The impacts of modelling prescribed vs. dynamic land cover in a high CO<sub>2</sub> future scenario &ndash; greening of the Arctic and Amazonian dieback

Kou-Giesbrecht, Sian; Arora, Vivek; Seiler, Christian; Wang, Libo

doi:https://doi.org/10.5194/egusphere-2023-2711

Preprints

https://doi.org/10.5194/egusphere-2023-2711

Preprints

29 Nov 2023

| 29 Nov 2023

The impacts of modelling prescribed vs. dynamic land cover in a high CO₂ future scenario – greening of the Arctic and Amazonian dieback

Sian Kou-Giesbrecht, Vivek Arora, Christian Seiler, and Libo Wang

Abstract. Terrestrial biosphere models are a key tool in investigating the role played by the land surface in the global climate system. However, few models simulate the geographic distribution of biomes dynamically, opting to prescribe them instead using remote sensing products. While prescribing land cover still allows for the simulation of the impacts of climate change on vegetation growth as well as the impacts of land use change, it prevents the simulation of climate change-driven biome shifts, with implications for projecting the future terrestrial carbon sink. Here, we isolate the impacts of prescribed vs. dynamic land cover implementations in a terrestrial biosphere model. We first introduce a framework for evaluating dynamic land cover (i.e., the spatial distribution of plant functional types across the land surface), which can be applied across terrestrial biosphere models alongside standard benchmarking of energy, water, and carbon cycle variables. After establishing confidence in simulated land cover, we then show that the simulated terrestrial carbon sink differs significantly between simulations with dynamic vs. prescribed land cover for a high CO₂ future scenario. This is because of important range shifts that are only simulated when dynamic land cover is implemented: tree expansion into the Arctic and Amazonian transition from forest to grassland. In particular, the projected net land-atmosphere CO₂ flux at the end of the 21^st century is twice as large in simulations with dynamic land cover than in simulations with prescribed land cover. Our results illustrate the importance of climate change-driven biome shifts for projecting the future terrestrial carbon sink.

Received: 17 Nov 2023 – Discussion started: 29 Nov 2023

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Sian Kou-Giesbrecht, Vivek Arora, Christian Seiler, and Libo Wang

Status: closed

RC1:
'Comment on egusphere-2023-2711', Anonymous Referee #1, 05 Feb 2024

The study investigates the impact of dynamically modeling land cover versus prescribing it in future high CO2 scenarios, focusing on Arctic greening and Amazonian dieback. It emphasizes the significance of simulating dynamic land cover to project future terrestrial carbon sinks accurately. Comparing simulations with prescribed and dynamic land cover, the study reveals substantial differences in predicting terrestrial carbon sinks, with dynamic simulations indicating a larger sink due to biome shifts such as tree expansion into the Arctic and the Amazon transitioning from forest to grassland. This highlights the crucial role of climate change-driven biome shifts in forecasting future carbon dynamics.
This issue reflects the reason why Dynamic Global Vegetation Models (DGVMs) were developed. There has been considerable research in this area, and the novelty claimed by the authors may not be as significant as suggested. However, there is value in accumulating such studies, and I do not oppose the publication of this manuscript. Below are some technical comments.
Minor issues:

(1) Line157 "Mortality when a PFT exists outside its bioclimatic limits"

Can PFTs exist outside of their bioclimatic limits?
(2) Line165 "m_bioclim,n"

There is no explanation for how to use this index.
(3) Line175 "c_n"

How is this variable calculated?
(4) Lines230-231 "We conducted different pre-industrial spin ups for each simulation"

How were they different?
(5) Figure 1a

This figure compares the temporal changes in the ratio of natural vegetation areas. Is it correct to understand that the ratio of natural vegetation area mentioned here excludes deserts, ice sheets, and croplands from the terrestrial area? It is necessary to define this term in the figure caption or the main text.

Citation: https://doi.org/10.5194/egusphere-2023-2711-RC1
- AC1:
  'Reply on RC1', Sian Kou-Giesbrecht, 06 Feb 2024
  We thank the reviewer for their review! We agree that this is a well-studied research area, and we hope that we have effectively described the background and previous research conducted. We think that we have applied a novel approach (in comparing the same model with prescribed vs. dynamic land cover thereby isolating the influence of dynamic land cover) and by introducing a new framework to evaluate model performance in simulating land cover itself to study this topic. As the reviewer points out, our results align with the results of previous studies, building on and contributing to the accumulation of such studies.
  
