the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Feb 2024
An 1200-year multi-proxy dendrochronological temperature reconstruction for the area of Austrian Alps
Abstract. Temperature reconstruction was carried out on the basis of a centuries-long dendrochronological scales from Austrian Alps. Chronologies of growth-ring width, maximum density of latewood and stable isotope content of carbon and oxygen were applied for the research. Subfossil wood and living trees originating from the area of Schwarzensee Lake were used for the construction of the chronologies. All measurements were performed with annual resolution. A very good match was found between the results obtained and the meteorological data, making it possible to precisely reconstruct the temperature of the growing season (May–September) over the years 800–2000 CE. The proportion of temperature variance explained by independent variables accounted for 52 % in the period common for the growth-ring chronologies and meteorological data. The statistics calculated during calibration and verification tests indicated that chronologies have high reconstruction skills and that the accuracy of reconstruction is good. Obtained data show the existence of significant cooling in the periods 900–1100 CE, 1275–1325 CE, 1450–1600 CE and 1800–1890 CE and evident warming around the years 1150 CE, 1250 CE, 1325–1425 CE, 1625–1775 CE. The strongest increasing trend in temperature has been observed since the beginning of the 20th century and is clearly indicative of an ongoing climate warming.
- Preprint
(1448 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on cp-2024-4', Anonymous Referee #1, 26 Feb 2024
The present work provides an interesting multi-proxy summer temperature reconstruction for the Austrian Alps. The reconstruction is based on a unique combination of tree-ring width (TRW), maximum latewood density (MXD), and carbon and oxygen stable isotope measurements that represent an impressive measurement effort. This unique dataset has merit for publication. The Temperature reconstruction shows good agreement with the measured temperature, using an exceptionally long temperature measurement. However, the data analysis and interpretation require improvement for publication.
General comments:
Figures should be enhanced by including units on the y-axis. Several statements need to be supported by citations to existing literature or the current results. The introduction requires more background detail and citations of previous dendroclimatological studies in the region. Currently, only correlations with temperature are presented for each proxy. The analysis should be expanded to include an examination of the relationships with other climate variables, such as precipitation and drought indices.
While the comparison with other temperature reconstructions is interesting, the unique contribution of this multi-proxy chronology compared to reconstructions based on a single proxy needs to be clarified. Namely, what specifically does the combination of TRW, MXD and stable isotopes provide beyond TRW or MXD alone?
The climate analysis and comparison with other records should be improved. References to major climate periods or events (e.g. Late Antique Little Ice Age, Medieval Climate Anomaly, Tambora Eruption) could put the temperature fluctuations in a broader context. Stable isotopes have the potential to capture lower frequency variability compared to TRW and MXD, but this potential is not fully explored. Comparisons could be made with additional non-tree ring reconstructions, such as Casty et al. 2005 (https://doi.org/10.1002/joc.1216) and Meier et al. 2007 (https://doi.org/10.1029/2007GL031381).
Specific comments:
Lines 38-39: Note that soil moisture is also an important driver of tree growth.
Lines 43-55: Consider adding a figure with a map showing the geographical location to provide more context.
Line 46: Clarify the elevation site.
Lines 85-86: Explain why Figures 2, 3, and 4 end in the year 2000 when the study goes to 2018.
Lines 135-137: It would be helpful to show the EPS, r-bar statistics in a supplementary figure.
Lines 215-216: The finding of higher correlations between TRW and temperature in June and September agrees with previous work such as Leal et al. 2007.
Lines 230-240: The moderate correlation coefficients may be due to the long-time windows used. Consider analysing 1900-2000 to demonstrate stronger correlations, using the remaining period for verification. Correlations further back in time can be less precise also for a decrease of the observation quality. Also, I suggest examining correlations with precipitation, humidity, scPDSI, and SPEI using interpolated CRU data for 1900-2000.
Lines 234-235: Show the climate correlation analysis between temperature and the final combined proxy chronology.
Line 247: Show precipitation and hydroclimate correlation analyses to support this statement.
Line 265: Consider showing this correlation analysis in a supplementary figure.
Lines 263-280: Good explanation of temperature signal, however previous Alpine isotopic studies found stronger drought signals. Show the correlations with all variables to prove temperature signal.
Line 283: Show the summer precipitation correlation analyses.
Line 314: State "significant correlations" to emphasize if the correlations are significant or not.
Line 234: In the method the authors should explain how the individual proxies have been merged into a final multi-proxy methodology. In Figure 2, it looks like the multi-centennial variability of the carbon isotope differs from the other proxies, driving the warm phase from 1650 to 1750 CE in the temperature reconstruction. This should be better explained, since this warm phase overlaps with the Little Ice Age, known to be a cold period in Europe and the Alps.
Line 330: Explain why the climate correlation time window analysis was narrowed for the final chronology but not for the individual proxies.
Line 424: Specify which panel of Figure 3 is being referred to.
