REPORT FOR THE NHESS_2021-298:

GENERAL COMMENTS

The paper presents an assessment of the spatio-temporal variability and trends of hot and dry summers, over the last fifty years, analysing the physical mechanisms driving the occurrence of hot summers in Romania. For this, the heatwave duration index (HWDI), the Standardized and Precipitation index (SPI) and the compound hot and dry index (DHD) are computed for this region. I consider that this work is interesting, however, I also need to say, that I find the manuscript difficult to read, especially because the reader is constantly referred to supplementary material. Many of the figures in the supplementary material are necessary to follow the results. In this sense, I consider that a reorganization of the Methodology and Results sections is necessary. I think that an improvement of the paper is need previous to publication in order to reach the expected international standards requested by the journal.

SPECIFIC COMMENTS

1. Firstly, I think the authors are wrong in their attempt to extend their work on eastern Europe. All the calculations of the indices are made considering only data from Romania, and all the results shown in the manuscript are based on these indices. Although it is true that Romania is part of eastern Europe, the results obtained for just a country cannot be generalized to the complete eastern Europe. In my opinion this is an error, because from the title of the article the reader expects to find results referring to a much broader region. However, this fact does not detract from the value of the work, since Romania's geographical position, as well as its topographic characteristics, make it a very interesting region from a climatological point of view.

2. Other important point is about the use of the standardized precipitation index (SPI) to analyse drought events. I know that the SPI is a robust index widely used since it has a clear computation procedure and multi-scalar character. Nevertheless, the SPI only uses precipitation data to detect drought events. However, in the context of global warming is important to consider the effects of the temperature on drought. In this sense there is a new drought index, similar to SPI, that has the additional benefit of taking it into account. This is the Standardized Precipitation-Evapotranspiration Index (SPEI; Vicente Serrano et al., 2010), which combines the benefit of using the reference evapotranspiration with the simplicity, robustness, and the multi-scalar properties of the SPI. The increasing pattern of evaporation by global warming is not a negligible factor for drought assessment. So, SPEI is relatively better for drought monitoring compared with SPI. Taking this into account, I consider that the comparative study of regional applicability of these indices is highly required for suitable applications.

Vicente-Serrano, S. M., S. Beguería, and J. I. López-Moreno (2010), A multiscalar drought index sensitive to global warming: The Standardized Precipitation Evapotranspiration Index, J. Clim., 23, 1696–1718, doi:10.1175/2009JCLI29091.

3. About the use of ROCADA dataset, I don´t understand the advantage of using it because it has the same 0.1° x 0.1° spatial resolution than EOBs and shorter temporal cover.

4. Page 1, lines 24-26: Authors literally conclude “that our study can help improve our understanding of the spatio-temporal variability of hot and dry summers, especially at the regional scale, as well as their driving mechanisms which might lead to a better predictability of these extreme events”. I think that this cannot be a specific conclusion of this work. I suggest to change this with:

“The results from this study can help improve our understanding of the spatio-temporal variability of hot and dry summer over Romania, as well as their driving mechanisms which might lead to a better predictability of these extreme events in the region.”

5. Page 2, lines 82-88: A first summary about the main objective of the paper is made, and then this sound repeated in the description of the two main objectives. I suggest rewriting this by linking the two paragraphs.

6. Page 3, lines 97-98: Figure S1, which shows the temporal evolution of the heat wave duration index (HWDI) averaged for Romania for different durations, is introduced without explain the specific definition used for HWDI. Along with this, Figure S1 results are not relevant for the study, so I would suggest not showing this figure, especially considering the high number of figures in the manuscript (plus 12 figures in the supplementary material).

7. Page 4, line 11: “(values <-1)” is referring to the values of SPI, which is cited later in the sentence. I suggest to eliminate this parentheses.

8. Page 4, line 121: the text in the parentheses is redundant. I suggest to change this with only (August SPI3 < -1).

9. Page 5, line 139: Figure 2 shows the HWDI averaged at the country level. ¿What is the meaning of that? ¿Is this the heatwave duration index averaged for Romania? If this is the case, the title of section 3.1 (summer heat waves in eastern Europe), must be changed by summer heat waves in Romania. I think that the complete analysis is centred in Romania, not using data from the rest of the countries of eastern Europe. So I think that even the title of the manuscript must be changed in order not to confuse to the reader.

