The morphology of experimentally produced charcoal distinguishes fuel types in the Arctic tundra

Abstract

Wildfires in the Arctic tundra have become increasingly frequent in recent years and have important implications for tundra ecosystems and for the global carbon cycle. Lake sediment–based records are the primary means of understanding the climatic influences on tundra fires. Sedimentary charcoal has been used to infer climate-driven changes in tundra fire frequency but thus far cannot differentiate characteristics of the vegetation burnt during fire events. In forested ecosystems, charcoal morphologies have been used to distinguish changes in fuel type consumed by wildfires of the past; however, no such approach has been developed for tundra ecosystems. We show experimentally that charcoal morphologies can be used to differentiate graminoid (mean = 6.77; standard deviation (SD) = 0.23) and shrub (mean = 2.42; SD = 1.86) biomass burnt in tundra fire records. This study is a first step needed to construct more nuanced tundra wildfire histories and to understand how wildfire will impact the region as vegetation and fire change in the future.

Keywords

Arctic fuel type graminoid morphology paleofire shrub tundra

Introduction

The effects of anthropogenic climate change are becoming increasingly evident, particularly at high latitudes (Serreze and Barry, 2011). The Arctic is one of the regions that will be most significantly impacted by climate change (Rogelj, 2018), but its remoteness has limited research on its changing landscape and ecosystems (Cordova, 2016). Tundra wildfires have important ecological and atmospheric consequences and have increased in frequency in recent years (Mack et al., 2011; Masrur et al., 2018; Rocha et al., 2012). Lightning is the sole natural agent for igniting wildfires in the Arctic tundra (Vachula et al., 2019), and limited lightning frequency makes tundra fires a rare phenomenon (Christian et al., 2003). However, hotter and drier summers projected for the future are expected to foster rainless thunderstorms, increasing the likelihood of lightning-sparked fires (Molinari et al., 2019). Indeed, during the 20th century, increasing summer temperatures correlate with greater wildfire frequency (Hoecker and Higuera, 2019).

As Arctic temperatures warm, shrub-type plants are becoming more prevalent – a process called shrubification (Zhang et al., 2013). Shrubification alters the fuel depth, fuel load, and fuel moisture content that predict the probability of ignition, fire intensity, and rate of spread. Shrubs are conducive to more intense wildfires because they consist of wood and leaves, both live and dead, which are highly flammable materials that burn much slower and hotter than graminoids (Anderson, 1982; Fuentes-Ramirez et al., 2016). The intensity of shrub-driven wildfires enables them to burn the organic layer covering the tundra’s permafrost, which, once exposed, may melt and release sequestered carbon dioxide to the atmosphere (Tarnocai et al., 2009). Because Arctic wildfires are promoted by the warming climate and also contribute to climate change through greenhouse gas release, they represent a prime example of a climate-related positive feedback. To understand the impact that wildfires will have in the Arctic tundra, long-term sediment records of charcoal are needed to determine how fire conditions vary with changes in climate. Experimentally produced charcoal will help us determine the physical markers of distinct fuel types.

Charcoal is produced by incomplete combustion of plants under low oxygen conditions, and when preserved in lake sediment cores can indicate the occurrence and frequency of past wildfires in the tundra (Chipman et al., 2015; Higuera 2008; Higuera et al., 2009, 2011; Hu et al., 2015; Sae-Lim et al., 2019). Charcoal records therefore can document the impacts of past environmental and ecological conditions on fire regimes in the Arctic tundra and thus offer insight into their controls and sensitivities. Classification methods are needed to identify the types of plants that compose the charcoal produced by wildfire events and could, for instance, help to identify the role of shrubification in past Arctic fire regimes. Previous work by Umbanhowar and McGrath (1998) showed that charcoal derived from wood, leaves, and grass in Minnesota can be distinguished based on their length:width ratios. Subsequently, length:width ratios of charcoal particles have been used to distinguish fuel types in forests, savannahs, and grasslands (Aleman et al., 2013; Crawford and Belcher, 2014; Enache and Cumming, 2006; Jensen et al., 2007; Leys et al., 2017). In addition, morphological measurements have been shown to be especially useful for cross-site comparisons (Halsall et al., 2018). However, to date, no such investigation has been undertaken in tundra ecosystems.

