Regional flood hazard mapping for February 2023 Ex-Tropical Cyclone Gabrielle, New Zealand

Pelmard, Joë; Lane, Emily Margaret; Harang, Alice; Bosserelle, Cyprien; Xu, Zhonghou; Pearson, Rose; Carey-Smith, Trevor; Smart, Graeme; Bind, Jochen; Kumar, Raghunathan Senthil; Cheng, Elvira; Bui, Giang; Paulik, Ryan; Zorn, Conrad; Wotherspoon, Liam

doi:10.5194/essd-2026-56

Preprints

https://doi.org/10.5194/essd-2026-56

Preprints

31 Mar 2026

| 31 Mar 2026

Status: this preprint is currently under review for the journal ESSD.

Regional flood hazard mapping for February 2023 Ex-Tropical Cyclone Gabrielle, New Zealand

Joë Pelmard, Emily Margaret Lane, Alice Harang, Cyprien Bosserelle, Zhonghou Xu, Rose Pearson, Trevor Carey-Smith, Graeme Smart, Jochen Bind, Raghunathan Senthil Kumar, Elvira Cheng, Giang Bui, Ryan Paulik, Conrad Zorn, and Liam Wotherspoon

Abstract. On February 13^th to 14^th 2023 ex-Tropical Cyclone Gabrielle (ETC Gabrielle) caused widespread flooding in Aotearoa New Zealand. The dataset presented is a collection of simulations representing the hydrodynamics response of 16 locations across the Hawke’s Bay and Tairāwhiti regional districts to ETC Gabrielle. Using gridded hourly rainfall derived from rainfall gauges, an end-to-end hydrodynamics modelling framework was applied to generate 16 inundation models with variable grid resolutions from 64 m to 4 m. The framework includes: (i) the identification of contributing river networks and catchments from an initial outline of the regions of interest, (ii) reconstruction of the DEM and surface roughness map, (iii) extraction of the rainfall, (iv) hydrological modelling in the upper catchments to estimate input river hydrographs, and (v) hydrodynamic modelling on the lower catchment. These models were calibrated and validated against hydrograph records (temporal accuracy), as well as post-event aerial imagery and in-situ high-water mark surveys (spatial accuracy) where available. With the increasing frequency of extreme weather events, this spatio-temporally detailed and methodologically consistent dataset offers valuable insights for researchers and stakeholders on the implications of large-scale flood events and will support efforts to enhance protective and preventive measures as well as post hoc analysis of direct and indirect impacts (e.g., building and infrastructure physical damage), while the model framework can be employed to simulate flood maps for future events. The dataset is available at https://doi.org/10.5281/zenodo.17986628 (Pelmard, 2025a) for the Tairāwhiti regional district and https://doi.org/10.5281/zenodo.18001575 (Pelmard, 2025b) for the Hawke’s Bay regional district.

Received: 22 Jan 2026 – Discussion started: 31 Mar 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.

Download & links

Status: open (until 27 Jun 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on essd-2026-56', Anonymous Referee #1, 20 May 2026 reply
This manuscript presents a dataset of hydrodynamic flood inundation simulations reconstructing the flood hazard characteristics associated with ex-Tropical Cyclone Gabrielle across 16 locations in Aotearoa New Zealand. The dataset is potentially valuable due to its high spatial resolution, regional coverage, and integration of hydrological and hydrodynamic modelling workflows. Validation against hydrographs, aerial imagery, and surveyed high-water marks further strengthens the potential usefulness of the dataset.
While the dataset itself is potentially valuable, the manuscript does not sufficiently articulate its specific scientific and practical relevance beyond the reconstruction of a past flood event. Section 1 focuses extensively on the description of the flood event and associated impacts, but the broader applicability, reuse potential, and added value of the dataset relative to existing flood inundation products remain insufficiently discussed. In particular, Sections 1 and 8 would benefit from a clearer and more concrete discussion of the intended applications, target users, and unique contributions of the dataset.
Additionally, the manuscript would benefit from substantial editorial revision. Several Sections (3,4,6) are difficult to follow due to verbose writing, excessive subsection fragmentation, and the mixing of conceptual workflow descriptions with implementation details. Reducing the number of subsections and improving the conceptual organization of the methodology would substantially improve readability and reproducibility. For example, the distinction between Sections 4 and 6.2 is not always clear, as validation methodology, calibration procedures, validation results, and uncertainty interpretation are distributed across multiple sections. This fragmentation makes the overall modelling and evaluation workflow difficult to follow. A clearer separation between methodology, validation protocol, and validation results/discussion would substantially improve readability. Finally, some aspects of the calibration procedure require stronger methodological justification, particularly regarding the scaling of rainfall and river inputs and the dynamic modification of topography to introduce levee breaches, as these choices may substantially influence the simulated inundation patterns and validation results.
Specific comments:
P2L49: The term “spectacular flash floods” appears informal and subjective for scientific writing. A more precise and neutral description of the event severity would be preferable.