10. Page 5, line 148: In table S1 results of the trend analysis for HWDI are shown. The trend analysis uses de Mann-Kendall test to detect the trend, but what method is used for trend estimation? All this information should be described in Methodology Section. A review of the methodology section is necessary.

11. Page 5, lines: 160-161: the average duration of HWs during the period 1950-1970 shows in Fig. 2g is lesser than 10 days.

12. Page 7, lines 220-223: This is repeated and was already explained in the methodology section.

13. Page 7, lines 228-235: I think that there are some errors in Figure 5. For example, in Figure 5a is stated June 2002 as one of the driest years. However I find from Fig. 5a that is 2003. ¿Is this correct? Similarly, from Figure 5g for SPI3, years 2002 and 2018 are stablished as driest summers. I find in this Figure that the years correspond to 2003 and 2012, respectively. Also, the quality of the Figures should be improved.

14. Page 7, line 233: Again authors are referring to the eastern part of Europe. However, the analysis is just for Romania.

15. Page 8, lines 264-268: I consider that this paragraph should be in Introduction section, and not in the results.

16. Page 9, line 307: The methodology used for ranking maps is explained in the supplementary material. I suggest to change it to the methodology section.

17. Page 9, line 318: In Figure S6 the location of the 2D atmospheric blocking is shown. The algorithm for the 2D atmospheric blocking index is also described in the supplementary material. I suggest to change it into the Methodology section.

18. Page 10, lines 316-324: In this paragraph is stablished that the pattern resulting from the atmospheric conditions is an increase in the number of hot days, especially in the southern and eastern regions of Romania. I cannot see this from figures 10e and S5. The evolution of the Tx anomaly (Figure S5) shows that this is maximum for the northern Romania.

19. Page 11, line 374: I suggest to introduce a briefly description of the stability map methodology in the Methodology section. In summary, almost all the methodology used is explained in the supplementary material, which presents almost the same number of figures as the manuscript itself. Also, all the figures in the supplementary material are described in detail in the manuscript text, because they are supporting the results, so it is logical to think that they should be a specific part of the manuscript, and not supplementary material. In this sense, I consider that a restructuring of the manuscript is necessary.

20. Page 11, lines 363-368: If Figure S11 shows the composites maps of Z500 and wind for the years when the HWDI index (averaged for Romania) was > 5 days I consider not appropriate the figure caption for it, which establishes the occurrence of monthly heat waves in the central part of Europe.

21. Page 11, lines 380-382: The spatial structure of Z500 anomalies (Figure S11a) is indicating advection of air from the north-eastern part of Europe into Romania, not from the south-eastern.

22. Page 12, lines 420-427: Conclusion section begins with a paragraph with conclusions from other studies and for other regions in Europe. I think that this information could be appropriate to introduce the need of making this study in Romania, in the introduction section, but not here. Additionally, later in the conclusions the Figures showing the different results found are again mentioned. These figures have been previously described in detail in the Results section, so I consider that they must not be mentioned here again. Also, I suggest to change this section by Conclusion and Discussion section.

TECHNICAL CORRECTIONS

Page 1, line 17: “2020and” should be “2020 and”.

Page 1, line 18: “HWs” should be “Heat Waves (HWs)”.

Page 2, line 40: “favors” should be “favour”.

Page 2, lines 80-90: “The paper is structured as follow in Section 2 we give a detailed description of the data and methods used in this study. In Section 3 we…” should be “The paper is structured as follow: in Section 2 we give a detailed description of the data and methods used in this study; in Section 3 we…”

Page 5, line 158: “centuries” must be changed by “decades”

Page 7, line 216: To add (Figure 4c).

Page 7, line 217: to add (Figure 4d)

Page 7, lines 238-239: (Figure 6a) was already indicated at the beginning of the sentence.

Page 10, line 322: Figure S6d-S6g should be Figure S5d-S5g.

Page 10, line 335: In figure 11e contours indicating the countries are in white colour and this does not permit to visualize them correctly.

Page 11, line 370: “map s” should be “maps”.

Supplementary material:

In Figures from S6 to S12 the longitude and latitude labels must be indicated, at least in the maps of the lowest row.

