Assessing the prevalence, timing, and rapidity of transitions between hydrological extremes and their relation to meteorological extremes in the conterminous United States
Caelan E. Simeoneand John C. Hammond
Publisher's note: the preprint and supplement were substituted by new versions, since there were a few instances where "Fig. 2" was replaced by "Error! Reference source not found". Also, the following sentence was added to the preprint's title page: This information product has been peer reviewed and approved for posting as a preprint by the U.S. Geological Survey.
Abstract. Rapid shifts between droughts and floods, termed hydrological whiplash, challenge water management, yet their timing and drivers remain poorly understood at continental scales. While drought-to-flood (DtF) transitions have received growing attention, flood-to-drought (FtD) transitions — though rarer — pose distinct operational challenges that are less well characterized. These wet-to-dry shifts can disrupt post-flood recovery, strain warm-season water demands, and create compounding risks for infrastructure and water quality. We analyzed daily streamflow records from 3,219 USGS streamgages (1981-2024) to characterize both DtF and FtD transitions across CONUS, with particular emphasis on understanding why these transitions are not symmetric inverses of each other. We test a wide variety of hydrological extreme transition definitions to examine the sensitivity of the number of transitions identified and their rapidity. Additionally, we identify a subset of transitions that may be impactful based on the maximum change in percentile magnitude during a transition. DtF transitions are faster than FtD transitions, and short-term (<= 30-days) transitions in both directions are concentrated in the Northeast, Northwest, and Rocky Mountains regions. Short-term DtF transitions are additionally concentrated in southern California and along the line from North Dakota down to Texas where precipitation approximately equals potential evapotranspiration. We find direction-specific controls: snow-dominated, urban, regulated, and minimally disturbed basins show the most frequent impactful DtF transitions, while regulated basins are most prone to impactful FtD transitions. Rapid and impactful transitions make up a substantially larger fraction of DtF transitions than FtD transitions across nearly all basin types. A key finding is that hydrological and meteorological whiplash rarely coincide: only 19-24% of hydrological extreme transitions co-occur with hydroclimate whiplash, revealing that basin storage, regulation, and routing processes create a fundamental decoupling between climate forcing and streamflow response. Our findings highlight the need to better understand quick hydrological transitions under increasing hydroclimatic volatility, particularly the understudied FtD direction, and the mechanisms by which anthropogenic modifications reshape the hydrological whiplash risk.
Received: 12 May 2026 – Discussion started: 02 Jun 2026
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Publisher's note: the preprint and supplement were substituted by new versions, since there were a few instances where "Fig. 2" was replaced by "Error! Reference source not found". Also, the following sentence was added to the preprint's title page: This information product has been peer reviewed and approved for posting as a preprint by the U.S. Geological Survey.
Status: final response (author comments only)
Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor
| : Report abuse
This manuscript addresses an important topic and provides a useful national-scale analysis of drought-to-flood and flood-to-drought streamflow transitions. I particularly appreciate the comprehensive sensitivity analysis and the discussion of implications for water resources management. However, several methodological details need substantial clarification. I also encourage the authors to sharpen the literature review and more explicitly state the scientific gap and contribution of the study.
Major comments
1. The introduction does a good job motivating the management challenges associated with drought-to-flood and flood-to-drought transitions. However, the literature review on related work is currently limited. For example, the manuscript refers to “impactful transitions” (line 175) and the overlap between meteorological and hydrological whiplash (lines 178-179). First, meteorological whiplash and hydrological whiplash need to be defined more clearly. Second, they need to be situated within existing literature. It would help to explain where previous studies stand in identifying impactful transitions, quantifying meteorological versus hydrological whiplash, and assessing their overlap. In addition, the novelty currently reads mainly as the “exploration of both DtF and FtD transitions” (line 171), but the manuscript should more clearly state the specific knowledge gap and how this study fills it. For example, is the key contribution the national-scale streamflow-based characterization, the comparison of DtF and FtD asymmetry, the sensitivity to event definitions, or the linkage to meteorological transitions?
