the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 May 2024
Spatiotemporal variability of flash floods and their human impacts in the Czech Republic during the 2001–2023 period
Abstract. Flash floods, characterized by their sudden onset, extreme discharges, short duration, material damage, and human loss, represent a significant natural hazard. Not well covered by standard hydrological observations, flash floods data can primarily be derived from various types of documentary evidence. This evidence served as the main data source for creating a flash flood database for the Czech Republic from 2001 to 2023. This database enabled detailed analysis of different aspects of flash floods. The annual series of the numbers of flash flood events, flash flood days, and affected municipalities showed significant inter-annual variability but no linear trends. The triggering rainfalls that generate flash floods were analyzed with respect to 1–3-hourly and daily precipitation totals and circulation types from the objective classification. While flash floods can occur anywhere, they were more frequently recorded at the foots of mountain slopes, often coinciding with "critical points" where built-up areas meet concentrated surface runoff pathways. The division of material damage caused by flash floods into eight categories indicated that the highest proportions of damage were to streets and communications, as well as to houses, their cellars, and basements. There were also several recorded fatalities. The understanding of flash floods in the Czech Republic aligns generally well with studies of flash floods in other European regions.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(2025 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2025 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-1467', Lubomir Solin, 31 Jul 2024
Flash floods, because they are unexpected, have severe consequences and there is not enough time to prepare for their occurrence, are becoming quite a serious societal problem. Therefore, the creation of a database on their occurrence and the negative consequences caused is an essential prerequisite for the systematic management of flood risk in river basins.
I have a few comments on the work.
- Note on flash flood definition:
Consider flash floods wihout pluvial flooding (Kaiser et al.2021) is not correct. During flash flood overland flow from a catchment, which occurs when the rainfall intensity is greater than the infiltration capacity of the land surface (pluvial flooding), is the critical component that contributes to a sudden and significant increase in the flow in the river channel and causes the water to overflow out of the channel (fluvial flooding). Flash flood is a mutual combination of fluvial and pluvial flooding.
- Note on data collection.
Data collection on flash floods based on information in newspapers and web portals as the authors note in the discussion is not exhaustive. As a rule, only major, catastrophic events are reported in these sources. Events who do not have a significant socio-economic impact go unnoticed. This type of uncertainty is considered by the authors to be a standard feature in such data collection. That is true, but how to deal with it, the authors do not give an answer. One way to get the most complete information about flash flooding is to analyse the frequency of declarations of Level 3 flood activity, which are usually declared by mayors of municipalities, in relation to meteorological and circulation patterns.
- Note on hydrological and geographical factors influencing the occurrence of flash floods
Compared to the meteorological and climatic aspects, the section on the influence of hydrological and geographical factors on spatial variability is treated in a very general way. Only sites with flash floods are listed and shown, and only a general statement is made that factors such as catchment size, land use, average slope, and relief fragmentation, river network characteristics lithology are considered to be key in terms of their influence on flash floods. However, any analysis of the geographic attributes of the catchments in which flash floods have occurred, or the hydrographic attributes of their watercourses in relation to, for example, the frequency of flash floods, is entirely lacking. A key hydrological characteristic in relation to the occurrence of flash flooding is the base flow index, but this is not mentioned at all.
- Note to victims of flash floods.
For the sake of completeness, in the discussion of flash flood victims, it would be appropriate to also mention the victims that occurred in Slovakia. In July 1998, a storm accompanied by strong winds and hailstorms occurred in the basin of the Mala Svinka. In the affected area, more than 100 millimetres fell in about 120 minutes. Fifty Roma from Jarovnice, mainly children, were victims of the torrential wave.
Citation: https://doi.org/10.5194/egusphere-2024-1467-RC1 - AC1: 'Reply on RC1', Rudolf Brazdil, 22 Aug 2024
-
RC2: 'Comment on egusphere-2024-1467', Gaston Demarée, 04 Aug 2024
Remarks to the Editor
The expertise of the present Reviewer is mainly in the field of historical floods having occurred in past centuries. He is less familiar with documentary data and the published literature on FFs (= Flash Floods) of the present 21st century period. This Reviewer appreciated that the authors of the present submitted manuscript have used the exhaustive data collection made available in the IT-era. The lead author, prof. Dr. Rudolf Brázdil is a world-known expert in historical floods and has published numerous papers in this context.
