically when I read "land cover" it refers to the areas of different types of vegetation, regardless of whether e.g. a hectare of grassland is "natural" or grazed by livestock.) I think a more complete explanation of what you mean by "land use" vs. "land cover" is warranted; note that they seem to be conflated in Table 3 Column 2. I also suggest changing the name of the LU90 experiment to something like SE90 (socioeconomic conditions 1990) instead.
- Still, the LU90 counterfactual doesn't seem able to distinguish between what I understand as "land use intensification" vs., say, increasing regulatory control of burning.
- How were the free parameters for vegetation constraint and AFR effect parameterized?
- Surprising trends in Africa:
- Lines 411-412: How does the modeled increase in Africa pasture burning square with Andela et al. (2014)?
- Fig. 9: Same question for crop fires—or are they even really considered in Andela et al. (2014)?Comments
- Lines 95-96 say that "the parameterisation of WHAM! presented in the main text takes relevant biophysical input variables from JULES model outputs," and Table 3 includes what seem to be JULES outputs from Best et al. (2011) and Clark et al. (2011) (although citations for those are missing from the reference list, so I can't be sure). But then that seems to be contradicted in some places. I think you're saying that the presented results are with WHAM! "offline" (or as you say, standalone), whereas in the future you hope to run JULES and INFERNO together at the same time. I might be getting confused because I consider what you have here to be a "one-way" coupling, whereas you hope to move to a "two-way" coupling. See Robinson et al. (2018). Modelling feedbacks between human and natural processes in the land system. Earth System Dynamics, 9(2), 895–914. doi: 10.5194/esd-9-895-2018. Anyway, here's where I got confused:
- Lines 334-335 (Table 6 caption): "These initial values will be updated during calibration of the planned coupled model with JULES-INFERNO."
- Lines 359-360 say that the coupling hasn't happened yet.
- Lines 448-449: "The relationship of suppression intensity to numbers of unmanaged fires will be determined through the planned coupling with JULES-INFENRO." (Note "INFENRO" typo, and that this sentence is confusing anyway—not sure what it's trying to say.)
- Table 3: All sources are missing from reference list except for Hurtt et al. (2020).
- Lines 401-407: It seems good that WHAM!'s burned area is almost always less than the GFED5-minus-GFED4 difference, since not all of that difference will be due to managed fires. (Although some managed fires will have been captured in GFED4 already.)
- Fig. 4: Interpretation of this figure seems limited to evaluating regional trends, so it might make more sense to just show the 1990-2014 difference rather than separate 1990 and 2014 maps. However, since some inputs to WHAM! (namely potential evapotranspiration and ecosystem NPP) can exhibit strong interannual variation, it might be more appropriate to map either (a) the difference between means of 5-year periods or (b) trends determined from a linear regression in each gridcell.
- Lines 439-441:
- I don't think Fig. 6b is informative as to the clustering of unmanaged fires in the WUI; the spatial resolution is just too coarse.
- How would a DGVM help determine industrialized firefighting effort? Isn't that what WHAM! is supposed to do?
- It would be helpful to briefly mention what is needed from a DGVM (could use JULES-INFERNO as an example) to translate from number of fires to burned area.
- Fig. 7: Same comments as for Fig. 4.
- Lines 460-461 mention global trends and refer to Fig. 9, but that figure has no "Global" panel.
- Fig. 10b: The caption should probably read "difference" instead of "change," the latter of which implies a change over time—especially when juxtaposed with 10a which does show change over time.
- Lines 496-498: Asia and South America don't seem to have any dots around HDI 0.6.
- I think it would be easier on the reader to have "performance evaluation" type results (lines 401-407/Fig. 3b, Sect. 3.2) in a single section. This should probably be the first section of the Results, before scientific evaluation begins. So, e.g., Sect. 3.1.1 would be comparison against GFED4-5 difference, Sect. 3.1.2 would be comparison against GFED5 crop burned area.
- It would be really helpful to have figures in the Supplement illustrating various parts of the model. E.g.:
- Geographic distribution of AFRs, AFTs, and LUSs.
- It would be great to have a section on characteristics of DGVMs that are amenable to use of WHAM! E.g., "written in Python/able to call out to Python" or "calculate fire annually".
- Higher-resolution figures would be nice, as the text in e.g. Fig. 3 looks funky.Typos etc.:
- Line 364: "Teckentrup et al, (2018) pearson's" should be "Teckentrup et al. (2018), Pearson's"
- Line 429 (Fig. 5 caption): Colon is red for some reason.
- Line 566: Comma should be deleted.