In this study, we sought to distinguish the fuel types of Arctic tundra charcoal by gathering qualitative and quantitative morphological data of charcoal particles produced by burning. Our study aimed to test whether morphological distinctions exist between shrub species and Arctic graminoid (sedge) species because these two types show different susceptibility to burning. Our study is an important first step needed to construct more nuanced tundra wildfire histories and to understand how wildfire will impact the region in the present and future.

Methods

Samples of seven plant species and one sample of aggregated mossy peat were collected from moist acidic tundra south of Toolik Field Station, located in the northern foothills of the Brooks Range, Alaska (exact sampling location = 68.62°N, 149.61°W; 755 m a.s.l.; Figure 1). Samples were transported from the field and stored at −20°C in sealed Whirlpak bags prior to being thawed and used for the study. The seven plant species sampled for this study were Ledum palustre, Salix pulchra, Betula nana, Rubus chamaemorus, Vaccinium vitis-idaea, Eriophorum vaginatum, and Carex bigelowii. These species were chosen as they are abundant tundra plants comprising a range of potential fuel types (e.g. shrubs, graminoids, forbs). For the shrub species, a mixture of leaves and wood was collected. The peat sample consisted of a mixture of partially decomposed plant matter and bryophytes excavated from the study site with a trowel.

Figure 1.

Study site: Toolik Field Station and Lake E5, in the northern foothills of the Brooks Range, Alaska.

Samples were placed in ceramic crucibles, weighed, and then dried overnight at 120°C in a muffle furnace. The crucibles were weighed again, covered with aluminum foil caps to limit oxygen availability, and were heated at 500°C for 2 h to thoroughly pyrolyze the plant matter. The mass of the residual charred plant matter was then measured to evaluate potential biases in fuel type that might be found in fossil records, as fuel types with high mass retention might be overrepresented in sediment samples.

The charred samples were disaggregated with a mortar and pestle for about 10 s and then washed through a 125-µm sieve with deionized water. Grinding of the samples was performed to mimic natural breakage that the charcoal particles would incur over time in the sediment (Crawford and Belcher, 2014; Umbanhowar and McGrath, 1998). Charcoal fragments caught by the sieve were suspended in water in a small dish and viewed under a binocular dissection microscope at 6.4× magnification. The 6.4× magnification was high enough because this level of magnification was large enough to distinguish and highlight individual charcoal particles, but not so large that particles would overfill the field of view. The samples were isolated by species in each dish (see Supplemental Material).

Photos taken at 6.4× magnification were used to quantify individual charcoal particles for each plant species. Particle length, width, and area were measured using the Adobe Photoshop measuring tool, and the length:width ratios were subsequently calculated. For each plant species, >80 measurements were taken in duplicate to ensure random selection of charcoal particles. These data were aggregated and displayed using box plots. The mean ratios for each plant type were calculated by combining the ratios found for individual species based on graminoid/shrub classification. ANOVA tests, conducted using Microsoft Excel, were used to test the significance of differences in the charcoal produced by each species and plant type.

To demonstrate how this proxy might be applied, three samples were taken from a sediment core from Lake E5 (Figure 1) (Vachula et al., 2019). Following Vachula et al. (2019), 4–5 cm³ of sediment was analyzed for sedimentary charcoal content by chemically treating (1:1 by volume mixture of bleach and sodium pyrophosphate; overnight) and washing samples in a 125-µm sieve. Table 1 describes the age and depth, and the percentage of pollen for each vegetation type found in each sample (Eisner and Colinvaux, 1992). Pollen data describe the primary plant types in this region during each of the three time periods in question. Graminoid plant types are estimated by the upland herbs category, while shrubs are estimated by the trees/shrubs category. The average charcoal length:width ratio for each sample was compared with pollen data to determine whether similar vegetation profiles were produced by both charcoal morphology and pollen count methods.

Table 1.

Age and pollen data of E5 samples.

