Separating the climatic signal from tree-ring width and maximum latewood density records

We propose a technique for separating the climatic signal which is contained in two tree-ring parameters widely used in dendroclimatology. The method is based on the removal of the relationship between tree-ring width (TRW) and maximum latewood density (MXD) observed for narrow tree rings from high latitudes. The new technique is tested on data from three larch stands located along the northern timberline in Eurasia. Correlations were calculated between the temperatures of pentads (five consecutive days), TRW chronologies and MXD chronologies calculated according to the standard and proposed methods. The analysis confirms the great importance of summer temperature for tree radial growth and tree-ring formation. TRW is positively correlated with the temperature of four to eight pentads (depending on the region) at the beginning of the growth season, but MXD as obtained by the standard technique is correlated with temperature over a much longer period. For maximum density series from which the relationship between MXD and TRW has been removed (MXD′), there is a clear correlation with temperatures in the second part of the growing season. These results are consistent with the known dynamics of tree-ring growth in high latitudes and mechanisms of tree-ring formation.     

A tree-ring densitometric transect from Alaska to Labrador

We describe a recently completed network of densitometric tree-ring time series representing various aspects of tree-growth for up to 200 years at 69 sites spread across the northern North American conifer zone from Yukon to Labrador. Duplicate cores, from 12 to 15 trees per site, provide time series for a suite of growth parameters including earlywood (spring), latewood (summer) and total (annual) ring widths and mean earlywood, mean latewood, minimum and maximum ring density. These data form the basis for extensive analyses of intra- and inter-site parameter comparisons and regional climate/tree-growth comparisons. Five large-scale regional chronologies do not suggest that any anomalous growth increases have occurred in recent decades, at least on these regional scales, despite the observed changes in atmospheric composition and climate.     

Yearly maps of summer temperatures in Western Europe from A.D. 1750 to 1975 and Western North America from 1600 to 1982:

The primary purpose of this publication is to make available two series of historical maps showing summer (April–September mean) temperatures over western Europe from A.D. 1750 and western North America from A.D. 1600. These maps have been reconstructed using networks of temperature-sensitive tree-ring chronologies as part of a continuing collaboration between the Swiss Federal Institute for Forest, Snow and Landscape Research (FSL) at Birmensdorf, Switzerland and the Climatic Research Unit (CRU) at the University of East Anglia, Norwich, U.K. The maps were reconstructed using relatively long maximum latewood density chronologies selected from the extensive networks of tree-ring data produced at the FSL. The reconstructions were produced using a spatial regression technique developed at CRU.The basis for the reconstructions are wood samples from various conifers growing at cool-moist sites close to the alpine or boreal timberlines. This material was analysed by x-ray densitometry and the resulting maximum latewood density chronologies were calibrated against meteorological data over a 100-year calibration period.These results emphasize the value of dendroclimatological research which can provide continentalscale reconstructions of past climate which represent a potential resource for climatologists, biologists, and historians with an interest in the spatial variability of summer temperatures over the middle to high latitudes.     

Prominent role of volcanism in Common Era climate variability and human history

Climate reconstructions for the Common Era are compromised by the paucity of annually-resolved and absolutely-dated proxy records prior to medieval times. Where reconstructions are based on combinations of different climate archive types (of varying spatiotemporal resolution, dating uncertainty, record length and predictive skill), it is challenging to estimate past amplitude ranges, disentangle the relative roles of natural and anthropogenic forcing, or probe deeper interrelationships between climate variability and human history. Here, we compile and analyse updated versions of all the existing summer temperature sensitive tree-ring width chronologies from the Northern Hemisphere that span the entire Common Era. We apply a novel ensemble approach to reconstruct extra-tropical summer temperatures from 1 to 2010 CE, and calculate uncertainties at continental to hemispheric scales. Peak warming in the 280s, 990s and 1020s, when volcanic forcing was low, was comparable to modern conditions until 2010 CE. The lowest June–August temperature anomaly in 536 not only marks the beginning of the coldest decade, but also defines the onset of the Late Antique Little Ice Age (LALIA). While prolonged warmth during Roman and medieval times roughly coincides with the tendency towards societal prosperity across much of the North Atlantic/European sector and East Asia, major episodes of volcanically-forced summer cooling often presaged widespread famines, plague outbreaks and political upheavals. Our study reveals a larger amplitude of spatially synchronized summer temperature variation during the first millennium of the Common Era than previously recognised.     

The dendroclimatic value of oak stable isotopes

Tree-ring stable oxygen and carbon isotope ratios (δ¹⁸O and δ¹³C) are an important archive for climate reconstructions. However, it remains unclear whether the polyvinyl acetate emulsion, often used for the preservation and fixation of wood samples, influences δ¹⁸O and δ¹³C signals. Further uncertainties are associated with the possible effects of geographical origin and cambial age of historical samples. Here, we present annually-resolved and absolutely-dated δ¹⁸O and δ¹³C measurements of 21 living oaks (Quercus robur and Q. petraea) from the Czech Republic. We find that the δ¹⁸O and δ¹³C signals in the extracted alpha-cellulose are not affected by polyvinyl acetate treatment. Covering the entire 20^th century and reaching until 2018 CE, our dataset reveals spatial and temporal coherency within and between the individual δ¹⁸O and δ¹³C chronologies of different oak species, sample locations, and tree ages. Highly significant (p < 0.01) Pearson’s correlation coefficients of the site-specific δ¹³C and δ¹⁸O chronologies range from 0.48–0.77 and 0.36–0.56, respectively. The isotopic inter-series correlations of Q. robur and Q. petraea from the same site are 0.75 and 0.43 for the mean δ¹³C and δ¹⁸O values, respectively. Significant (p < 0.01) correlations of 0.49 and 0.84 are found for δ¹³C and δ¹⁸O, respectively, when all measurements from all sampling locations and tree ages are included. Our study shows that non-pooled oak δ¹⁸O and δ¹³C measurements from both species, different locations, and diverse tree ages can be combined into robust isotopic chronologies for climate reconstructions.     