  Thank you for pointing out that this was unclear. “Mortality when a PFT exists outside its bioclimatic limits” ensures that PFTs do not venture outside of their bioclimatic envelopes. Each grid cell is “seeded” with a small fractional coverage of each natural PFT. Whether a given PFT persists is determined by its bioclimatic limits. If the PFT is within its bioclimatic limits, there is no mortality associated with bioclimatic limits () and the PFT can persist and expand. If the PFT is outside of its bioclimatic limits, the PFT is “killed off” by mortality associated with bioclimatic limits (). We will explain this better in the revised manuscript. The detailed explanations of the mortality calculations are given in the Supplementary Information. We will move important points to the main text and reference the Supplementary Information clearly in the revised manuscript.
  
  Thank you for pointing out that this variable was undefined. is the mortality when a PFT exists outside its bioclimatic limits for PFT n. We will explain this in the revised manuscript.
  
  This is explained in depth in the Supplementary Information. We will move important points to the main text and reference the Supplementary Information clearly in the revised manuscript.
  
  We conducted a different pre-industrial spin-up for simulations with prescribed landcover and for simulations with dynamic landcover. Within the simulations for prescribed landcover, we conducted different pre-industrial spin-ups for simulations with GLC2000 land cover and ESACCI land cover. These are described in Table 1. We will describe this clearly and reference Table 1 in the revised manuscript.
  
  Thank you for pointing out that this was unclear. This figure only shows natural vegetation area (i.e., it excludes croplands). Antarctica and Greenland are also excluded. We will clarify this in the revised manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2023-2711-AC1
RC2:
'Comment on egusphere-2023-2711', Anonymous Referee #2, 16 Apr 2024

General comments:
Kou Giesbrecht et al. utilize a dynamic land cover model to understand how changes in land cover can change the carbon sink in high CO2 future scenarios. The authors argue that implementing a dynamic land cover model changes the carbon fluxes where at the end of the century the net biome productivity is twice as large with the dynamic land cover than without it at the end of the century. While other studies have included dynamic land models, this research is still highly relevant and important to emphasize as many CMIP6 models still do not include dynamic vegetation and/or fires which can have impacts on the future land carbon sink as the authors have described in this paper. While this reviewer believes that this study is worth publishing, there are a few points this reviewer thinks would be important to address prior to the publication. These points of discussion include more discussions on the role of fire in the changing land cover, carbon dynamics as well as the changes in future carbon storage. There are also questions regarding the statistical significance of changes of vegetation/PFT which this reviewer thinks may be important to note in their study. This reviewer believes there should also be some clarifications in the methods regarding the year of data used from various products such as ESA CCI and the MODIS.
Specific comments:

1) When creating the bioclimatic index, the authors note on L171-2 that this assumes the current biome ranges are in equilibrium with the 1900-1920 climate. There have been many studies recently, particularly focused on the high latitudes in Alaska and Canada, showing that the land cover has changed substantially since 1985, as increasing burned areas from wildfire continue to change the land cover (Macander et al. 2022, Wang et al., 2019). Could the authors please elaborate/justify this assumption for the readers or have a small discussion around this issue.

Macander, M. J., Nelson, P. R., Nawrocki, T. W., Frost, G. V., Orndahl, K. M., Palm, E. C., Wells, A. F., & Goetz, S. J. (2022). Time-series maps reveal widespread change in plant functional type cover across Arctic and boreal Alaska and Yukon. Environmental Research Letters, 17(5), 054042. https://doi.org/10.1088/1748-9326/ac6965

Wang, J. A., Sulla-Menashe, D., Woodcock, C. E., Sonnentag, O., Keeling, R. F., & Friedl, M. A. (2020). Extensive land cover change across Arctic–Boreal Northwestern North America from disturbance and climate forcing. Global Change Biology, 26(2), 807–822. https://doi.org/10.1111/gcb.14804

2) This reviewer noticed that C4 grasses increases in the Arctic boreal region in Figure B5. These areas usually only contain C3 grasses as seen by Still et al. 2003, which the authors cite, and a recent paper by Luo et al. 2024. One question this reviewer has is if the authors believe the increases in C4 grasses is a reasonable change in PFTs for the Arctic boreal region. This leads to another question of if the area change in PFTs is significant within the model. This reviewer thinks it would be helpful to the readers for the authors to indicate areas of significant changes in their maps.