Line 430-435. Nice result, but it should be noted that the high correlation pattern is probably driven by the high frequency, as all the other correlations are based on TRW and MXD, which point to a common high frequency signal. The difference may be in the low frequency or long-term trend. The new reconstruction seems to have a long temperature rise that the other reconstructions are not able to capture as they are based on TRW and MXD.
Figures:
A new introductory figure should be added showing the location of the study area and meteorological stations to provide an important geographical context.
In Figure 1, the y-axis scale should be included to allow proper interpretation of the data.
Figure 2 shows a very good agreement between the measurements and reconstruction, clearly demonstrating the proxy record accuracy.
Table 3 is a bit confusing in its current form. Consider adding the time windows analyzed in the first row to clarify the periods represented. Also use either R2 or r consistently for the correlation metrics.
For Figure 4, adding y-axis scales and directly labelling the different temperature reconstructions on the plots would improve clarity and interpretation. Relating the observed temperature fluctuations to known climate periods, such as the Medieval Climate Anomaly and Little Ice Age, would provide helpful context for the trends. The reconstructions in Büntgen et al. 2011 and Büntgen et al. 2016 show the patterns as they are the same reconstruction, (Figure 2 Büntgen et al. 2016)
Figure 4, adding y-axis scales and directly labelling the different tree-ring based temperature reconstructions on the plots would improve clarity and interpretation. Relating the observed temperature fluctuations in these reconstructions to known climate periods, such as the Medieval Climate Anomaly and Little Ice Age, would provide helpful context. The reconstructions Büntgen et al. 2011 and Büntgen et al. 2016 show similar patterns, as they reflect the same tree-ring reconstruction (Figure 2 of the Büntgen et al. 2016 paper).
Citation: https://doi.org/10.5194/cp-2024-4-RC1 -
RC2: 'Comment on cp-2024-4', Anonymous Referee #2, 27 Mar 2024
General comments:
The Authors present a 1200-year May-September air temperature reconstruction derived from tree-ring width, latewood density, and stable carbon and oxygen isotopes from the Schwarzensee area in the Austrian Alps. The work lacks of methodological and paleoclimatological novelty but can be of interest at the regional scale. The paper has major flows. Hypotheses, background, and motivation for the conduct of this study are unclear. Why this study site was selected and why study was conducted? What is the advantage of using stable isotope proxies compared to classical tree-ring isotope analysis? Please provide citations relevant to the stable isotope studies.
The main output of the study highlighted cooling and warming periods with increasing air temperature trend since the beginning of the 20th century compared to the past.
The novelty (approach, region, and testing hypothesis) of this study compared to existing published studies must be specified. Citations and comparisons with existing studies in the region and globally should be provided (e.g., Kress et al. 2014; Büntgen et al., 2021; Churakova-Sidorova et al., 2022; Kuhl et al., 2024). It is unclear how the multi-proxy chronology was built and how it corresponds with other chronologies from Europe or with other proxies from the region like lake sediments.
Büntgen, U., Urban, O., Krusic, P.J. et al. Recent European drought extremes beyond Common Era background variability. at. Geosci. 14, 190–196 (2021). https://doi.org/10.1038/s41561-021-00698-0
Churakova-Sidorova, O.V., Myglan, V.S., Fonti, M.V. et al. Modern aridity in the Altai-Sayan mountain range derived from multiple millennial proxies. Sci Rep 12, 7752 (2022). https://doi.org/10.1038/s41598-022-11299-1
Kress, A., S. Hangartner, H. Bugmann, U. Büntgen, D. C. Frank, M. Leuenberger, R. T. W. Siegwolf, and M. Saurer (2014), Swiss tree rings reveal warm and wet summers during medieval times, Geophys. Res. Lett., 41, 1732–1737, doi:10.1002/ 2013GL059081.
Kuhl, E., Esper, J., Schneider, L. et al Revising Alpine summer temperatures since 881 CE. Clim Dyn (2024). https://doi.org/10.1007/s00382-024-07195-1
Specific comments:
Title: The title can be shortened. Suggestion: “A 1200-year multi-proxy May-September temperature reconstruction from the Austrian Alps” or even revised based on the findings and case study.
Abstract:
L. 15-16 it will be good to provide clarification at the beginning of the manuscript about averaged May-September air temperature reconstruction.
L. 16 and throughout the whole manuscript “growth-ring width” replace with tree-ring width; “maximum density of latewood” replace with maximum latewood density; “stable isotope content of carbon and oxygen” replace with stable carbon and oxygen isotopes or isotopic composition of stable carbon and oxygen isotopes. Please also add a small delta symbol for stable carbon and oxygen isotopes. Basic dendrochronological characteristics should be provided in a standard way: as wood density, tree-ring (see Methods of Dendrochronology https://link.springer.com/book/10.1007/978-94-015-7879-0, E. R. Cook &L.A. Kairiukstis, 1990).