2. Section 2.2 is central to the study, but many of the definitions and implementation choices are difficult to follow. Several terms are introduced before they are defined. The authors should reorganize this section so that event-definition components are introduced first, followed by using the definitions and then the sensitivity settings. Major points that need clarification include:
a) The manuscript uses minimum drought durations of 5, 15, and 30 days. It is worth reconsidering whether 5-day or even 15-day low-flow periods should be called “droughts,” or whether “dry periods” would be more accurate. Also, the duration may represent different lengths under fixed and variable-threshold methods. In the variable-threshold approach, it seems each daily value represents a 7-day averaged condition, so it has to be clarified whether the effective duration equals the nominal duration under the two approaches.
b) The phrase “the interevent period is […] less than 10% of the drought volume” (line 241) is confusing because a period is being compared with a volume. The authors should define drought volume and explain how this criterion is calculated.
c) Line 242, does “pool drought events” mean if drought days have inter-event period <=5 days, those days are merged as one event? It is not until here that I understand why you need an inter-event period, which was introduced in line 223. Similarly, is the “15-day” for flood separation (lines 246-247) a similar concept of “inter-event period” used in drought? If so, why not introduce it earlier in the first paragraph in section 2.2?
d) Lines 240-247, aren’t these the same as in lines 232-236? Again, it would be clearer to introduce all pooling, separation, and duration criteria before using the definitions.
3. The manuscript uses both fixed and variable percentile approaches, but the implementation is difficult to follow in several places.
a) The zero-flow adjustment is described very briefly (lines 212-215). While reference is cited, the authors should explain how days with different numbers of preceding zero-flow days are combined with positive-flow days when assigning percentiles, and whether this procedure is applied before or after 7-day averaging in the variable approach. It is also unclear whether zero-flow duration is calculated using the full time series (e.g., fixed approach) or only within the comparison window for a given day-of-year (e.g., variable approach).
b) The authors should acknowledge that changing the reference period can change the percentile values themselves (line 210). This matters because percentiles are central to identifying droughts and transitions.
c) The main text appears to identify droughts using variable percentiles but then evaluate “impactful” transitions using a fixed full-record percentile space (line 266). This hybrid framework is difficult to follow.
d) Line 329 states that the maximum increase and decrease in streamflow percentiles are calculated during each season and the preceding 90 days. Does this correspond to a 120-day window in total? However, the following sentence says these maximum changes are compared against the “seasons” where there were meteorological or hydrological transitions, which seems to imply a 30-day window. The wording is confusing and should be clarified.
e) The authors should also clarify how the maximum change in streamflow percentile is calculated. How do they ensure that the identified maximum change corresponds to the actual transitional event, rather than another large streamflow fluctuation occurring within the same window? An example would be very helpful to illustrate the procedure.
f) The authors should also specify which event values are compared to calculate the 50-percentile shift, such as drought minimum to flood maximum, event mean to event peak, or event end to event start, etc.
4. The manuscript compares hydrological DtF/FtD transitions with meteorological dry-to-wet and wet-to-dry transitions, but the temporal matching criteria need more explanation.
a) The manuscript should clarify the timescale mismatch between hydrological and meteorological transitions. Hydrological transitions are currently identified using relatively short event-scale criteria (e.g., 5-, 15-, and 30-day for drought, and no stated duration for flood), while meteorological transitions are based on seasonal timestep (line 317; is it 3-month?). The co-occurrence rate likely depends on the selected timescale, and the timescale mismatch should be acknowledged.
b) The authors should clarify how event order is considered. For example, must a hydrological DtF transition occur after a meteorological dry-to-wet transition? If order is not required, then the manuscript should avoid causal language such as meteorological transitions “produce” hydrological transitions (line 575).