- Supplement A, Sect. 1: "...firstly to WHAM! more readily transferrable..."Citation: https://doi.org/10.5194/egusphere-2023-2162-CC2 -
RC1: 'Comment on egusphere-2023-2162', carolina ojeda leal, 22 Feb 2024
First of all, the acronym WHAM is an excellent catchphrase, congrats on that. This paper presents an updated version of DGVM to understand human-fire dynamics globally, which is a valuable source of future fire trends. I suggest adding the most recent paper of Fischer et al. (https://doi.org/10.1016/j.crm.2023.100573) and the paper of Smith et al. (https://doi.org/10.1093/biosci/biv182) to discuss the difference between the terms proposed by those authors: Global Adaptation Mapping Initiative (GAMI) fire-human adaptation database and firescapes.
Furthermore, I wonder what is the difference between the land-fire systems (LFS) proposed here to classify the anthropogenic fire regimes (AFRs) with the categories proposed by Blondel in his paper (https://doi.org/10.1007/s10745-006-9030-4) or traditional land use land cover obtained with GIS used in other articles (https://doi.org/10.5194/nhess-21-3663-2021).
Also, please add in line 525 another explanation of land use intensification in South America, it is not just clearing the land for cattle. For example, in Chile, we have increased over the decades the available land for pyrophitic species (like pine and eucalyptus) and urban sprawl.
Lastly, the paper was written in a highly technical language that was hard to read, and therefore, WHAM will be hard to replicate for scientists in other countries.
Citation: https://doi.org/10.5194/egusphere-2023-2162-RC1 -
RC2: 'Comment on egusphere-2023-2162', Sam Rabin, 23 Feb 2024
I read this manuscript with great interest, as I'm pondering how to include management fires in a DGVM myself. DAFI and WHAM! represent huge leaps forward in this area. I say this because I don't want my questions and comments to come off as critical for no reason! Rather, I have so many because I'm so interested in this subject and think WHAM! is really important.
Questions
- Lines 193-196: What exactly is the difference between tendency and extent? How were they separately parameterized?- Line 305: What "initial model outputs" are being referred to here?
- Lines 323-324: It might be worth mentioning here that fire *control* applies only to managed fires whereas fire *suppression* applies only to unmanaged fires. (This is clear from the very helpful Fig. 1, but it could use reiterating.)
- Lines 359-360: This says that "evaluation of model outputs here focuses on managed fire only," but Sects. 3.1.2 and 3.1.3 look at, respectively, unmanaged fire and its suppression.
- Lines 382-383: Isn't fire return interval typically considered just the inverse of annual burned fraction? In which case this wouldn't be an independent evaluation.
- Lines 485-487: Does this mean that "extensive livestock farming" occurs outside the LUH2 "pasture" area? If not, how does land cover affect managed fire used there?
- Fig. 11:
- Subplot A shows "crop" (presumably residue burning + field prep) whereas subplot B shows only crop residue burning. Why?
- What are the individual dots in subplot B? Each continent in each year? If so, what year range? and does subplot A also include individual years?
- Caption says that "increased HDI consistently leads to decreased pasture fire," but this isn't true for Africa—a pretty important continent when it comes to pasture burning! (Similar issue with lines 495-496.)
- Lines 481-482: What do you mean by "land use intensification"? Same for lines 525-531. LU90 doesn't seem able to distinguish between extensification (which is usually defined as increasing area of a land use type) vs. intensification (increasing management inputs/technology on existing cropland/pasture/etc.).
- ... Actually, I think I may have misinterpreted LU90. Typically when I read "land use," it's in the context of land use *area*, but I think that's included in LC90 instead. (Typically when I read "land cover" it refers to the areas of different types of vegetation, regardless of whether e.g. a hectare of grassland is "natural" or grazed by livestock.) I think a more complete explanation of what you mean by "land use" vs. "land cover" is warranted; note that they seem to be conflated in Table 3 Column 2. I also suggest changing the name of the LU90 experiment to something like SE90 (socioeconomic conditions 1990) instead.
- Still, the LU90 counterfactual doesn't seem able to distinguish between what I understand as "land use intensification" vs., say, increasing regulatory control of burning.
- How were the free parameters for vegetation constraint and AFR effect parameterized?
- Surprising trends in Africa:
- Lines 411-412: How does the modeled increase in Africa pasture burning square with Andela et al. (2014)?