It is unclear from the abstract how the reconstruction of May-September was performed. Is it 52% of temperature variance explained in combined multi-proxy reconstruction based on tree-ring width, stable carbon and oxygen isotopes, and latewood density? How much (in %) is it explained by each parameter (carbon, oxygen, tree-ring width, maximum latewood density)? Please specify.
L. 22-23 “..the statistic is good..” Please provide numbers and confirm.
L.25 “The strongest increasing trend» How strong and significant this trend is compared to the past? Please specify and provide values or estimates in %.
- Introduction
More citations and comparisons with other studies from the region and the globe are recommended.
L 35-40 Please provide citations
The "Introduction" and "Materials and Methods" sections are mixed up with the introducing region in the introduction part.
L. 59 …ring width (TRW) replaced with tree-ring width (TRW)
2. Material and Methods
L. 100 “.. were omitted during densitometric and isotopic research” replace with ..” were excluded from densitometric and stable isotope measurements”.
L. 110-115 Please provide an original citation to X-ray densitometrical tool by Schweingruber FH and Briffa KR 1995 and Schweingruber FH 1988. The citation to Kłusek and Grabner, 2016 is not the original one.
L. 140-141 “Then, the material obtained from four trees was pooled (combined) for a particular year.” This sentence contradicts the sentence above that 5 living trees were used for the stable isotope analyses (L. 139). Please clarify how many trees as a pooled material were used for the stable isotope analyses.
L. 140-150 The method for cellulose extraction described is originally developed by Loader 1997 https://doi.org/10.1016/S0009-2541(96)00133-7. If this protocol and description were modified, this should be explicitly mentioned.
L. 143 if cellulose was extracted and analyzed, it should be mentioned at the beginning of the article.
L. 164 Why was analysis performed until 2000, while samples were available until 2015? The 15 years can make a big difference. Please clarify.
L. 165-166 Please clarify what kind of correction was performed for stable carbon isotopes (pin? δ13C atm CO2 only?) and if any corrections were applied for oxygen isotope chronology.
L. 167-170 Why climate analysis was performed with temperature only and not with or additionally with other climatic parameters (e.g., precipitation, vapor pressure deficit, relative humidity)? Please clarify.
L. 171-172 What is the reason for taking the overlap period from September of the previous year to October of the current one? In this case correlation coefficients due to overlap will be higher, but not necessarily will provide reasonable eco-physiological information, which is supposed to be derived from the stable isotopes and tree-ring parameters. Were the 14 months to all tree-ring width and stable isotope parameters applied? In Table 1 later is even from January of the previous year to December of the current one? Please explain.
3. Results and Discussion
L. 182 „ ...various physicochemical parameters of wood..” What is ment here? Stable isotopes as biogeochemical tools?
L. 184 It is unclear why until 2000 only, while living trees were collected until 2015.
L. 187-190 should be moved to the Material and Methods section.
L. 216 What about August?
Results of Table 1 showed that TRW negatively correlated with summer temperatures of the previous year. Please explain why the statistical analysis was performed from January of the previous year to December of the current one. Why not consider from September of the previous year to August of the current one? Overlaps with the months can give better statistical values but do not make any eco-physiological sense. Otherwise, please explain. Why only the temperature parameter was considered and not other climatic parameters like precipitation, vapor pressure deficit, or relative humidity? Did you check correlations with other climatic parameters? If yes, please provide information in supplementary material or at least check it or cite, if it was published before.
L. 245 – References are needed.
L. 214 “Péclet effect..” please provide relevant citations (e.g., Craig and Gordon 1965)
L. 310-314 Not all parameters captured temperature signal from May to September. This should be taken into consideration and discussed.
L. 349 it is unclear why the period from 1960 to 2000 was not included in reconstruction. If there is a reason, please explain.
L. 388 “These results reflects the temperature conditions at a relatively small territory localised close to Lake Schwarzensee.”. It would make sense to rephrase the title as a case study from the Austrian Alps or specify A 1200-year Schwarzensee May-September air temperature reconstruction derived from tree-ring parameters and stable isotopes.
Figures:
Figure 1: It will be interesting to see not a standard deviation, but rather raw data. Here is also unclear if is it z-scored data or not. If yes, should be specified. Please add the symbol small delta (δ) to 13C and 18O.
Figure 2: It is unclear how multi-proxy reconstruction was obtained and / or for which chronology the comparison with averaged May-September air temperature is shown in Fig 2. Please clarify in the Figure legend.
Figure 3: capture: Reconstats program – please add a citation.
Table:
Table 1: Please add symbol small delta (δ) to 13C and 18O.
Citation: https://doi.org/10.5194/cp-2024-4-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 257 | 63 | 18 | 338 | 15 | 13 |
- HTML: 257
- PDF: 63
- XML: 18
- Total: 338
- BibTeX: 15
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1