c) Lines 325–326 state that the co-occurrence between hydrological DtF transitions and meteorological dry-to-wet transitions is evaluated during the same season or the immediately preceding or following season. Does this correspond to a total window of approximately 270 days? However, the caption of Figure 7 states that the co-occurrence is evaluated within one season. It was not until I reached Table S1 that I realized both metrics (same season only, and same or adjacent season) were analyzed. I suggest clarifying this distinction in the main text.
d) Lines 320–321: The authors may also want to clarify the rationale for the different timing choices used throughout the analysis. For the co-occurrence analysis, meteorological and hydrological transitions are assigned to the season in which the transition ends, whereas the rose plots use the month in which the first event ends. Is there a reason for using different temporal reference points in these analyses? A brief explanation would help readers understand the methodology.
5. Several statements should be interpreted more carefully. For example, when the manuscript reports the fraction of transitions occurring within 90 days or 30 days (Lines 431-433), this is more of a measure of the prevalence, not transition speed itself. Claims that DtF transitions are “faster” (line 431) should be supported by direct comparison of transition-time, such as Fig3 c, h.
Similarly, where the manuscript compares percentages across regions or groups, the term “fraction” may be more accurate than “frequency” (line 517), unless actual event counts are being compared. Small differences, such as 21.9% versus 20.8% (line 519), should not be overinterpreted as indicating that transitions are meaningfully more common in one region than another.
6. Several watershed characteristics used in the analysis need clearer definitions and data sources. The authors should specify the PET method and data source (line 292), define PET before first use (line 305), provide the data source for snow water equivalent (line 293), and clarify whether the “more than 25%” snow criterion is based on a long-term average. The calculation of “input seasonality” (line 309) also needs to be described.
There is also inconsistency in the use of “concentration.” On line 308, it appears to refer to flow concentration, while lines 533 and 536 describe rainfall and snowmelt concentration. Please clarify which variable you use to calculate the concentration.
Minor comments
Line 200: The 95% completeness threshold still allows approximately 18 missing days per year. Because the whiplash definition is daily, please check whether missing days are clustered.
Line 201: Clarify whether rolling decades are updated every year or every 10 years.
Lines 225–227: Clarify whether flood percentiles are calculated using daily percentiles or 7-day percentiles like drought identification.
Line 252: The manuscript uses both “transition duration” and “transition time.” Please use one term consistently.
Line 319: “not” should be “note.”
Line 352: Please specify what the transition time here refers to. It is FtD, not DtF. Right?
Line 364: “Fig. A13” should be “Fig. S13.”
Line 374: The sentence appears to be missing a word.
Line 451: Figure 6 and Figure 7 show DtF transitions under 30 days and FtD transitions under 90 days. What about DtF under 90 days and FtD under 30 days?
Line 502: Figure 4 appears to indicate prevalence rather than timing unless seasonality is shown using the combined methods.
Line 517, these are not only “sub-seasonal DtF” but also impactful DtF that a 50% percentile shift is required. Same thing throughout this section.
Line 522: “Fig. 5a” should likely be “Fig. 6a.”
Lines 549-552, this sentence is very long and difficult to read. Consider breaking it into two shorter sentences.
Line 569, is the 24.0% a national median?
Lines 667–668: Please clarify how Figure 6 and Figure S10 support the claim that FtD transitions within 30 days are mostly impactful. Fig. 6a,b,c shows the fraction of impactful events as a function of watershed climates, land use, and hydrologic regions, and there is no classification of transition time.
Line 701: Remove redundant “may.”
Lines 738–740: Please briefly describe the Huge Creek watershed for readers unfamiliar with it, such as whether it is urbanized or experienced major modifications in early 2009.
Lines 824–825: Clarify how observational case studies “align with projections,” since the case studies are from historical records while projections refer to the future?
Lines 827–830: Are you trying to compare between DtF (or dry-to-wet) versus WtD (or wet-to-dry)? The current sentence reads like comparing fixed-percentile and variable percentile, or correlation between FtD and WtD transitions.
