Evidence for fire in the Pliocene Arctic in response to amplified temperature

The mid-Pliocene is a valuable time interval for investigating equilibrium climate at current atmospheric CO2 concentrations because atmospheric CO2 concentrations are thought to have been comparable to the current day and yet the climate and distribution of ecosystems were quite different. One intriguing, but not fully understood, feature of the early to mid-Pliocene climate is the amplified Arctic temperature response and its impact on Arctic ecosystems. Only the most recent models appear to correctly estimate the degree of warming in the Pliocene Arctic and validation of the currently proposed feedbacks is limited by scarce terrestrial records of climate and environment. Here we reconstruct the summer temperature and fire regime from a subfossil fenpeat deposit on west–central Ellesmere Island, Canada, that has been chronologically constrained using cosmogenic nuclide burial dating to 3.9+ 1.5/− 0.5 Ma. The estimate for average mean summer temperature is 15.4± 0.8 C using specific bacterial membrane lipids, i.e., branched glycerol dialkyl glycerol tetraethers. This is above the proposed threshold that predicts a substantial increase in wildfire in the modern high latitudes. Macro-charcoal was present in all samples from this Pliocene section with notably higher charcoal concentration in the upper part of the sequence. This change in charcoal was synchronous with a change in vegetation that included an increase in abundance of fire-promoting Pinus and Picea. Paleo-vegetation reconstructions are consistent with warm summer temperatures, relatively low summer precipitation and an incidence of fire comparable to fire-adapted boreal forests of North America and central Siberia. To our knowledge, this site provides the northernmost evidence of fire during the Pliocene. It suggests that ecosystem productivity was greater than in the present day, providing fuel for wildfires, and that the climate was conducive to the ignition of fire during this period. The results reveal that interactions between paleo-vegetation and paleoclimate were mediated by fire in the High Arctic during the Pliocene, even though CO2 concentrations were similar to modern values. Published by Copernicus Publications on behalf of the European Geosciences Union. 1064 T. L. Fletcher et al.: Evidence for fire in the Pliocene Arctic

constrained using radionuclide dating to 3.9 +1.5/-0.5 Ma. 23 The estimate for average mean summer temperature is 15.4±0.8°C using specific bacterial membrane lipids, i.e. 24 branched glycerol dialkyl glycerol tetraethers. Macro-charcoal was present in all samples from this Pliocene section 25 with notably higher charcoal concentration in the upper part of the sequence. This change in charcoal was synchronous 26 with a change in vegetation that saw fire-promoting Pinus and Picea increase in abundance. Paleovegetation 27 reconstructions are consistent with warm summer temperatures, relatively low summer precipitation and an incidence 28 of fire comparable to fire adapted boreal forests of North America, or potentially central Siberia. 29 To our knowledge, this site provides the northern-most evidence of fire during the Pliocene. It suggests that ecosystem 30 productivity was much greater, providing fuel for wildfires, and that the climate was conducive to the ignition of fire 31 during this period. This study indicates that interactions between paleovegetation and paleoclimate were mediated by 32 fire in the High Arctic during the Pliocene, even though CO2 concentrations were similar to modern. 33 5 (99:1, v:v) at a concentration of 10 mg ml -1 and subsequently passed through a 0.45 µm PTFE filter. Finally, the polar 144 fractions were analyzed for GDGTs by ultra-high performance liquid chromatography -atmospheric pressure positive 145 ion chemical ionization -mass spectrometry (UHPLC-APCI-MS) using the method described by Hopmans et al., 146 (2016). The polar fractions of some samples were re-run on the UHPLC-APCI-MS multiple times and the average 147 fractional abundances of the brGDGTs was determined. 148 For the calculation of brGDGT-based proxies, the brGDGTs are specified by the Roman numerals as indicated in 149  The square brackets denote the fractional abundance of the brGDGT within the bracket relative to the total brGDGTs. 157 The distributions of aquatically produced brGDGTs in the lake calibration developed by Pearson et al. (2011) were 158 used to determine MST. When this calibration is used the fractional abundances of IIa and IIa′ must be summed 159 because these two isomers co-eluted under the chromatographic conditions used by Pearson  Surface water pH = 8.95 + 2.65 * CBT′ RMSE of 0.80 (5) 168

Vegetation and Fire Reconstruction 169