This Reviewer is a non-paid consultant of the Royal Meteorological Institute of Belgium (RMIB) at Brussels, Belgium. He has published extensively on hydrological catchment modeling, probability distributions of rainfall depths, historical climatology and hydrology.
Review Procedure
All co-authors are members of scientific institutions (Masaryk University, Global Change Research Institute of the Czech Academy of Sciences, Masaryk Water Research Institute, Czech Hydrometeorological Institute) located at Brno, Moravia, Czech Republic what indicates a strong coherence in their research activities and in producing the manuscript under review.
The extensive data base used in this manuscript challenges the scarcity of ready-available information of FFs which are typically events presenting a limited spatiotemporal context. First of all, the printed newspaper Právo, the main national newspaper, and the online Novinky.cz, were scanned. Did local publications mainly dealing with advertising and providing local events were also taken into account? What was the proportion of newspaper data, internet sources, CHMI Reports, professional papers and Other Data Sources? Are there events mentioned in several of these sources at the same time? What are the minor case letters a (Rozhovice) and b (Nový Jičín) in Figure 1? Although the frontier region of Šumava Mountains contains many FFs, none is mentioned in figure 1 – is it a coincidence?
In the database of Flash Floods (section 3.1) the item ‘elevation (in meter above sea level)’ was not selected. However, it might not have any interest as events happened maybe mostly in hilly areas. Maybe this factor might show up in the annual totals of FFs in the administrative regions of the CR (see Figure 8).
This Reviewer would suggest to include the abbreviation FFD in the legend of Figure 2 as was already done with the abbreviation FFE: … and flash flood days (FFDs); …
The probability distributions of the FFEs and FFDs in the summer half-year seem to be Gaussian.
Line 470: Is there a hydrometeorological reason why the flood formation predominantly occurs in the late evening and in the night? Most probable the temporal occurrence of ‘thunderstorms‘are a basic argument (seen section 4.2.1).
Significant FFEs (see section 4.4): no remarks
Statistical evaluation of deadly events during FFEs is difficult as the reference periods are very different. Some reference periods even include the late 19th century which might even question their data base collection techniques. Strictly speaking the rates in a broader geographical context are only comparable on the time frame of the present data base. However, previous studies by Brázdil, the lead-author, using similar data collection techniques and methodology show clearly no significant change in fatality rates over time in the Czech Republic.
Conclusion: This manuscript is clear, well written and has a large exhaustive reference list dealing with FFs. The authors were very successful in exploiting the data bases mentioned in section 2 ‘Data’ in the context of FFE and FFD-related occurrences. Finally, it might be suggested to the authors in a potential future paper to extend the results of the present manuscript dealing with FFs in the Czech Republic in the reference period 2001-2023 to earlier periods.
This Reviewer suggests publication as it stands leaving the authors, if they wish, taking care of the few minor suggestions and remarks mentioned by this Reviewer.
Citation: https://doi.org/10.5194/egusphere-2024-1467-RC2 - AC2: 'Reply on RC2', Rudolf Brazdil, 22 Aug 2024
-
RC3: 'Comment on egusphere-2024-1467', Anonymous Referee #3, 14 Aug 2024
The work on flash flood is an interesting methodological effort to understand the behavior and evolution of these phenomena. Its interest is growing due to the impacts it causes in a short time and in very limited areas. Hence the complexity of its study and the need to make approximations on a detailed scale and proposing specific methodologies.
General comments
- It would be interesting if the abstract contained a preview of specific results of the research carried out.
- Would it be possible to define the phenomenon with instrumental or at least quantitative variables? Not only related to hydrological criteria.
- Defining flash floods based on hydrological behavior does not completely define flash floods. That is correct, but future research should also consider effects outside the river system. That is, pluvial floods. It's a more complex or diverse approach to be promoted in next steps of research.
