Inconsistencies between observed, reconstructed, and simulated precipitation indices for England since   the year 1650&thinsp;CE

Bothe, Oliver; Wagner, Sebastian; Zorita, Eduardo

doi:https://doi.org/10.5194/cp-15-307-2019

Articles | Volume 15, issue 1

https://doi.org/10.5194/cp-15-307-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/cp-15-307-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 15, issue 1

Research article

|

15 Feb 2019

Research article |

| 15 Feb 2019

Inconsistencies between observed, reconstructed, and simulated precipitation indices for England since the year 1650 CE

Oliver Bothe, Sebastian Wagner, and Eduardo Zorita

Download

Final revised paper (published on 15 Feb 2019)
Supplement to the final revised paper
Preprint (discussion started on 09 May 2018)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review', Anonymous Referee #1, 22 May 2018
- AC2: 'Final Response to referee reports', Oliver Bothe, 16 Jul 2018
RC2: 'Reviewer comments on Bothe et al. Inconsistencies between observed, reconstructed and simulated precipitation', Anonymous Referee #2, 22 May 2018
- AC2: 'Final Response to referee reports', Oliver Bothe, 16 Jul 2018
RC3: 'Review', Anonymous Referee #3, 12 Jun 2018
- AC2: 'Final Response to referee reports', Oliver Bothe, 16 Jul 2018
AC1: 'Initial response to the three referees' reports', Oliver Bothe, 15 Jun 2018

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (28 Jul 2018) by Jürg Luterbacher

Dear authors,
thanks very much for addressing the questions/critics/suggestions of the reviewers. All reviewers mention similar major points with regard to the clarity of the introduction, aim of the study, the presentation of the results, critical aspects related to the choice of instrumental data, reconstruction and models, their interpretation, visualisation of the results as well as the conclusions. Thus, it should be straight forward to revise the manuscript accordingly in agreement with the suggested changes of the authors.
The authors use SPI as the measure for their analysis. As it is only based on precipitation, I agree that analysis that use temperature do not need to be shown. The revised version thus should include more studies that deal with precipitation and as suggested by the authors in their last section of reviews, incorporate additional precipitation/drought sensitive information. This would mean that the revised version provides a comprehensive analysis of all available precipitation and drought information from parts of the British Isles. The (in)consistencies between the various information sources need to be made more clearly as the reviewers suggested. It would be important to explain also in more detail why SPI is used and not other drought related measures. SPI is a specific index for meteorological drought with strengths and limitations that need to be discussed (see information for instance in: http://climatedataguide.ucar.edu/climate-data/standardized-precipitation-index-spi
About the choice of an appropriate drought index from a model point of view, the authors might refer to Raible et al. (2017) on “Drought indices revisited-improving and testing of drought indices in a simulation of the last two millennia for Europe", Tellus, 69, 1287492
For instrumental data/reconstruction data the authors might do a detailed screening on the literature for this area and then choose.
Further, the use of RCM data for comparisons, interpretation, comments that have been risen all reviewers. This point needs to be more clearly addressed, what is the purpose of showing RCM analysis in comparison with instrumental/reconstruction data and how do the analysis go beyond the current state of the art. The authors mention that they will make this point more clear and also move parts in a SOM. In this context one open issue refers to the suggestion of reviewer 1 to write two different papers rather than implementing everything in one paper, thus removing for instance the model/data comparison and leave it for a new publication. I leave the decision up to the authors, but any decision to include models or exclude them should be clearly explained.
I agree with Reviewer 1 and 2 that have concerns using the tree ring reconstructions, either as they may not reflect precipitation or the season under consideration or that the Trouet et al. jet reconstruction is not an appropriate measure for circulation purposes. With respect to circulation analysis, for the past centuries there are monthly to seasonal large scale SLP and Z500 reconstructions that are based on instrumental pressure and ship log books information. They are more suitable and trustworthy than natural proxy reconstructions and should be used instead.
I think it is a valuable comment by reviewer 2 concerning the Rinne et al paper. I am sure that the authors would provide the data for analysis.

