• climate change;
  • dendrochronology;
  • ecological collapse;
  • Scandinavia;
  • serial sectioning;
  • temperature warming;
  • tree-ring standardization

Hallinger et al. (2010) suggest that Scandinavian summer warming in recent decades was strong enough to trigger an altitudinal and northward expansion of shrubs. The authors introduce juniper (Juniperus nana Willd.) ring-width measurements taken along an elevational transect in northwest Sweden (Abisko), apply traditional dendrochronological methods (serial sectioning, cross-dating, chronology development, response functioning) and argue for a distinct link between rising summer temperatures and increasing radial and vertical growth rates. They postulate a northward expansion of shrubs since the 1970s.

Their findings and similar studies enter the zeitgeist and give birth to popular notions of ‘ecological collapse’. Putative environmental responses to climate variations have drawn the attention of ecologists, biologists, epidemiologists and even economists as a result of their effects on the dynamic structure of both terrestrial and marine ecosystems and the natural resources therein (Stenseth et al., 2002). Reported ecological consequences of a recent Scandinavian temperature rise include: dampening of population cycles (Ims et al., 2008; Kausrud et al., 2008), shifts in species ranges and treelines (Kullman, 2001, 2002; Beaugrand et al., 2002; Parmesan & Yohe, 2003; Thuiller, 2004), changes in phenology and match–mismatch relationships (Beaugrand et al., 2003; Kauserud et al., 2008), amplification of trophic interaction (Edwards & Richardson, 2004; Kirby & Beaugrand, 2009), increases in epidemiological diseases (Patz et al., 2005; McMichael et al., 2006), and declines in wildlife productivity and fish stocks (Perry et al., 2005; Mysterud & Ostbye, 2006; Beaugrand et al., 2009).

Here, we address three pitfalls relevant to the conclusions drawn by Hallinger et al. (2010): (1) non-trending Scandinavian summer temperatures, (2) effects of serial sectioning and (3) artificial index inflation.


  1. Top of page
  2. Pitfalls
  3. Conclusions
  4. Acknowledgements
  5. References

(1) Instrumental station measurements provide reliable information on northern Scandinavian temperature variability as far back as the early 19th century (Tornedalen; Klingbjer & Moberg, 2003). The warmest and coldest June–July–August means occurred in 1937 and 1902, respectively. Lines of evidence for significant long-term warming from cooler Little Ice Age to warmer recent summers, however, remain absent. A similar course is obtained from December to February winter means: cooler conditions before c. 1910, warmth between c. 1930 and 1950, cooling from c. 1950 to 1980 and increasing temperatures from the 1980s to present (Fig. 1a). Linear trends of summer and winter temperatures are similarly free of any long-term trends. Independent of the data providers considered (GHCN (Global Historical Climatology Network) and GISS (Goddard Institute for Space Studies)) and the versions used (raw and homogenized), station readings resemble grid-box means (CRUTEM3v; Brohan et al., 2006) of the past 150 yr (Fig. 1a). The Abisko record used in Hallinger et al. correlates at 0.91 with gridded CRU (Climatic Research Unit) summer temperatures back to 1869, and also does not indicate any long-term warming.


Figure 1.  (a) Summer (June–July–August (JJA)) temperature measurements of eight Scandinavian stations (> 65°N) considering raw and homogenized data from the GHCN (Global Historical Climatology Network) and GISS (Goddard Institute for Space Studies) (32 series; orange), the Tornedalen composite back to 1816 (pink), and the gridded CRUTEM3v mean, averaged over 65–70°N and 20–30°E (red). The blue curve shows the gridded December–February winter mean, and corresponding linear trends (1860–2008) are described at the left side. (b) Twenty-five chronologies (green) and their mean (dark green). (c) Reconstructed summer temperatures back to 500 AD with the 20 warmest decades being superimposed (horizontal bars).

Download figure to PowerPoint

To best understand past changes in northern Scandinavian forest growth, we herein utilize a tree-ring network of 1179 series containing c. 190 000 annual maximum latewood density measurements from 25 conifer sites in Norway, Sweden, Finland and Russia at > 65°N. This unique compilation appears to be most representative for vegetation dynamics across boreal Europe. Nonclimatic age trends were removed from the raw density series using Regional Curve Standardization (RCS; Esper et al., 2003). Each of the site records perfectly tracks inter-annual to multi-decadal variations in northern Scandinavian summer temperature (Fig. 1b). Correlation with air temperature back to 1860 is 0.83.

