Wood density provides new opportunities for reconstructing past temperature variability from southeastern Australian trees

https://doi.org/10.1016/j.gloplacha.2016.03.010Get rights and content

Highlights

  • 500-year reconstruction of early summer temperatures in Tasmania

  • Density and tree-ring width used to reconstruct high- and low-frequency variability

  • Unusually warm over the past 60 years; consistent with Mt Read reconstruction

  • Warm period likely related to trends in Southern Hemisphere circulation features

Abstract

Tree-ring based climate reconstructions have been critical for understanding past variability and recent trends in climate worldwide, but they are scarce in Australia. This is particularly the case for temperature: only one tree-ring width based temperature reconstruction – based on Huon Pine trees from Mt Read, Tasmania – exists for Australia. Here, we investigate whether additional tree-ring parameters derived from Athrotaxis cupressoides trees growing in the same region have potential to provide robust proxy records of past temperature variability.

We measured wood properties, including tree-ring width (TRW), mean density, mean cell wall thickness (CWT), and tracheid radial diameter (TRD) of annual growth rings in Athrotaxis cupressoides, a long-lived, high-elevation conifer in central Tasmania, Australia. Mean density and CWT were strongly and negatively correlated with summer temperatures. In contrast, the summer temperature signal in TRW was weakly positive. The strongest climate signal in any of the tree-ring parameters was maximum temperature in January (mid-summer; JanTmax) and we chose this as the target climate variable for reconstruction. The model that explained most of the variance in JanTmax was based on TRW and mean density as predictors. TRW and mean density provided complementary proxies with mean density showing greater high-frequency (inter-annual to multi-year) variability and TRW showing more low-frequency (decadal to centennial-scale) variability. The final reconstruction model is robust, explaining 55% of the variance in JanTmax, and was used to reconstruct JanTmax for the last five centuries (1530–2010 C.E.). The reconstruction suggests that the most recent 60 years have been warmer than average in the context of the last ca. 500 years. This unusually warm period is likely linked to a coincident increase in the intensity of the subtropical ridge and dominance of the positive phase of the Southern Annular Mode in summer, which weaken the influence of the band of prevailing westerly winds and storms on Tasmanian climate. Our findings indicate that wood properties, such as mean density, are likely to provide significant contributions toward the development of robust climate reconstructions in the Southern Hemisphere and thus toward an improved understanding of past climate in Australasia.

Introduction

Tree-ring based reconstructions of climate have been critical for understanding past climate variability and placing recent climatic trends in a long-term context. However, for most of the Southern Hemisphere, including Australia, there are few tree-ring based climate reconstructions, which constrains our understanding of recent and potential future climatic changes. In Australia, the limited number of tree-ring based climate reconstructions is largely attributable to a lack of tree species that are suitable for dendrochronology. The dominant angiosperm genera do not generally produce visually distinct or strictly annual rings suitable for dendrochronological methods (Schweingruber, 1992), but recent progress points to the potential some of these genera may hold (Brookhouse and Brack, 2006, Brookhouse et al., 2008, Heinrich et al., 2009, Whinder et al., 2013).

In addition to this, the primary focus of Australian dendrochronology has been on tree-ring width (TRW) – the most commonly used tree-ring parameter in chronology development globally – which has failed to provide clear climatic signals in many of the species and sites examined to date. Recent successes in the reconstruction of past rainfall and/or drought indices have used TRW measurements from the mainland conifer Callitris columellaris (including the previously named C. intratropica, Farjon, 2005) in the north and west of the continent (Cullen and Grierson, 2009, D et al., 2008, O'Donnell et al., 2015). However, to date, the only temperature reconstruction derived from TRW is based on Huon Pine (Lagarostrobos franklinii) from one high-elevation site (Mount Read, 900 m.a.s.l.) in Tasmania (Cook et al., 1991, Cook et al., 1992, Cook et al., 2000). Globally, this is also one of the longest tree-ring based climate reconstructions (1600 B.C.E.–1991 C.E.). Despite considerable efforts over the past two decades, high-quality reconstructions of temperature based on low-elevation Huon Pine TRW (Buckley et al., 1997) and TRW of other long-lived conifer species i.e., Pencil Pine (Athrotaxis cupressoides, Allen et al., 2011) and Celery Top Pine (Phyllocladus aspleniifolius, Allen et al., 1999) in Tasmania have so far been unsuccessful.