I am convinced, that the suggestions of the reviewers, their constructive criticisms as well as the way the authors intend to revise the paper accordingly, will strongly improve the quality of the paper

With my best wishes and thanks very much and looking forward to the revised version

Jürg

Hide

AR by Oliver Bothe on behalf of the Authors (21 Sep 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (26 Sep 2018) by Jürg Luterbacher

RR by Anonymous Referee #2 (29 Oct 2018)

Suggestions for revision or reasons for rejection

Second Review of Bothe et al.

The authors have improved the paper in this iteration. Some further work remains to be done however. The paper is longer than it needs to be, especially the discussion. There is also work to be done on teasing out the key contributions. At least the paper should draw out where there is agreement and not between sources, ie. Provide the key periods. This would be usefully done in a table and would help focus future research to further understand the issues raised.

The integration of tree rings (based on width) is done so in the full knowledge that previous work by the authors of these data sets have shown that there is a lack of stability in terms of the relationship with station based precipitation. This work confirms a well established fact, that there is inconsistency between tree ring reconstructions and station based observations. I am therefore not sure what this aspect adds. This is well known. Surely this paragraph from the discussion negates the use of the tree ring series you use for the purpose you use them for.

“Young et al. (2015) conclude that these differences make it unlikely that the tree-ring based works and their δ 18O based work represent the same environmental parameter, and they emphasize the lack of a calibration against regional precipitation data. They further discuss the reasons given by Wilson et al. (2013) for the lacking stability of the Wilson et al. reconstruction, namely, different climate-sensitivity of the trees, unreliable instrumental data, and pollution. Young et al. (2015) conclude that their own data reliably reflects precipitation while the tree-ring widths most likely represent the combination of various environmental influences on tree growth instead of a single climate parameter.”

To help tease out inconsistencies I am not sure why you don’t use series such as the Paris London Index, London Sea Level Pressure (both by Richard Cornes) or perhaps the Westerly Index to help point to which data sources that potential errors derive from. There is more data available to help this work. It seems a missed opportunity not to use them.

Also, regarding the data, EWP is a composite drawing on stations as far north as Edinburgh. Yet the analysis, reconstructions and modelling is based on southern England. It is not surprising to me that inconsistencies exist given the study design.

Simulation results are based on a single RCM model chain. Can this provide a robust estimate for comparison with observations and reconstructions? You have not convinced me.

The discussion is far too long and contains some repetition, making it difficult to read. It needs to be considerably shortened and sharpened in its focus on key points. It needs to distil the key contributions of this work. The structure is also odd, there are concluding remarks followed by conclusions, and section 5.3 is Discussion of Results? The way it is organised re-traces the entire paper again. It needs to get to the job it is supposed to do, i.e. discuss key points from the findings.

It is my opinion that you give too great a credit to the early observations. These are likely to be highly uncertain. Indeed for many of the series you use the early parts are composites taken from weather diaries and very early instrumentation. In addition do you use the corrected Oxford series of Burt and Howden who corrected for gauge being on a roof and changes in gauge design? Or is it the original series? You could look more critically upon the observations.

Given that we know the reconstructions have issues, it would be useful to highlight to the reader the particular periods where we have agreement between the different sources and those that we do not. This might help focus and provide a priority for future work. What I am saying is that the results need to be consolidated and key insights distilled. What you are saying at the moment is that there is disagreement between the observations and reconstructions, disagreement between both of these and model simulations but the latter is not as different to the observations as the reconstructions are. I need more insight on this - are there periods where all agree/disagree etc.

Please note that the negative relationship between temperature and precipitation does not include winter where the relationship is warm and wet. The way it is stated in the text gives the sense that it is year round. It is my understanding that temperature/precipitation relationships are strongest in winter and summer and much more variable in spring, which is why the signal may be variant in your comparisons. Again, this is not helped by the choice of your study season, which is different from the tree ring season and makes the entire analysis hard to follow at times.