Line 895: Be more specific about which event characteristics are most sensitive to event definitions. Earlier in the main text it says the fraction of short-term transitions (<30 days) remains consistently higher for DtF transitions across different definitions.
Figure 1b: Add a description of the blue and red shading, similar to the Figure 2 caption.
Figure 2: Clarify the use of “rapid and sub-seasonal transitions.” Some events have transition times longer than 14 days but shorter than 30 days (e.g., 20-54 days on line 384), which are short-term event, right?
Figure 2: Some panels appear to show drought and flood overlapping at the same time. Please explain how the same flow value can be classified as both drought and flood.
Figure 2 / Line 413–415: Clarify what is meant by “only use drought percentile thresholds.” Does it mean that you didn’t consider the inter-event, minimum drought length, etc.? Also explain why one site uses flood stage rather than flow in the main text or methods, not in Figure S13. Currently the method only states that streamflow percentiles are calculate based on daily flow (line 206).
Figure 4b: If boxes are sorted by median, please state this.
Figure 6: Some boxes appear truncated before the 1.5×IQR range, while the caption says only outliers outside 1.5×IQR are omitted. Please check the plotting. The same issue appears in Figures S8–S9.
Figure 6: Consider using abbreviations for hydrologic regions or introducing the region numbers and names in the main text so readers do not need to refer to the supplement.
Figure 7 caption: Browner or bluer colors indicate sign, not necessarily more extreme or smaller changes. Also, if negative values indicate drier-than-average conditions, the color should correspond to brown rather than blue.
Figure S1: Make the figure-panel order consistent across percentile thresholds. The order of the figure panel based on the 10% (start from 5d iet and 15d min drought) is currently different from the order based on the 15% and 20% (start from 15d iet and 15 min drought).
Figure S2 caption: The first semicolon should be a colon, and “food” should be “flood.”
Figure S5: Add a space in “Fig.3definitions.” “Fig. A4” should be “Fig. S4.” Same issue for Figure S8 caption.
Figure S8/S9: The sentence referring to region names in Fig. A4 is redundant with the final caption sentence.
Figure S9: The red horizontal dotted lines are difficult to read. Consider increasing spacing between boxes or otherwise improving visibility.
Figure S10: The phrase “transitions relative and 90 preceding days relative” appears broken.
Figure S11: In the third sentence, “x-axes” should be “y-axes.”
Figure S13, “data are only available from 2000 to 2025”. Does this site still satisfy the criteria described in line 201 that at least 8 of 10 complete years for decades from 1981–2020? Also, Fig S13 says because of the missing data, the reference period is 2000-2020. This is not mentioned in line 211 where it indicates all the reference periods are 1981-2020.
Lines 1098–1100: The paper seems published in 2017 instead of 2021. Also, the doi is not found. It seems the paper’s doi is https://doi.org/10.1038/nclimate3225. Please check.
Text S1: The sentence “Our flood identification using a peaks-over-threshold approach calibrated to yield approximately one event per year represents a middle ground between overly restrictive and overly permissive flood definitions” is difficult to read. Consider breaking it into two shorter sentences.
This manuscript provides a valuable large-sample analysis of hydrological transitions between drought and flood conditions across the conterminous United States. The use of 3,219 USGS streamgages and multiple event definitions is a major strength, and the paper makes a useful contribution by comparing both drought-to-flood and flood-to-drought transitions. The finding that DtF transitions are generally faster and more common than FtD transitions appears robust and important. The analysis of basin characteristics and the comparison with meteorological whiplash are also valuable.
I recommend that the authors clarify several points and consider softening some interpretations.
First, I think it would be good to define what an impactful transition is. Since it is based on large changes in streamflow percentiles, the word “impactful” may sound like the authors are referring to real societal, ecological, or infrastructure impacts.