Other physical wood properties (i.e., density, cell wall thickness (CWT), tracheid radial diameter (TRD), and microfibril angle (MFA); e.g., Drew et al., 2013) as well as chemical wood properties (i.e., stable isotope concentrations, e.g., Treydte et al., 2006, Brienen et al., 2012, and trace element concentrations, e.g., Poussart et al., 2006) are also known to record climatic information. In particular, strong temperature signals have been identified in the wood density of conifers across Europe (e.g., Briffa et al., 2002, Büntgen et al., 2010, Trouet, 2014) and North America (e.g., Briffa et al., 1992, D et al., 1992, Davi et al., 2003, Luckman and Wilson, 2005, Wilson et al., 2007). In many cases, wood density, particularly maximum latewood density (MXD), has been more strongly correlated with temperature and over a longer summer season than TRW (Briffa et al., 2002, Grudd, 2008, Tuovinen et al., 2009, Trouet et al., 2012). Consequently, MXD has been widely used in the Northern Hemisphere to build temperature-sensitive chronologies (e.g., Schweingruber and Briffa, 1996, Frank and Esper, 2005, Grudd, 2008) and to reconstruct summer temperatures over the last several centuries to millennia (e.g., Briffa et al., 1992, Luckman and Wilson, 2005). Recent work has also demonstrated the potential of various wood properties (e.g., CWT, TRD, MFA, and mean ring density) of several long-lived Tasmanian conifers as sources of past climatic information (Allen et al., 2012, Drew et al., 2013, Allen et al., 2013) and for reconstructing stream flow (Allen et al., 2015). Despite this potential, temperature reconstructions based on these wood properties have not yet been developed.

Here, we investigate the potential of several of these wood properties for reconstructing past temperatures in Australia. We focus on the native conifer, Athrotaxis cupressoides (Pencil Pine), which is endemic to high-elevation (700–1300 m.a.s.l.) areas of Tasmania (Farjon, 2005). In addition to TRW, we measured mean density, TRD, and CWT. Given the strength of climatic signals previously identified in these wood properties, we expect that chronologies based on wood properties, particularly mean density, will allow us to produce the first robust A. cupressoides-based temperature reconstruction in Tasmania.

Section snippets

Site description

We collected A. cupressoides samples at two high elevation sites (~ 1200 m.a.s.l.) in central Tasmania, Australia (41.742°S, 146.703°E; Fig. 1a). The first site is adjacent to Pine Lake. The other site is on a southwest-facing slope adjacent to Mickey Creek. These two sites are ca. 1.5–2 km apart on Tasmania’s Central Plateau. The Pine Lake-Mickey Creek (PLMC) site is approximately 100 km east of the Mt Read site (~ 900 m.a.s.l.) that was sampled to develop the only existing tree-ring based

Chronology and model development

Both the mean density and CWT chronologies were 655 years long (1355–2010 C.E.). CWT was calculated from mean density and the mean density and CWT time series were strongly correlated over the full period of overlap (r = 0.91; p < 0.001). Both the mean density and CWT chronologies showed high common signal strength over the length of the chronology (RBAR > 0.3; for mean density Fig. 3a–b; CWT data not shown). EPS values for the mean density chronology were higher than the 0.85 threshold from 1489 C.E.

