Evolutionary Characteristics of Daytime and Nocturnal Precipitation Heterogeneity in Gansu Province, Northwest China

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents a study on the characteristics of precipitation in a region of China, using different indices and daily data from “23” stations over 50 years. The study does not present a new methodology, but analyzes the data for a specific region, so, the study, based on those data, must be as clear and useful for other researchers as possible, so that the whole article is useful and its publication makes sense. I think that, in the current state of the manuscript, the presentation of the results and the corresponding conclusions are not very clear, making it difficult to follow the paper and to understand the provided results. 

 

Some specific comments are provided below:

 

1.- The introduction should be better organized, starting with global, national and then local results. Currently, they refer to global results, then to results for China, then global results again, ... 

 

2.- Lines 73-74 PCD and PCP were previously defined but not PCI, CL and Q.

 

3.- Lines 111-112 (... all of which use 0.1 mm as the lower limit to judge whether precipitation occurs) and 162-63 (The average annual precipitation days (that is, precipitation frequency) of the entire province …). 

Usually, “wet days” are defined as those with precipitation > 1 mm (https://www.wcrp-climate.org/images/observations/etccdi/Extract-WG1AR5_Chapter02_FINAL.pdf by https://www.wcrp-climate.org/etccdi ). If 0.1 mm is used as a threshold to classify wet days in this study, it should be clearly specified and also justified. 

Moreover, is 0.1 mm also considered for the “entire province”? I mean, is 0.1 mm the threshold applied to the average precipitation for each day? I assume that to create the time series the average of the 23 stations is calculated for each day. It should be also clarified.

 

Perhaps the term precipitation frequency (days) is not the most appropriate for the number of wet days in a year, because one would expect units in percentage or something similar.

 

4.- The conversion of precipitation data (for each pentad) to vectors should be better explained, although the corresponding reference is provided (in which it is explained very clearly). Mainly the idea that the angle (direction) corresponds to the pentad and the amount of precipitation to the vector magnitude.

 

5.- Line 165 “... the evolution trend of precipitation intensity (4.2±0.5 mm/d; Figure 2d)” 

Figure 2d does not correspond to the time series of precipitation intensity and that trend is not indicated.

 

6.- In several plots axis titles are mixed with labels. The title location should be corrected.

 

7.-  Figure 2d and corresponding text. Lines 170-171 “... the annual precipitation was mainly dominated by precipitation frequency (r=0.58, p<0.001; Figure 2d), followed by precipitation intensity (r=0.48, p<0.001 …”

Precipitation intensity is not defined. I assume the authors are using SDII (Simple daily intensity index), ratio of annual total precipitation to the number of wet days. So, 

 

Annual precipitation (mm) = Precipitation intensity (mm/day) x Precipitation frequency (days)

 

This simple relationship could be used to estimate the importance of the changes in each of the  factors on the change in annual precipitation. Perhaps partial correlations could be also used to study these contributions. However, the authors made two linear fits (Precipitation intensity vs. Annual precipitation and Precipitation frequency vs. Annual precipitation) separately. 

 

8.- Table 1: Same as in comment 7. The correlations coefficients are computed separately. 

 

What is the meaning of the shadowed rows? It should be indicated in the table caption.

How are the stations sorted in this table? Why?

The altitude is indicated, but the mean annual precipitation could be also useful to facilitate interpretation of some comments in the text.

 

In  the text, for example in the abstract, the authors mentioned 23 meteorological stations, however in this table only 20 stations are shown. 

 

Lines 104-107: “... 23 stations (Figure 1) with complete data and strict quality control were selected for the period 1970-2019. “ Based on what? Data gaps…?. It should be briefly commented.

 

9.- Lines 192-193. The  mean precipitation frequencies for daytime and nighttime are not significantly different, as can be seen from the confidence intervals for each one.

 

10.- Lines 223-224: “ … the proportion of precipitation belonging to arid, semiarid, and semihumid areas was roughly… “ Are these three regions indicated in any figure or table?

 

11.- LInes 242-247: Results commented are difficult to follow if they are not indicated in tables or figures, or summarized in some averages or values. For example, when the authors say “... the more prominent performance in high-altitude areas…” Are the differences with lower altitudes statistically significant? What are the stations considered to be high altitude? Where are they labeled?

 

12.- Lines 250-256: Changes in precipitation could be due to thermodynamic changes, but also to modifications in synoptic patterns, causing certain conditions to occur more or less frequently. It seems like a biased explanation and impossible to verify with the data provided. The conclusions point to more in-depth studies of this type, but they have not been carried out.