Luo, X., Zhou, H., Satriawan, T.W. et al. Mapping the global distribution of C4 vegetation using observations and optimality theory. Nat Commun 15, 1219 (2024). https://doi.org/10.1038/s41467-024-45606-3

3) The authors mention that in the discussion that changes L407-419 the role that warming climate has on the expansion of vegetation. However as mentioned in an earlier comment, fires can accelerate this expansion allowing shrubs/deciduous trees to expand into these areas. For example, fires can replace the needleleaf evergreen conifers with more deciduous vegetation such as shrubs and/or deciduous forests (Baltzer et al. 2021, Liu et al. 2022, Lucash et al. 2023, Weiss et al. 2023). Can the authors expand on the effect that fires have within the model to explain the change in distribution of PFTs since fires are one of the primary mechanisms of change in the Arctic-boreal region?

Baltzer, J. L., Day, N. J., Walker, X. J., Greene, D., Mack, M. C., Alexander, H. D., Arseneault, D., Barnes, J., Bergeron, Y., Boucher, Y., Bourgeau-Chavez, L., Brown, C. D., Carrière, S., Howard, B. K., Gauthier, S., Parisien, M.-A., Reid, K. A., Rogers, B. M., Roland, C., … Johnstone, J. F. (2021). Increasing fire and the decline of fire adapted black spruce in the boreal forest. Proceedings of the National Academy of Sciences, 118(45), e2024872118. https://doi.org/10.1073/pnas.2024872118
Liu, Y., Riley, W.J., Keenan, T.F. et al. Dispersal and fire limit Arctic shrub expansion. Nat Commun 13, 3843 (2022). https://doi.org/10.1038/s41467-022-31597-6
Lucash, M. S., Marshall, A. M., Weiss, S. A., McNabb, J. W., Nicolsky, D. J., Flerchinger, G. N., Link, T. E., Vogel, J. G., Scheller, R. M., Abramoff, R. Z., & Romanovsky, V. E. (2023). Burning trees in frozen soil: Simulating fire, vegetation, soil, and hydrology in the boreal forests of Alaska. Ecological Modelling, 481, 110367. https://doi.org/10.1016/j.ecolmodel.2023.110367
Weiss, S. A., Marshall, A. M., Hayes, K. R., Nicolsky, D. J., Buma, B., & Lucash, M. S. (2023). Future transitions from a conifer to a deciduous-dominated landscape are accelerated by greater wildfire activity and climate change in interior Alaska. Landscape Ecology. https://doi.org/10.1007/s10980-023-01733-8
4) This reviewer thinks that since the paper is describing the impacts of modeling prescribed vs. dynamic land cover, that the paper can benefit from authors discussing the changes in the future carbon storage and how changes in PFT change the overall carbon storage, expanding on the discussion of changes in productivity that is already included in the paper. The authors show that there is enhanced net biome productivity which would be anticipated with expansion of PFTs in high-latitudes. One question is if PFT conversions from one type to another, such as evergreen needleleaf to deciduous broadleaf, in high latitudes contribute to an enhancement of NBP within the model as well. How do the changes in NBP and PFT affect the distribution of carbon in the various pools from the beginning of the run if fires are combusting the vegetation/soils? There have been many studies showing an increase in primary productivity after fires from the flux tower level to remote sensing (Rocha and Shaver 2011, Coursolle et al. 2012, Kim et al. 2024), but a suppression of aboveground carbon sink (Wang et al. 2021). However, this can be counteracted with warming induced growth (Wang et al. 2023), as well as reduction in fire frequency with a change in PFT (Mack et al 2021). It would be interesting to expand on the discussion with the context of fire within the model.
Coursolle, C., Margolis, H. A., Giasson, M.-A., Bernier, P.-Y., Amiro, B. D., Arain, M. A., Barr, A. G., Black, T. A., Goulden, M. L., McCaughey, J. H., Chen, J. M., Dunn, A. L., Grant, R. F., & Lafleur, P. M. (2012). Influence of stand age on the magnitude and seasonality of carbon fluxes in Canadian forests. Agricultural and Forest Meteorology, 165, 136–148. https://doi.org/10.1016/j.agrformet.2012.06.011
Kim, J. E., Wang, J. A., Li, Y., Czimczik, C. I., & Randerson, J. T. (2024). Wildfire-induced increases in photosynthesis in boreal forest ecosystems of North America. Global Change Biology, 30(1), e17151. https://doi.org/10.1111/gcb.17151
Mack, M. C., Walker, X. J., Johnstone, J. F., Alexander, H. D., Melvin, A. M., Jean, M., & Miller, S. N. (2021). Carbon loss from boreal forest wildfires offset by increased dominance of deciduous trees. Science, 372(6539), 280–283. https://doi.org/10.1126/science.abf3903
Rocha, A. V., & Shaver, G. R. (2011). Burn severity influences postfire CO2 exchange in arctic tundra. Ecological Applications, 21(2), 477–489. https://doi.org/10.1890/10-0255.1
Wang, J., Taylor, A. R., & D’Orangeville, L. (2023). Warming-induced tree growth may help offset increasing disturbance across the Canadian boreal forest. Proceedings of the National Academy of Sciences, 120(2), e2212780120. https://doi.org/10.1073/pnas.2212780120