A master chronology of 12 ring-width and density chronologies from Swedish Torneträsk (Briffa et al., 1992; Grudd et al., 2002; Grudd, 2008), central Sweden (Gunnarson et al., 2010), Finnish Lapland (Helama et al., 2009), coastal Norway (Kirchhefer, 2001) and regional-scale networks (Gouirand et al., 2008; Linderholm et al., 2009; this Study) allows Scandinavian temperatures to be reconstructed back to AD 500 (Fig. 1c). Comparably warm summers occurred in the 8th century, between c. 900 and 1200, in the 15th century and again in the early-mid 20th century. The reduced warm/cool/warm amplitude associated with the Medieval Warm Period/Little Ice Age/Anthropogenic Era is indicative of internal ocean–atmosphere coupling and resulting climate inertia strong enough to override external forcing (U. Büntgen et al., unpublished).

(2) Cross-dating constraints of prostrate shrubs that grow slowly, eccentrically and irregularly necessitate the sampling of various disks from different stem heights within a single individual (Kolishchuk, 1990; Woodcock & Bradley, 1994). This method, known as serial sectioning, facilitates the annually precise dating of each ring. Application of serial sectioning is particularly important when annual growth rings are characterized by low visibility and poor measurability of discontinuous and even missing rings, as well as by the high internal growth variability of (dwarf) shrubs (Schweingruber & Dietz, 2001). Hallinger et al. successfully applied the serial section technique in their study. They sampled 4–12 disks at a distance of 10 cm along each stem, measured 2–4 radii per disk, averaged all radii per sample and finally averaged 6–24 radii per individual plant.

We believe this to be a delicate issue, as the continuous inclusion of additional juvenile wood with increasing stem height might systematically exaggerate the outermost ring-width values. We composed a simple sketch to highlight the principle of, and data derived from, serial sectioning (Fig. 2). Nevertheless, we are aware that our example might not be representative for growth trends at all sites and among all species at high-northern latitudes. From a conceptual perspective, various disks from different stem heights contain different segment lengths and growth rates. Their mean is thus biased towards the wider outermost rings, as radii from higher stem sections contain more juvenile wood. In fact, averaging different measurement series from different stem heights results in an artificial dominance of juvenile growth phases. This effect might even be amplified, because juvenile phases obtained from upper stem sections contain less prolonged periods of suppressed growth, which are commonly reported near the root collar (Esper & Schweingruber, 2004). At the same time, the suppressed juvenile growth rates that characterize the first decades of plant growth near altitudinal and latitudinal distribution limits would impact the radial growth-change assessment performed by Hallinger et al.: division of the arithmetic ring-width mean of the first (suppressed) 25 yr by the last (released) 25 yr of shrub life. Estimates of vertical growth changes further remain questionable because the same data and methods were reconsidered and thus prolonged possible shortcomings.


Figure 2.  Schematic diagram of (a) the serial section technique, (b) the corresponding ring-width radii and (c) the systematically biased mean record when simply averaging the individual radii from different stem heights (ARW, average radii width). It should be noted that this sketch is not necessarily representative for growth trends at all sites and among all species at high-northern latitudes.

Download figure to PowerPoint

(3) An optimal method for removing biological age trends inherent in raw tree-ring width, density and isotopic measurement series before any valid interpretation of external signals (still) does not exist (Briffa et al., 2001; Esper et al., 2003; Helama et al., 2005; Melvin & Briffa, 2008). Crucial for investigating phenomena related to global change is the possible occurrence of ‘end-effect’ issues in the process of tree-ring detrending and chronology development (Frank et al., 2009). The risk of artificial index inflation (as a result of dividing annual growth increments by a standardization curve) increases if the fitted standardization curve decreases towards zero, especially below c. 0.5 mm (1/0.5 = 2.0), and if the synthetic growth function underestimates the absolute measurements (see Cook & Peters, 1997 for details). Adaptive power-transformation can remove heteroscadastic properties by raising time-series to a fractional power, permitting differencing (instead of dividing) of the power-transformed growth values from the estimated standardization curves (Cook & Peters, 1997). The expected effect in the data used by Hallinger et al. might be even more severe, as their average ring widths range between 0.050 and 0.370 mm (at 800 and 1100 m asl), small enough to boost recent ring-width indices when calculating ratios instead of the residuals after power-transformation (see Cook & Peters, 1997).