 

13. Lines 258-265 and Figure 3. Given the average PCD values and confidence intervals (to one standard deviation), is there any station that really differs from another? It seems that the behavior is very similar. They point out 3 stations that have a medium PCD, but the segmentation is arbitrary and the differences are very small. I think the only thing that can be said is that there are no significant differences between stations.

 

14. Lines 284-285: “From the perspective of spatial variation, it is easy to recognize that PCD increases gradually with decreasing annual precipitation (Figure 3a).” 

Figure 3b?

 

15. Figure 4 and lines 314-315: The four most representative weather stations have been chosen. What makes them the most representative? For the figure, is the rainfall of all of them averaged? 

Figure 4 is, perhaps, one of the easiest to understand. I think it has been included because the indices used, PCD and PCP, although they condense the main characteristics into a value, are complicated to interpret. The main problem is that a change in one of the indices could be due to many causes, there may even be changes that compensate each other and would not be detected by the indices. Finally, perhaps it would be necessary to use annual cycles to really explain what is happening.

 

16.- “As a result, PCP in arid areas tends to have a wider range of variations, with an advance to mid-June or a delay to mid-August being normal.” 

What stations belong to arid zones? They have not been labeled that way in any table or figure.

 

17.- Lines 234-235 “... revealed that at least one of them had reached a significant increase level (p<0.05) in most areas (87%), and only 3 stations (13%) did not have a clear trend (Table 1).”

Table 1 does not summarize the trends but the linear correlations between some variables.

 

In summary, although the data and the idea are interesting, I think a thorough revision of the writing, methods, and presentation of the data is necessary for the paper to be of interest to other researchers.

Comments on the Quality of English Language

None

Author Response

Dear reviewer:

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: [The introduction should be better organized, starting with global, national and then local results. Currently, they refer to global results, then to results for China, then global results again, ... ]

Response 1: [Thank you for pointing this out. We've improved it. Given that global warming is an important cause of precipitation response, we first briefly describe the key outcomes of rising temperatures and the reality of increased hydrological hazards. Then, the second section systematically elaborates on the related research results of precipitation change (the core research object of this paper), starting from the global scale, transitioning to the Chinese scale, and finally focusing on the local and surrounding areas. It can be found on Lines 37-91 of the revised manuscript.]

Comments 2: [Lines 73-74 PCD and PCP were previously defined but not PCI, CI and Q.]

Response 2: [We agree with this comment. Therefore, the full names of the above indicators have already been written out when they first appear. It can be found on Lines 75-76 of the revised manuscript.]

Comments 3: [Lines 111-112 (... all of which use 0.1 mm as the lower limit to judge whether precipitation occurs) and 162-63 (The average annual precipitation days (that is, precipitation frequency) of the entire province …). Usually, “wet days” are defined as those with precipitation > 1mm(https://www.wcrp-climate.org/images/observations/etccdi/Extract-WG1AR5_Chapter02_FINAL.pdf by https://www.wcrp-climate.org/etccdi ). If 0.1 mm is used as a threshold to classify wet days in this study, it should be clearly specified and also justified. Moreover, is 0.1 mm also considered for the “entire province”? I mean, is 0.1 mm the threshold applied to the average precipitation for each day? I assume that to create the time series the average of the 23 stations is calculated for each day. It should be also clarified.Perhaps the term precipitation frequency (days) is not the most appropriate for the number of wet days in a year, because one would expect units in percentage or something similar.]

Response 3: [Thank you for pointing this out. We've improved it, which can be found on Lines 116-123 of the revised manuscript. As mentioned above, the criteria for determining wet days are indeed different. 1 mm is suitable for calculating the extreme index, while 0.1 mm is stipulated in the Chinese national standard "Grade of Precipitation", GB/T 28592-2012. The calculation process of statistical results has also been described.]

Comments 4: [The conversion of precipitation data (for each pentad) to vectors should be better explained, although the corresponding reference is provided (in which it is explained very clearly). Mainly the idea that the angle (direction) corresponds to the pentad and the amount of precipitation to the vector magnitude.]

Response 4: [Thank you for pointing this out. We've improved it, which can be found on Lines 155-157 of the revised manuscript. The result that the amount of precipitation corresponds to the vector magnitude can only be calculated according to the specific data through Formulas 3-6.]

Comments 5: [Line 165 “... the evolution trend of precipitation intensity (4.2±0.5 mm/d; Figure 2d)” Figure 2d does not correspond to the time series of precipitation intensity and that trend is not indicated.]