In Figure S8, the contour lines indicating persistent atmospheric blocking system are very difficult to appreciate.

General Overview:

The authors analyze the hotspots for warm and dry summers in Romania using E-OBS and a regional dataset covering Romania. The authors intend to study the spatio-temporal variability and trends of hot and dry summers in the eastern part of Europe, focusing on Romania, between 1950 and 2020 and to study the relationship between the frequency of hot summers and the prevailing large-scale atmospheric circulation.

The manuscript fails in different aspects. Please find my major comments below:

• The authors should use in their analysis the SPEI over SPI. The SPEI is designed to consider both precipitation and potential evapotranspiration (PET) in determining drought. Thus, unlike the SPI, the SPEI captures the main impact of increased temperatures on water demand.

Vicente-Serrano et al. (2012)

• I don´t understand the use of ROCADA database. One can argue that ROCADA use more weather station in the computation of the gridded dataset and therefore finer spatial scales will be resolved. However, throughout the manuscript it’s not clear the different between EOBS and ROCADA neither the conclusion is different when using one or another. Therefore, I would go with the long -term dataset EOBS.

• Section 3.3, this section intends to analyze the compound events in terms of hot and dry extremes. Are SPI < -1 really extreme? I don´t agree with the method used for defining compound event. They are only based on a month-to-month comparison and don´t go into further detail. What led to what? Pre-conditioning of soil moisture probably plays a role in the major Heat waves in the region. Have the authors though in using bi-variate methods to analyze the compound events? Or even to do a lag analysis between the dry and month summers?

Zscheischler and Fischer, 2020 : 10.1016/j.wace.2020.100270

Ribeiro at al., 2020 : 10.1016/j.wace.2020.100279

• Section 3.4., there is some lack of novelty in analyzing the synoptic meteorological patterns of the specific droughts years. No statistically significance is presented. 

Sousa et al., 2021 : 10.1175/JCLI-D-20-0658.1

Therefore, all the changes need to be made, in order to the paper goes for a second round of revision.

The manuscript does not have enough scientific merit to be published in the journal. It does not provide significantly new information which go beyond the current state of the art. It is descriptive and does not add new elements in current understanding of compound extremes in the area. Further a few parts are inconsistent and also not scientifically sound. Thus, I suggest rejection. A detailed review is provided below.

The title is misleading as it only deals with Romania, not Eastern Europe.

The abstract mentions that compound extremes are considered. However, it only reports on changes in extreme temperature and precipitation/drought separately. There is a lot of methodological papers out in the literature that deal with compound extremes, how they are modelled using sophisticated methods.

Introduction: There is a lot of literature that deals with extremes in southeastern Europe, including Romania, for instance also in the form of reviews:

Kuglitsch, F. et al. 2010: Heat Wave Changes in the Eastern Mediterranean since 1960, Geophys. Res. Lett., 37, L04802.

Ulbrich, U. et al. 2013: Climate of the Mediterranean: synoptic patterns, temperature, precipitation, winds and their extremes. Future Climate Projections. In: Regional Assessment of Climate Change in the Mediterranean: A. Navarra, L. Tubiana (eds.), Springer

The introduction does not provide a clear justification why this work is needed, does not show gaps in current understanding and does not formulate a clear hypothesis.

The choice of more than 5 days defining a heatwave is not objectively based (lines 109/110). My suggestion would be to consult the latest literature that deal with more objective measures how heatwaves are defined.

Heatwave results reported (lines 149-150, 4 to 5 days) are in disagreement with the definition provided in lines 109/110 that state more than 5 days.

Lines 160/161, there is an overlap of having the year 1970 in both periods. In addition, the period 1970-1985 has a different length compared to others that makes the comparison difficult.

SPI is not the most appropriate measure for drought. For the area, SPEI is a better index that combines temperature and precipitation. For a review please consult Raible et al. (2017): Drought indices revisited – improving and testing of drought indices in a simulation of the last two millennia for Europe, Tellus A: Dynamic Meteorology and Oceanography, 69, 1287492.

The comparison between EOBs and Rocada does not provide new evidence, it could be skipped and the analysis could be concentrated on EOBs.

Sentence on lines 122-123 is not clear.