  1. Top of page
  2. Pitfalls
  3. Conclusions
  4. Acknowledgements
  5. References

(1) Cooler northern Scandinavian temperatures offset large-scale warming trends. The warmest summer in 1937 and a missing long-term trend from generally cooler Little Ice Age to warmer recent conditions, bring into question the link between a recent increase in radial and vertical growth of northern European shrubs. At the same time, Scandinavian forest growth perfectly tracks variations in summer temperature, and contemporary ring-width and density rates range below earlier highs. We believe that peak warming in the 1930s was strong enough to trigger the establishment of new shrub seedlings. This early germination phase would subsequently explain the average shrub age of c. 80 yr observed by Hallinger et al., indicating favourable colonization conditions between c. 1920 and 1940. This early-to-mid 20th century recruitment phase was also characterized by warm winters, and further coincides with the average germination time of dwarf shrubs in northern Norway (Bär et al., 2007) and sapling germination across western Siberia (Esper & Schweingruber, 2004). Further evidence of new seedlings associated with early-to-mid 20th century warming is reported from central Sweden (Kullman, 2001), northern Quebec (Payette & Filion, 1985) and the Polar Urals (Shiyatov, 1993). The establishment and survival of seedlings has been described as a decades-long process of suppressed annual increments before an abrupt growth release finally occurs (Esper & Schweingruber, 2004). As mortality of saplings in the harsh northern environment is high, only the fittest individuals will survive (see Schweingruber & Poschlod, 2005 for a dendroecological review of herbs and shrubs). Additional sampling of dead shrubs and trees could provide information on the causes and consequences of their dieback.

(2) Serial sectioning describes a useful technique for accurately cross-dating several ring-width measurement series within individual shrubs, characterized by extremely slow, eccentric and abnormal growth rates. Serial sectioning has been mainly applied to the dating of shrubs; however, the risk of ring-width bias is inherent when chronologies are derived from simple averaging radii of different stem heights from the same individual. Artificially inflated raw measurements may have affected not only the results of Hallinger et al., but also possibly the dwarf shrub chronologies developed by Bär et al. (2006, 2007). The consequences of climate change predicted from such measurements may therefore be amplified by the fact that juvenile growth at higher stem positions appears to be less suppressed (Esper & Schweingruber, 2004). To overcome such methodological constraints, we suggest individual standardization of measurement radii from different stem heights before averaging measurements at the shrub/tree level. Subsequent chronology development must be performed on dimensionless indices.

(3) Tree-ring standardization is essential to distinguish between external signals and internal variations. However, it also carries the risk of artificially inflating the indices when annual growth increments are divided by a standardization curve that either decreases towards zero or undershoots the absolute measurements. Power-transformation, in combination with residuals, might help to reduce artificial index value inflation in the tree-ring detrending process. Other ‘end-effect’ issues associated with chronology development processes and data compositions may remain, however (e.g. Büntgen et al., 2005, 2006, 2007, 2008; Melvin & Briffa, 2008; Frank et al., 2009). Nevertheless, we believe that some of the recent growth increases described by Hallinger et al. may be a simple detrending relic, as shrub increments range from 0.01 to 0.3 mm yr−1. This would explain the relationship of stronger growth increases at higher elevations where average ring width is lowest. It should further be noted that Hallinger et al. employed the standardized ring-width series not only to assess radial enlargement but also to extrapolate vertical growth. Optimized correlation between the standardized shrub master-chronology and June–July temperature (= 0.4; 1913–2004) is fairly low in comparison to correlations of > 0.8 obtained from a northern Scandinavian tree-ring density compilation. Subsequent hypotheses on elevational effects on climate–growth relationships thus remain challenging.