A hydroclimatic regionalization of central Mongolia as inferred from tree rings

In arid and semi-arid regions of the world, such as Mongolia, the future of water resources under a warming climate is of particular concern. The influence of increasing temperatures on precipitation is difficult to predict because precipitation trends in coming decades could have a high degree of spatial variability. In this study, we applied a rotated principal component analysis (RPCA) to a network of 20 tree-ring chronologies across central Mongolia from 1790 to 1994 to evaluate spatial hydroclimatic variability and to place recent variability in the context of the past several centuries. The RPCA results indicate that the network consists of four tree-growth anomaly regions, which were found to be relatively stable through time and space. Correlation analyses reveal spatial linkages between the tree-growth anomalies and instrumental data, where annual streamflow variability was strongly associated with tree-growth anomalies from their respective regions from 1959 to 1994 (r = 0.52–0.64, p < 0.05). This study highlights the extent of spatial variability in hydroclimate across central Mongolia and emphasizes the value of using tree-ring networks in locations with limited instrumental records.     

A double bootstrap approach to Superposed Epoch Analysis to evaluate response uncertainty

The association between climate variability and episodic events, such as the antecedent moisture conditions prior to wildfire or the cooling following volcanic eruptions, is commonly assessed using Superposed Epoch Analysis (SEA). In SEA the epochal response is typically calculated as the average climate conditions prior to and following all event years or their deviation from climatology. However, the magnitude and significance of the inferred climate association may be sensitive to the selection or omission of individual key years, potentially resulting in a biased assessment of the relationship between these events and climate. Here we describe and test a modified double-bootstrap SEA that generates multiple unique draws of the key years and evaluates the sign, magnitude, and significance of event-climate relationships within a probabilistic framework. This multiple resampling helps quantify multiple uncertainties inherent in conventional applications of SEA within dendrochronology and paleoclimatology. We demonstrate our modified SEA by evaluating the volcanic cooling signal in a Northern Hemisphere tree-ring temperature reconstruction and the link between drought and wildfire events in the western United States. Finally, we make our Matlab and R code available to be adapted for future SEA applications.     

Different climate response of three tree ring proxies of Pinus sylvestris from the Eastern Carpathians,Romania

The aim of this study was to compare the climatic responses of three tree rings proxies: tree ring width (TRW), maximum latewood density (MXD), and blue intensity (BI). For this study, 20 cores of Pinus sylvestris covering the period 1886–2015 were extracted from living non-damaged trees from the Eastern Carpathian Mountains (Romania). Each chronology was compared to monthly and daily climate data. All tree ring proxies had a stronger correlation with the daily climate data compared to monthly data. The highest correlation coefficient was obtained between the MXD chronology and daily maximum temperature over the period beginning with the end of July and ending in the middle of September (r = 0.64). The optimal intervals for the temperature signature were 01 Aug – 24 Sept for the MXD chronology, 05 Aug – 25 Aug for the BI chronology, and both 16 Nov of the previous year – 16 March of the current year and 15 Apr – 05 May for the TRW chronology. The results from our study indicate that MXD can be used as a proxy indicator for summer maximum temperature, while TRW can be used as a proxy indicator for just March maximum temperature. The weak and unstable relationship between BI and maximum temperature indicates that BI is not a good proxy indicator for climate reconstructions over the analysed region.     

Temporal instability in tree-growth/climate response in the Lower Bavarian Forest region: implications for dendroclimatic reconstruction

This paper explores the temporal stability of growth/climate relationships in ring-width chronologies of Norway spruce [ Picea abies (L.) Karst] and silver fir ( Abies alba Mill) in the Lower Bavarian Forest region in southern Germany. These chronologies were compiled, using both historic and living tree-ring data, with the main aim of developing a dendroclimatic reconstruction for the region covering the last 500 years. Moving window correlation analysis shows that prior to the twentieth century, both species co-vary in a similar way (1480–1899 mean r =0.66). There is no significant correlation between the species chronologies since ca. 1930, which partly reflects anomalous growth trends in the fir chronology since ca. 1960. Multiple regression analysis was utilised to assess the ability of both species chronologies to model March–August precipitation. The precipitation signal of the spruce data was found to be both stronger than the fir data (1872–1930 calibration: r ²=0.45 vs 0.25) and more time stable. After ca. 1930, the fir chronology loses it ability to model March–August precipitation until there is no climate signal at all in the fir data in recent decades. The spruce data also express a later weakening in their climate signal in the mid 1970s. We present compelling evidence indicating that the anomalous trends observed in the fir data, since the mid 1960s, appear to be predominantly related to local SO₂ emissions from power plants and refineries. It is also likely that this local anthropogenic forcing is the cause of the weakening of the climate signal in the spruce data since the mid 1970s. The conclusions from this study are: (1) The fir tree-ring data cannot be used for traditional dendroclimatic calibration, although prior to the twentieth century the decadal variability in the fir data is very similar to spruce and so these data could be used to extend potential reconstructions in the future; (2) The recent decline and recovery event in the fir data appears to be unique to the twentieth century and is not part of a natural episodic phenomenon; (3) Traditional dendroclimatic calibration of March–August precipitation will be made using solely the spruce ring-width data. However, due to SO₂ forcing in recent decades, the calibration period will be shortened to the 1871–1978 period.