Second, the comparison between hydrological and meteorological whiplash is interesting, but the two datasets use different time scales. Hydrological transitions are based on daily streamflow, while meteorological transitions are based on seasonal events. Because of this, some mismatch is expected. The statement that hydrological and meteorological whiplash “rarely coincide” may be a little too strong. It would be safer to say that they show limited overlap under the definitions and timing criteria used in this study.
Third, FtD transitions are much less common than DtF transitions. The authors should clarify whether the conclusions about rapid FtD transitions are based on enough events, or whether they may be affected by a small number of sites or cases. Adding event counts, uncertainty ranges, or a minimum-event-count rule would make this part stronger.
Fourth, the paper uses ≤90 days for sub-seasonal transitions, ≤30 days for short-term transitions, and ≤14 days for rapid transitions. These thresholds are useful, but the authors should explain why the same thresholds are suitable for all regions and basin types. For example, a 30-day transition may mean something different in a small urban basin compared with a snowmelt basin or a large regulated river.
Finally, Lines 220–226 should better explain why these specific drought and flood definitions were chosen. The authors later say that the selected definitions help balance FtD and DtF transitions, but that reason alone is not fully convincing. It would help to also explain why these definitions make hydrological sense.
Some detailed suggestions:
Figures
All figure sizes and font sizes could be bigger.
In the caption, some subplot labels are put in front of the subplot description, while some are not. Please make the formatting consistent across all captions.
Ensure all the legends in all figures don’t overlap with other parts.
Figure captions should be more clearly separated from the main text. For example, lines 491-499 could easily be mistaken for main text rather than a figure caption.
For Figure 3, the legends are not well aligned with their corresponding subplots and overlap with labels or figure elements. This is especially noticeable for the legends associated with panels c/h, d/i, and e/j.
For Figure 6, the figure is difficult to read. The numbers and decimals overlaid on the boxplots are hard to see and may need to be repositioned.
Abstract
Line 16: Please define CONUS when it first appears.
Lines 15, 18, and 20: The verbs “analyzed,” “test,” and “identify” use mixed tense. Please make the tense consistent throughout the abstract and manuscript.
Line 24: The phrase “the line” should be defined more clearly. For example, specify that this refers to the region where precipitation is approximately equal to potential evapotranspiration, or P/PET ≈ 1.
Introduction
The Introduction feels somewhat long. It may be beneficial to have some subheaders to guide the reader.
Terms such as “wet-to-dry,” “rapid wet-to-dry shifts,” and “rapid FtD and DtF transitions” should be defined more clearly and clarify the difference.
Lines 69–70 and Lines 88–90 seem to discuss similar ideas regarding hydrological intensification and increased climate volatility. These sentences could be shortened or combined to reduce repetition.
Methods
Line 198 should specify where these streamgages are located when they first appear in the main text.
Line 234, the term “conterminous” is technically correct, but for a broader audience, “contiguous United States” may be more immediately understandable.
Line 242, the phrase “a fraction of 0.1” seems unnecessary unless this exact fraction is used directly in the analysis. Consider simplifying the sentence.
Line 286, the phrase “HCDN; Lins, 2012” may make readers think HCDN is part of the citation. Consider revising for clarity, for example: “the Hydro-Climatic Data Network (HCDN), as described by Lins (2012).”
Publisher's note: the preprint and supplement were substituted by new versions, since there were a few instances where "Fig. 2" was replaced by "Error! Reference source not found". Also, the following sentence was added to the preprint's title page: This information product has been peer reviewed and approved for posting as a preprint by the U.S. Geological Survey.
Publisher's note: the preprint and supplement were substituted by new versions, since there were a few instances where "Fig. 2" was replaced by "Error! Reference source not found". Also, the following sentence was added to the preprint's title page: This information product has been peer reviewed and approved for posting as a preprint by the U.S. Geological Survey.