Response 5: [Thank you for pointing this out. We have corrected this statement, which can be found on Lines 174-176 of the revised manuscript.]

Comments 6: [In several plots axis titles are mixed with labels. The title location should be corrected.]

Response 6: [Agree. We have corrected these mistakes, which can be found in Figure 2 (Lines 189-199) - Figure 5 (Lines 355-360) of the revised manuscript.]

Comments 7: [Figure 2d and corresponding text. Lines 170-171 “... the annual precipitation was mainly dominated by precipitation frequency (r=0.58, p<0.001; Figure 2d), followed by precipitation intensity (r=0.48, p<0.001 …”Precipitation intensity is not defined. I assume the authors are using SDII (Simple daily intensity index), ratio of annual total precipitation to the number of wet days. So, Annual precipitation (mm) = Precipitation intensity (mm/day) x Precipitation frequency (days). This simple relationship could be used to estimate the importance of the changes in each of the  factors on the change in annual precipitation. Perhaps partial correlations could be also used to study these contributions. However, the authors made two linear fits (Precipitation intensity vs. Annual precipitation and Precipitation frequency vs. Annual precipitation) separately.]

Response 7: [Thank you for pointing this out. We've improved it, which can be found on Lines 173-182 of the revised manuscript. There is no doubt that annual precipitation is the product of the number of precipitation days and the intensity of precipitation (SDII). However, which is the primary factor is not fixed from year to year. So through two linear regressions, we can compare which one has the greater overall effect.]

Comments 8: [Table 1: Same as in comment 7. The correlations coefficients are computed separately. What is the meaning of the shadowed rows? It should be indicated in the table caption. How are the stations sorted in this table? Why? The altitude is indicated, but the mean annual precipitation could be also useful to facilitate interpretation of some comments in the text. In  the text, for example in the abstract, the authors mentioned 23 meteorological stations, however in this table only 20 stations are shown. Lines 104-107: “... 23 stations (Figure 1) with complete data and strict quality control were selected for the period 1970-2019. “ Based on what? Data gaps…?. It should be briefly commented.]

Response 8: [Thank you for pointing this out. We've improved it, which can be found in the new Table 1 (Lines 200-204) of the revised manuscript. Response 7 may also be helpful. Shading is used to visually distinguish the number of indicators that pass the significance test. From top to bottom, the number decreases and the trend weakens. The three stations where no indicator passed the significance test were not marked in the original table.]

Comments 9: [Lines 192-193. The mean precipitation frequencies for daytime and nighttime are not significantly different, as can be seen from the confidence intervals for each one.]

Response 9: [We agree with this comment. Therefore, we first conclude that the precipitation frequency is generally slightly higher during the daytime, and then give a result of 1.05 times based on the average. It can be found on Lines 204-206 of the revised manuscript.]

Comments 10: [Lines 223-224: “ … the proportion of precipitation belonging to arid, semiarid, and semihumid areas was roughly… “ Are these three regions indicated in any figure or table?]

Response 10: [Thank you for pointing this out. We have improved it, which can be found in the new Table 1 (Lines 200-204) and Lines 236-237 of the revised manuscript.]

Comments 11: [Lines 242-247: Results commented are difficult to follow if they are not indicated in tables or figures, or summarized in some averages or values. For example, when the authors say “... the more prominent performance in high-altitude areas…” Are the differences with lower altitudes statistically significant? What are the stations considered to be high altitude? Where are they labeled?]

Response 11: [Thank you for pointing this out. These lines belong to the summary, so we have not added chart comments in the original text. To improve it, we have put a lot of effort into the results and analysis as well as the related charts, which can be found in the new Table 1 (Lines 200-204) and Lines 256-261 of the revised manuscript.]

Comments 12: [Lines 250-256: Changes in precipitation could be due to thermodynamic changes, but also to modifications in synoptic patterns, causing certain conditions to occur more or less frequently. It seems like a biased explanation and impossible to verify with the data provided. The conclusions point to more in-depth studies of this type, but they have not been carried out.]

Response 12: [Thank you for pointing this out. As mentioned above, the formation mechanism of precipitation change is extremely complex, so we can only carry out some superficial discussions based on the existing results to explore potential directions for further work. We have done our best to improve it, which can be found on Lines 265-271 of the revised manuscript.]

Comments 13: [Lines 258-265 and Figure 3. Given the average PCD values and confidence intervals (to one standard deviation), is there any station that really differs from another? It seems that the behavior is very similar. They point out 3 stations that have a medium PCD, but the segmentation is arbitrary and the differences are very small. I think the only thing that can be said is that there are no significant differences between stations.]

