the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Drought evolution characteristics of the Qinghai-Tibet Plateau over the last 100 years based on SPEI
Abstract. The standardized precipitation evapotranspiration index (SPEI) of the Tibetan Plateau was calculated using the CRU4.03 gridded dataset from 1901 to 2018 in this paper. Then, based on the SPEI data, drought on the Qinghai-Tibet Plateau was studied in terms of its spatial and temporal distributions and its changing characteristics over the last 100 years. The results revealed that the precipitation in the southeastern part of the Qinghai-Tibet Plateau has been steadily rising over the last 100 years, in conjunction with only minor temperature shifts. In the northwestern part of the plateau, precipitation has decreased significantly, accompanied by a significant increase in temperature. The drought on the Tibetan Plateau showed a clear gradual increase in aridity from southeast to northwest over the last hundred years. The SPEI also showed distinct seasonal patterns, steadily increasing in spring and summer and decreasing significantly in autumn and winter. In addition, each season had its own spatial characteristics. The northeastern part of the plateau, except the Qaidam Basin, showed a significant aridity trend in all seasons. A wet trend prevailed in the southeastern and southern areas. Drought on the Tibetan Plateau exhibits apparent cyclical oscillations with a main period of 54 years and has different cyclical characteristics in different seasons.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(1474 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on nhess-2021-73', Anonymous Referee #1, 24 Apr 2021
The paper of Wang et al. addresses drought trends in the Tibetan Plateau using the SPEI index and attempts to provide some indications about the main temporal and spatial trends in the years 1901-2018. Nevertheless, the manuscript is very far from achieving its objectives, because it has many drawbacks leading me to suggest its rejection. The main problems are:
- The paper does not provide any novelty either concerning methodology or results. It mainly looks like a technical report, but rather confused.
- The paper is very poorly organized, with several serious drawbacks in every Section (please refer to specific comments).
- Methodology explanation has many unclear aspects. In particular, the spatial scale of the problem is not declared. If the authors use a spatially distributed database, I don’t understand why they talk about spatial interpolation. Furthermore, please note that a global SPEI database, based on CRU TS dataset v4.03, already exists: https://spei.csic.es/database.html
- There’s practically no discussion. The authors never try to explain the reason for the results they get. Furthermore, results are formally inaccurate (e.g., the term “significant” is used with excessive ease).
Therefore, I strongly suggest that the authors deeply rethink both research aims and the structure of their paper and try to take more advantage of the only original part of their study, consisting of retrieving drought hazard data, which could be more fruitfully utilized for validating global datasets.
Please find below my specific comments.
- Abstract
It is quite confusing concerning results.
L13: “The drought on the Qinghai-Tibetan Plateau showed a clear gradual increase in aridity from southeast to northwest over the last hundred years” I guess the increase concerns the aridity trend
- Introduction
It is poor in many details. A more thorough discussion is needed concerning the choice of the drought index. Recent studies on climate trend over such an important and wide area must be described in much more detail. Even the choice of the climate dataset must be adequately justified, highlighting the (not few) drawbacks of the selected dataset and other datasets available. E.g., how does the selected dataset work on the vast study area? In what subregions are it more or less reliable? Is it more performing in the most recent years and less in the past? Why is it preferable to others? Still concerning the dataset, please consider that the last version of CRU TS is the 4.05 released on 17 March 2021. Even though I acknowledge that it has been released very recently, some discussion about the differences with the selected version (released in May 2019) is needed.
- Methods (not Method)
Many ready-to-use tools exist to calculate SPEI and to perform Mann-Kendall tests and wavelet analyses. If the authors used some of these tools, as I guess, they should explicitly state it.
- Section 2.1
Explanation of SPEI is not clear, though it is a quite well-known index. The time scale used for SPEI is never declared.
Sometimes present tense is used, other times past tense. This needs to be checked.
Equations: Equations numbers are all (0.0)
L59: lowercase gamma
L61: why interpolation? It’s just a difference
L63: PETi
L67: parameters b and c?
L73: what about W0 and W2? Are all the Wi terms calculated using the equation below?
- Section 2.2
The authors abruptly introduce the Mann-Kendall test without explaining why they need it
- Data and processing
3.2 Data preprocessing
Why the bilinear interpolation algorithm was used? Was the CRU spatial resolution modified? I can’t find any statement related to this point throughout the text so far. What are the “points to be measured”?
