Rainfall Threshold for Shallow Landslides Initiation and Analysis of Long-Term Rainfall Trends in a Mediterranean Area

Round 1

Reviewer 1 Report

1. The application of ID curve is mainly restricted by local hydrological, geological, morphological and land use conditions. The contents of Table 9 and Figure 9 completely show this local characteristics of ID curve. From the point of statistics, the collected data must be in the same population to avoid Type two error. The data of rainfall intensity and duration with the similar conditions of geology, morphology and land use are suggested to put together for exploring new ID curve. Especially, the study extent with special geological condition of many faults and presumed faults is different from other area.
2. In case of discussing the effect of climate change, the ID curve of specified location with different year term is suggested to increase the contribution of this study.
3. The word of “ant” in Line 178 is suggested to be revised.
4. The word of “50 10%” in Lines 281 and 282 is suggested to be revised.

Author Response

Reviewer #1

We wish to thank the anonymous reviewer for the improvement of the manuscript after comments and request.

The application of ID curve is mainly restricted by local hydrological, geological, morphological and land use conditions. The contents of Table 9 and Figure 9 completely show this local characteristic of ID curve. From the point of statistics, the collected data must be

in the same population to avoid Type two error. The data of rainfall intensity and duration with the similar conditions of geology, morphology and land use are suggested to put together for exploring new ID curve. Especially, the study extent with special geological condition of many faults and presumed faults is different from other area.

We are thankful to the reviewer for his suggestions. We think that a more detailed analysis in terms of geological, morphological or land use conditions is certainly a very interesting question. But it would require to greatly extend this paper, which is focused not only on the development of a rainfall threshold for the occurrence of shallow landslides and debris-flows, but also on the analysis of the long-term trend in the rainfall patterns and their possible effects on landslides occurrence in the study area

2. In case of discussing the effect of climate change, the ID curve of specified location with different year term is suggested to increase the contribution of this study.

As suggested, we implemented rainfall thresholds defined for different periods and compared them with rainfall trends to analyze the effects of changes in long-term rainfall trends on rainfall conditions triggering landslides in the study area.

3. The word of “ant” in Line 178 is suggested to be revised.

The word “ant” has been revised

4. The word of “50 10%” in Lines 281 and 282 is suggested to be revised.

The sentence “ 5 o 10%” has been revised in “5%”

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript established a rainfall threshold of landslides and analyzed long-term rainfall trends for the Portofino promontory. The authors have collected a large amount of data for analysis, and these researches are of great significance for understanding the changes in the precipitation, the triggering factor of landslides, and establishing landslide warnings in the future. However, come major issues are remained in current manuscript. Major revisions are needed before it can be considered for acceptance.

1. About the rainfall threshold of landslides established by this manuscript, it is good to compare it with the threshold of other regions or the world from previous studies. However, it is inappropriate for the authors to attribute the lower threshold obtained in this study to the complex lithological, structural, morphological and land use settings, and peculiar climate conditions. The areas the authors compare are all prone to landslides, where have conditions that are prone to landslides. Before reaching such a conclusion, the authors should compare the conditions of the study area with those areas. Only when statistically significant differences in this area can be seen, the explanation can be made.

2. About the long-term rainfall trends, I agree that the authors use some statistical methods to observe the rainfall trend in this area from the past to the present. However, the authors also want to use those trends to explain possible future changes. This is unreasonable. The past trend does not mean that the future will be like that. What's more, in a large number of climate change researches, there have been many possible future scenarios using joint physical models of the atmosphere and ocean. Rainfall and other data under climate change scenarios are also available through dynamic downscaling to obtain regional-scale data. Authors should consider using such data for analysis, or at least citing some previous studies using these data to assist in explanations.

3. The relationship between the rainfall threshold of landslides established by the authors and the analyzed long-term rainfall trend is very unclear. The current manuscript looks like it has two topics. The authors should add an extra section at the end of the discussion chapter to link the two topics together. How will changes in long-term rainfall trends affect the rainfall threshold of landslides? Or to establish rainfall thresholds for different periods from the past to the present to compare with rainfall trends. What impact will future changes in rainfall have on the frequency and scale of future landslides? These need to be discussed.

Please find other comments from the attached file.

Comments for author File: Comments.pdf

Author Response

Reviewer #2

We wish to thank the anonymous reviewer for the improvement of the manuscript after comments and request.

1. About the rainfall threshold of landslides established by this manuscript, it is good to compare it with the threshold of other regions or the world from previous studies. However, it is inappropriate for the authors to attribute the lower threshold obtained in this study to the complex lithological, structural, morphological and land use settings, and peculiar climate conditions. The areas the authors compare are all prone to landslides, where have conditions that are prone to landslides. Before reaching such a conclusion, the authors should compare the conditions of the study area with those areas. Only when statistically significant differences in this area can be seen, the explanation can be made.

We think that a comparison between the geological, topographical, land use and climatic settings of the study area and the conditions of the territories for which bibliographic curves have been established, is certainly a very interesting question to analyze. But it would require to greatly extend this paper, which is focused not only on the development of a rainfall threshold for the occurrence of shallow landslides and debris-flows, but also on the analysis of the long-term trend in the rainfall patterns and their possible effects on landslides occurrence in the study area

2. About the long-term rainfall trends, I agree that the authors use some statistical methods to observe the rainfall trend in this area from the past to the present. However, the authors also want to use those trends to explain possible future changes. This is unreasonable. The

past trend does not mean that the future will be like that. What's more, in a large number of climate change researches, there have been many possible future scenarios using joint physical models of the atmosphere and ocean. Rainfall and other data under climate change

scenarios are also available through dynamic downscaling to obtain regional-scale data. Authors should consider using such data for analysis, or at least citing some

previous studies using these data to assist in explanations.