Is there a mistake in stating that Rinne et al calibrated to Oxford? My memory is that it was Kew they used. I may be wrong but please double check.

A key objective of the paper is to show the value of the SPI over moving windows allows a better comparison of datasets. You need to spell out this contribution more, be explicit. I have not been convinced of its merits above other approaches from this paper. For instance, in the abstract you state that SPI is a valuable tool for bridging part of the problem in assessing agreement and disagreement – what are the problems it addressed. I think that this is an important part of the paper as mentioned above we know there is disagreement between reconstructions and observations – what new does the SPI analysis bring, what further insight?

Again in the abstract, what do you mean by the most recent historical period. This is not defined.

The table on date sources used needs to include the years covered by each.

Hide

RR by Anonymous Referee #3 (31 Oct 2018)

Suggestions for revision or reasons for rejection

General Remarks: This is an update to a previously reviewed manuscript. The writing and clarity are significantly improved, however, there are a number of changes that must be made before publication. Please see the specific comments below. The one more general comment that must be addressed is that the discussion and conclusion sections need to be more clearly written. In particular, there are parts of these sections that read like methodology and results. These parts should be replaced with clear and concise statements on what can and cannot be taken away from the results—what do the results tell us and why is that important?

Page 1, Line 1: Suggest changing “The Standardized Precipitation Index is a valuable tool for bridging part of the problems in assessing agreement and disagreement between the different sources of information. We assess the agreement in the temporal evolution of percentiles of the precipitation distributions” to “The Standardized Precipitation Index is a valuable tool for assessing agreement between the different sources of information, as it allows for a comparison of the temporal evolution of percentiles of the precipitation distributions”.

Page 1, Line 8: Suggest changing “consistent relations” to “consistency”.

Page 1, Line 9: Suggest changing “purely” to “impact of”.

Page 1, Line 9: I do not understand what is meant by the sentence that starts on this line.

Page 1, Line 22: What is meant by “supports”, can you be more specific here?

Page 2, Line 3: Suggest removing “and inconsistency”.

Page 2, Line 21: Suggest removing “complementing the current suite of statistical diagnostics for data-model comparisons”.

Page 3, Line 25: Suggest removing “regional instrumental series”.

Page 3, Line 28: Add “at” between “looking how”.

Page 4, Line 4: I have never heard the term “document asset”. My understanding is that Climate of the Past uses the term supplementary material, although that may have changed. This note is just to make sure to use the correct term throughout.

Page 4, Line 7: Change “most” to “mostly”.

Page 5, Line 16: Suggest removing “long”.

Page 5, Line 26: Suggest adding “records” after “temperature”.

Page 6, Line 9: Suggest adding “observations” after “precipitation”.

Page 6, Line 14: Suggest removing “in the past”.

Page 6, Line 14: Suggest changing “relations” to “the relationship”.

Page 7, Line 31: Suggest changing “resolved” to “resolution”.

Page 8, Line 10: Suggest changing “of the methods” to “associated with this method”.

Page 8, Line 24: What was the rational for choosing 40 data points. If it was arbitrary it would be useful to briefly defend this choice.

Page 9, Line 2: Suggest changing “which” to “what”.

Page 9, Line 8: Suggest adding “can” before “point out”, changing “that” to “if”, and removing “completely”.

Page 9, Line 14: Some background on what the Weibull standard deviation can tell us would be useful here.

Page 9, Line 26: Change “Figures” to “figures”.

Page 11, Line 4: Suggest adding “sections” after “following”.

Page 12, Line 14: Why the choice of non-overlapping 11-year averages?

Page 12, Line 16: Suggest changing “gives also” to “also gives”.

Page 14, Line 2-3: Can you quantify this?

Page 14, Line 13: Suggest changing “relation” to “relationship”.

Page 14, Line 16: Suggest removing “though possibly not surprisingly”.