Viewed
Total article views: 344 (including HTML, PDF, and XML)
HTML
PDF
XML
Total
Supplement
BibTeX
EndNote
230
106
8
344
32
8
8
HTML: 230
PDF: 106
XML: 8
Total: 344
Supplement: 32
BibTeX: 8
EndNote: 8
Views and downloads (calculated since 02 Jun 2026)
Cumulative views and downloads
(calculated since 02 Jun 2026)
Viewed (geographical distribution)
Total article views: 307 (including HTML, PDF, and XML)
Thereof 307 with geography defined
and 0 with unknown origin.
This study examines how quickly rivers across the U.S. shift between very dry and very wet conditions, an emerging challenge for water managers because rapid changes compress decision windows and create operational stress. Drought-to-flood transitions occur more quickly and more often than flood-to-drought transitions, with short events concentrated in certain regions. Flood-to-drought transitions are less common but can involve large and rapid declines in flow.
This study examines how quickly rivers across the U.S. shift between very dry and very wet...
This manuscript addresses an important topic and provides a useful national-scale analysis of drought-to-flood and flood-to-drought streamflow transitions. I particularly appreciate the comprehensive sensitivity analysis and the discussion of implications for water resources management. However, several methodological details need substantial clarification. I also encourage the authors to sharpen the literature review and more explicitly state the scientific gap and contribution of the study.
Major comments
1. The introduction does a good job motivating the management challenges associated with drought-to-flood and flood-to-drought transitions. However, the literature review on related work is currently limited. For example, the manuscript refers to “impactful transitions” (line 175) and the overlap between meteorological and hydrological whiplash (lines 178-179). First, meteorological whiplash and hydrological whiplash need to be defined more clearly. Second, they need to be situated within existing literature. It would help to explain where previous studies stand in identifying impactful transitions, quantifying meteorological versus hydrological whiplash, and assessing their overlap. In addition, the novelty currently reads mainly as the “exploration of both DtF and FtD transitions” (line 171), but the manuscript should more clearly state the specific knowledge gap and how this study fills it. For example, is the key contribution the national-scale streamflow-based characterization, the comparison of DtF and FtD asymmetry, the sensitivity to event definitions, or the linkage to meteorological transitions?
2. Section 2.2 is central to the study, but many of the definitions and implementation choices are difficult to follow. Several terms are introduced before they are defined. The authors should reorganize this section so that event-definition components are introduced first, followed by using the definitions and then the sensitivity settings. Major points that need clarification include:
a) The manuscript uses minimum drought durations of 5, 15, and 30 days. It is worth reconsidering whether 5-day or even 15-day low-flow periods should be called “droughts,” or whether “dry periods” would be more accurate. Also, the duration may represent different lengths under fixed and variable-threshold methods. In the variable-threshold approach, it seems each daily value represents a 7-day averaged condition, so it has to be clarified whether the effective duration equals the nominal duration under the two approaches.
b) The phrase “the interevent period is […] less than 10% of the drought volume” (line 241) is confusing because a period is being compared with a volume. The authors should define drought volume and explain how this criterion is calculated.
c) Line 242, does “pool drought events” mean if drought days have inter-event period <=5 days, those days are merged as one event? It is not until here that I understand why you need an inter-event period, which was introduced in line 223. Similarly, is the “15-day” for flood separation (lines 246-247) a similar concept of “inter-event period” used in drought? If so, why not introduce it earlier in the first paragraph in section 2.2?
d) Lines 240-247, aren’t these the same as in lines 232-236? Again, it would be clearer to introduce all pooling, separation, and duration criteria before using the definitions.