- Fig. 9: Same question for crop fires—or are they even really considered in Andela et al. (2014)?Comments
- Lines 95-96 say that "the parameterisation of WHAM! presented in the main text takes relevant biophysical input variables from JULES model outputs," and Table 3 includes what seem to be JULES outputs from Best et al. (2011) and Clark et al. (2011) (although citations for those are missing from the reference list, so I can't be sure). But then that seems to be contradicted in some places. I think you're saying that the presented results are with WHAM! "offline" (or as you say, standalone), whereas in the future you hope to run JULES and INFERNO together at the same time. I might be getting confused because I consider what you have here to be a "one-way" coupling, whereas you hope to move to a "two-way" coupling. See Robinson et al. (2018). Modelling feedbacks between human and natural processes in the land system. Earth System Dynamics, 9(2), 895–914. doi: 10.5194/esd-9-895-2018. Anyway, here's where I got confused:
- Lines 334-335 (Table 6 caption): "These initial values will be updated during calibration of the planned coupled model with JULES-INFERNO."
- Lines 359-360 say that the coupling hasn't happened yet.
- Lines 448-449: "The relationship of suppression intensity to numbers of unmanaged fires will be determined through the planned coupling with JULES-INFENRO." (Note "INFENRO" typo, and that this sentence is confusing anyway—not sure what it's trying to say.)
- Table 3: All sources are missing from reference list except for Hurtt et al. (2020).
- Lines 401-407: It seems good that WHAM!'s burned area is almost always less than the GFED5-minus-GFED4 difference, since not all of that difference will be due to managed fires. (Although some managed fires will have been captured in GFED4 already.)
- Fig. 4: Interpretation of this figure seems limited to evaluating regional trends, so it might make more sense to just show the 1990-2014 difference rather than separate 1990 and 2014 maps. However, since some inputs to WHAM! (namely potential evapotranspiration and ecosystem NPP) can exhibit strong interannual variation, it might be more appropriate to map either (a) the difference between means of 5-year periods or (b) trends determined from a linear regression in each gridcell.
- Lines 439-441:
- I don't think Fig. 6b is informative as to the clustering of unmanaged fires in the WUI; the spatial resolution is just too coarse.
- How would a DGVM help determine industrialized firefighting effort? Isn't that what WHAM! is supposed to do?
- It would be helpful to briefly mention what is needed from a DGVM (could use JULES-INFERNO as an example) to translate from number of fires to burned area.
- Fig. 7: Same comments as for Fig. 4.
- Lines 460-461 mention global trends and refer to Fig. 9, but that figure has no "Global" panel.
- Fig. 10b: The caption should probably read "difference" instead of "change," the latter of which implies a change over time—especially when juxtaposed with 10a which does show change over time.
- Lines 496-498: Asia and South America don't seem to have any dots around HDI 0.6.
- I think it would be easier on the reader to have "performance evaluation" type results (lines 401-407/Fig. 3b, Sect. 3.2) in a single section. This should probably be the first section of the Results, before scientific evaluation begins. So, e.g., Sect. 3.1.1 would be comparison against GFED4-5 difference, Sect. 3.1.2 would be comparison against GFED5 crop burned area.
- It would be really helpful to have figures in the Supplement illustrating various parts of the model. E.g.:
- Geographic distribution of AFRs, AFTs, and LUSs.
- It would be great to have a section on characteristics of DGVMs that are amenable to use of WHAM! E.g., "written in Python/able to call out to Python" or "calculate fire annually".
- Higher-resolution figures would be nice, as the text in e.g. Fig. 3 looks funky.Typos etc.:
- Line 364: "Teckentrup et al, (2018) pearson's" should be "Teckentrup et al. (2018), Pearson's"
- Line 429 (Fig. 5 caption): Colon is red for some reason.
- Line 566: Comma should be deleted.
- Supplement A, Sect. 1: "...firstly to WHAM! more readily transferrable..."ReplyCitation: https://doi.org/10.5194/egusphere-2023-2162-RC2 - AC1: 'Response to reviewers', Oliver Perkins, 22 Mar 2024
Peer review completion
Journal article(s) based on this preprint
Data sets
Data for running WHAM! v1.0 Oli Perkins, James Millington, Matt Kasoar, Apostolos Voulgarakis, Cathy Smith, and Jay Mistry https://zenodo.org/record/8363979
Model code and software
Wildfire_Human_Agency_Model: v1.0 Oli Perkins, James Millington, Matt Kasoar, Apostolos Voulgarakis, Cathy Smith, and Jay Mistry https://zenodo.org/records/10142828
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 428 | 138 | 24 | 590 | 33 | 18 | 24 |
- HTML: 428
- PDF: 138
- XML: 24
- Total: 590
- Supplement: 33
- BibTeX: 18
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Matthew Kasoar
Apostolos Voulgarakis
Cathy Smith
Jay Mistry
James Millington
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1993 KB) - Metadata XML
-
Supplement
(994 KB) - BibTeX
- EndNote
- Final revised paper