We summarize that recent Scandinavian temperatures range within their envelope of natural variability, and that methodological pitfalls may be responsible for some growth increases during the past two-to-three decades. Besides those data- and methodological-induced constraints have more general ecological effects, including topography, micro-climatology, wind conditions, snow cover, soil properties, moisture availability, nutrition supply, mycorrhiza symbiosis, fungal diseases, insect defoliation, browsing pressure and land-use/land-cover change not carefully been discussed by Hallinger et al. (2010). The assumption of a linear relationship between average plant age and elevation, as proposed by Hallinger et al. (2010), lacks scientific grounds. Moreover, we believe that better replicated data sets collected at broader spatial scales would be required to overcome local disturbance factors that probably alter the conclusions reached regarding large-scale ecosystem changes, such as a pan-Arctic shrub expansion postulated by Hallinger et al. (2010). To date, ample evidence suggests caution before establishing causal links between summer warming and vegetation expansion (e.g. Vaganov et al., 1999; Kullman, 2001, 2002; Holtmeier, 2003; Esper & Schweingruber, 2004; Bär et al., 2006, 2007, 2008; Babst et al., 2010). Nevertheless, we are convinced by the dendroecological novelty (and relevance) of studies on nontree woody life forms and clearly see their enormous potential to learn more about vegetation dynamics and the possible impacts of environmental/climate change on ecosystems that cover huge areas beyond the alpine and the arctic tree limits.

Future studies at the interface of climate variability and ecological susceptibility should be particularly cautious to not fuel the ongoing ‘global change’ debate without unambiguous evidence. Premature conclusions jeopardize the long-term reliability of the scientific community as a whole. We believe in the strength of interdisciplinary efforts towards better understanding past, present and future interactions of climate and ecology. We are aware of the importance of the late 20th century, during which most of the globe warmed and for which most of the ecological records contain information (Stenseth et al., 2002). However, we are also concerned about the tendency to which information on climatic variations is often hastily translated in a nexus of environmental threats.


  1. Top of page
  2. Pitfalls
  3. Conclusions
  4. Acknowledgements
  5. References

We thank H. Grudd, B. Gunnarson, H. Linderholm, A. Kirchhefer and S. Helama for making their published tree-ring data available. A. Bräuning, A. Kirdyanov and E. Gleeson provided comments on an earlier version of the manuscript. Supported by the EC project Millennium (grant 017008) and the SNSF project NCCR (grant Extract).