Page 14, Line 31: Suggest changing the part of the sentence after “period of” to “overlap”.

Page 14, Line 32: I think you mean “shorter timescale” here, “smaller scale” might be confusing since it could also refer to spatial scales.

Page 15, Line 1: Suggest changing “lack the clear overall trend” to “do not have a long term trend”.

Page 15, Line 7: Suggest changing the part of the sentence after “indicate” to “the opposite”.

Page 15, Line 9: I would reference the most relevant figure for the “smoothed evolution” at the end of this sentence.

Page 15, Line 13: Add “us” before “to”.

Page 15 Line 14: Why define an acronym for standard deviation here? If you wish to do so, it makes sense to do it at first mention in the methods section. You also do not use “SD” in the manuscript text, only in the figure labels—perhaps put the acronym definition in the figure caption.

Page 15, Line 15: The Southern-Central reconstruction appears to have an increasing trend.

Page 16, Line 12: It is worth being more specific here, something like “…do not agree on the time evolution of precipitation percentiles”.

Page 16, Line 12: There really does not seem to be much agreement. I think that this sentence confuses that point. Suggest reformatting to “Any agreement between the simulation and reconstructions appears random, with a tendency instead towards an opposite evolution—particularly for the dry percentile”. You can then go into a bit more detail and the explanation in the following paragraph.

Page 17, Line 3: Suggest changing to “…would show a range of trajectories and thus these results do not preclude that the model is capturing…”

Page 17, Line 7: Suggest changing to “…more or less in opposition (Figure 3).”

Figure 7 caption: I was not able to understand what this caption means.

Page 18, Line 7: Why 1815?

Page 19, Line 14: Suggest removing “prior to approximately the year 1850 CE”.

Page 20, Line 11: I am not sure that I understand why this is specifically relevant to a comparison of precipitation and temperature. It seems more relevant to the comparisons of, for instance, model simulations to reconstructions.

Page 20, Line 12: Suggest adding “thus” after “We” and removing “the” before “precipitation”.

Page 22, Line 12: Add “the” before “environment”.

Page 22, Line 17: Suggest adding “inherent to SPI” after “standardization”.

Page 22, Line 21: Suggest removing the part of the sentence after “as well”.

Page 23, Line 16: Add “the” between “in” and “form”.

Page 23, Line 17: Suggest putting a period after the Parker reference then starting a new sentence: “In addition, there are long instrumental station observations of precipitation…”.

Page 23, Line 18: I would remove these two sentences as there is not much value in discussing what you do not analyze.

Page 24, Line 5: Suggest moving “ACRE” into the parentheses and the full written out version into the main part of the sentence.

Page 24, Line 10: You have already used PDSI at this point.

Sections 5.1 and 5.2: Much of these sections is redundant. I think the manuscript would be fine with all of this removed.

Page 27, Line 1: Suggest removing the period and reformatting as “, however, there is occasional temporal consistency between the simulations and observation based data.”

Page 27, Line 6 to 5.3.3: This is just more results and should either be moved to the results section or removed entirely.

Page 28, Line 3: Again this paragraph is largely results.

Page 28, Line 16: But it is really the reconstructions that look inconsistent with both the observations and simulations and thus I would focus predominantly on that in this paragraph (at the moment only the final sentence discusses this disagreement).

Page 28, Line 14: I would reformat this as “If we expect temporal consistency among the different sources of information, then we are assuming that all the sources of information are responding to the impact of external climate forcing”.

Page 28, Line 27: I would revisit this sentence with any eye towards clarity.

Page 28, Line 31: Change “frequent” to “frequency”.

Page 28, Line 34: Suggest changing to “However, the analysis period includes relatively large…”

Page 29, Line 12: Does this not require that the opposite behavior is “significant” relative to randomness? This statement seems too speculative. Likewise, the observations and reconstructions often do not agree…

Page 30, Line 1: Change “frequent” to “frequency”.