Wang, J. A., Baccini, A., Farina, M., Randerson, J. T., & Friedl, M. A. (2021). Disturbance suppresses the aboveground carbon sink in North American boreal forests. Nature Climate Change, 11(5), 435–441. https://doi.org/10.1038/s41558-021-01027-4
5) Have the authors compared the MODIS collection 6/6.1 to MODIS collection 5 to see if there are significant differences between the two datasets, and is there a specific reason the authors chose to use collection 5 to create the bioclimatic index instead of using collection 6?
6) How well do the dynamic models match observational trends in annual land cover during the MODIS/ ESA CCI during the overlap of the model and observational records?
Technical corrections/comments:
L197 Please provide citations/html sources for the GLC2000 product
L202 - Please provide citations/html sources for the ESA CCI product and which version you are using.
L202 - What years are utilized for the ESA CCI land cover product when setting up simulation 2 (S2)? Are each of the years for ESA CCI land cover utilized? Same for MODIS when creating the bioclimatic index.
L137-8 Typo, “When crop area decreases, natural vegetation area proportionally increase”

Citation: https://doi.org/10.5194/egusphere-2023-2711-RC2
- AC2:
  'Reply on RC2', Sian Kou-Giesbrecht, 01 May 2024
  We thank the reviewer for their review! We have addressed their individual points below:
  Thank you for drawing our attention to this issue. As the reviewer points out, we assumed that current biome ranges are in equilibrium with the 1900-1920 climate when parameterising bioclimatic limits. We made this assumption because of the lag that occurs between climate change and observed differences in established plant species ranges, especially in long-lived plant species such as trees (Corlett and Wescott 2013). In the revised manuscript, we will discuss the limitations of our approach. Namely that, as the reviewer points out, there have been shifts in land cover since 1900-1920. While our approach captures slow shifts due to climate change, it does not capture rapid shifts due to wildfire disturbance.
  
  Thank you for this keen observation and for pointing us to this recent paper. While high-latitude regions are currently dominated by C3 grasses, the increase in temperature that occurs in SSP585 could support the growth of C4 grasses previously not common at these latitudes. In alignment with both the crossover-temperature model and the optimality model which includes temperature in Luo et al. 2024, the temperature envelope occupied by C4 grasses shifts northwards to encompass high-latitude regions in SSP585. In the revised manuscript, we will include a plot of both temperature (and precipitation) for the present-day and the 2081-2100 average for SSP585 (in addition to temperature and precipitation change, which is shown in Figure B6). In the revised manuscript, we can also highlight areas of substantial change in our maps using a cut-off (e.g., > 10% difference).
  
  Thank you for drawing our attention to the role of fire. The reviewer is correct that fire plays an important role in ecosystem transitions. The calculation of net biome productivity (NBP) does include CO2 emissions from fires (Figure 7a). In the revised manuscript, we will explain this, and we will also include an explicit discussion of fire. Briefly, while in the prescribed land cover implementation, fire only reduces vegetation biomass, in the dynamic land cover implementation, fire both reduces vegetation biomass and creates bare ground that can be colonized by a new plant functional type (such as grasses or deciduous trees, which are favoured by fire weather conditions). Furthermore, the fire spread rate is greater for grasses than for trees. In the revised manuscript, we will explain these differences, and we will show differences in both change in burned area and fire CO2 emissions between prescribed vs. dynamic land cover implementations. We will also examine the differences in the land cover of specific PFTs (especially evergreen needleleaf and broadleaf deciduous in boreal regions) between prescribed vs. dynamic land cover implementations.
  
  Thank you for this suggestion. In the revised manuscript, we will examine C stocks in vegetation and soil. As described in our response to your previous point, we will also examine the role of fire and specific PFTs (especially evergreen needleleaf and broadleaf deciduous in boreal regions).
  
  We used the MODIS Collection 5 dataset because it had been previously used and processed for analyses. We briefly recalculated the bioclimatic limits with the MODIS Collection 6.1 dataset and there were only minor differences. This is a helpful suggestion and in future analyses, we will use the newest dataset.
  
  Thank you for this suggestion. While we agree that this analysis would yield interesting results, we think that this is beyond the scope of our paper, which we want to focus on future simulations of land surface model. While we do use the present-day average to evaluate model output (Figures 2 and 3), evaluating the annual trends would be a significant additional analysis.
  