Sample name	Sample abbreviation	Top age (cal. yr BP)	Trees/shrubs (%)	Upland herbs (%)
TOOL14-E5-1A-1, 81.5–83.5cm	A	7913–8182	76.0	23.5
TOOL14-E5-1A-2, 6–8 cm	B	13,962–14,133	37.9	61.0
TOOL14-E5-1A-3, 24–26 cm	C	20,962–21,068	30.2	69.4

Results

Among the four shrubs (Ledum palustre, Salix pulchra, Betula nana, and Vaccinium vitis-idaea), two graminoids (Eriophorum vaginatum and Carex bigelowii), and one shrub/forb (Rubus chamaemorus) we burned, we found that graminoid charcoal tended to be more rectangular in shape, whereas shrub and forb charcoal had varied polygonal shapes with more curved edges (Figure 2). Graminoid charcoal was consistently more elongate, while shrub and forb charcoal tended to have more similar lengths and widths. Indeed, the burnt shrubs produced charcoal with length:width ratios ranging from 2.09 to 2.50 (Table 2; Figure 3). In comparison, the burnt graminoids Eriophorum vaginatum and Carex bigelowii produced charcoal with length:width ratios of 5.46 and 8.09, respectively (Table 2; Figure 3). Grouped by plant type, graminoid charcoal had a much higher mean length:width ratio (6.77) than shrub charcoal (2.42). ANOVA test results confirmed that the mean length:width ratios are statistically significantly different between shrubs and graminoids (f = 198.00 and f-critical = 3.86; p <0.05).

Figure 2.

Morphological differences in the sedimentary charcoal produced by tundra plants (images are resized to display differences in morphology).

Table 2.

Summary of the length:width and mass retained from plant species analyzed in this study.

Sample ID	Fuel type	Mean length:width ratio	Mean surface area (µm²)	Mean mass retained (%)
Ledum palustre	Shrub	2.47	38,715.48	30.29
Salix pulchra	Shrub	2.09	20,8571.46	32.47
Betula nana	Shrub	2.34	68,127.28	25.56
Vaccinium vitis-idaea	Shrub	2.50	35,756.02	32.55
Rubus chamaemorus	Shrub/forb	2.70	90,060.98	25.05
Average: shrubs	Shrub	2.42	88,246.24	26.05
Carex bigelowii	Graminoid	8.09	96,727.42	26.74
Eriophorum vaginatum	Graminoid	5.46	78,221.70	25.05
Average: graminoids	Graminoid	6.77	87,474.56	25.90
Peat	Mixture	2.06	–	49.02

Figure 3.

Boxplots of the distributions of length:width ratios of charcoal particles produced by shrubs and graminoids in the tundra.

Among the seven plant species, the average surface area of the charred and crashed charcoal particles was as follows: Ledum palustre = 38,715.48 µm², Salix pulchra = 208,571.46 µm², Betula nana = 68,127.28 µm², Rubus chamaemorus = 90,060.98 µm², Vaccinium vitis-idaea = 35,756.02 µm², Eriophorum vaginatum = 78,221.70 µm², and Carex bigelowii = 96,727.42 µm². The average surface area for shrub species was 88,246.24 µm², and the average surface area for graminoid species was 87,474.56 µm² (Table 2).

Among the seven plant species, the percent mass retained was fairly consistent from the dried state to the charred state, with an average of 29.72% of mass retained (Ledum palustre = 30.29%, Salix pulchra = 32.47%, Betula nana = 25.56%, Rubus chamaemorus = 25.05%, Vaccinium vitis-idaea = 32.55%, Eriophorum vaginatum = 25.05%, and Carex bigelowii = 26.74%). The peat sample showed a much smaller loss in mass than any of the plant species, with 49.02% of mass retained.

The average length:width ratio for each of the three sediment samples taken from E5 was A = 1.59, B = 11.62, and C = 1.53, respectively.