Response 13: [Thank you for pointing this out. Because PCD values generally fluctuate strongly from year to year, statistical differences between sites are not easy to achieve. However, relative differences in PCD can be obtained employing averages, percentages, or contrasts year by year. This is very similar to many of the results of Climate Change 2021: The Physical Science Basis, and although there are few statistically significant differences, support for the findings is widespread.]

Comments 14: [Lines 284-285: “From the perspective of spatial variation, it is easy to recognize that PCD increases gradually with decreasing annual precipitation (Figure 3a).” Figure 3b?.]

Response 14: [We agree with this comment. Therefore, We have corrected this statement, which can be found on Lines 299-300 of the revised manuscript.]

Comments 15: [Figure 4 and lines 314-315: The four most representative weather stations have been chosen. What makes them the most representative? For the figure, is the rainfall of all of them averaged? Figure 4 is, perhaps, one of the easiest to understand. I think it has been included because the indices used, PCD and PCP, although they condense the main characteristics into a value, are complicated to interpret. The main problem is that a change in one of the indices could be due to many causes, there may even be changes that compensate each other and would not be detected by the indices. Finally, perhaps it would be necessary to use annual cycles to really explain what is happening.]

Response 15: [Thank you for pointing this out. According to Table 2 (Lines 189-199), the results of these four stations are the most significant. Combined with factors such as distance and altitude, a case study is carried out on them. The results are averaged over the four stations. I couldn't agree more. But it is not easy to solve these problems quickly. All we can do is keep trying.]

Comments 16: [As a result, PCP in arid areas tends to have a wider range of variations, with an advance to mid-June or a delay to mid-August being normal.” What stations belong to arid zones? They have not been labeled that way in any table or figure.]

Response 16: [Thank you for pointing this out. We've improved it, which can be found in the new Table 1 (Lines 200-204) and Lines 369-370 of the revised manuscript. ]

Comments 17: [Lines 234-235 “... revealed that at least one of them had reached a significant increase level (p<0.05) in most areas (87%), and only 3 stations (13%) did not have a clear trend (Table 1).” Table 1 does not summarize the trends but the linear correlations between some variables.]

Response 17: [Thank you for pointing this out. For Table 1 (Lines 200-204), the last six columns of data are all correlation coefficients of the evolution trend of precipitation amount or precipitation frequency from 1970 to 2019. We make sure they're not linear correlations between some variables.]

Reviewer 2 Report

Comments and Suggestions for Authors

The anomaly of precipitation concentration is the key cause of drought and flood extreme events. In the context of climate change, precipitation concentration is likely to have significant impacts on ecosystems, agriculture, water resources and human life. In short, understanding and mastering the characteristics and laws of precipitation concentration is of great significance for coping with the challenges brought by global warming. This manuscript makes an effort to quantify the Evolution Characteristics of Daytime and Nocturnal Precipitation Heterogeneity in Gansu Province, Northwest China. Compared with the research on precipitation in Gansu Province, this manuscript does have many new highlights and findings, which is helpful to the study of precipitation in the region with complex climatic conditions. The climatic conditions in Gansu Province are extremely special, and the research of PCD and PCP of precipitation for this region is indeed very rare. This manuscript provides a more comprehensive analysis of the evolution characteristics of Precipitation at different time scales and altitudes in the region, which is of great interest for understanding the water-cycle and climate change in Gansu Province. Therefore, I feel that this has strong potential to become a published paper after making minor revisions to the detailed comments below.

Comments

1. Line 2~3: The first is the title, which is generally considered to correspond to DAYTIME and NIGHTTIME, and DIURNAL and NOCTURNAL, so I would suggest that the authors reconsider whether the title needs to be revised.

2. Line 73~76: What do the five indicators PCD, PCP, PCI, CI, and Q mean? I think the authors should explain or define them as necessary in the introduction section.

3. Line 93: How should the word RICH be understood in this sentence? Is the author trying to convey the idea that Gansu Province has a complex and diverse range of climate types? Does it need to be corrected?

4. Line 98: I don't think this is a correct citation, please revise the citation here as required by the journal of WATER.

5. Line 104: Please revise the citation here.

6. In Figure 2, there should be some spacing between the titles and values of all the subfigures, please make corrections.

Author Response

Dear reviewer:

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: [Line 2~3: The first is the title, which is generally considered to correspond to DAYTIME and NIGHTTIME, and DIURNAL and NOCTURNAL, so I would suggest that the authors reconsider whether the title needs to be revised.]