  1. Top of page
  2. Pitfalls
  3. Conclusions
  4. Acknowledgements
  5. References
  • Babst F, Esper J, Parlow E. 2010. Landsat TM/ETM+ and tree-ring based assessment of spatiotemporal patterns of the autumnal moth (Epirrita autumnata) in northernmost Fennoscandia. Remote Sensing of Environment 114: 637646.
  • Bär A, Bräuning A, Löffler J. 2006. Dendroecology of dwarf shrubs in the high mountains of Norway – A methodological approach. Dendrochronologia 24: 1727.
  • Bär A, Pape R, Bräuning A, Löffler J. 2007. Ring-width chronologies of the Alpine dwarf shrub Empetrum hermaphroditum from the Norwegian Mountains. IAWA Journal 28: 325338.
  • Bär A, Pape R, Bräuning A, Löffler J. 2008. Growth-ring variations of dwarf shrubs reflect regional climate signals in alpine environments rather than topoclimatic differences. Journal of Biogeography 35: 625636.
  • Beaugrand G, Brander KM, Lindley JA, Souissi S, Reid PC. 2003. Plankton effect on cod recruitment in the North Sea. Nature 426: 661664.
  • Beaugrand G, Luczak C, Edwards M. 2009. Rapid biogeographical plankton shifts in the North Atlantic Ocean. Global Change Biology 15: 17901803.
  • Beaugrand G, Reid PC, Ibanez F, Lindley JA, Edwards M. 2002. Reorganization of North Atlantic marine copepod biodiversity and climate. Science 296: 16921694.
  • Briffa KR, Jones PD, Bartholin TS, Eckstein D, Schweingruber FH, Karlén W, Zetterberg P, Eronen M. 1992. Fennoscandian summers from AD 500: temperature changes on short and long timescales. Climate Dynamics 7: 111119.
  • Briffa KR, Osborn TJ, Schweingruber FH, Harris IC, Jones PD, Shiyatov SG, Vaganov EA. 2001. Low-frequency temperature variations from a northern tree ring density network. Journal of Geophysical Research 106: 29292941.
  • Brohan P, Kennedy JJ, Harris I, Tett SFB, Jones PD. 2006. Uncertainty estimates in regional and global observed temperature changes: a new dataset from 1850. Journal of Geophysical Research. doi: 10.1029/2005JD006548.
  • Büntgen U, Esper J, Frank DC, Nicolussi K, Schmidhalter M. 2005. A 1052-yr tree-ring proxy for Alpine summer temperatures. Climate Dynamics 25: 141153.
  • Büntgen U, Frank DC, Kaczka RJ, Verstege A, Zwijacz-Kozica T, Esper J. 2007. Growth/climate response of a multi-species tree-ring network in the Western Carpathian Tatra Mountains, Poland and Slovakia. Tree Physiology 27: 689702.
  • Büntgen U, Frank DC, Nievergelt D, Esper J. 2006. Summer temperature variations in the European Alps, AD 755–2004. Journal of Climate 19: 56065623.
  • Büntgen U, Frank DC, Wilson R, Carrer M, Urbinati C, Esper J. 2008. Testing for tree-ring divergence in the European Alps. Global Change Biology 14: 24332453.
  • Cook ER, Peters K. 1997. Calculating unbiased tree-ring indices for the study of climatic and environmental change. The Holocene 7: 361370.
  • Edwards M, Richardson AJ. 2004. Impact of climate change on marine pelagic phenology and trophic mismatch. Nature 430: 881884.
  • Esper J, Cook ER, Krusic PJ, Peters K, Schweingruber FH. 2003. Tests of the RCS method for preserving low-frequency variability in long tree-ring chronologies. Tree-Ring Research 59: 8198.
  • Esper J, Schweingruber FH. 2004. Large-scale treeline changes recorded in Siberia. Geophysical Research Letters 31, L06202, doi: 10.1029/2003GL019178.
  • Frank D, Büntgen U, Esper J. 2009. Response to “Late 20th century growth acceleration in greek firs (Abies cephalonica) from Cephalonica Island, Greece: a CO2 fertilization effect? Dendrochronologia 27: 223227.
  • Gouirand I, Linderholm HW, Moberg A, Wohlfarth B. 2008. On the spatiotemporal characteristics of Fennoscandia tree-ring based summer temperature reconstructions. Theoretical and Applied Climatology 91: 125.
  • Grudd H. 2008. Tornetra¨sk tree-ring width and density AD 500–2004: a test of climatic sensitivity and a new 1500-year reconstruction of north Fennoscandian summers. Climate Dynamics 31: 843857.
  • Grudd H, Briffa KR, Karlén W, Bartholin TS, Jones PD, Kromer B. 2002. A 7400-year tree-ring chronology in northern Swedish Lapland: natural climatic variability expressed on annual to millennial timescales. The Holocene 12: 657665.
  • Gunnarson BE, Linderholm HW, Moberg A. 2010. Improving a tree-ring reconstruction from west-central Scandinavia: 900 years of warm-season temperatures. Climate Dynamics. doi: 10.1007/s00382-010-0783-5.
  • Hallinger M, Manthey M, Wilmking M. 2010. Establishing a missing link: warm summers and winter snow cover promote shrub expansion into alpine tundra in Scandinavia. New Phytologist 186: 890899.