Discussion

Tundra charcoal morphology

The morphologies of charcoal particles produced by plant species found in the tundra vary between different fuel types and are unique relative to charcoal produced in temperate ecosystems. Among seven different species of plants, Arctic graminoid species had a mean length:width ratio two to four times higher than that of shrub species. This difference is higher than that found in central North America, where the length:width ratios of graminoids were less than twice (1.8×) those of trees (Umbanhowar and McGrath, 1998). The mean graminoid species length:width ratio was 6.77, which exceeds the values previously reported in the literature. For example, a length:width ratio of 2.00 was used to distinguish graminoids in the Afrotropics, where graminoids had a ratio larger than 2.00 and shrubs had a ratio less than 2.00 (Aleman et al., 2013). Furthermore, Umbanhowar and McGrath (1998) reported a length:width ratio of 3.62 for grass species in Minnesota. An ANOVA test between our graminoid length:width data and Umbanhowar and McGrath’s (1998) graminoid length:width data revealed that this is a statistically significant difference (p < 0.05). There are a few potential explanations for this mismatch. Umbanhowar and McGrath (1998) found an average surface area of 65,630 µm² for grasses, 64,946 µm² for leaves, and 50,150 µm² for wood. In this study, the average surface areas of 88,246 µm² for graminoids and 87,475 µm² for shrubs were found. The fact that our charcoal particles are on average larger than those measured in Umbanhowar and McGrath (1998) may explain why our length:width ratios are also higher; generally, it appears that our particles underwent less breakage. An alternative explanation could be that tundra graminoids simply produce more elongate charcoal than do temperate graminoids. Both the graminoid species used in our study were in the sedge family, while the graminoid species used by Umbanhowar and McGrath (1997) were of the grass family, which could also explain this discrepancy. These graminoid families may have different breakage tendencies because of their relative durability.

Average length:width ratios for burnt woody matter were also statistically significantly different between our study of Arctic vegetation and Umbanhowar and McGrath’s (1997) study of Minnesota vegetation (p < 0.05). A possible explanation for this difference could be the fact that wood and leaves come primarily from trees in central North America and from shrubs in the Arctic tundra. Further research into the differences in charcoal morphology between trees and shrubs would be necessary to conclude whether this discrepancy typifies Arctic and temperate vegetation. Regardless of the magnitude of the ratio, however, both studies clearly showed that graminoid species produce charcoal with greater length:width ratios than their woody species counterparts.

Application

Pollen data indicate that sample A was deposited during a time with increased tree and shrub abundances, whereas samples B and C correspond to times with increased graminoid abundances. Charcoal length:width ratios agreed with these pollen results for sample A, where the average length:width = 1.59, and for sample B, the average length:width = 11.62. Sample C displayed a low average length:width (1.52), even though this sample reportedly had a large proportion of graminoid species. It may be that many graminoids were present at this point in history, but shrubs were still the primary fuel type (i.e. shrubs were more likely to burn and contribute to charcoal deposits than graminoids). Similarly, shrubs may be overrepresented in the sediment record relative to graminoids in terms of charcoal production. Indeed, our measurements of mass retention indicate that shrubs generally produce more charcoal per unit of biomass than graminoids (Table 2), which would be expected to bias sedimentary charcoal profiles. The average length:width ratio for sample B was also significantly larger than any of the results we have seen already for graminoid charcoal particles. It may be that our method of breaking down charcoal particles using mortar and pestle to model natural breakage was too thorough, and realistically charcoal particles in the sediment would not have been subjected to that level of physical degradation. It is valuable to recognize that pollen data is representative of what plant types were growing at a particular point in time, while charcoal data is representative of what plant types were burning. Perhaps our application does not serve its purpose in testing the value of this method because it is based on a misguided assumption that these two proxies give us the same information about vegetation types. This study supports continued research of charcoal proxies to develop accurate wildfire histories, and shows that a deeper discussion is needed to assess the differing uses of pollen and charcoal proxies. It seems that they may tell fundamentally different stories.

Implications: Paleofires and impacts of climate change

The ability to distinguish plant types from charcoal particles is an important tool for understanding the impact of climate change on Arctic tundra fires. Prior paleofire studies have focused primarily on total aggregate charcoal as a tool to study fire events and regimes in the tundra, but total charcoal accumulation rates (CHAR) only offer general records of fire history (Chipman et al., 2015; Higuera, 2008; Hu et al., 2015; Sae-Lim et al., 2019). We show that the measurement of charcoal particle morphology can offer insight into the fuel type being burned in tundra fires. The significance of charcoal morphology for paleofire studies was also shown in a 2017 study by Leys BA, Commerford JL, and McLauchlan KK, in this case in a non-tundra environment. This study thus enables paleofire research to take a step further than both CHAR analysis, which identifies past fires, and to combine new charcoal and traditional palynological methods to identify what types of vegetation burned in past fires. Using charcoal length:width ratios to isolate past shrubification events may help us to gain a clearer idea of what ecosystem changes we may see in the Arctic in the coming years. However, charcoal morphologies alone cannot identify the role of shrubification events in tundra fire regimes. Multiple paleoecological proxies may be required to fully understand the fire-vegetation dynamics, in part due to limitations for the charcoal proxy (e.g. uncertainties of source area, differences of charcoal mass retention during fires).