Response 1: [Thank you for pointing this out. We have carefully considered this proposal and looked up similar paper titles. The existing statement is the most common, so we tend to leave it unchanged. It can be found on Lines 2-4 of the revised manuscript.]

Comments 2: [Line 73~76: What do the five indicators PCD, PCP, PCI, CI, and Q mean? I think the authors should explain or define them as necessary in the introduction section.]

Response 2: [Agree. The full names of the above indicators have already been written out when they first appear. It can be found on Lines 75-76 of the revised manuscript.]

Comments 3: [Line 93: How should the word RICH be understood in this sentence? Is the author trying to convey the idea that Gansu Province has a complex and diverse range of climate types? Does it need to be corrected?]

Response 3: [We agree with this comment. Therefore, We have corrected this statement, which can be found on Lines 98-99 of the revised manuscript.]

Comments 4: [Line 98: I don't think this is a correct citation, please revise the citation here as required by the journal of WATER.]

Response 4: [Thank you for pointing this out. We have corrected this citation, which can be found on Line 103 of the revised manuscript.]

Comments 5: [Line 104: Please revise the citation here.]

Response 5: [Thank you for pointing this out. We have corrected this citation, which can be found on Line 109 of the revised manuscript.]

Comments 6: [In Figure 2, there should be some spacing between the titles and values of all the subfigures, please make corrections.]

Response 6: [Thank you for pointing this out. We have corrected these mistakes, which can be found in Figure 2 (Lines 189-199) of the revised manuscript.]

Reviewer 3 Report

Comments and Suggestions for Authors

Comments

l  The article is dealing with spatial and temporal distribution characteristics of precipitation in Gansu province and specifically at daytime and nocturnal scales based on long time series. The results offer a useful statistical evaluation of various significant precipitation parameters and in this frame the paper can be accepted for publication after considering some minor suggestions in the following.

l  In Abstract line 12, some correction is needed in the meaning of phrase “mechanism of the formation of climate change”, since climate change affects precipitation occurrence and in a much lesser degree precipitation affects mechanisms of climate change formation.

l  In 1. Introduction line 44, in the value “0.16 °C/10a”, the meaning “10a” should be given where is first written in the manuscript.

l  In 1. Introduction, lines 47 to 49, for the Beijing 2 Aug 2023 example, some reference or some relevant data for this event are necessary.

l  In 2. Materials and Methods, 2.2. Data Acquisition, given the large area studied some comment is needed about the data representatives based on the arrangement and spacing of the 23 weather stations, Specifically, a comment is needed about how adequately this weather station number is able for an adequate sampling of rainfalls and especially thunderstorm rainfall amounts.

l  In 3. Results and Discussion, 3.1. Basic Characteristics of Precipitation, lines 255-256, the mechanisms which global warming may affects high elevation rainfalls need more research and the results in the present study should be considered as preliminary. Local scale factors are primary responsible for increasing or decreasing temperature and precipitation that sometimes may balance or overcome the global warming effect. Global warming and climate change mainly affects large scale weather systems responsible for rainfall occurrence and it is under intensive research on how climate change re-distributes rainfalls in space and time around the world.

l  In 3. Results and Discussion, 3.2. Spatiotemporal Variation of PCD, lines 299-302, an interesting result, that can be also characterized as indicative and preliminary given than there is a need for more studies with longer times series precipitation data to establish this possible feature of high-altitude regions.

Comments for author File: Comments.pdf

Author Response

Dear reviewer:

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: [In Abstract line 12, some correction is needed in the meaning of phrase “mechanism of the formation of climate change”, since climate change affects precipitation occurrence and in a much lesser degree precipitation affects mechanisms of climate change formation.]

Response 1: [Thank you for pointing this out. We have corrected this statement, which can be found on Lines 11-12 of the revised manuscript.]

Comments 2: [Introduction line 44, in the value “0.16 °C/10a”, the meaning “10a” should be given where is first written in the manuscript.]

Response 2: [Agree. The meaning of “10a” has already been written and can be found on Line 44 of the revised manuscript.]

Comments 3: [Introduction, lines 47 to 49, for the Beijing 2 Aug 2023 example, some reference or some relevant data for this event are necessary.]

Response 3: [We agree with this comment. Therefore, We have corrected this statement, which can be found on Lines 48-49 of the revised manuscript.]