As suggested, we have eliminated direct considerations about future scenario, descending from the trends. We have considered the coherence of analyzed trends with previous studies citing papers concerning extreme precipitation events scenario in the Mediterranean.

3. The relationship between the rainfall threshold of landslides established by the authors and the analyzed long-term rainfall trend is very unclear. The current manuscript looks like it has two topics. The authors should add an extra section at the end of the discussion chapter to link the two topics together. How will changes in long-term rainfall trends affect the rainfall threshold of landslides? Or to establish rainfall thresholds for different periods from the past to the present to compare with rainfall trends. What impact will future changes in rainfall have on the frequency and scale of future landslides? These need to be

discussed

As suggested, we defined rainfall threshold for different periods and compared them with the observe rainfall trends. In particular, we identified two periods (1910-1999 and 2000-2019) on the basis of the temporal distribution of the events, taking into account also the different type of pluviometric measurements available. Then, we discussed the effects of the observed trends on the rainfall conditions triggering landslides in the past and in the future, to further support findings of our trend analysis.

We split Discussion section into 3 sub-sections with the aim of evaluating thresholds, trends and the synthesis in terms of landslide triggering and future scenario. We added even an assessment of the rainfall rate coefficient of variation for 30 years periods, assessing its variation in time.

4. Please find other comments from the attached file.

All suggested revisions and changes have been implemented, and all required explanations have been provided. You can find replies to all comments from the pdf file.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors describe the development of a threshold for the occurrence of landslides for the Portofino promontory in Italy. They also discuss long-term trends in the rainfall patterns. This is an interesting and timely topic and the authors do a good job describing their methods. The paper is generally well written, but there are a few instances where the sentence structure can be improved. I have indicated some of those places in the manuscript in the accompanying .pdf file.

The methods used by the authors are generally well developed, and therefore there is less of a methodological innovation present in the paper. More interesting is the trend in rainfall patterns, and in this instance, I feel the authors need to do a little more to demonstrate the changes. Therefore, I would suggest, that they strengthen this section of their paper, showing more clearly the patterns that have undergone some significant alterations in the last decades.

In addition, when developing their threshold pattern for the landslide occurrence, I believe it would be quite fascinating to develop at least two thresholds at various times, which would also support their conclusion of changes in the precipitation patterns. It seems to me, looking at figure 5, a good date to separate would be pre and after 1990, but perhaps the authors can explore other dates, e.g. based on the rainfall data at fig 11 1920, 1940, or 1970 may be candidates.

Overall I would recommend this research to be published, but the additional analysis which shouldn’t take the authors too much time - I hope – should strengthen this manuscript and generate additional readership.

Some more minor detailed comments are in the attached .pdf file.

Comments for author File: Comments.pdf

Author Response

Reviewer #3

We wish to thank the anonymous reviewer for the improvement of the manuscript after comments and request.

The authors describe the development of a threshold for the occurrence of landslides for the Portofino promontory in Italy. They also discuss long-term trends in the rainfall patterns. This is an interesting and timely topic and the authors do a good job describing their methods. The paper is generally well written, but there are a few instances where the sentence structure can be improved. I have indicated some of those places in the manuscript in the accompanying .pdf file.

The methods used by the authors are generally well developed, and therefore there is less of a methodological innovation present in the paper. More interesting is the trend in rainfall patterns, and in this instance, I feel the authors need to do a little more to demonstrate the

changes. Therefore, I would suggest, that they strengthen this section of their paper, showing more clearly the patterns that have undergone some significant alterations

in the last decades.

As suggested, in order to strengthen the Discussion, we split the section into 3 sub-sections with the aim of evaluating thresholds, trends and the synthesis in terms of landslide triggering and future scenario. We added an assessment of the rainfall rate coefficient of variation for 30 years periods, assessing its variation in time.

In addition, when developing their threshold pattern for the landslide occurrence, I believe it would be quite fascinating to develop at least two thresholds at various times, which

would also support their conclusion of changes in the precipitation patterns. It seems to me, looking at figure 5, a good date to separate would be pre and after 1990, but perhaps the authors can explore other dates, e.g. based on the rainfall data at fig 11 1920, 1940, or 1970 may be candidates.

As suggested, we defined rainfall threshold for different periods and compared them with the observe rainfall trends. In particular, we identified two periods (1910-1999 and 2000-2019) on the basis of the temporal distribution of the events, taking into account also the different type of pluviometric measurements available. Then, we discussed the effects of the observed trends on the rainfall conditions triggering landslides in the past and in the future, to further support findings of our trend analysis.

Overall I would recommend this research to be published, but the additional analysis which shouldn’t take the authors too much time - I hope – should strengthen this manuscript and generate additional readership.

Some more minor detailed comments are in the attached .pdf file.

All suggested revisions and changes have been implemented, and all required explanations have been provided. You can find replies to all comments from the pdf file.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

This paper has revised. I have no more comment.

Reviewer 3 Report

Congratulations to the authors for a fine paper. I enjoyed reading it and I think it makes an important contribution to assessing climate change consequences.