P2L52-53: The statement regarding “multi-scale impacts” is currently too broad and would benefit from greater specificity. The manuscript should clarify what scales and impact types are being referred to and how infrastructure failures demonstrated these effects.

P3: Several acronyms used in Figure 1 (e.g., VCSN, HBRC, and GDC) are not defined upon first use, which may reduce readability for an international audience unfamiliar with regional New Zealand institutions and datasets. These abbreviations should be expanded either in the main text or in the figure caption.

P3: In Figure 1 (a) and (b) is missing.

P6L106: The rationale and methodology used to select the modelling domains are currently insufficiently described. In particular, the use of “population” and “economical exposure” as selection criteria requires clearer definition. Additionally, the term ‘economical exposure’ appears inappropriate in this context and should likely be replaced.

P7: The title “Numerical framework” appears somewhat misleading, as the section describes the overall modelling workflow and data-processing pipeline rather than the numerical framework itself.

P7L111: Several acronyms and institution names (e.g., NIWA) are used without being defined upon first use. This reduces readability for an international audience and should be addressed consistently throughout the manuscript.

P7L120: The opening description of the domain delineation workflow in Section 3.3 is difficult to follow. In particular, the phrase “through the compilation of geographical data” remains too vague, and the specific role of the different datasets (e.g., shoreline, LiDAR coverage…) in refining the area of interest polygon is not clearly explained.

P9L155: A few lines above, the framework is introduced as “GeoFabrics”, while here it is written as “Geofabrics”. Consistent naming should be used throughout the manuscript.

P9L162-165: This sentence is difficult to follow and would benefit from being split into shorter statements.

P11L226: Repetition (see line210).

P12L236: Acronym already introduced.

P13L261: The treatment of surveyed points that are dry in the model outputs is potentially problematic. Replacing dry cells with the nearest wet grid cell may artificially reduce prediction errors and obscure flood extent mismatches. This procedure requires stronger methodological justification and its potential influence on the reported validation statistics should be explicitly assessed.

P12L238: The title ‘Modelling strategy’ appears somewhat misleading, as the section primarily describes calibration, validation, and scenario simulations rather than the modelling strategy itself.

P13: The calibration methodology requires further clarification and justification. While the manuscript describes the calibration as ‘minimal’, the applied adjustments include scaling rainfall inputs, modifying river injections, and dynamically altering topography to introduce levee breaches, which may substantially influence the simulated inundation patterns. The criteria used to apply these corrections, their magnitude, and their potential impact on model reproducibility and event-specific tuning are currently insufficiently documented.

P20: Figure 8d is very hard to read. The manuscript states in the abstract that the models were “validated against hydrograph records”; however, the corresponding validation procedure remains insufficiently documented in the main text. While example comparisons are presented in this figure, it is unclear which quantitative criteria or metrics are used to assess model performance.

While the high-water mark validation procedure is described in detail, the corresponding validation approaches based on hydrograph records and aerial imagery remain comparatively under-documented. Although example comparisons are presented, it is often unclear how these validations were performed in practice, which variables or criteria were used, and whether the assessment was qualitative or quantitative.

Reply
Citation: https://doi.org/10.5194/essd-2026-56-RC1

Data sets

Multi-Scenario Flood Simulations Dataset for 2023 ex-Tropical Cyclone Gabrielle in Hawke's Bay, Aotearoa New Zealand J. Pelmard et al. https://doi.org/10.5281/zenodo.18001575

Multi-Scenario Flood Simulations Dataset for 2023 ex-Tropical Cyclone Gabrielle in Tairāwhiti, Aotearoa New Zealand J. Pelmard et al. https://doi.org/10.5281/zenodo.17986628

Viewed

Total article views: 520 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
335	164	21	520	19	28

HTML: 335
PDF: 164
XML: 21
Total: 520
BibTeX: 19
EndNote: 28

Views and downloads (calculated since 31 Mar 2026)

Month	HTML	PDF	XML	Total
Apr 2026	199	57	18	274
May 2026	136	107	3	246
Jun 2026	0

Cumulative views and downloads (calculated since 31 Mar 2026)

Month	HTML	PDF	XML	Total
Apr 2026	199	57	18	274
May 2026	136	107	3	246
Jun 2026	0

Viewed (geographical distribution)

Total article views: 520 (including HTML, PDF, and XML) Thereof 520 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 02 Jun 2026

Short summary

This study presents a dataset of simulations reconstructing February 2023 floods caused by ex-Tropical Cyclone Gabrielle in eastern Aotearoa New Zealand. Combining rainfall and landscape data and numerical modelling, the dataset gives insights into how flooding developed over space and time across the modeled regions. The results aim to support researchers, communities and decision-makers in post-event analysis, risk assessment and planning, to better prepare for extreme weather events.


Total:	0
HTML:	0
PDF:	0
XML:	0