  Thank you for catching the issues listed in the technical corrections/comments. We will fix these in the revised manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2023-2711-AC2

Status: closed

RC1:
'Comment on egusphere-2023-2711', Anonymous Referee #1, 05 Feb 2024

The study investigates the impact of dynamically modeling land cover versus prescribing it in future high CO2 scenarios, focusing on Arctic greening and Amazonian dieback. It emphasizes the significance of simulating dynamic land cover to project future terrestrial carbon sinks accurately. Comparing simulations with prescribed and dynamic land cover, the study reveals substantial differences in predicting terrestrial carbon sinks, with dynamic simulations indicating a larger sink due to biome shifts such as tree expansion into the Arctic and the Amazon transitioning from forest to grassland. This highlights the crucial role of climate change-driven biome shifts in forecasting future carbon dynamics.
This issue reflects the reason why Dynamic Global Vegetation Models (DGVMs) were developed. There has been considerable research in this area, and the novelty claimed by the authors may not be as significant as suggested. However, there is value in accumulating such studies, and I do not oppose the publication of this manuscript. Below are some technical comments.
Minor issues:

(1) Line157 "Mortality when a PFT exists outside its bioclimatic limits"

Can PFTs exist outside of their bioclimatic limits?
(2) Line165 "m_bioclim,n"

There is no explanation for how to use this index.
(3) Line175 "c_n"

How is this variable calculated?
(4) Lines230-231 "We conducted different pre-industrial spin ups for each simulation"

How were they different?
(5) Figure 1a

This figure compares the temporal changes in the ratio of natural vegetation areas. Is it correct to understand that the ratio of natural vegetation area mentioned here excludes deserts, ice sheets, and croplands from the terrestrial area? It is necessary to define this term in the figure caption or the main text.

Citation: https://doi.org/10.5194/egusphere-2023-2711-RC1
- AC1:
  'Reply on RC1', Sian Kou-Giesbrecht, 06 Feb 2024
  We thank the reviewer for their review! We agree that this is a well-studied research area, and we hope that we have effectively described the background and previous research conducted. We think that we have applied a novel approach (in comparing the same model with prescribed vs. dynamic land cover thereby isolating the influence of dynamic land cover) and by introducing a new framework to evaluate model performance in simulating land cover itself to study this topic. As the reviewer points out, our results align with the results of previous studies, building on and contributing to the accumulation of such studies.
  
  Thank you for pointing out that this was unclear. “Mortality when a PFT exists outside its bioclimatic limits” ensures that PFTs do not venture outside of their bioclimatic envelopes. Each grid cell is “seeded” with a small fractional coverage of each natural PFT. Whether a given PFT persists is determined by its bioclimatic limits. If the PFT is within its bioclimatic limits, there is no mortality associated with bioclimatic limits () and the PFT can persist and expand. If the PFT is outside of its bioclimatic limits, the PFT is “killed off” by mortality associated with bioclimatic limits (). We will explain this better in the revised manuscript. The detailed explanations of the mortality calculations are given in the Supplementary Information. We will move important points to the main text and reference the Supplementary Information clearly in the revised manuscript.
  
  Thank you for pointing out that this variable was undefined. is the mortality when a PFT exists outside its bioclimatic limits for PFT n. We will explain this in the revised manuscript.
  
  This is explained in depth in the Supplementary Information. We will move important points to the main text and reference the Supplementary Information clearly in the revised manuscript.
  
  We conducted a different pre-industrial spin-up for simulations with prescribed landcover and for simulations with dynamic landcover. Within the simulations for prescribed landcover, we conducted different pre-industrial spin-ups for simulations with GLC2000 land cover and ESACCI land cover. These are described in Table 1. We will describe this clearly and reference Table 1 in the revised manuscript.
  
  Thank you for pointing out that this was unclear. This figure only shows natural vegetation area (i.e., it excludes croplands). Antarctica and Greenland are also excluded. We will clarify this in the revised manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2023-2711-AC1
RC2:
'Comment on egusphere-2023-2711', Anonymous Referee #2, 16 Apr 2024

General comments:
Kou Giesbrecht et al. utilize a dynamic land cover model to understand how changes in land cover can change the carbon sink in high CO2 future scenarios. The authors argue that implementing a dynamic land cover model changes the carbon fluxes where at the end of the century the net biome productivity is twice as large with the dynamic land cover than without it at the end of the century. While other studies have included dynamic land models, this research is still highly relevant and important to emphasize as many CMIP6 models still do not include dynamic vegetation and/or fires which can have impacts on the future land carbon sink as the authors have described in this paper. While this reviewer believes that this study is worth publishing, there are a few points this reviewer thinks would be important to address prior to the publication. These points of discussion include more discussions on the role of fire in the changing land cover, carbon dynamics as well as the changes in future carbon storage. There are also questions regarding the statistical significance of changes of vegetation/PFT which this reviewer thinks may be important to note in their study. This reviewer believes there should also be some clarifications in the methods regarding the year of data used from various products such as ESA CCI and the MODIS.
Specific comments:

1) When creating the bioclimatic index, the authors note on L171-2 that this assumes the current biome ranges are in equilibrium with the 1900-1920 climate. There have been many studies recently, particularly focused on the high latitudes in Alaska and Canada, showing that the land cover has changed substantially since 1985, as increasing burned areas from wildfire continue to change the land cover (Macander et al. 2022, Wang et al., 2019). Could the authors please elaborate/justify this assumption for the readers or have a small discussion around this issue.

Macander, M. J., Nelson, P. R., Nawrocki, T. W., Frost, G. V., Orndahl, K. M., Palm, E. C., Wells, A. F., & Goetz, S. J. (2022). Time-series maps reveal widespread change in plant functional type cover across Arctic and boreal Alaska and Yukon. Environmental Research Letters, 17(5), 054042. https://doi.org/10.1088/1748-9326/ac6965

Wang, J. A., Sulla-Menashe, D., Woodcock, C. E., Sonnentag, O., Keeling, R. F., & Friedl, M. A. (2020). Extensive land cover change across Arctic–Boreal Northwestern North America from disturbance and climate forcing. Global Change Biology, 26(2), 807–822. https://doi.org/10.1111/gcb.14804

2) This reviewer noticed that C4 grasses increases in the Arctic boreal region in Figure B5. These areas usually only contain C3 grasses as seen by Still et al. 2003, which the authors cite, and a recent paper by Luo et al. 2024. One question this reviewer has is if the authors believe the increases in C4 grasses is a reasonable change in PFTs for the Arctic boreal region. This leads to another question of if the area change in PFTs is significant within the model. This reviewer thinks it would be helpful to the readers for the authors to indicate areas of significant changes in their maps.

Luo, X., Zhou, H., Satriawan, T.W. et al. Mapping the global distribution of C4 vegetation using observations and optimality theory. Nat Commun 15, 1219 (2024). https://doi.org/10.1038/s41467-024-45606-3

3) The authors mention that in the discussion that changes L407-419 the role that warming climate has on the expansion of vegetation. However as mentioned in an earlier comment, fires can accelerate this expansion allowing shrubs/deciduous trees to expand into these areas. For example, fires can replace the needleleaf evergreen conifers with more deciduous vegetation such as shrubs and/or deciduous forests (Baltzer et al. 2021, Liu et al. 2022, Lucash et al. 2023, Weiss et al. 2023). Can the authors expand on the effect that fires have within the model to explain the change in distribution of PFTs since fires are one of the primary mechanisms of change in the Arctic-boreal region?