Section 5.3.5: This discussion feels a bit too general and should be refocused towards discussing the results and implications of this study.

Hide

ED: Reconsider after major revisions (31 Oct 2018) by Jürg Luterbacher

AR by Oliver Bothe on behalf of the Authors (12 Dec 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (15 Dec 2018) by Jürg Luterbacher

RR by Anonymous Referee #2 (31 Dec 2018)

RR by Anonymous Referee #3 (18 Jan 2019)

Suggestions for revision or reasons for rejection

This is an update to a previously reviewed manuscript. The following comments are minor can be addressed without the need for another round of reviews.

Page 1, Line 10: Suggest changing “observational indices” to “observations”.

Page 1, Line 22: Suggest changing to “It strengthens our confidence if they prove to be consistent sources of information”.

Page 2, Line 6: Suggest changing “data fields, the grid resolution” to “observations, the spatial resolution”.

Page 3, Line 3: Suggest changing “the South of Great Britain” to “southern Great Britain” to be consistent with the more common usage in the manuscript.

Page 3, Line 9: Suggest removing “a simulation” after the comma.

Page 3, Line 23: Suggest changing “not least” to “also”.

Page 3, Line 26: Suggest moving the comma to before “especially”.

Page 3, Line 29: Suggest removing “compare”.

Page 4, Line 1: Suggest changing “on” to “of”.

Page 4, Line 10: Suggest changing “by” to “from”.

Page 5, Line 20: Suggest changing “the South of Great Britain” to “southern Great Britain” to be consistent with the more common usage in the manuscript.

Page 5, Line 33: Suggest changing “are, e.g.,” to “include”.

Page 6, Line 2: Suggest changing “allow assessing” to “allow for assessment of”.

Page 6, Line 7: Suggest changing “the South of Great Britain” to “southern Great Britain” to be consistent with the more common usage in the manuscript.

Page 7, Line 7: Suggest putting June to August in parentheses and an “a” before “scaling” and “a” on the following line before “supplement”.

Page 7, Line 12: Add the year of the reference in parentheses.

Page 7, Line 23: Suggest removing the sentence starting “However” as it does not appear to add useful information.

Page 7, Line 26: Suggest changing “the South of Britain” to “southern Great Britain” to be consistent with the more common usage in the manuscript.

Page 7, Line 30: Suggest starting the sentence “To our knowledge, this simulation…” and removing the part after the comma at the end of the sentence.

Page 8, Line 13: Suggest changing “degree” to “degrees”, also on Line 15.

Page 9, Line 12: Suggest removing “at least”.

Page 9, Line 16: Suggest changing “on” to “when”.

Page 10, Line 11: Suggest combining these two sentences “data and there are…”.

Page 10, Line 25: Suggest removing “at least”.

Page 10, Line 29: Suggest changing “this” to “the” and removing “interannually” on the following line.

Page 11, Line 5: Suggest adding “can” before “point”.

Page 11, Line 5: Suggest changing “calculating” to “for the calculation of”.

Page 11, Line 8: Suggest flipping “represent traditionally”.

Page 11, Line 15: Suggest adding “the” before “reference”.

Page 11, Line 30: I might add at the end of the sentence: “which have too much high-frequency variability and are too-smoothed, respectively”.

Figure 2 caption: Suggest removing the comma in the last sentence and replace it with “that”.

Page 13, Line 7: Suggest changing the end of the sentence to: “than the agreement amongst the observational data”.

Page 13, Line 10: Suggest removing “on interannual time-scales”.

Page 13, Line 16: Suggest changing the beginning of the sentence to “There is a strong relationship…”.

Page 14, Line 1: Suggest changing “relation” to “relationship” here and on Lines 3 and 7.

Page 14, Line 12: Suggest removing “for this resolution”.

Page 17, Line 4: Suggest changing “93.3th, i.e. wet, percentile” to “wet percentile” and a similar change with the dry percentile statement on Line 8.

Page 17, Line 10: Suggest adding “before transformation into a distributional form” at the end of the sentence.