Comments 4: [Materials and Methods, 2.2. Data Acquisition, given the large area studied some comment is needed about the data representatives based on the arrangement and spacing of the 23 weather stations, Specifically, a comment is needed about how adequately this weather station number is able for an adequate sampling of rainfalls and especially thunderstorm rainfall amounts.]

Response 4: [Thank you for pointing this out. We couldn't agree more. However, this dataset is currently available, with the largest number of meteorological stations, the highest observation accuracy, the longest record sequence and the most widely used. Therefore, in the face of the reality of data scarcity, it is the best option for now.]

Comments 5: [In 3. Results and Discussion, 3.1. Basic Characteristics of Precipitation, lines 255-256, the mechanisms which global warming may affects high elevation rainfalls need more research and the results in the present study should be considered as preliminary. Local scale factors are primary responsible for increasing or decreasing temperature and precipitation that sometimes may balance or overcome the global warming effect. Global warming and climate change mainly affects large scale weather systems responsible for rainfall occurrence and it is under intensive research on how climate change re-distributes rainfalls in space and time around the world.]

Response 5: [Thank you for your constructive comments. We have benefited a lot from them, as can be seen in Lines 266-271 of the revised manuscript.]

Comments 6: [In 3. Results and Discussion, 3.2. Spatiotemporal Variation of PCD, lines 299-302, an interesting result, that can be also characterized as indicative and preliminary given than there is a need for more studies with longer times series precipitation data to establish this possible feature of high-altitude regions.]

Response 6: [Thank you for pointing this out. We couldn't agree more. This is exactly what this work is trying to say.]

Reviewer 4 Report

Comments and Suggestions for Authors

At present, the increasing number and obvious intensification of extreme drought or flood events have posed a serious threat to humanity. Therefore, the evolution of precipitation pattern has attracted much attention, and the refined research is one of the important directions that need to be strengthened in the next few years. This paper describes the diurnal difference and evolution trend of precipitation concentration in Gansu province, northwest China. Compared with similar studies, this paper does have many new highlights and findings, which contribute to deepening the scientific understanding of hydrological response mechanisms in the context of climate change. Arid and semi-arid regions are particularly sensitive to precipitation changes, and subdaily scale studies in these regions are indeed very rare. In this manuscript, the temporal and spatial distribution characteristics and formation mechanism of precipitation concentration on the diurnal scale are comprehensively analyzed, which is of great significance for understanding and coping with regional climate change. Among them, some conclusions are very novel and attractive, and also provide new insights for related research. For example, the daytime performance is more prominent, and the trend of decreasing precipitation concentration is most obvious in the high altitude area, and so on. Overall, this manuscript is well-written and has a solid introduction and methodology, the results seem to be robust, and the flow of the manuscript makes logical sense. Much of the findings may be important for those studying precipitation concentrations in various world regions. In general, I feel that this has strong potential to become a published paper after making minor revisions to the comments below.

Comments

 

1. The statement of p≈0.05 (Line 355) should not be rigorous, it is recommended to give a specific value.

2. For the equation in Table 3, the number of digits is insufficient because of multiple occurrences of 0.000X.

 

3. Line 73，the first appearance of PCI and CI should not be abbreviated, it is recommended to write the full name.

Comments on the Quality of English Language

 Figure 2a, 2c, 3b, 5b (some text overlaps) and 3a, 5a (incomplete site display) need to be modified.

Author Response

Dear reviewer:

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: [The statement of p≈0.05 (Line 355) should not be rigorous, it is recommended to give a specific value.]

Response 1: [Thank you for pointing this out. We have corrected p≈0.05 to p<0.10. It can be found on Line 373 of the revised manuscript.] 

Comments 2: [For the equation in Table 3, the number of digits is insufficient because of multiple occurrences of 0.000X.]

Response 2: [We agree with this comment. Therefore, We have increased the number of digits from three to four or five. It can be found in Table 3 (Lines 377-379) of the revised manuscript.]

Comments 3: [Line 73，the first appearance of PCI and CI should not be abbreviated, it is recommended to write the full name.]

Response 3: [Agree. The full names of PCI and CI have already been written out when they first appear. It can be found on Lines 75-76 of the revised manuscript.]

Comments 4: [Figure 2a, 2c, 3b, 5b (some text overlaps) and 3a, 5a (incomplete site display) need to be modified.]

Response 4: [Thank you for pointing this out. We have corrected these mistakes, which can be found in Figure 2 (Lines 189-199), Figure 3 (Lines 290-296) and Figure 5 (Lines 355-360) corresponding to the revised manuscript.]