For charcoal, a total of thirty 2 cm 3 samples were taken at 5 cm intervals from depths from 380 and 381.45 MASL at 170 the BP site, with an additional 2cm -3 sample collected at 381.65 MASL. All samples were deflocculated using sodium 171 hexametaphosphate and passed through 500, 250 and 125 µm nested mesh sieves. The residual sample caught on each 172 sieve was then collected in a gridded petri dish and examined using a stereomicroscope at 20-40X magnification to 173 obtain charcoal concentration (fragments cm -3 ). Charcoal area (mm 2 cm -3 ) was measured for each sample using 174 specialized imaging software from Scion Corporation. For a detailed description of methods see Brown and Power 175 (2013). 176 Vegetation was reconstructed using pollen and spores (herein pollen) at selected elevations chosen to capture upper 177 and lower sections of the elevation profile, and that corresponded with changes in charcoal. The sample depths selected 6 processed using standard approaches (Moore et al., 1991)

Geochronology 197
The burial dating results with 26 Al/ 10 Be in quartz sand at 10 m below modern depth provides four individual ages. 198 From shallowest to deepest, the burial ages are 3.6 +1.5/-0.5 Ma, 3.9 +3.7/-0.5 Ma, 4.1 +5.8/-0.4 Ma, and 4.0 +1.5/-199 0.4 Ma (Table S2), with an unweighted mean age of 3.9 Ma. The convoluted probability distribution function yields 200 a maximum probability age of 4.5 Ma. Unfortunately, the positive tails of the probability distribution functions of two 201 of the samples exceeds the radiodecay saturation limit of the burial age. Therefore, their probability distributions do 202 not reflect the actual age probabilities and uncertainty. Given the positive tail in the probability distribution functions, 203 and the inability to convolve all samples, we recommend using the unweighted mean age, 3.9 Ma, with an uncertainty 204 of +1.5/-0.5 Ma as indicated by the two samples with unsaturated limits. Despite the apparent upward younging of the 205 individual burial ages, the 1σ-uncertainties overlap rendering the samples indistinguishable. 206

Provenance of branched GDGTs 208
Previously, brGDGT derived MAAT estimates (-0.6 ± 5.0 °C) from BP sediments were developed using the older 209 chromatography methods that did not separate the 5-and 6-methyl brGDGTs, and a soil calibration (Ballantyne et  210 al., 2010). In marine and lacustrine sediments, bacterial brGDGTs were thought to originate predominantly from 7 the temperature estimates based upon them (Warden et  2013). In this study, we examine the distribution of brGDGTs in an attempt to determine their origin and consequently 220 the most appropriate calibration to utilize in order to reconstruct temperatures from the BP sediments. 221 Branched GDGTs IIIa and IIIa′ on average had the highest fractional abundance of the brGDGTs detected in the BP 222 sediments (see Fig. S2 for structures; Table S4). A previous study established that when plotted in a ternary diagram 223 the fractional abundances of the tetra-, penta-and hexamethylated brGDGTs, soils lie within a distinct area (Sinninghe 224 Damsté, 2016). To assess whether the brGDGTs in the BP deposit were predominantly derived from soils, we 225 compared the fractional abundances of the tetra-, penta-and hexamethylated brGDGTs in the BP sediments to those 226 from modern datasets in a ternary diagram (Fig. 3). Since the contribution of brGDGTs from either peat or aquatic 227 production could affect the use of brGDGTs for paleoclimate application, in addition to comparing the samples to the imperative to only compare samples in a ternary diagram like this where all of the datasets were analyzed with the 231 novel methods that separate the 5-and 6-methyl brGDGTs since the improved separation can result in an increased 232 quantification of hexamethylated brGDGTs. Recently, samples from East African lake sediments were analyzed using 233 these new methods (Russell et al., 2018) and so these samples were included in the ternary plot for comparison (Fig.  234 3). Although the lakes from the East African dataset are all from a tropical area, they vary widely in altitude and, thus, 235 in MAAT. We separated them into three categories by MAAT (lakes >20°C, lakes between 10-20°C and lakes<10°C). 236 By comparing all the samples in the ternary plot, it was evident that the BP samples plotted closest to the lacustrine 237 sediment samples from regions in East Africa with a MAAT <10°C, suggesting that the provenance of the majority 238 of the brGDGTs from the BP sediments was not soil or peat but lacustrine aquatic production. 