Page 19, Line 9: Suggest removing “per-se”.

Page 20, Line 8: Remove “each window”.

Page 21, Line 2: Suggest changing “of” to “for”.

Page 21, Line 6: Suggest adding “the” before the list of percentiles.

Page 22, Line 4: Suggest removing “shortly”.

Page 22, Line 8: I understand what you are saying but the sentence is a bit confusing, I suggest you revisit this with an eye towards clarity.

Page 23, Line 11: Suggest changing “show again” to “again show”.

Page 23, Line 15: Suggest changing “only shortly” to “briefly”. I would also move up the first sentence of the next paragraph into this first paragraph, changing the beginning of that sentence to read “In particular, we show…for this season.”.

Page 24, Line 2: Suggest changing “high” to “large” and adding “such” before “small”.

Page 24, Line 9: Suggest combining these two sentences.

Page 24, Line 17: Suggest changing “relation” to “relationship”.

Page 24, Line 24: Suggest changing “shortly” to “briefly”.

Page 24, Line 27: Suggest changing “simulations” to “simulated”.

Page 24, Line 4: Suggest changing “relation” to “relationship”.

Page 25, Line 17: Suggest changing “shows again” to “again shows”.

Page 25, Line 21: Suggest changing “in” to “with”.

Page 25, Line 22: Suggest changing “relation” to “relationship”.

Page 25, Line 26: Suggest removing “of climate parameters”.

Page 25, Line 32: Suggest changing “which again” to “that” and removing “the input changed over time”. It is unclear what that sentence means—if you choose not to remove it you should revisit the sentence with an eye towards clarity.
¬¬
Page 25, Line. 33: Suggest changing “proxy” to “proxies”.

Page 26, Line 8: Suggest changing “relation” to “relationship”.

Page 26, Line 11: Suggest removing “a priori” and “on” to “as to” on the following line.

Page 26, Line 19: Suggest changing “to identify” to “the identification of”.

Page 26, Line 29: Suggest changing “differences to” to “differences as compared to”.

Page 26, Line 32: Suggest changing “the reconstructions relation to temperature” to “the relationship between the reconstructions and temperature” and “relate” to “reflect” on the following line.

Page 27, Line 6: Suggest changing “relations” to “relationships” and removing “from March to July”. I would also change “relation” to “relationship” throughout this paragraph and the next.

Page 27, Line 14: Suggest adding “the” before “simulation”. I would also clarify what relationship you are talking about on the following line.

Page 27, Line 18: Suggest adding “the” before “simulation”.

Page 27, Line 24: Suggest adding “magnitude” after “large”.

Page 27, Line 27: Suggest removing “large”.

Page 28, Line 9: Suggest adding “variability in” before “the North…”

Page 28, Line 19: What is meant by “increase our confidence in forced changes”.

Page 28, Line 25: Suggest changing “comparing” to “comparison of”.

Page 28, Line 30: Suggest removing “indeed”.

Page 28, Line 32: Suggest changing “Particularly” to “In particular,”. This sentence does not appear to be consistent with the previous sentence where you note that they have limited agreement. “Comparable changes” implies consistency while “limited agreement” implies inconsistency.

Page 28, Line 33: Suggest removing “mainly”.

Page 29, Line 2: This sentence is vague, I suggest that you expand on the “different processes”.

Page 29, Line 8: Suggest changing “relation” to “relationship” here and on the next line.

Hide

ED: Publish subject to technical corrections (21 Jan 2019) by Jürg Luterbacher

AR by Oliver Bothe on behalf of the Authors (24 Jan 2019) Author's response Manuscript

Short summary

Our understanding of future climate changes increases if different sources of information agree on past climate variations. Changing climates particularly impact local scales for which future changes in precipitation are highly uncertain. Here, we use information from observations, model simulations, and climate reconstructions for regional precipitation over the British Isles. We find these do not agree well on precipitation variations over the past few centuries.