Reviewer 5 Report

Comments and Suggestions for Authors

Review report on "Evolution Characteristics of Daytime and Nocturnal Precipitation Heterogeneity in Gansu Province, Northwest China"

General Opinion: 

The manuscript evaluated the spatiotemporal characteristics of precipitation concentration degree (PCD), and precipitation concentration period (PCP) at daytime and nocturnal scales. The study was performed in Gansu province using daily precipitation datasets from 23 national meteorological stations covering 1970-2019).. The following were confirmed:

1.     The annual distribution showed both differences in precipitation amounts at daytime and nocturnal.

2.     The nocturnal precipitation rate correlated moderately with annual precipitation, a wetting trend.

3.     PCD showed a correlation pattern with a moderate to strong relationship with values in daytime > nocturnal, especially in dry areas.

4.     PCP showed changes in the 39^th -41^st pentad, with daytime > nocturnal as a rule of thumb

5.     The PCD and PCP showed that arid regions should be given high priority. This certainly has implications for the decision-making process in the debate on climate change.

Specific Opinion and comments:  The manuscript is an original article, and the analysis contributes to the ongoing Discussion of climate change. An extended benefit is the choice of the study area, which may serve as a test bed for future studies in different tropical regions or as input for various operational decisions. The overall presentation of the abstract is good. The keywords are appropriate and would improve the paper search if published. The manuscript is fairly organized and provides a theoretical perspective to the intended audience. I find the research procedures and techniques are fairly well described and reproducible.

However, several major weaknesses were observed that require the authors to address them scientifically. The authors should address this concern throughout the manuscript. To guide the authors, I have concerns; I have itemized my comments and suggestions in the attached file.

o  

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Moderate language editing required

Author Response

Dear reviewer:

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: [Line 26: "earlier than". What does that mean? Please rephrase or clarify.]

Response 1: [Thank you for pointing this out. We've replaced "earlier" with "smaller". It can be found on Line 27 of the revised manuscript.] 

Comments 2: [The introduction provides much context or useful information for the readers. However, most paragraphs are wordy or in length, distorting the core message presented. Thus, it needs to be improved significantly. The novelty of the work needs to be highlighted as well. There is a need to clearly state the objective(s) of the study towards the end of the introduction.]

Response 2: [Thank you for pointing this out. We've improved it, which can be found on Lines 37-91 of the revised manuscript.]

Comments 3: [Line 44-53: The entire paragraph lacks clarity and is too wordy. Rephrase the entire sentence.]

Response 3: [Thank you for pointing this out. We've improved it, which can be found on Lines 46-55 of the revised manuscript.]

Comments 4: [Line 73-74: define abbreviation in the introduction section.]

Response 4: [Agree. Thank you for pointing this out. The full names of PCI and CI have already been written out when they first appear. It can be found on Lines 75-76 of the revised manuscript.]

Comments 5: [I noticed that some of the concepts used and justification are missing. A few examples include Lines 261-263: I expect this categorization to have also been highlighted in the methods sections (Lines 148-149). Also, provide a reference for PCD categorization.]

Response 5: [Thank you for pointing this out. We've improved it, which can be found on Lines 116-123, 143-163 of the revised manuscript.]

Comments 6: [The results section is complemented with significant figures to help visualize results. However, I observed that the figure's presentations are sloppy. A few examples: Figures 2-3 should be on one page. The discussion section is scanty and needs to be improved to make reading very interesting and useful.]

Response 6: [Thank you for pointing this out. We have corrected these deficiencies, which can be found in Figure 2 (Lines 189-199), Figure 3 (Lines 290-296) , Figure 4 (Lines 290-296), Figure 5 (Lines 355-360) and related paragraphs corresponding to the revised manuscript.]

Comments 7: [The conclusion section captures and summarises the entire work well. The practical implication of the findings as input for various operational decisions should be highlighted.]

Response 7: [Thank you for pointing this out. We've improved it, which can be found on Lines 433-436 of the revised manuscript.]

Comments 8: [The references cited are relevant to the study.]

Response 8: [Thank you for pointing this out. We have reconfirmed them.]

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have answered the specific questions that had been proposed as particular examples in the first review. In my opinion, not all of them have been adequately answered, but they could be corrected.

 

However, my major concern, as stated in the first review, is that the structure of the manuscript, the results and the analyses need to be clearer so that they are really useful to potential readers of the work once published. For that reason, in that first revision, I proposed to reject it and not to make major changes, given the short time periods of this journal, which do not allow for larger scale changes or in-depth revisions. I think it is more appropriate to do a resubmission, which provides authors enough time to think about what is the most important information they want to communicate and the best way to do it. In the present state, there is a lot of information, not all of it equally relevant, that needs to be simplified or analyzed more clearly.  

