Kathryn Allen

Wood properties in a long-lived conifer reveal strong climate signals where ring-width series do not

Although tree-ring-width chronologies have been widely used for temperature reconstructions, there are many sites around the world at which there is little evidence of a clear climate signal in the ring-width chronologies. This is the case with the long-lived conifer Huon pine (Lagarostrobos franklinii (Hook. F.) Quinn), endemic to Tasmania, Australia, when the species grows at low elevation. In this study, we developed chronologies of several wood properties (e.g., tracheid radial diameter, microfibril angle) from Huon pine growing at a low-elevation site. We found that despite the absence of a climate signal in the ring-width chronologies, there were significant correlations between wood density, tracheid radial diameter and microfibril angle and temperature, stream flow and a drought index, enabling the development of robust chronologies. This novel finding suggests that chronologies based on these wood properties may have important potential for climate reconstructions from sites and species that have not yet been realized. In particular, a relatively extensive resource of ancient, low-elevation Huon pine in western Tasmania, in which climate signals have not been found using ring widths, may now be useful as part of the broader effort to reconstruct Southern Hemisphere climate.

A global multiproxy database for temperature reconstructions of the Common Era

Reproducible climate reconstructions of the Common Era (1 CE to present) are key to placing industrial-era warming into the context of natural climatic variability. Here we present a community-sourced database of temperature-sensitive proxy records from the PAGES2k initiative. The database gathers 692 records from 648 locations, including all continental regions and major ocean basins. The records are from trees, ice, sediment, corals, speleothems, documentary evidence, and other archives. They range in length from 50 to 2000 years, with a median of 547 years, while temporal resolution ranges from biweekly to centennial. Nearly half of the proxy time series are significantly correlated with HadCRUT4.2 surface temperature over the period 1850–2014. Global temperature composites show a remarkable degree of coherence between high- and low-resolution archives, with broadly similar patterns across archive types, terrestrial versus marine locations, and screening criteria. The database is suited to investigations of global and regional temperature variability over the Common Era, and is shared in the Linked Paleo Data (LiPD) format, including serializations in Matlab, R and Python.

Validating putatively cross-dated Callitris tree-ring chronologies using bomb-pulse radiocarbon analysis

Nuclear weapons testing during the 1950s generated an atmospheric pulse of the carbon isotope, 14C. Worldwide, trees growing during that period and in subsequent decades assimilated 14C-enriched CO2, leaving a distinct isotopic signature that can be used to precisely date tree rings. Thirty single-ring samples were extracted for AMS 14C analysis from cores taken from living trees of five different Callitris species [C. endlicheri (Parl.) F.M. Bailey, C. glaucophylla Joy Thomps. & L.A.S. Johnson, C. intratropica Benth., C. preissii Miq., and C. rhomboidea R.Br. ex Rich. & A. Rich] at 13 sites. The ages of individual tree rings were determined by both 14C bomb-pulse dating and cross-dating (based on 20–30 cores from the same site) in order to (1) provide independent verification of tree-ring dates, (2) detect false or missing rings from sites with otherwise good chronologies, and (3) test whether growth rings were annual for cores from sites where cross-dating was not possible. Our approach confirmed dates on chronologies from monsoon tropical sites, provided checked chronologies in subtropical and temperate sites, and improved dating control on arid-zone ring counts. It was found that Callitris are more likely to form regular annual rings when growing in seasonally dry environments than in more arid sites with highly variable precipitation patterns.

Preliminary December–January inflow and streamflow reconstructions from tree rings for western Tasmania, southeastern Australia

Projected decreases and changes in the seasonal distribution of precipitation will have profound impacts on southeastern Australia, including its ability to generate renewable hydroelectricity. Recent decreases in precipitation over the region may be significant in the context of instrumental records, but the question of whether these decreases are within long-term natural variability remains. To help address this issue, we present December–January streamflow and dam inflow reconstructions for southeastern Australia. These reconstructions for the Tasmanian west coast are based solely on local tree ring chronologies and span up to 1600 years. Nonparametric estimates, however, indicate good model skill for the last 458 years (streamflow) and 478 years (dam inflow). The reconstructions indicate that twentieth century conditions were well within the range of historical variability, and were in fact relatively wet. The period from approximately 1600 to 1750 CE was one of the enhanced variability and a high proportion of low and high flow events occurred in the seventeenth century. There are significant relationships between streamflow and inflow reconstructions and large-scale ocean-atmosphere processes such as ENSO and the Southern Annular Mode. Critically, our two reconstructions rely heavily on new tree ring chronologies based on properties such as tracheid radial diameter, cell wall thickness, and density, underscoring the importance of these different types of chronologies in reconstructions.

The Temperature Response in the Ring Widths of Phyllocladus Aspleniifolius (Celery–top Pine) Along an Altitudinal Gradient in the Warra LTER Area, Tasmania

The temperature response of four Phyllocladus aspleniifolius (Celery-top Pine) sites along an altitudinal gradient within a cool temperate broad leaf forest environment in the southwest of Tasmania, Australia was examined. Although strong evidence of a systematically changing response with elevation could not be found, there was evidence that minimum temperature in particular may be important in determining the altitudinal extent of the species. Climatic responses of the Warra LTER (Long Term Ecological Research) area sites were representative of other known sites in southwestern Tasmania. A link between event years in P. aspleniifolius and warm/dry conditions indicates that these event years may provide a guide to the historical frequency of fire weather in Tasmania’s southwest.