Most important, no detailed information about the study area is provided (and, of course, no maps delimiting the study area).
- Results
Fig. 1 does not explain what the authors state in the text. Furthermore, it is not clear. Titles of vertical axes are not provided. The meaning of “Frequency of droughts (L)” per year is not clear at all.
Fig. 2: again, what do the authors mean with drought frequency? More generally, the reasons provided by authors to justify the use of SPEI compared to historical droughts (not clear how they were recorded) is weak and not objectively presented.
L156: so, I understand that the authors did not need to calculate potential evapotranspiration, but they used it directly from the CRU dataset.
Fig.3: not clear what is the difference between the SPEI line here and in Fig. 1. Furthermore, I guess that the overall linear trend is not significant.
LL157-159: here it looks like the authors start some reasoning about the overall time trend, then they abruptly shift to spatial heterogeneity.
Section 4.3: what do the authors mean by the term “significant” from the statistical point of view? This aspect is of the foremost importance to understand the main message. Not clear how the different time periods were subdivided.
L186: Aljinshan … Qaidam Basin … Ali Plateau… etc: these toponyms are unknown to most of the readers. As a consequence, this paragraph is almost incomprehensible. That’s why a map of the study area is needed.
Section 4.4: the periodic analysis and Figs. 6 and 7 are poorly presented and discussed. It is neither recalled later.
- Discussion
Section 5.1: actually, that’s not Discussion, but presenting further results concerning precipitation and temperature, which, if really needed, should have been presented before SPEI results (since SPEI depends on precipitation and temperature). As a matter of fact, in the Conclusions, the authors first recall precipitation and temperature results, then SPEI.
Fig. 8 is not well described: what are UB and UF? What are the units and what do the vertical axes mean?
L236: “Zhang and colleagues (Zhang Wangxiong et al., 2019)” must be rewritten Zhang et al. (2009).
As before, when the authors explain spatial changes referring to toponyms, the text is incomprehensible.
Section 5.2: that’s not Discussion, it is coming back to Results and presenting them in another (and possibly not consistent) way.
- Conclusions (not Conclusion and Discussion)
LL266-267: that’s not true, the main topic of the paper is the SPEI index, not precipitation and temperature that are presented only in Section 5.1.
LL277-281 and LL297-306: two long sentences are repeated twice. This is a sign of insufficient carefulness in the final drafting of the manuscript.
Citation: https://doi.org/10.5194/nhess-2021-73-RC1 -
AC1: 'Reply on RC1', Shengzhen Wang, 18 Jul 2021
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2021-73/nhess-2021-73-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Shengzhen Wang, 18 Jul 2021
-
RC2: 'Comment on nhess-2021-73', Anonymous Referee #2, 26 Apr 2021
I have reviewed the manuscript nhess-2021-73 "Drought evolution characteristics of the Qinghai-Tibet Plateau over the last 100 years based on SPEI" by Wang et al. The paper attempts at a description of drought features over the Tibet Plateau by means of a statistical analysis of the SPI index over a very long period of time. The topic of the research is up to date and coherent with the scope of the journal and the specific case study is potentially interesting but I regret to say that the current version of the manuscript represents for presentation, methodology, analysis and results commenting a poor contribution to the current and relevant scientific literature. I would not recommend the paper for publication and warmly invite authors for an in-depth revision of the research framework prior to a further submission.
In the following some specific comments.
1)Title: the title only refers to the SPEI index but in the end the paper also analyses the precipitation and temperature trends.
2)Abstract: it is a confuse description of the research idea and content of the paper, it highlights the fact authors need to revise the research framework.
2)Introduction: after a fair introduction of the current relevant literature, it confusedly presents the proposed research work.
3)Method: the description of the methodology is not exhaustive. In the presentation of the results, I only saw representation of one single time series. What does it represent? Is it the average over the region? If so, I do not believe the average over a large region is sufficient to explain a behaviour. Provided the fact that in the results also gridded maps are represented, I guess more time series should be available, one for each cell of the grid. How do authors deal with the spatial dimension of the problem then? And also how do authors deal with the temporal dimension of the problem? It is well known the SPEI, such as SPI, can be assessed over different accumulation time scales, each of which has a specific significance. This specific issue is never discussed in the paper.