  • Helama S, Timonen M, Holopainen J, Ogurtsov MG, Mielikäinen K, Eronen M, Lindholm M, Meriläinen J. 2009. Summer temperature variations in Lapland during the Medieval Warm Period and the Little Ice Age relative to natural instability of the thermohaline circulation on multi-decadal and multicentennial scales. Journal of Quaternary Science 24: 450456.
  • Helama S, Timonen M, Lindholm M, Meriläinen J, Eronen M. 2005. Extracting long-period climate fluctuations from tree-ring chronologies over timescales of centuries to millennia. International Journal of Climatology 25: 17671770.
  • Holtmeier F-K 2003. Mountain timberlines. Ecology, patchiness and dynamics. Dordrecht, the Netherlands: Kluwer, 369.
  • Ims RA, Henden J-A, Killengreen ST. 2008. Collapsing population cycles. Trends in Ecology and Evolution 23: 7986.
  • Kauserud H, Stige LC, Vik JO, Okland RH, Hoiland K, Stenseth NC. 2008. Mushroom fruiting and climate change. Proceedings of the National Academy of Sciences, USA 105: 38113814.
  • Kausrud KL, Mysterud A, Steen H, Vik JO, Østbye E, Cazelles B, Framstad E, Eikeset AM, Mysterud I, Solhøy T et al. 2008. Linking climate change to lemming cycles. Nature 456: 9397.
  • Kirby RR, Beaugrand G. 2009. Trophic amplification of climate warming. Proceedings of the Royal Society-B 276: 40954103.
  • Kirchhefer AJ. 2001. Reconstruction of summer temperatures from tree-rings of Scots pine (Pinus sylvestris L.) in coastal northern Norway. The Holocene 11: 4152.
  • Klingbjer P, Moberg A. 2003. A composite monthly temperature record from Tornedalen in Northern Sweden, 1802–2002. International Journal of Climatology 23: 14651494.
  • Kolishchuk VG. 1990. Dendroclimatological study of prostrate woody plants. In: CookER, KairiukstisLA, eds. Methods of dendrochronology. Applications in the environmental sciences. London, UK: Kluwer, 5155.
  • Kullman L. 2001. 20th-century climate warming and treelimit rise in the southern Scandes of Sweden. Ambio 30: 7280.
  • Kullman L. 2002. Rapid recent range-margin rise of tree and shrub species in the Swedish Scandes. Journal of Ecology 90: 6877.
  • Linderholm HW, Björklund JA, Seftigen K, Gunnarson BE, Grudd H, Jeong JH, Drobyshev I, Liu Y. 2009. Dendroclimatology in Fennoscandia – from past accomplishments to future potential. Climate of the Past 6: 93114.
  • McMichael AJ, Woodruff RE, Hales S. 2006. Climate change and human health: present and future risks. The Lancet 367: 859869.
  • Melvin TM, Briffa KR. 2008. A “signal-free” approach to dendroclimatic standardisation. Dendrochronologia 26: 7186.
  • Mysterud A, Ostbye E. 2006. The effect of climate and density on individual and population growth of roe deer Capreolus capreolus at northern latitudes – the Lier valley, Norway. Wildlife Biology 12: 321329.
  • Parmesan C, Yohe G. 2003. A globally coherent fingerprint of climate change impacts across natural systems. Nature 421: 3742.
  • Patz JA, Campbell-Lendrum D, Holloway T, Foley JA. 2005. Impact of regional climate change on human health. Nature 438: 310317.
  • Payette S, Filion L. 1985. White spruce expansion at the tree line and recent climatic change. Canadian Journal of Forest Research 15: 241251.
  • Perry AL, Low PJ, Ellis JR, Reynolds JD. 2005. Climate change and distribution shifts in marine fishes. Science 308: 19121915.
  • Schweingruber FH, Dietz H. 2001. Annual rings in the xylem of dwarf shrubs and perennial dicotyledonous herbs. Dendrochronologia 19: 115126.
  • Schweingruber FH, Poschlod P. 2005. Growth rings in herbs and shrubs: life span, age determination and stem anatomy. Forest, Snow and Landscape Research 79: 195415.
  • Shiyatov SG. 1993. The upper timberline dynamics during the last 1100 years in the Polar Ural Mountains. Palaeoclimatic Research 9: 195203.
  • Stenseth NC, Mysterud A, Ottersen G, Hurrell JW, Chan K, Lima M. 2002. Ecological effects of climate fluctuations. Science 297: 12921296.
  • Thuiller W. 2004. Patterns and uncertainties of species’ range shifts under climate change. Global Change Biology 10: 20202027.
  • Vaganov EA, Hughes MK, Kirdyanov AV, Schweingruber FH, Silkin PP. 1999. Influence of snowfall and melt timing on tree growth in subarctic Eurasia. Nature 400: 149151.
  • Woodcock H, Bradley RS. 1994. Salix arctica (PALL.): its potential for dendroclimatological studies in the high arctic. Dendrochronologia 12: 1122.