In the Mediterranean region, the effects of torrential rain are currently directed in large proportion towards this type of phenomenon, linked to drainage problems, poor urban planning, or to the effect of the great minute intensity in which the precipitation occurs. Distribution of these events is more extense than fluvial system, affecting areas in which historial or instrumental records cannot describe similar previous situations.
- The work only covers a period of 23 years. It would have been much better to generate robust results for the purposes of climate analysis, to be able to extend the study period, at least up to 30 years.
Specific comments
Lines 100-105. Figure 1.
The location map is insufficient. Showing only the study region hinders viewing the entire river systems. A reference to the European continent and Central Europe would be very convenient. Considering foreign readers, it's important aspect to be considered.
Lines 110 and following
The use of local press sources has positive aspects. But in itself it leaves information unrecorded because it is not systematic and filters the published information by editorial criteria. It should be complemented with reports from emergency management authorities, administrations, insurance companies and social networks where amateur observers generate records of notable quality.
Do the authors plan to deepen the research using these more objective and systematic sources of information?
Regarding the use of information obtained from social networks, of course, its use would only be as a complement as a secondary source, but it can provide a very good level of detail.
Lines 210 and following
It would be interesting to explain the availability of precipitation intensity records. Current pluvial flooding situations have a direct relationship with precipitation in millimeters per minute. Minute intensity data, if it exists, would be very explanatory.
Unfortunately, this flash flood phenomenon "escapes" from the most common records by daily total.
For example, in some regions we perceive an increase in flash floods due to episodes that do not reach a large total magnitude. 100 or 200mm of total daiy precipitation are not necessary. On the other hand, the effects are very serious due to rain events that reach or exceed 2/4 millimeters per minute for periods of 10 to 20 minutes.
Line 235 and following
Figures 5a and 5b. The data on prevailing winds and weather types in these figures would perhaps be better served with a pie chart style display, or a "compass rose" simulation.
Suggested graphic resource is already applied in figure 7a.
Line 277.
Concept "preliminary financial damage estimates". It would be better "preliminary economic damage estimates" ?
Line 280. Figure 7c.
This important display of data by categories, it would be better not only a graphic display, but also showing detailed data in a table. To make easier comparative evaluations between concepts and evolution in time. This resource is already used in Table 1.
Citation: https://doi.org/10.5194/egusphere-2024-1467-RC3 - AC3: 'Reply on RC3', Rudolf Brazdil, 22 Aug 2024
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-1467', Lubomir Solin, 31 Jul 2024
Flash floods, because they are unexpected, have severe consequences and there is not enough time to prepare for their occurrence, are becoming quite a serious societal problem. Therefore, the creation of a database on their occurrence and the negative consequences caused is an essential prerequisite for the systematic management of flood risk in river basins.
I have a few comments on the work.
- Note on flash flood definition:
Consider flash floods wihout pluvial flooding (Kaiser et al.2021) is not correct. During flash flood overland flow from a catchment, which occurs when the rainfall intensity is greater than the infiltration capacity of the land surface (pluvial flooding), is the critical component that contributes to a sudden and significant increase in the flow in the river channel and causes the water to overflow out of the channel (fluvial flooding). Flash flood is a mutual combination of fluvial and pluvial flooding.
- Note on data collection.
Data collection on flash floods based on information in newspapers and web portals as the authors note in the discussion is not exhaustive. As a rule, only major, catastrophic events are reported in these sources. Events who do not have a significant socio-economic impact go unnoticed. This type of uncertainty is considered by the authors to be a standard feature in such data collection. That is true, but how to deal with it, the authors do not give an answer. One way to get the most complete information about flash flooding is to analyse the frequency of declarations of Level 3 flood activity, which are usually declared by mayors of municipalities, in relation to meteorological and circulation patterns.
- Note on hydrological and geographical factors influencing the occurrence of flash floods
Compared to the meteorological and climatic aspects, the section on the influence of hydrological and geographical factors on spatial variability is treated in a very general way. Only sites with flash floods are listed and shown, and only a general statement is made that factors such as catchment size, land use, average slope, and relief fragmentation, river network characteristics lithology are considered to be key in terms of their influence on flash floods. However, any analysis of the geographic attributes of the catchments in which flash floods have occurred, or the hydrographic attributes of their watercourses in relation to, for example, the frequency of flash floods, is entirely lacking. A key hydrological characteristic in relation to the occurrence of flash flooding is the base flow index, but this is not mentioned at all.