4)Data and preprocessing: overall the data are very poorly described. The climate data are poorly presented, probably a graphical comparison between rain gauge and grid data could have been important to explain the need for the use of CRU 4.03. The geographical context is not described at all. It is not clear what drought hazard data are. Data processing: how? But more importantly why?
5) Results and discussion: these sections are very poor for both the content and the interpretation provided. The feeling is that the authors are not familiar with statistical analysis of time series or at least with the interpretation of the relevant results. The time scale and the spatial scale of the problem are not considered properly, as previously mentioned. Figures are not well described, captions are not complete, frequently the reader does not know what is looking at and sometimes the interpretation of the results are wrong.
What is Figure 1 representing? The average SPEI? What is a typical annual drought frequency? How do we guess from Figure 1 and 2 there is a good agreement between SPEI and drought frequency (who is it assessed?) for the case study? Unless the legend of Figure 2 is wrong, the two panels (upper and lower) describe different spatial patterns. Drought frequency is large in the northern area which is instead the region with the lower SPEI (again what index are we looking at?).
Figure 3: again, which index is this?
Figure 4: we cannot assess temporal trends over short period of time (or at least we can but there is no meaning), so how authors assessed the interplay for an increasing and decreasing trend in the single seasonal time series? This makes no sense to my opinion.
Figure 5: the legend should illustrate number not just quality of the mapped attributes.
Figure 6-7: the wavelet analysis, which authors mentioned to engage in the abstract is very poorly undertaken. What is the physical meaning? How does it compare or add to the previous results of the time series analysis?
Figure 8: what does Figure 8 illustrate? According to the caption and axes titles, should be the trend of temperature and precipitation. But this should be an univocal number for each single time series. Additionally in the methodology authors never say how they assess the magnitude of the trend, which is not provided by the Mann-Kendall test. In the first and third panels, what does the pattern illustrate? Why are they depicted in blue and red?
Figure 9: again the legend should illustrate number not just quality of the mapped attributes.
Citation: https://doi.org/10.5194/nhess-2021-73-RC2 -
AC2: 'Reply on RC2', Shengzhen Wang, 18 Jul 2021
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2021-73/nhess-2021-73-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Shengzhen Wang, 18 Jul 2021
Interactive discussion
Status: closed
-
RC1: 'Comment on nhess-2021-73', Anonymous Referee #1, 24 Apr 2021
The paper of Wang et al. addresses drought trends in the Tibetan Plateau using the SPEI index and attempts to provide some indications about the main temporal and spatial trends in the years 1901-2018. Nevertheless, the manuscript is very far from achieving its objectives, because it has many drawbacks leading me to suggest its rejection. The main problems are:
- The paper does not provide any novelty either concerning methodology or results. It mainly looks like a technical report, but rather confused.
- The paper is very poorly organized, with several serious drawbacks in every Section (please refer to specific comments).
- Methodology explanation has many unclear aspects. In particular, the spatial scale of the problem is not declared. If the authors use a spatially distributed database, I don’t understand why they talk about spatial interpolation. Furthermore, please note that a global SPEI database, based on CRU TS dataset v4.03, already exists: https://spei.csic.es/database.html
- There’s practically no discussion. The authors never try to explain the reason for the results they get. Furthermore, results are formally inaccurate (e.g., the term “significant” is used with excessive ease).
Therefore, I strongly suggest that the authors deeply rethink both research aims and the structure of their paper and try to take more advantage of the only original part of their study, consisting of retrieving drought hazard data, which could be more fruitfully utilized for validating global datasets.
Please find below my specific comments.
- Abstract
It is quite confusing concerning results.
L13: “The drought on the Qinghai-Tibetan Plateau showed a clear gradual increase in aridity from southeast to northwest over the last hundred years” I guess the increase concerns the aridity trend
- Introduction
It is poor in many details. A more thorough discussion is needed concerning the choice of the drought index. Recent studies on climate trend over such an important and wide area must be described in much more detail. Even the choice of the climate dataset must be adequately justified, highlighting the (not few) drawbacks of the selected dataset and other datasets available. E.g., how does the selected dataset work on the vast study area? In what subregions are it more or less reliable? Is it more performing in the most recent years and less in the past? Why is it preferable to others? Still concerning the dataset, please consider that the last version of CRU TS is the 4.05 released on 17 March 2021. Even though I acknowledge that it has been released very recently, some discussion about the differences with the selected version (released in May 2019) is needed.
