Three well-preserved mummified wood specimens have been excavated from three earliest Eocene (similar to 55.5, 55.2 and 53.3 Ma) kimberlite pipes in the subarctic Northwest Territories, Canada (similar to 64 degrees N). Each specimen contained multi-decadal length tree-ring series and allowed measurements of stable hydrogen isotope ratios (delta H-2 values) of the lignin methoxyl groups (commonly used cellulose was largely degraded). We used these delta H-2 signatures for the reconstruction of three representative delta H-2 values of the local precipitation by applying calibrated isotope fractionations to investigate deep-time paleoclimatology. Our reconstructions indicate unprecedented low values for the subarctic early Eocene which, however, show within similar to 2.2 Myr an increasing trend from 206 17, 202 17 to 168 17 parts per thousand. These values were interpreted along with other Arctic, subarctic and mid-latitudinal proxy records of that time period indicating in total a period of low stable hydrogen and oxygen isotope ratios in precipitation (onset between 55.7 and 54.9 Ma and a recovery at 53.3 +/- 0.6 Ma) which is assumed to have been primarily caused by a continental cold period. An increased magnitude of this cold period was noted for our subarctic reconstructions pointing to a polar amplification where surface air temperature changes in the Arctic exceed the global trend (in response to climate forcing). In an attempt to quantify the polar amplification magnitude we estimated Arctic temperature changes using our delta H-2 results in combination with existing relationships between early Eocene temperatures and stable water isotopes. Comparing the Arctic temperature change with a global estimation proposes a polar amplification of magnitude <4. Our estimate for the earliest Eocene shows either agreement or indicates a lower magnitude when compared to previously described Cenozoic polar amplifications commonly ranging between 2 and 4.

Stable isotope ratios of carbon (δ13C) and oxygen (δ18O) from tree ring cellulose can provide valuable paleoclimatic information at annual and subannual resolution, from time periods long before instrumental climate records. Recently, mummified (non-permineralized) wood was discovered within Canadian Subarctic kimberlites, inviting paleoclimatic investigations of the time in which the trees grew. In the mummified wood, polysaccharides (hemicellulose, α-cellulose) are frequently preferentially degraded, in the low-silica anaerobic burial environment of the kimberlites, resulting in a lignin-rich material. However, some samples from the Ekati Panda kimberlite pipe (ca. 53.3Ma) contain remnant cellulose, demonstrating the extraordinary preservation potential of kimberlites. Preservation of α-cellulose is important because it allows for the construction of stable isotopic proxy records of deep-time paleoclimates at annual to subannual resolution. Established α-cellulose extraction methods [i.e., using a 17.5% sodium hydroxide (NaOH) solution] were unsuitable for this material because all holocelluloses were dissolved. Therefore, we tested variants of two cellulose extraction methods [i.e., Brendel et al. (2000) and Jayme-Wise (Leavitt and Danzer 1993, Loader et al. 1997)] to optimize a procedure for extraction of a consistent yield of mummified wood cellulose for stable isotopic analysis. Stable carbon (δ13C) and oxygen (δ18O) isotopes were measured from cellulose from each extraction method, as well as Extractive-Free Wood and unextracted mummified wood. vitrinite reflectance, used to assess thermal alteration of the unextracted material to estimate post-burial temperatures within the kimberlite, suggested low post-burial peak temperatures (Tpeak=60°C). We detected no mineral contaminants (i.e., iron oxides) in the material using Energy-Dispersive X-ray Spectroscopy (EDX). Despite low cellulose yield (<5%), the Attenuated Total Reflection Fourier-Transform Infrared (ATR-FTIR) spectra of mummified cellulose samples strongly resembled modern α-cellulose. All cellulose treatments were similar in stable isotope values and ATR-FTIR peaks, but significantly different from unextracted wood and Extractive-Free Wood. As, reported by other sources (Anchukaitis et al. 2008, Brookman and Whittaker 2012), use of the Brendel method may cause cellulose acetylation, introducing an FTIR peak near 1720cm−1, thus complicating hemicellulose detection at a peak near 1725cm−1. For this reason, the Jayme-Wise method is recommended for detection of hemicelluloses in mummified wood. If hemicelluloses are not present in holocellulose, due to groundwater hydrolysis, the NaOH step may be omitted or reduced in concentration and still produce α-cellulose.