- Note to victims of flash floods.
For the sake of completeness, in the discussion of flash flood victims, it would be appropriate to also mention the victims that occurred in Slovakia. In July 1998, a storm accompanied by strong winds and hailstorms occurred in the basin of the Mala Svinka. In the affected area, more than 100 millimetres fell in about 120 minutes. Fifty Roma from Jarovnice, mainly children, were victims of the torrential wave.
Citation: https://doi.org/10.5194/egusphere-2024-1467-RC1 - AC1: 'Reply on RC1', Rudolf Brazdil, 22 Aug 2024
-
RC2: 'Comment on egusphere-2024-1467', Gaston Demarée, 04 Aug 2024
Remarks to the Editor
The expertise of the present Reviewer is mainly in the field of historical floods having occurred in past centuries. He is less familiar with documentary data and the published literature on FFs (= Flash Floods) of the present 21st century period. This Reviewer appreciated that the authors of the present submitted manuscript have used the exhaustive data collection made available in the IT-era. The lead author, prof. Dr. Rudolf Brázdil is a world-known expert in historical floods and has published numerous papers in this context.
This Reviewer is a non-paid consultant of the Royal Meteorological Institute of Belgium (RMIB) at Brussels, Belgium. He has published extensively on hydrological catchment modeling, probability distributions of rainfall depths, historical climatology and hydrology.
Review Procedure
All co-authors are members of scientific institutions (Masaryk University, Global Change Research Institute of the Czech Academy of Sciences, Masaryk Water Research Institute, Czech Hydrometeorological Institute) located at Brno, Moravia, Czech Republic what indicates a strong coherence in their research activities and in producing the manuscript under review.
The extensive data base used in this manuscript challenges the scarcity of ready-available information of FFs which are typically events presenting a limited spatiotemporal context. First of all, the printed newspaper Právo, the main national newspaper, and the online Novinky.cz, were scanned. Did local publications mainly dealing with advertising and providing local events were also taken into account? What was the proportion of newspaper data, internet sources, CHMI Reports, professional papers and Other Data Sources? Are there events mentioned in several of these sources at the same time? What are the minor case letters a (Rozhovice) and b (Nový Jičín) in Figure 1? Although the frontier region of Šumava Mountains contains many FFs, none is mentioned in figure 1 – is it a coincidence?
In the database of Flash Floods (section 3.1) the item ‘elevation (in meter above sea level)’ was not selected. However, it might not have any interest as events happened maybe mostly in hilly areas. Maybe this factor might show up in the annual totals of FFs in the administrative regions of the CR (see Figure 8).
This Reviewer would suggest to include the abbreviation FFD in the legend of Figure 2 as was already done with the abbreviation FFE: … and flash flood days (FFDs); …
The probability distributions of the FFEs and FFDs in the summer half-year seem to be Gaussian.
Line 470: Is there a hydrometeorological reason why the flood formation predominantly occurs in the late evening and in the night? Most probable the temporal occurrence of ‘thunderstorms‘are a basic argument (seen section 4.2.1).
Significant FFEs (see section 4.4): no remarks
Statistical evaluation of deadly events during FFEs is difficult as the reference periods are very different. Some reference periods even include the late 19th century which might even question their data base collection techniques. Strictly speaking the rates in a broader geographical context are only comparable on the time frame of the present data base. However, previous studies by Brázdil, the lead-author, using similar data collection techniques and methodology show clearly no significant change in fatality rates over time in the Czech Republic.
Conclusion: This manuscript is clear, well written and has a large exhaustive reference list dealing with FFs. The authors were very successful in exploiting the data bases mentioned in section 2 ‘Data’ in the context of FFE and FFD-related occurrences. Finally, it might be suggested to the authors in a potential future paper to extend the results of the present manuscript dealing with FFs in the Czech Republic in the reference period 2001-2023 to earlier periods.