- Methods (not Method)
Many ready-to-use tools exist to calculate SPEI and to perform Mann-Kendall tests and wavelet analyses. If the authors used some of these tools, as I guess, they should explicitly state it.
- Section 2.1
Explanation of SPEI is not clear, though it is a quite well-known index. The time scale used for SPEI is never declared.
Sometimes present tense is used, other times past tense. This needs to be checked.
Equations: Equations numbers are all (0.0)
L59: lowercase gamma
L61: why interpolation? It’s just a difference
L63: PETi
L67: parameters b and c?
L73: what about W0 and W2? Are all the Wi terms calculated using the equation below?
- Section 2.2
The authors abruptly introduce the Mann-Kendall test without explaining why they need it
- Data and processing
3.2 Data preprocessing
Why the bilinear interpolation algorithm was used? Was the CRU spatial resolution modified? I can’t find any statement related to this point throughout the text so far. What are the “points to be measured”?
Most important, no detailed information about the study area is provided (and, of course, no maps delimiting the study area).
- Results
Fig. 1 does not explain what the authors state in the text. Furthermore, it is not clear. Titles of vertical axes are not provided. The meaning of “Frequency of droughts (L)” per year is not clear at all.
Fig. 2: again, what do the authors mean with drought frequency? More generally, the reasons provided by authors to justify the use of SPEI compared to historical droughts (not clear how they were recorded) is weak and not objectively presented.
L156: so, I understand that the authors did not need to calculate potential evapotranspiration, but they used it directly from the CRU dataset.
Fig.3: not clear what is the difference between the SPEI line here and in Fig. 1. Furthermore, I guess that the overall linear trend is not significant.
LL157-159: here it looks like the authors start some reasoning about the overall time trend, then they abruptly shift to spatial heterogeneity.
Section 4.3: what do the authors mean by the term “significant” from the statistical point of view? This aspect is of the foremost importance to understand the main message. Not clear how the different time periods were subdivided.
L186: Aljinshan … Qaidam Basin … Ali Plateau… etc: these toponyms are unknown to most of the readers. As a consequence, this paragraph is almost incomprehensible. That’s why a map of the study area is needed.
Section 4.4: the periodic analysis and Figs. 6 and 7 are poorly presented and discussed. It is neither recalled later.
- Discussion
Section 5.1: actually, that’s not Discussion, but presenting further results concerning precipitation and temperature, which, if really needed, should have been presented before SPEI results (since SPEI depends on precipitation and temperature). As a matter of fact, in the Conclusions, the authors first recall precipitation and temperature results, then SPEI.
Fig. 8 is not well described: what are UB and UF? What are the units and what do the vertical axes mean?
L236: “Zhang and colleagues (Zhang Wangxiong et al., 2019)” must be rewritten Zhang et al. (2009).
As before, when the authors explain spatial changes referring to toponyms, the text is incomprehensible.
Section 5.2: that’s not Discussion, it is coming back to Results and presenting them in another (and possibly not consistent) way.
- Conclusions (not Conclusion and Discussion)
LL266-267: that’s not true, the main topic of the paper is the SPEI index, not precipitation and temperature that are presented only in Section 5.1.
LL277-281 and LL297-306: two long sentences are repeated twice. This is a sign of insufficient carefulness in the final drafting of the manuscript.
Citation: https://doi.org/10.5194/nhess-2021-73-RC1 -
AC1: 'Reply on RC1', Shengzhen Wang, 18 Jul 2021
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2021-73/nhess-2021-73-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Shengzhen Wang, 18 Jul 2021
-
RC2: 'Comment on nhess-2021-73', Anonymous Referee #2, 26 Apr 2021
I have reviewed the manuscript nhess-2021-73 "Drought evolution characteristics of the Qinghai-Tibet Plateau over the last 100 years based on SPEI" by Wang et al. The paper attempts at a description of drought features over the Tibet Plateau by means of a statistical analysis of the SPI index over a very long period of time. The topic of the research is up to date and coherent with the scope of the journal and the specific case study is potentially interesting but I regret to say that the current version of the manuscript represents for presentation, methodology, analysis and results commenting a poor contribution to the current and relevant scientific literature. I would not recommend the paper for publication and warmly invite authors for an in-depth revision of the research framework prior to a further submission.