Wood density as a proxy for temperature in southeastern Australia

Wood density contains a strong inter-annual climate signal, but failed to reproduce a long-term trend evident in the observed temperature. The absence of a long-term trend in the mean density chronology is largely attributable to a series of three years in the last decade (2005, 2006, 2007) when mean density values were much higher than average, albeit not extreme. However, if mean density is used as the sole predictor of temperature, the resulting reconstruction underestimates observed

Conclusions

Our findings highlight the potential to use the complementary climate signals in TRW and mean density to provide a more complete picture of inter-annual to centennial-scale variation in past temperatures in southeastern Australia. The relative weakness of the TRW signal alone meant, until now, it has not been possible to reconstruct temperatures from A. cupressoides. Wood density provides new opportunities for successful temperature reconstructions in southeastern Australia from tree species in

Acknowledgements

This research was funded by an Australian Research Council Discovery Project grant (DP120104320 to PJB). We are grateful to Michael Goddard for assistance in preparing core samples for analysis, Scott Nicholls for assistance in preparing and analysing samples, and the participants of the Dendroclimatology Masterclass as part of WorldDendro2014: Anders Brundin, Binod Dawadi, Nathan English, Maarit Kalela-Brundin, Robert Kennedy, Kathelyn Paredes, and Meritxell Ramirez-Olle. We are also grateful

References (86)

  • K.J. Allen et al.

    The potential to reconstruct broadscale climate indices associated with southeast Australian droughts from Athrotaxis species, Tasmania.

    Clim. Dyn.

    (2011)
  • G.F. Antonova et al.

    Effects of environmental factors on wood formation in larch (Larix sibirica Ldb.) stems

    Trees

    (1997)
  • G.F. Antonova et al.

    Effects of environmental factors on wood formation in Scots pine stems

    Trees

    (1993)
  • Australian Bureau of Meteorology, 2008. Australian Climate Influences [WWW Document]. URL...
  • R.J.W. Brienen et al.

    Oxygen isotopes in tree rings are a good proxy for Amazon precipitation and El Niño-Southern Oscillation variability

    Proc. Natl. Acad. Sci.

    (2012)
  • K. Briffa et al.

    Tree-ring density reconstructions of summer temperature patterns across western North America since 1600

    J. Clim.

    (1992)
  • K.R. Briffa et al.

    Tree-ring width and density data around the Northern Hemisphere: part 1, local and regional climate signals

    The Holocene

    (2002)
  • M. Brookhouse et al.

    Crossdating and analysis of eucalypt tree rings exhibiting terminal and reverse late wood

    Trees

    (2006)
  • M. Brookhouse et al.

    The potential of tree rings in Eucalyptus pauciflora for climatological and hydrological reconstruction

    Geogr. Res.

    (2008)
  • B.M. Buckley et al.

    A Changing Temperature Response With Elevation for Lagarostrobos franklinii in Tasmania, Australia, in: Climatic Change at High Elevation

    (1997)
  • U. Büntgen et al.

    Diverse climate sensitivity of Mediterranean tree-ring width and density

    Trees

    (2010)
  • W. Cai et al.

    SAM and regional rainfall in IPCC AR4 models: can anthropogenic forcing account for southwest Western Australian winter rainfall reduction?

    Geophys. Res. Lett.

    (2006)
  • E.R. Cook

    A time-series analysis approach to tree-ring standardization

    (1985)
  • E. Cook et al.

    Climatic change over the last millennium in Tasmania reconstructed from tree-rings

    The Holocene

    (1992)
  • E. Cook et al.

    Climatic Change in Tasmania inferred from a 1089-year tree-ring chronology of Huon Pine

    Science

    (1991)
  • E.R. Cook et al.

    Warm-season temperatures since 1600 BC reconstructed from Tasmanian tree rings and their relationship to large-scale sea surface temperature anomalies

    Clim. Dyn.

    (2000)
  • E.R. Cook et al.

    Methods of Dendrochronology — Applications in the Environmental Sciences

    (1990)
  • Cook, E.R., Krusic, P.J., n.d. Program ARSTAN, A Tree-ring Standardization Program Based on Detrending and...
  • E.R. Cook et al.

    Drought reconstructions for the continental United States*

    J. Clim.

    (1999)
  • E.R. Cook et al.

    Calculating unbiased tree-ring indices for the study of climatic and environmental change

    The Holocene

    (1997)
  • L.E. Cullen et al.