Baltzer, J. L., Day, N. J., Walker, X. J., Greene, D., Mack, M. C., Alexander, H. D., Arseneault, D., Barnes, J., Bergeron, Y., Boucher, Y., Bourgeau-Chavez, L., Brown, C. D., Carrière, S., Howard, B. K., Gauthier, S., Parisien, M.-A., Reid, K. A., Rogers, B. M., Roland, C., … Johnstone, J. F. (2021). Increasing fire and the decline of fire adapted black spruce in the boreal forest. Proceedings of the National Academy of Sciences, 118(45), e2024872118. https://doi.org/10.1073/pnas.2024872118
Liu, Y., Riley, W.J., Keenan, T.F. et al. Dispersal and fire limit Arctic shrub expansion. Nat Commun 13, 3843 (2022). https://doi.org/10.1038/s41467-022-31597-6
Lucash, M. S., Marshall, A. M., Weiss, S. A., McNabb, J. W., Nicolsky, D. J., Flerchinger, G. N., Link, T. E., Vogel, J. G., Scheller, R. M., Abramoff, R. Z., & Romanovsky, V. E. (2023). Burning trees in frozen soil: Simulating fire, vegetation, soil, and hydrology in the boreal forests of Alaska. Ecological Modelling, 481, 110367. https://doi.org/10.1016/j.ecolmodel.2023.110367
Weiss, S. A., Marshall, A. M., Hayes, K. R., Nicolsky, D. J., Buma, B., & Lucash, M. S. (2023). Future transitions from a conifer to a deciduous-dominated landscape are accelerated by greater wildfire activity and climate change in interior Alaska. Landscape Ecology. https://doi.org/10.1007/s10980-023-01733-8
4) This reviewer thinks that since the paper is describing the impacts of modeling prescribed vs. dynamic land cover, that the paper can benefit from authors discussing the changes in the future carbon storage and how changes in PFT change the overall carbon storage, expanding on the discussion of changes in productivity that is already included in the paper. The authors show that there is enhanced net biome productivity which would be anticipated with expansion of PFTs in high-latitudes. One question is if PFT conversions from one type to another, such as evergreen needleleaf to deciduous broadleaf, in high latitudes contribute to an enhancement of NBP within the model as well. How do the changes in NBP and PFT affect the distribution of carbon in the various pools from the beginning of the run if fires are combusting the vegetation/soils? There have been many studies showing an increase in primary productivity after fires from the flux tower level to remote sensing (Rocha and Shaver 2011, Coursolle et al. 2012, Kim et al. 2024), but a suppression of aboveground carbon sink (Wang et al. 2021). However, this can be counteracted with warming induced growth (Wang et al. 2023), as well as reduction in fire frequency with a change in PFT (Mack et al 2021). It would be interesting to expand on the discussion with the context of fire within the model.
Coursolle, C., Margolis, H. A., Giasson, M.-A., Bernier, P.-Y., Amiro, B. D., Arain, M. A., Barr, A. G., Black, T. A., Goulden, M. L., McCaughey, J. H., Chen, J. M., Dunn, A. L., Grant, R. F., & Lafleur, P. M. (2012). Influence of stand age on the magnitude and seasonality of carbon fluxes in Canadian forests. Agricultural and Forest Meteorology, 165, 136–148. https://doi.org/10.1016/j.agrformet.2012.06.011
Kim, J. E., Wang, J. A., Li, Y., Czimczik, C. I., & Randerson, J. T. (2024). Wildfire-induced increases in photosynthesis in boreal forest ecosystems of North America. Global Change Biology, 30(1), e17151. https://doi.org/10.1111/gcb.17151
Mack, M. C., Walker, X. J., Johnstone, J. F., Alexander, H. D., Melvin, A. M., Jean, M., & Miller, S. N. (2021). Carbon loss from boreal forest wildfires offset by increased dominance of deciduous trees. Science, 372(6539), 280–283. https://doi.org/10.1126/science.abf3903
Rocha, A. V., & Shaver, G. R. (2011). Burn severity influences postfire CO2 exchange in arctic tundra. Ecological Applications, 21(2), 477–489. https://doi.org/10.1890/10-0255.1
Wang, J., Taylor, A. R., & D’Orangeville, L. (2023). Warming-induced tree growth may help offset increasing disturbance across the Canadian boreal forest. Proceedings of the National Academy of Sciences, 120(2), e2212780120. https://doi.org/10.1073/pnas.2212780120

Wang, J. A., Baccini, A., Farina, M., Randerson, J. T., & Friedl, M. A. (2021). Disturbance suppresses the aboveground carbon sink in North American boreal forests. Nature Climate Change, 11(5), 435–441. https://doi.org/10.1038/s41558-021-01027-4
5) Have the authors compared the MODIS collection 6/6.1 to MODIS collection 5 to see if there are significant differences between the two datasets, and is there a specific reason the authors chose to use collection 5 to create the bioclimatic index instead of using collection 6?
6) How well do the dynamic models match observational trends in annual land cover during the MODIS/ ESA CCI during the overlap of the model and observational records?
Technical corrections/comments:
L197 Please provide citations/html sources for the GLC2000 product
L202 - Please provide citations/html sources for the ESA CCI product and which version you are using.
L202 - What years are utilized for the ESA CCI land cover product when setting up simulation 2 (S2)? Are each of the years for ESA CCI land cover utilized? Same for MODIS when creating the bioclimatic index.
L137-8 Typo, “When crop area decreases, natural vegetation area proportionally increase”

Citation: https://doi.org/10.5194/egusphere-2023-2711-RC2
- AC2:
  'Reply on RC2', Sian Kou-Giesbrecht, 01 May 2024
  We thank the reviewer for their review! We have addressed their individual points below:
  Thank you for drawing our attention to this issue. As the reviewer points out, we assumed that current biome ranges are in equilibrium with the 1900-1920 climate when parameterising bioclimatic limits. We made this assumption because of the lag that occurs between climate change and observed differences in established plant species ranges, especially in long-lived plant species such as trees (Corlett and Wescott 2013). In the revised manuscript, we will discuss the limitations of our approach. Namely that, as the reviewer points out, there have been shifts in land cover since 1900-1920. While our approach captures slow shifts due to climate change, it does not capture rapid shifts due to wildfire disturbance.
  