Future work is needed to solidify charcoal morphologies as tundra fire proxies, including how to better understand the expected ratios of elongate to non-elongate (graminoid to shrub) charcoal particles that one would find in a mixed sample from a lake sediment core. The samples measured in this experiment were isolated by species, but it is unlikely that sedimentary charcoal from natural wildfires would contain only one plant type. It would thus be valuable to determine the charcoal production rates and relative biomass of shrubs and graminoids alike in order to contextualize what portion of a sedimentary charcoal sample should be shrub-type and graminoid-type particles under typical ecological conditions and shrubification conditions.

In addition, more data on the relative flammability of different plant types in the tundra biome is needed to understand tundra fire dynamics. Vegetation types show different fire behavior based on their fuel model: graminoids tend to burn faster and cooler, while shrubs burn slower, hotter, and with higher intensity (Anderson, 1982; Fuentes-Ramirez et al., 2016). However, these models were developed in more temperate ecosystems. Beyond these general characteristics, little is known about the specific burning habits of different plant types found in the tundra. It is key to know how much more severe we can expect wildfires to be in a shrub dominated ecosystem relative to an ecosystem dominated by graminoid plant types, both to better inform paleofire interpretations and to better prepare for future fires.

The next step in testing this method will involve mixing plant types in known ratios and finding the average length:width ratio of these mixed samples. We will then plot the mix ratio (shrub:graminoid ratio) against the average length:width ratio to calibrate expected changes in the observed length:width ratio with changes in the graminoid/shrub ratio. This research will also help us understand how mass retention rates of different plant types may play into their observed representation in the charcoal sample.

Supplemental Material

Supplement_v2 – Supplemental material for The morphology of experimentally produced charcoal distinguishes fuel types in the Arctic tundra

Supplemental material, Supplement_v2 for The morphology of experimentally produced charcoal distinguishes fuel types in the Arctic tundra by Eleanor MB Pereboom, Richard S Vachula, Yongsong Huang and James Russell in The Holocene

Footnotes

Acknowledgements

We thank C. Umbanhowar for his time responding to our questions about his data and helpful reviews. We also thank an anonymous reviewer for the helpful comments and improvements to this manuscript.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: EMBP and this work were supported by an Undergraduate Teaching and Research Award granted by Brown University.

ORCID iDs

Eleanor MB Pereboom

Richard S Vachula

Supplemental material

Supplemental material for this article is available at

References

Aleman

Blarquez

Bentaleb

et al. (2013) Tracking land-cover changes with sedimentary charcoal in the Afrotropics. The Holocene 23(12): 1853–1862.

Anderson

(1982) Aids to determining fuel models for estimating fire behavior. General Technical Report INT-22, April. Forest Service, U.S. Department of Agriculture. Available at: https://www.fs.fed.us/rm/pubs_int/int_gtr122.pdf

Chipman

Hudspith

Higuera

et al. (2015) Spatiotemporal patterns of tundra fires: Late-Quaternary charcoal records from Alaska. Biogeosciences 12(13): 4017–4027. Available at: http://www.biogeosciences.net/12/4017/2015/bg-12-4017-2015.html

Christian

Blakeslee

Boccippio

et al. (2003) Global frequency and distribution of lightning as observed from space by the Optical Transient Detector. Journal of Geophysical Research: Atmospheres 108: ACL 4-1–ACL 4-15.

Cordova

(2016) NSF’s 10 big ideas. Special Report, NSF: National Science Foundation. Available at: https://www.nsf.gov/news/special_reports/big_ideas/

Crawford

Belcher

(2014) Charcoal morphometry for paleoecological analysis: The effects of fuel type and transportation on morphological parameters. Applications in Plant Sciences 2(8): 1400004.

Eisner

Colinvaux

(1992) Late Quaternary pollen records from Oil Lake and Feniak Lake, Alaska, USA. Arctic and Alpine Research 24(1): 56–63. Available at: https://www-jstor-org.web.bisu.edu.cn/stable/1551320?origin=crossref

Enache

Cumming

(2006) Tracking recorded fires using charcoal morphology from the sedimentary sequence of Prosser Lake, British Columbia (Canada). Quaternary Research 65(2): 282–292. Available at: https://www-sciencedirect-com.web.bisu.edu.cn/science/article/pii/S0033589405001432

Fuentes-Ramirez

Veldman

Holzapfel

et al. (2016) Spreaders, igniters, and burning shrubs: plant flammability explains novel fire dynamics in grass-invaded deserts. Ecological Applications 26(7): 2311–2322.