239 The average estimated surface water pH for the BP sediments (8.6±0.2) calculated using eq. (5), is within the 6 -9 240 range typical of lakes and rivers (Mattson, 1999). This value is near the upper limit of rich fens characterized by the 241 presence of S. scorpioides (Kooijman and Westhoff, 1995;Kooijman and Paulissen, 2006) and is higher than what 242 would be expected for peat-bog sediments that are acidic (pH 3-6; Clymo, 1964) and which constitute most of the 243 peats studied by Naafs et al. (2017). A predominant origin from lake aquatic production is in keeping with previous 244 interpretation of the paleoenvironment of the BP site, which was at least at times covered by water as evidenced by 245 fresh water diatoms, fish remains and gnawed beaver sticks in the sediment (Mitchell et al., 2016). . A concern when applying this calibration is that it is based on lakes from an equatorial region that does not 253 experience substantial seasonality, whereas, the Pliocene Arctic BP site did experience substantial seasonality 254 (Fletcher et al., 2017). Biological production (including brGDGT production) in BP was likely skewed towards 255 summer and, therefore, summer temperature has a larger influence on the reconstructed MAAT. Unfortunately, no 256 global lake calibration set using individually quantified 5-and 6-methyl brGDGTs is yet available. Therefore were relatively high. In contrast, low numbers of Picea (3%), Pinus (3%) and fern spores were recorded. Additional 275 wetland taxa like Myrica (5%) and Cyperaceae (6%) were also noted. Overall, the non-arboreal (23%) signal was well 276 developed. Crumpled and/or ruptured inaperturate grains with surface sculpturing that varied from scabrate to 277 verricate were noted in the assemblage (12%), but could not be definitely identified. It is possible that these grains 278 represent Populus, Cupressaceae or additional Cyperaceae pollen. Between 381.10-381.25 MASL, Larix (38%) and 279 Betula (21%) increased in abundance, followed by ferns (7%). Cyperaceae remained at similar levels (6%) whereas 280 Picea and Pinus decreased to 2% and 1%, respectively. Unidentified inaperturate types collectively averaged 14%. 281 Fennoscandia. Of these areas, the western coast of northern North America and eastern coast of southern Sweden has 289 the most similarity to the reconstructed BP climate in terms of MST (Fig. 7b) and summer precipitation (Fig. 7c). 290 While high counts of active fire days are common in the western part of the North American boreal forest, it is not 291 as common in the eastern part of the North American boreal forest (Fig. 7d), likely due to the differences in the 292 precipitation regime. There were also low fire counts in Fennoscandia likely due to historical severe fire suppression 293 (Brown and Giesecke, 2014; Niklasson and Granström, 2004). Therefore, based on our reconstruction of the climate 294 and ecology of the BP site, our results suggest that BP most closely resembled a boreal-type forest ecosystem shaped 295 by fire, similar to those of Washington, British Columbia, Northwest Territories, Yukon and Alaska (but see Sect. The plant and animal fossil assemblages observed at BP suggest a depositional age between 3 and 5 Ma (Matthews Jr 300 and Ovenden, 1990; Tedford and Harington, 2003). This biostratigraphic age was corroborated with an amino-acid 301 racemization age (>2.4 ± 0.5 Ma) and Sr-correlation age (2.8-5.1 Ma) on shells (Brigham-Grette and Carter, 1992) in 302 biostratigraphically correlated sediments on Meighen Island, situated 375 km to the west-north-west. The previously 303 calculated burial age of 3.4 Ma for the BP site is a minimum age because no post-depositional production of 26 Al or 304 10 Be by muons was assumed. If the samples are considered to have been buried at only the current depth (ca. 10 m, 305 see supplemental data) then the ages plot to the left and outside of the burial field, indicating that the burial depth was 306 significantly deeper for most of the post-depositional history. The revised cosmogenic nuclide burial age is 3.9 +1.5/-307 0.5 Ma. It is the best interpretation of burial age data based on improved production rate systematics (e.g. Lifton et 308 al., 2014), and more reasonable estimates of erosion rate and ice cover since the mid-Pliocene (see Fig. S3; Table S5). 309 As the stratigraphic position of the cosmogenic samples is very close to the BP peat layers, we interpret the age to 310 represent the approximate time that the peat was deposited. The abundance of charcoal at BP demonstrates that climatic conditions were conducive to ignition and that sufficient 320 biomass available for combustion existed across the landscape. brGDGTs-derived temperature estimates suggest mean 321 summer temperatures at BP exceeded the ~13.5 °C threshold (Young et al., 2017) that drastically increases the chance 322 of wildfire. Indeed, the estimate of ~15.4°C suggests summer temperatures is ~11°C higher than modern day Eureka, 323 Canada (~4.1°C; Fig. 2). Given a global mean increase of 3°C for the Pliocene compared to modern (see fig. 1) this 324 11°C increase represents 3.6x arctic amplification of temperature (NB. although comparing summer temperatures to 325 mean global temperature increase is likely imprecise, given much increase of arctic warmth in Pliocene climate models 326 is from winter warming (see Ballantyne at al. 2013) 3.6x is likely an underestimate rather than an overestimate.) 