  Thank you for this keen observation and for pointing us to this recent paper. While high-latitude regions are currently dominated by C3 grasses, the increase in temperature that occurs in SSP585 could support the growth of C4 grasses previously not common at these latitudes. In alignment with both the crossover-temperature model and the optimality model which includes temperature in Luo et al. 2024, the temperature envelope occupied by C4 grasses shifts northwards to encompass high-latitude regions in SSP585. In the revised manuscript, we will include a plot of both temperature (and precipitation) for the present-day and the 2081-2100 average for SSP585 (in addition to temperature and precipitation change, which is shown in Figure B6). In the revised manuscript, we can also highlight areas of substantial change in our maps using a cut-off (e.g., > 10% difference).
  
  Thank you for drawing our attention to the role of fire. The reviewer is correct that fire plays an important role in ecosystem transitions. The calculation of net biome productivity (NBP) does include CO2 emissions from fires (Figure 7a). In the revised manuscript, we will explain this, and we will also include an explicit discussion of fire. Briefly, while in the prescribed land cover implementation, fire only reduces vegetation biomass, in the dynamic land cover implementation, fire both reduces vegetation biomass and creates bare ground that can be colonized by a new plant functional type (such as grasses or deciduous trees, which are favoured by fire weather conditions). Furthermore, the fire spread rate is greater for grasses than for trees. In the revised manuscript, we will explain these differences, and we will show differences in both change in burned area and fire CO2 emissions between prescribed vs. dynamic land cover implementations. We will also examine the differences in the land cover of specific PFTs (especially evergreen needleleaf and broadleaf deciduous in boreal regions) between prescribed vs. dynamic land cover implementations.
  
  Thank you for this suggestion. In the revised manuscript, we will examine C stocks in vegetation and soil. As described in our response to your previous point, we will also examine the role of fire and specific PFTs (especially evergreen needleleaf and broadleaf deciduous in boreal regions).
  
  We used the MODIS Collection 5 dataset because it had been previously used and processed for analyses. We briefly recalculated the bioclimatic limits with the MODIS Collection 6.1 dataset and there were only minor differences. This is a helpful suggestion and in future analyses, we will use the newest dataset.
  
  Thank you for this suggestion. While we agree that this analysis would yield interesting results, we think that this is beyond the scope of our paper, which we want to focus on future simulations of land surface model. While we do use the present-day average to evaluate model output (Figures 2 and 3), evaluating the annual trends would be a significant additional analysis.
  
  Thank you for catching the issues listed in the technical corrections/comments. We will fix these in the revised manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2023-2711-AC2

Sian Kou-Giesbrecht, Vivek Arora, Christian Seiler, and Libo Wang

Viewed

Total article views: 416 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
316	72	28	416	24	22

HTML: 316
PDF: 72
XML: 28
Total: 416
BibTeX: 24
EndNote: 22

Views and downloads (calculated since 29 Nov 2023)

Month	HTML	PDF	XML	Total
Dec 2023	41	16	7	64
Jan 2024	12	7	3	22
Feb 2024	42	12	3	57
Mar 2024	24	9	0	33
Apr 2024	79	12	6	97
May 2024	59	10	4	73
Jun 2024	59	6	5	70

Cumulative views and downloads (calculated since 29 Nov 2023)

Month	HTML	PDF	XML	Total
Dec 2023	41	16	7	64
Jan 2024	12	7	3	22
Feb 2024	42	12	3	57
Mar 2024	24	9	0	33
Apr 2024	79	12	6	97
May 2024	59	10	4	73
Jun 2024	59	6	5	70

Viewed (geographical distribution)

Total article views: 417 (including HTML, PDF, and XML) Thereof 417 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 29 Jun 2024

Short summary

Terrestrial biosphere models can either prescribe the geographical distribution of biomes or simulate them dynamically, capturing climate change-driven biome shifts. We isolate and examine the differences between these different land cover implementations. We find that the simulated terrestrial carbon sink at the end of the 21st century is twice as large in simulations with dynamic land cover than in simulations with prescribed land cover due to important range shifts in the Arctic and Amazon.


Total:	0
HTML:	0
PDF:	0
XML:	0

The impacts of modelling prescribed vs. dynamic land cover in a high CO2 future scenario – greening of the Arctic and Amazonian dieback

Viewed

Viewed (geographical distribution)

The impacts of modelling prescribed vs. dynamic land cover in a high CO₂ future scenario – greening of the Arctic and Amazonian dieback