10.

Halsall

Ellingsen

Asplund

et al. (2018) Fossil charcoal quantification using manual and image analysis approaches. The Holocene 28: 1345–1353.

11.

Higuera

(2008) MCAgeDepth 0.1: Probabilistic Age-Depth Models for Continuous Sediment Records: User’s Guide. Bozeman, MT: Montana State University.

12.

Higuera

Brubaker

Anderson

et al. (2009) Vegetation mediated the impacts of postglacial climate change on fire regimes in the south-central Brooks Range, Alaska. Ecological Monographs 79(2): 201–219. Available at: https://esajournals.onlinelibrary.wiley.com/doi/full/10.1890/07-2019.1

13.

Higuera

Chipman

Barnes

et al. (2011) Variability of tundra fire regimes in Arctic Alaska: Millennial-scale patterns and ecological implications. Ecological Applications 21(8): 3211–3226.

14.

Hoecker

Higuera

(2019) Forest succession and climate variability interacted to control fire activity over the last four centuries in an Alaskan Boreal landscape. Landscape Ecology 34(2): 227–241.

15.

Higuera

Duffy

et al. (2015) Arctic tundra fires: Natural variability and responses to climate change. Frontiers in Ecology and the Environment 13(7): 369–377.

16.

Jensen

Lynch

Calcote

et al. (2007) Interpretation of charcoal morphotypes in sediments from Ferry Lake, Wisconsin, USA: Do different plant fuel sources produce distinctive charcoal morphotypes? The Holocene 17(7): 907–915.

17.

Leys

Commerford

McLauchlan

(2017) Reconstructing grassland fire history using sedimentary charcoal: Considering count, size and shape. PLoS ONE 12(4): e0176445.

18.

Mack

Bret-Harte

Hollingsworth

et al. (2011) Carbon loss from an unprecedented Arctic tundra wildfire. Nature 475(7357): 489–492.

19.

Masrur

Petrov

DeGroote

(2018) Circumpolar spatio-temporal patterns and contributing climatic factors of wildfire activity in the Arctic tundra from 2001–2015. Environmental Research Letters 13(1): 014019.

20.

Molinari

Lehsten

Blarquez

et al. (2018) The climate, the fuel and the land use: Long-term regional variability of biomass burning in boreal forests. Global Change Biology 24(10): 4929–4945.

21.

Rocha

Loranty

Higuera

et al. (2012) The footprint of Alaskan tundra fires during the past half-century: Implications for surface properties and radiative forcing. Environmental Research Letters 7(4): 044039.

22.

Rogelj

(2018) Special Report: Global Warming of 1.5°C (ed G

Flato

). Geneva; Cambridge: IPCC; Cambridge University Press.

23.

Sae-Lim

Russell

Vachula

et al. (2019) Temperature-controlled tundra fire severity and frequency during the last millennium in the Yukon-Kuskokwim delta, Alaska. The Holocene 29: 1223–1233.

24.

Serreze

Barry

(2011) Processes and impacts of Arctic amplification: A research synthesis. Global and Planetary Change 77(1): 85–96. Available at: https://www-sciencedirect-com.web.bisu.edu.cn/science/article/pii/S0921818111000397

25.

Tarnocai

Canadell

Schuur

EAG

et al. (2009) Soil organic carbon pools in the northern circumpolar permafrost region. Global Biogeochemical Cycles 23(2): 003327.

26.

Umbanhowar

McGrath

(1998) Experimental production and analysis of microscopic charcoal from wood, leaves and grasses. The Holocene 8(3): 341–346.

27.

Vachula

Huang

Longo

et al. (2019) Evidence of Ice Age humans in eastern Beringia suggests early migration to North America. Quaternary Science Reviews 205: 35–44.

28.

Zhang

Miller

Smith

et al. (2013) Tundra shrubification and tree-line advance amplify arctic climate warming: Results from an individual-based dynamic vegetation model. Environmental Research Letters 8(3): 034023.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

5.98 MB