    Multi-decadal scale variability in autumn-winter rainfall in south-western Australia since 1655 AD as reconstructed from tree rings of Callitris columellaris (Cupressaceae)

    Clim. Dyn.

    (2009)
  • R. D'Arrigo et al.

    Experimental reconstruction of monsoon drought variability for Australasia using tree rings and corals

    Geophys. Res. Lett.

    (2008)
  • R. D'Arrigo et al.

    Tree-ring width and maximum latewood density at the North American tree line: parameters of climatic change

    Can. J. For. Res.

    (1992)
  • M.P. Denne

    Temperature and tracheid development in Pinus sylvestris seedlings

    J. Exp. Bot.

    (1971)
  • D.M. Drew et al.

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

    Tree Physiol.

    (2013)
  • W. Drosdowsky

    The latitude of the subtropical ridge over eastern Australia: the L index revisited

    Int. J. Climatol.

    (2005)
  • R. Evans

    Rapid measurement of the transverse dimensions of tracheids in radial wood sections from Pinus radiata

    Holzforschung

    (1994)
  • A. Farjon

    A Monograph of Cupressaceae and Sciadopitys

    (2005)
  • A.O. Fierro et al.

    Links between central west Western Australian rainfall variability and large-scale climate drivers

    J. Clim.

    (2013)
  • J.H. Friedman

    A Variable Span Smoother

  • H.C. Fritts

    Tree Rings and Climate

    (1976)
  • N.P. Gillett et al.

    Regional climate impacts of the Southern Annular Mode

    Geophys. Res. Lett.

    (2006)
  • T. Gouhier

    biwavelet: conduct univariate and bivariate wavelet analyses. R Package Version 0.17.04.

  • Cited by (13)

    • Characteristics of a multi-species conifer network of wood properties chronologies from Southern Australia

      2022, Dendrochronologia
      Citation Excerpt :

      Generally stronger signals in anatomical wood parameters may be related to these smaller elements within a ring capturing climate information at the time of their formation whereas total ring width can only reflect the growing season as whole (Martin-Benito et al., 2012; Ziaco et al., 2016; Guada et al., 2021). Chronologies based on these properties have formed the foundation of an increasing number of regional climate reconstructions (e.g. Briffa et al., 2002b; Panyushkina et al., 2003; Grudd, 2008; O’Donnell et al., 2016; Pritzkow et al., 2016; Ziaco et al., 2016; Rydval et al., 2017; Souto-Herrero et al., 2017; Buckley et al., 2018; Wilson et al., 2019; Linderholm et al., 2015; Allen et al., 2015, 2018 amongst others). Despite rapid progress in the development and understanding of these proxies, thus far the majority of work has been confined to tree species in the Northern Hemisphere.

    • Palaeoclimate potential of New Zealand Manoao colensoi (silver pine) tree rings using Blue-Intensity (BI)

      2020, Dendrochronologia
      Citation Excerpt :

      The RW chronology produced by D’Arrigo et al. (1998) achieved a maximum correlation coefficient of r = 0.27 against November-April regional and local data, a coefficient that EBI, LBI and MXBI all exceeded here (r = (-0.42) to (-0.54) before first-differencing, and (-0.61) to (-0.70) after; albeit in the narrower climatic window of December-February (Tables 3–4). This increased ability to capture summer temperatures mimics findings from the NH (Wilson et al., 2014, 2017a; Fuentes et al., 2018) and suggests similar success could be achieved in the SH, particularly from species that have documented relationships between density and temperature such as Anthrotaxis cupressoides (Allen et al., 2012; O’Donnell et al., 2016), Lagarostrobos franklinnii (Drew et al., 2013) and Phyllocladus aspleniifolius (Allen et al., 2012) from Tasmania and Halocarpus biformis (Xiong et al., 1998) from New Zealand. Further research into the differing ecophysiology of SH and NH conifers and pine trees could clarify some of our other results.

    View all citing articles on Scopus
    View full text