The recent discovery of well-preserved mummified wood buried within a subarctic kimberlite diamond mine prompted a paleoclimatic study of the early Eocene "hothouse" (ca. 53.3 Ma). At the time of kimberlite eruption, the Subarctic was warm and humid producing a temperate rainforest biome well north of the Arctic Circle. Previous studies have estimated that mean annual temperatures in this region were 4–20 °C in the early Eocene, using a variety of proxies including leaf margin analysis and stable isotopes (δ13C and δ18O) of fossil cellulose. Here, we examine stable isotopes of tree-ring cellulose at subannual- to annual-scale resolution, using the oldest viable cellulose found to date. We use mechanistic models and transfer functions to estimate earliest Eocene temperatures using mummified cellulose, which was well preserved in the kimberlite. Multiple samples of Piceoxylon wood within the kimberlite were crossdated by tree-ring width. Multiple proxies are used in combination to tease apart likely environmental factors influencing the tree physiology and growth in the unique extinct ecosystem of the Polar rainforest. Calculations of interannual variation in temperature over a multidecadal time-slice in the early Eocene are presented, with a mean annual temperature (MAT) estimate of 11.4 °C (1 σ = 1.8 °C) based on δ18O, which is 16 °C warmer than the current MAT of the area (−4.6 °C). Early Eocene atmospheric δ13C (δ13Catm) estimates were −5.5 (±0.7) ‰. Isotopic discrimination (Δ) and leaf intercellular pCO2 ratio (ci/ca) were similar to modern values (Δ = 18.7 ± 0.8 ‰; ci/ca = 0.63 ± 0.03 %), but intrinsic water use efficiency (Early Eocene iWUE = 211 ± 20 μmol mol−1) was over twice the level found in modern high-latitude trees. Dual-isotope spectral analysis suggests that multidecadal climate cycles somewhat similar to the modern Pacific Decadal Oscillation likely drove temperature and cloudiness trends on 20–30-year timescales, influencing photosynthetic productivity and tree growth patterns.

Abstract The discovery of exceptionally well-preserved Paleogene wood fossils (ca. 55–53 Ma) within Canadian Arctic diamond-bearing kimberlites prompted a paleoclimatic study of the Paleocene-Eocene Transition. The samples are not petrified, but have been “mummified” by their inclusion in pyroclastic debris and still contain primordial wood material. However, preferential cellulose loss has rendered the wood very fragile, precluding the use of standard dendrochronological methods of surface preparation. Similar to archaeological charcoal, breaking the mummified wood allows superior visualization of tree-ring boundaries and wood anatomy, but often produces irregular surfaces making microscopic examination difficult. Therefore, a simple aluminum clamp was constructed to break radial wood transects in a controlled manner for the purpose of collecting dendrochronological and wood-anatomical data for paleoclimatic reconstructions. Because it does not require the use of chemical treatments or stabilizing resins, the wood remains chemically unaltered, allowing chemical and isotopic analyses to be undertaken. Future studies of fragile woods may benefit from this method of controlled breaking if sanding is ineffective. Résumé La découverte de fossiles bois du Paléogène (55–53 Ma), exceptionnellement bien conservés, dans des kimberlites diamantifères dans l'Arctique canadienne a incité à une etude paléoclimatique de la transition Paléocène-Eocène. Les échantillons ne sont pas petrifies. Ils ont été «momifié» de par leur inclusion dans des débris pyroclastiques et contiennent toujours du matériel du bois primordial. Toutefois, la perte de cellulose préférentiel a rendu le bois très fragile, ce qui exclus l'utilisation de méthodes dendrochronologiques usuelles afin de preparer la surface. Comme pour les charbons de bois archéologiques, en brisant le bois momifié, il est possible de visualiser avec acuité les frontières des arbres-anneau et l'anatomie du bois. Cependant, ceci produit souvent des surfaces irrégulières qui rendant l'examen microscopique difficile. Par conséquent, une pince en aluminium simple a été construite afin de briser des transects radiaux de bois de manière contrôlée, dans le but de recueillir des données dendrochronologiques et des données de l'aspect anatomique du bois, afin d'établir des reconstructions paléoclimatiques. Parce que l'utilisation de traitements chimiques ou de résines de stabilisation n'est pas néccessaire, le fossile de bois reste chimiquement inchangé, ce qui permet des analyses chimiques et isotopiques. Les études futures de bois fragiles peuvent bénéficier de notre méthode de rupture contrôlée si le ponçage est inefficace.