Fire frequency variation in south-eastern Tasmanian dry eucalypt forest 1740-2004 from fire scars

Summary An understanding of fire history is important in determining appropriate fire management regimes for biodiversity conservation in fire-prone ecosystems, such as the dry eucalypt forests of temperate Australia. We tested whether ring counts and evidence of fire in the stumps of felled eucalypts could be used to construct fire chronologies in the dry forests of south-eastern Tasmania. Given that the dates of fires derived from this method were consistent with other evidence of fire years, we constructed chronologies for 13 sites in the region. We applied a conversion factor for fires per decade per site based on the relationship between fire detection and sample size for all sites. Between 1740 and 1819 when indigenous people were managing the region, decadal fire frequency averaged 0.7. Between 1820 and 1849 fires were very infrequent in the region, with a mean decadal fire frequency of 0.4. An upward transition to a higher fire frequency took place between the 1840s and 1850s. Between 1850 and 1909 decadal fire frequency varied between 0.8 and 1.2 then sharply increased again. Between 1910 and 1989 it varied between 1.3 and 1.7. Extensive fire years were strongly related to annual precipitation <0.75 standard deviations below the mean. Variation in annual precipitation, however, could not explain the sharp transitions in decadal fire frequency that took place in 1820, 1850, 1910 and 1990 and the constancy of fire frequency between these dates. The relationships of these transitions to land use changes are described.

A structural time series approach to the reconstruction of Tasmanian maximum temperatures

Structural time series modelling was applied to the problem of reconstructing maximum temperatures from three regional Phyllocladus aspleniifolius tree-ring data sets from Tasmania, Australia. The resulting maximum temperature reconstructions span the past 190 years, and show considerable improvement over reconstructions produced by the more traditional technique of Principal Component regression. Reconstructions covering the period 1810‐1990 reveal increased amplitude in all three time series in the 20th as compared to the 19th century. Consistency between reconstructions also decreases in the 20th century. Such changes are suggestive of differences in atmospheric circulation patterns of the two centuries. At a general level similarities between the Phyllocladus aspleniifolius and the much publicised Lagarostrobos franklinii reconstructions are apparent. However, some significant differences exist between the two, and are most likely attributable to biological differences between species and to the higher elevation of Lagarostrobos franklinii tree-ring sites.  1999 Published by Elsevier Science Ltd. All rights reserved.

The paleoclimate context and future trajectory of extreme summer hydroclimate in eastern Australia

Eastern Australia recently experienced an intense drought (Millennium Drought, 2003-2009) and record-breaking rainfall and flooding (austral summer 2010-2011). There is some limited evidence for a climate change contribution to these events, but such analyses are hampered by the paucity of information on long-term natural variability. Analyzing a new reconstruction of summer (December-January-February) Palmer Drought Severity Index (the Australia-New Zealand Drought Atlas; ANZDA, 1500-2012 CE), we find moisture deficits during the Millennium Drought fall within the range of the last 500 years of natural hydroclimate variability. This variability includes periods of multi-decadal drought in the 1500s more persistent than any event in the historical record. However, the severity of the Millennium Drought, which was caused by autumn (March-April-May) precipitation declines, may be underestimated in the ANZDA because the reconstruction is biased towards summer and antecedent spring (September-October-November) precipitation. The pluvial in 2011, however, which was characterized by extreme summer rainfall faithfully captured by the ANZDA, is likely the wettest year in the reconstruction for Coastal Queensland. Climate projections (RCP 8.5 scenario) suggest that eastern Australia will experience long-term drying during the 21st century. While the contribution of anthropogenic forcing to recent extremes remains an open question, these projections indicate an amplified risk of multi-year drought anomalies matching or exceeding the intensity of the Millennium Drought.

A 277 year cool season dam inflow reconstruction for Tasmania, southeastern Australia

Seasonal variability is a significant source of uncertainty in projected changes to precipitation across southeastern Australia (SEA). While existing instrumental records provide seasonal data for recent decades, most proxy records (e.g., tree rings, corals, speleothems) offer only annual reconstructions of hydroclimate. We present the first cool-season (July–August) reconstruction of dam inflow (Lake Burbury) for western Tasmania in SEA based on tree-ring width (Athrotaxis selaginoides) and mean latewood cell wall thickness (Phyllocladus aspleniifolius) chronologies. The reconstruction, produced using principal component regression, verifies back to 1731 and is moderately skillful, explaining around 23% of the variance. According to the reconstruction, relatively low inflow periods occurred around 1860, the early 1900s and 1970, while relatively high inflows occurred in the 1770s and 1810s. Highest reconstructed inflows occurred in 1816, and lowest in 1909. Comparison with available documentary and instrumental records indicates that the reconstruction better captures high rather than low flow events. There is virtually no correlation between our reconstruction and another for December–January inflow for the same catchment, a result consistent with the relationship between seasonal instrumental data. This suggests that conditions in one season have not generally reflected conditions in the other season over the instrumental record, or for the past 277 years. This illustrates the value of obtaining reconstructions of regional hydroclimatic variability for multiple individual seasons in regions where dry and wet seasons are not strongly defined. The results also indicate that the hydroclimate of the southeastern Australian region cannot be adequately represented by a single regional reconstruction.

Data Descriptor: A global multiproxy database for temperature reconstructions of the Common Era

PAGES, a core project of Future Earth, is supported by the U.S. and Swiss National Science Foundations. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Some of this work was conducted as part of the North America 2k Working Group supported by the John Wesley Powell Center for Analysis and Synthesis, funded by the U.S. Geological Survey. B. Bauer, W. Gross, and E. Gille (NOAA National Centers for Environmental Information) are gratefully acknowledged for helping assemble the data citations and creating the NCEI versions of the PAGES 2k data records. We thank all the investigators whose commitment to data sharing enables the open science ethos embodied by this project.