In the following some specific comments.
1)Title: the title only refers to the SPEI index but in the end the paper also analyses the precipitation and temperature trends.
2)Abstract: it is a confuse description of the research idea and content of the paper, it highlights the fact authors need to revise the research framework.
2)Introduction: after a fair introduction of the current relevant literature, it confusedly presents the proposed research work.
3)Method: the description of the methodology is not exhaustive. In the presentation of the results, I only saw representation of one single time series. What does it represent? Is it the average over the region? If so, I do not believe the average over a large region is sufficient to explain a behaviour. Provided the fact that in the results also gridded maps are represented, I guess more time series should be available, one for each cell of the grid. How do authors deal with the spatial dimension of the problem then? And also how do authors deal with the temporal dimension of the problem? It is well known the SPEI, such as SPI, can be assessed over different accumulation time scales, each of which has a specific significance. This specific issue is never discussed in the paper.
4)Data and preprocessing: overall the data are very poorly described. The climate data are poorly presented, probably a graphical comparison between rain gauge and grid data could have been important to explain the need for the use of CRU 4.03. The geographical context is not described at all. It is not clear what drought hazard data are. Data processing: how? But more importantly why?
5) Results and discussion: these sections are very poor for both the content and the interpretation provided. The feeling is that the authors are not familiar with statistical analysis of time series or at least with the interpretation of the relevant results. The time scale and the spatial scale of the problem are not considered properly, as previously mentioned. Figures are not well described, captions are not complete, frequently the reader does not know what is looking at and sometimes the interpretation of the results are wrong.
What is Figure 1 representing? The average SPEI? What is a typical annual drought frequency? How do we guess from Figure 1 and 2 there is a good agreement between SPEI and drought frequency (who is it assessed?) for the case study? Unless the legend of Figure 2 is wrong, the two panels (upper and lower) describe different spatial patterns. Drought frequency is large in the northern area which is instead the region with the lower SPEI (again what index are we looking at?).
Figure 3: again, which index is this?
Figure 4: we cannot assess temporal trends over short period of time (or at least we can but there is no meaning), so how authors assessed the interplay for an increasing and decreasing trend in the single seasonal time series? This makes no sense to my opinion.
Figure 5: the legend should illustrate number not just quality of the mapped attributes.
Figure 6-7: the wavelet analysis, which authors mentioned to engage in the abstract is very poorly undertaken. What is the physical meaning? How does it compare or add to the previous results of the time series analysis?
Figure 8: what does Figure 8 illustrate? According to the caption and axes titles, should be the trend of temperature and precipitation. But this should be an univocal number for each single time series. Additionally in the methodology authors never say how they assess the magnitude of the trend, which is not provided by the Mann-Kendall test. In the first and third panels, what does the pattern illustrate? Why are they depicted in blue and red?
Figure 9: again the legend should illustrate number not just quality of the mapped attributes.
Citation: https://doi.org/10.5194/nhess-2021-73-RC2 -
AC2: 'Reply on RC2', Shengzhen Wang, 18 Jul 2021
The comment was uploaded in the form of a supplement: https://nhess.copernicus.org/preprints/nhess-2021-73/nhess-2021-73-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Shengzhen Wang, 18 Jul 2021
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
855 | 447 | 47 | 1,349 | 41 | 49 |
- HTML: 855
- PDF: 447
- XML: 47
- Total: 1,349
- BibTeX: 41
- EndNote: 49
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1
Cited
3 citations as recorded by crossref.
- Temperature contributes more than precipitation to the greening of the Tibetan Plateau during 1982–2019 T. Bafitlhile & Y. Liu 10.1007/s00704-021-03882-9
- Distribution Patterns of Gymnosperm Species along Elevations on the Qinghai–Tibet Plateau: Effects of Climatic Seasonality, Energy–Water, and Physical Tolerance Variables M. Umair et al. 10.3390/plants12234066
- Appropriated protection time and region for Qinghai–Tibet Plateau grassland S. Qian et al. 10.1515/geo-2022-0383