This Reviewer suggests publication as it stands leaving the authors, if they wish, taking care of the few minor suggestions and remarks mentioned by this Reviewer.
Citation: https://doi.org/10.5194/egusphere-2024-1467-RC2 - AC2: 'Reply on RC2', Rudolf Brazdil, 22 Aug 2024
-
RC3: 'Comment on egusphere-2024-1467', Anonymous Referee #3, 14 Aug 2024
The work on flash flood is an interesting methodological effort to understand the behavior and evolution of these phenomena. Its interest is growing due to the impacts it causes in a short time and in very limited areas. Hence the complexity of its study and the need to make approximations on a detailed scale and proposing specific methodologies.
General comments
- It would be interesting if the abstract contained a preview of specific results of the research carried out.
- Would it be possible to define the phenomenon with instrumental or at least quantitative variables? Not only related to hydrological criteria.
- Defining flash floods based on hydrological behavior does not completely define flash floods. That is correct, but future research should also consider effects outside the river system. That is, pluvial floods. It's a more complex or diverse approach to be promoted in next steps of research.
In the Mediterranean region, the effects of torrential rain are currently directed in large proportion towards this type of phenomenon, linked to drainage problems, poor urban planning, or to the effect of the great minute intensity in which the precipitation occurs. Distribution of these events is more extense than fluvial system, affecting areas in which historial or instrumental records cannot describe similar previous situations.
- The work only covers a period of 23 years. It would have been much better to generate robust results for the purposes of climate analysis, to be able to extend the study period, at least up to 30 years.
Specific comments
Lines 100-105. Figure 1.
The location map is insufficient. Showing only the study region hinders viewing the entire river systems. A reference to the European continent and Central Europe would be very convenient. Considering foreign readers, it's important aspect to be considered.
Lines 110 and following
The use of local press sources has positive aspects. But in itself it leaves information unrecorded because it is not systematic and filters the published information by editorial criteria. It should be complemented with reports from emergency management authorities, administrations, insurance companies and social networks where amateur observers generate records of notable quality.
Do the authors plan to deepen the research using these more objective and systematic sources of information?
Regarding the use of information obtained from social networks, of course, its use would only be as a complement as a secondary source, but it can provide a very good level of detail.
Lines 210 and following
It would be interesting to explain the availability of precipitation intensity records. Current pluvial flooding situations have a direct relationship with precipitation in millimeters per minute. Minute intensity data, if it exists, would be very explanatory.
Unfortunately, this flash flood phenomenon "escapes" from the most common records by daily total.
For example, in some regions we perceive an increase in flash floods due to episodes that do not reach a large total magnitude. 100 or 200mm of total daiy precipitation are not necessary. On the other hand, the effects are very serious due to rain events that reach or exceed 2/4 millimeters per minute for periods of 10 to 20 minutes.
Line 235 and following
Figures 5a and 5b. The data on prevailing winds and weather types in these figures would perhaps be better served with a pie chart style display, or a "compass rose" simulation.
Suggested graphic resource is already applied in figure 7a.
Line 277.
Concept "preliminary financial damage estimates". It would be better "preliminary economic damage estimates" ?
Line 280. Figure 7c.
This important display of data by categories, it would be better not only a graphic display, but also showing detailed data in a table. To make easier comparative evaluations between concepts and evolution in time. This resource is already used in Table 1.
Citation: https://doi.org/10.5194/egusphere-2024-1467-RC3 - AC3: 'Reply on RC3', Rudolf Brazdil, 22 Aug 2024
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 408 | 143 | 139 | 690 | 18 | 13 |
- HTML: 408
- PDF: 143
- XML: 139
- Total: 690
- BibTeX: 18
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 279 | 40 |
| Czech Republic | 2 | 75 | 10 |
| China | 3 | 29 | 4 |
| Germany | 4 | 26 | 3 |
| Netherlands | 5 | 22 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 279
Rudolf Brázdil
Dominika Faturová
Monika Šulc Michalková
Jan Řehoř
Martin Caletka
Pavel Zahradníček
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(2025 KB) - Metadata XML