3. The manuscript uses both fixed and variable percentile approaches, but the implementation is difficult to follow in several places.
a) The zero-flow adjustment is described very briefly (lines 212-215). While reference is cited, the authors should explain how days with different numbers of preceding zero-flow days are combined with positive-flow days when assigning percentiles, and whether this procedure is applied before or after 7-day averaging in the variable approach. It is also unclear whether zero-flow duration is calculated using the full time series (e.g., fixed approach) or only within the comparison window for a given day-of-year (e.g., variable approach).
b) The authors should acknowledge that changing the reference period can change the percentile values themselves (line 210). This matters because percentiles are central to identifying droughts and transitions.
c) The main text appears to identify droughts using variable percentiles but then evaluate “impactful” transitions using a fixed full-record percentile space (line 266). This hybrid framework is difficult to follow.
d) Line 329 states that the maximum increase and decrease in streamflow percentiles are calculated during each season and the preceding 90 days. Does this correspond to a 120-day window in total? However, the following sentence says these maximum changes are compared against the “seasons” where there were meteorological or hydrological transitions, which seems to imply a 30-day window. The wording is confusing and should be clarified.
e) The authors should also clarify how the maximum change in streamflow percentile is calculated. How do they ensure that the identified maximum change corresponds to the actual transitional event, rather than another large streamflow fluctuation occurring within the same window? An example would be very helpful to illustrate the procedure.
f) The authors should also specify which event values are compared to calculate the 50-percentile shift, such as drought minimum to flood maximum, event mean to event peak, or event end to event start, etc.
4. The manuscript compares hydrological DtF/FtD transitions with meteorological dry-to-wet and wet-to-dry transitions, but the temporal matching criteria need more explanation.
a) The manuscript should clarify the timescale mismatch between hydrological and meteorological transitions. Hydrological transitions are currently identified using relatively short event-scale criteria (e.g., 5-, 15-, and 30-day for drought, and no stated duration for flood), while meteorological transitions are based on seasonal timestep (line 317; is it 3-month?). The co-occurrence rate likely depends on the selected timescale, and the timescale mismatch should be acknowledged.
b) The authors should clarify how event order is considered. For example, must a hydrological DtF transition occur after a meteorological dry-to-wet transition? If order is not required, then the manuscript should avoid causal language such as meteorological transitions “produce” hydrological transitions (line 575).
c) Lines 325–326 state that the co-occurrence between hydrological DtF transitions and meteorological dry-to-wet transitions is evaluated during the same season or the immediately preceding or following season. Does this correspond to a total window of approximately 270 days? However, the caption of Figure 7 states that the co-occurrence is evaluated within one season. It was not until I reached Table S1 that I realized both metrics (same season only, and same or adjacent season) were analyzed. I suggest clarifying this distinction in the main text.
d) Lines 320–321: The authors may also want to clarify the rationale for the different timing choices used throughout the analysis. For the co-occurrence analysis, meteorological and hydrological transitions are assigned to the season in which the transition ends, whereas the rose plots use the month in which the first event ends. Is there a reason for using different temporal reference points in these analyses? A brief explanation would help readers understand the methodology.
5. Several statements should be interpreted more carefully. For example, when the manuscript reports the fraction of transitions occurring within 90 days or 30 days (Lines 431-433), this is more of a measure of the prevalence, not transition speed itself. Claims that DtF transitions are “faster” (line 431) should be supported by direct comparison of transition-time, such as Fig3 c, h.
Similarly, where the manuscript compares percentages across regions or groups, the term “fraction” may be more accurate than “frequency” (line 517), unless actual event counts are being compared. Small differences, such as 21.9% versus 20.8% (line 519), should not be overinterpreted as indicating that transitions are meaningfully more common in one region than another.
6. Several watershed characteristics used in the analysis need clearer definitions and data sources. The authors should specify the PET method and data source (line 292), define PET before first use (line 305), provide the data source for snow water equivalent (line 293), and clarify whether the “more than 25%” snow criterion is based on a long-term average. The calculation of “input seasonality” (line 309) also needs to be described.
There is also inconsistency in the use of “concentration.” On line 308, it appears to refer to flow concentration, while lines 533 and 536 describe rainfall and snowmelt concentration. Please clarify which variable you use to calculate the concentration.
Minor comments