The manuscript states at various places “statistical significant” changes. However, no information on the underlying statistics to test significance is provided. A few maps show significant areas related to trends, however it is missing how those regions are calculated. Further, the synoptical maps do not have a field significance information and thus they are difficult to interpret. Further, the maps are not unexpected and the processes that lead to drought or heat extremes are well documented in the literature elsewhere.

The conclusions include information from the introduction and duplicate the results. As such, a lot of information is irrelevant and the last paragraph is not a conclusion from the analysis shown.

General comment

The authors present a comprehensive analysis regarding the spatial-temporal variability of hot and dry summer in Romania, as well as their combined effect (compound events) that can be considered a novelty for this region. Various characteristics of the heat waves (duration, spatial extent and frequency) and drought index (represented by the standardized precipitation index) have been analysed in terms of their long term trends, decadal variability and driving factor triggering their occurrence. The title “Hotspots for warm and dry summers in eastern Europe, with a focus on Romania” is not fit very well with the analysis carried out in this article that is done only for Romania and can not be extended over the entire eastern Europe. Therefore, the title could be changed. Same comment for the title of the sections (3.1, 3.2,....).

The manuscript represent a substantial contribution to the understanding of large-scale mechanisms controlling the variability of extreme clime events (heatwaves and droughts) in a region with complex topographic features such as Romania. The scientific approaches and the applied methods are valid at international standards. The results are discussed in an appropriate and balanced way, including appropriate references. The tables and figures are appropriate to present the results. The conclusions are clear presented and supported by the obtained results.

Considering all these aspects, the manuscript can be published with some minor corrections presented in my specific comments and technical corrections.

Specific comments

1) Pag 13, L429, “in the eastern part of Europe” should be completed with “focusing on Romania” since the heatwaves have been analysed only over Romania.

2) Pag 13, L431-432: The conclusion “i) the length, spatial extent and frequency of HWs in Romania has increased significantly over the last 70 years, for all summer month” is not entirely true; there is a strong decadal variability with a slight decreasing before ‘70 years (a correct change point could be find by using the Pettit test) and a real (accelerated) increase after 1990. I) and ii) could be combined. This is also in agreement with the author’s comments presented in sections 3.1.

3) Pag 13, L438-439, “A significant increase in the frequency of hot extremes has been found at country level, with the most affected regions being in the north-western part and the Dobrogea region (Figure 4)”. As I see in Fig. 4, this conclusion is true only on seasonal scale (JJA) and the most affected regions are more extended than those mentioned by the authors. There are differences between the individual months: for example, in July, only a small part (eastern regions and some very small south-western areas) exhibit significant trends. In Table S1 is not clear if the trend is computed for the time series of spatial average over the country. This should be clear mentioned.

4) Pag 13, L459, “The occurrence of HWs in the eastern part of Europe”, “ the eastern part of Europe” should be replaced by “Romania”, see also my previous comments

Technical corrections

- Pag3, L86 “to determined” change in “to determine”

- Pag4, L111-112 “values < 1”, insert “SPI1 after “1-month”

- Pag 5, L158, change “two centuries” with “two decades”.

- Pag 7, L233, replace “Europe” with “Romania”.

- Pag 12, L414: “August 2015 is also the month with the longest July heatwave (Figure 2c)” something is not correct here: .”July heatwave (Figure 2c)” should be replaced by “August heatwave (Fig. 2e)”

- Pag 12, L417, “periods (i.e. 1950 – 200”: 1950 – 200 should be replaced by 1950-2020.

- Pag 12, L427: I think that “To extended the overview also” should be replaced by “To extend....”

- Pag 13, L429, “in the eastern part of Europe” should be completed with “focusing on Romania” since the heatwaves have been analysed only over Romania.

- Pag 14, L475, “IPP report (IPCC, 2021)”, IPP should be replaced by IPCC.

- References: please check all references, there are some duplications in the authors’ list: example, Bădăluță, C.-A., Perșoiu, A., Ionita, M., Nagavciuc, V., Bistricean, P.-I. P.-I., Persoiu, A., Ionita, M., Nagavciuc, V. and Bistricean, P.-I. P.-I.: Stable H and O isotope-based investigation of moisture sources and their role in river and groundwater recharge in the NE Carpathian Mountains, East-Central Europe, Isot. Environ. Heal. Stud., 55(2), 1–18,......