 

Some notes regarding the responses to the comments on the first version:

 

* Response 4: 

The authors added the sentence “θj represents the corresponding azimuth angle of the jth pentad(j=1,0°≤θj<5°;j=2,5°≤θj<10°;...;j=72,355°≤θj<360°),while the year can be seen as 360 °.”

 

Is each value assigned to an angle interval? How is it done? I think it is assigned to the center point of each interval (2.5, 7.5, …). But it should be specified. 

 

* Response 6:

In some plots axis titles are over the labels.

 

* Response 7:

The authors, as they comment, perform the linear regressions, separately, of total precipitation with precipitation frequency (PF) and with precipitation intensity (PI). From the comparison of both regressions they decide which of the two variables has a greater effect on annual precipitation. So, do they assume that there is no correlation between PI and PF? Partial correlations are not taken into account. 

 

* Response 8: 

Total annual precipitation has been added, which facilitates interpretation. However, it still does not appear to which region (arid, etc.) each station belongs, although it is mentioned in the new text added and in the response 10. 

 

In the table caption, "Results that did not pass the significance test are not listed." was added to indicate why three of the 23 stations had been removed, ... but the stations have also been added in the last rows of the table. Either one or the other.

 

* Response 10:

Indeed, in many cases the differences between several observations or simulations are noticeable, but not statistically significant. It is not that the results cannot be presented, but the related comments should be in that sense. 

 

* Response 15:

The reason for this selection must be specified in the manuscript.

 

Comments on the Quality of English Language

None

Author Response

Dear reviewer:

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

Comments 1: [Response 4: The authors added the sentence“θj represents the corresponding azimuth angle of the jth pentad(j=1, 0°≤θj<5°; j=2, 5°≤θj<10°; ...; j=72, 355°≤θj<360°), while the year can be seen as 360 °.”Is each value assigned to an angle interval? How is it done? I think it is assigned to the center point of each interval (2.5, 7.5, …). But it should be specified.]

Response 1: [Thanks again for pointing this out. We've improved it. It can be found on Lines 153-154 of the revised manuscript.]

Comments 2: [Response 6: In some plots axis titles are over the labels.]

Response 2: [We apologize for our oversight. All the errors caused by different versions of OFFICE have been corrected, which can be found in Figure 2- Figure 5 of the revised manuscript.]

Comments 3: [Response 7: The authors, as they comment, perform the linear regressions, separately, of total precipitation with precipitation frequency (PF) and with precipitation intensity (PI). From the comparison of both regressions they decide which of the two variables has a greater effect on annual precipitation. So, do they assume that there is no correlation between PI and PF? Partial correlations are not taken into account.]

Response 3: [Thanks again for pointing this out. We've improved it by adding partial correlation analysis results, which can be found on Lines 179-180 of the revised manuscript. However, the results of partial correlation coefficient with annual precipitation show that the difference between PF (0.993，the detailed results can be found in the annex) and PI (0.992) is very weak. Therefore, we tend to retain the results of pearson correlation analysis.]

Comments 4: [Response 8: Total annual precipitation has been added, which facilitates interpretation. However, it still does not appear to which region (arid, etc.) each station belongs, although it is mentioned in the new text added and in the response 10. In the table caption, "Results that did not pass the significance test are not listed." was added to indicate why three of the 23 stations had been removed, ... but the stations have also been added in the last rows of the table. Either one or the other.]

Response 4: [Thanks again for pointing this out. We've improved it, which can be found on Lines 233-235 of the revised manuscript. "Results that did not pass the significance test are not listed." This only applies to the results of the last six columns. Because the first three columns do not involve significance tests, they are for ease of interpretation.]

Comments 5: [Response 10: Indeed, in many cases the differences between several observations or simulations are noticeable, but not statistically significant. It is not that the results cannot be presented, but the related comments should be in that sense.]

Response 5: [Thank you for pointing this out. We agree with this comment. Therefore, whether analyzing the causes of PCD reduction or PCP advance, we choose representative sites that pass the significance test as much as possible.]

Comments 6: [Response 15: The reason for this selection must be specified in the manuscript.]

Response 6: [Thank you for pointing this out. We've improved it, which can be found on Lines 330-332 of the revised manuscript.]

Round 3

Reviewer 1 Report

Comments and Suggestions for Authors

None