327 Without increasing arctic amplification of temperature that accompanies increasing CO2, mean summer temperatures 328 would fall below the ~13.5°C threshold. This is evidence that Pliocene arctic amplification of temperatures was a 329 direct feedback to increased wildfire activity, but also an indirect feedback as the increased extent of boreal forest into 330 the higher latitudes, also possible due to arctic amplification of temperatures, provided the fuel (Fig. 6) 331 An increase in atmospheric convection has been simulated in response to diminished sea-ice during warmer intervals 332  (12-18°C, Fig. 7B). The 352 eastern coast of North America has higher rainfall during the summer (>270 mm), than the west coast and Alaska 353 (Fig. 7C), which correlates to the timing of western fires. The low summer precipitation for much of the west (<200 354 mm), is consistent with previously published summer precipitation estimates for BP (~190 mm). As a result, the fire 355 regime of the west coast ~50°N may be a better analogue for BP than the east coast of North America. In central 356 the region experiencing higher summer precipitation (252->288 mm), than the more similar eastern, Swedish part of 358 the region (~198 mm). 359 Investigation of the modern fire detection data (Fig. 7D) suggests that the two regions most climatically similar to 360 BP, ~50°N western North America and central Sweden, have radically different fire regimes. It is likely this is caused 361 by historical fire suppression in Sweden that limits the utility of modern data for comparison with this study (Brown 362 and Giesecke, 2014; Niklasson and Granström, 2004). To understand the fire regimes, as shaped by climate and species 363 composition rather than human impacts, we considered both the modern and recent Holocene reconstructions for these 364 regions (Table 1). This shows that, a) within any region variation arises from the complex spatial patterning of fire 365 across landscapes, and b) that the regions most similar to BP (~50°N western North American and eastern 366 Fennoscandian reconstructions for the recent Holocene) have shorter fire return intervals than the cooler Alaskan 367 tundra or wetter summer ~50°N region of the eastern North American coast.  (Table 1). Based on similarity of the climate variables, the more southerly 374 MFRIs (~80 years) may be a better analogue. Key differences between boreal fires in North America compared to 375 Russia are a higher fire frequency with more burned area in Russia, but a much lower crown fire and a difference in 376 timing of disturbance, with spring fires prevailing in Russia compared to mid-summer fires in western Canada The CRACLE analyses also suggest that slightly drier conditions may have prevailed during the three wettest months 406 (249-285mm vs. 192-219mm). While the interaction between climate, vegetation and fire is complex, small changes 407 in MST and precipitation could have directly altered both the vegetation and fire regime, which in turn further 408 promoted fire adapted taxa. In addition to regional climatic factors, community change at the site may have been 409 further influenced by local hydrological conditions, such as channel migration, pond infilling and ecosystem 410 engineering by beaver (Cantor spp.). 411 The high charcoal content of the upper portion (~ Unit IV) of the sequence has three potential explanations: 412 reworking of previously deposited charcoal, decreased sedimentation, or increased wildfire production of charcoal. 413 We consider the first unlikely because there is no difference in the shape of the macrocharcoal between the upper and 414 lower portions of the sequence, whereas we would anticipate a change in the dimensions of the charcoal if it had 415 undergone additional physical breakdown from reworking (see Fig. S4). The second, decreased sedimentation, may 416 occur if the deposition is a result of infrequent, episodic flooding intermixed with long periods during which charcoal 417 was deposited. The recorded sedimentology does not support this explanation, but due to the complexity of flooding 418 processes, also does not disprove this explanation. We, however, favour the third explanation of increased wildfire 419 due to the change in plant composition consistent with a greater influence of fire. If accepted, it is likely that frequent, 420 Critically, the charcoal record at BP suggests substantial biomass burning that could have acted as a feedback 449 mechanism amplifying or dampening warming during the Pliocene. Its potential role as a feedback to climate is 450 suggested by its prevalence through time, and forest fire's complex direct impacts on the surface radiative budget (e.g. Arctic sites and modelling experiments using varying fire regimes and extent is warranted to better characterize the 454 fire regime in order to improve accuracy of fire simulations in earth system models of Pliocene climate. 455

CONCLUSION 456
The novel temperature estimates presented here confirm that summer temperatures were considerably warmer during 457 the Pliocene (15.4 ± 0.8 °C) compared to the modern Arctic. The ~11°C higher summer temperatures at Beaver Pond 458 support an increasing influence of arctic amplification of temperatures when CO2 reaches and exceeds modern levels.