Pith. sign in

REVIEW 4 major objections 5 minor 38 references

One fitted model — a gauge function built from the river network's block graph — estimates any joint flood probability at 10 stations, including simultaneous floods at four stations with a return period over 5,000 years.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Using block-graph gauge functions, the paper fits the first geometric extremal graphical model to 10 river gauging stations, enabling single-model estimates of simultaneous flood probabilities.

T0 review reviewed 2026-08-02 challenge →

load-bearing objection First real implementation of geometric extremal graphical models; sound methods and honest diagnostics, but the headline flood probability rests on a peak-matching preprocessing step that is not stress-tested. the 4 major comments →

arxiv 2607.15000 v2 pith:HXBLABTK submitted 2026-07-16 stat.AP stat.ME

Flood risk estimation via geometric extremal graphical models

Kristina Grolmusova , Jennifer L. Wadsworth , Thordis L. Thorarinsdottir This is my paper
classification stat.AP stat.ME MSC 62G3260G7062P12
keywords geometric extremesgauge functionextremal graphical modelsblock graphsflood risk estimationjoint exceedance probabilitiesriver flow extremestruncated gamma model
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper aims to bring the geometric approach to multivariate extremes to a scale that matters for flood risk: river networks typically have 10–20+ gauging stations, whereas geometric models had previously been fitted only in up to four dimensions, or up to ten with less interpretable machine-learning methods. The central proposal is to build the gauge function — the boundary curve describing how extreme flows can co-occur — out of the river network's block-graph structure, yielding a parsimonious 12-parameter model for the 10-station Preston network. The paper argues that one such fitted model can estimate any joint flood probability on the network, avoiding the awkward choice of a conditioning site that the conditional-extremes approach forces. As an illustration, it estimates the annual probability of simultaneous flooding at four stations as 1.8 × 10^-4, a return period beyond 5,000 years. If the approach is right, a single fit replaces a whole family of site-specific models for flood risk assessment.

Core claim

The paper's central claim is that geometric extremal graphical models can be made statistical. The gauge function g — the 1-homogeneous boundary function whose unit set is the limit set of the scaled extreme cloud — is assembled from the river network's block graph as a sum of clique gauges minus separator coordinates; for the Preston network this becomes six bivariate Gaussian gauges for adjacent flow-connected pairs plus one four-dimensional Gaussian gauge for the flow-unconnected clique. Fitted through the truncated-gamma model for radial exceedances above an angle-dependent threshold, this 12-parameter gauge reproduces the observed tail dependence, and with a new correction coefficient i

What carries the argument

The gauge function g(x) — the 1-homogeneous boundary of the limit set G = {x : g(x) ≤ 1}, whose shape dictates which variables can be extreme together. The paper builds g from a block graph (a decomposable graph with single-vertex separators, matching a river network's branching) as g(x) = Σ_C g_C(x_C) − Σ_D x_D; for Preston, six bivariate Gaussian gauges plus one four-dimensional clique gauge, minus x_3, x_4, x_5 and 3x_7. The truncated-gamma model R | {W = w, R > r_0(w)} ~ truncated Gamma(α, g(w)) makes the gauge a likelihood. The correction coefficient C_corr = (1/|S|) Σ_s exp(−b_s)/Pr̂(X_s > b_s) rescales estimates to offset marginal bias. The gauge carries the argument: it encodes depen

Load-bearing premise

The load-bearing premise is that the ±2-day peak-matching window turns raw daily flow records into genuine joint flood events: the paper itself notes the matched data show stronger dependence than the original series, so if the window is wrongly specified the fitted gauge and every extrapolated probability — including the 1.8 × 10^-4 annual flood probability — are biased.

What would settle it

Re-fit the dependence model on the original 15,459 daily observations under several alternative alignment rules — no matching at all, ±1-day and ±3-day windows, and lags based on river travel times — and compare the resulting gauge parameters and the estimated annual probability of simultaneous flooding at stations (3), (4), (5) and (7). If that probability moves by orders of magnitude across the choices, the central claim collapses; if it stays within the reported 0.8–2.4 × 10^-4 interval, the matching assumption is vindicated.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • A single fitted model answers joint flood probability questions for any combination of stations, replacing the four conditioning-site models that the conditional-extremes framework would need for the same four-station event — and it can handle combinations (such as stations 1 and 2) that no single conditioning model covers.
  • The annual probability of simultaneous flooding at stations (3), (4), (5) and (7) is estimated at 1.8 × 10^-4 (95% CI 0.8–2.4 × 10^-4), corresponding to a return period over 5,000 years; the extended model puts it at 1.6 × 10^-4.
  • The correction coefficient improves the model-based joint-tail coefficient χ_S(u) — the probability that all stations in a set S exceed a high quantile together — moving it closer to empirical values and narrowing the uncertainty as u → 1.
  • The graphical construction keeps the parameter count at 12 for 10 stations (six edge gauges plus six clique parameters) instead of the 45 parameters of a full 10-dimensional Gaussian gauge, which is what makes networks of 10–20+ stations tractable.
  • The exponential-Gaussian extension lets the data decide whether adjacent flow-connected pairs can experience joint extremes, with γ ≥ 1 estimated for pairs (3,7), (4,7) and (5,7); the paper notes the extrapolation differences from the simpler model appear only very far in the tail.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper: the same block-graph recipe should transfer to other river networks of 10–20+ stations, because the parameter count grows with the number of edges and cliques rather than with the square of the dimension; the flow-unconnected clique is the bottleneck, since each unconnected station added there costs more clique parameters.
  • Beyond the paper: the correction coefficient is not specific to flood risk — any geometric-extremes application (rainfall, sea levels, spatial fields) that simulates from a fitted gauge inherits the same marginal bias, so C_corr can be lifted directly into those settings.
  • Beyond the paper: the single-fit property suggests a queryable flood-risk surface — joint probabilities for any demanded set of sites and thresholds — but its far-tail calibration rests on the matched-event construction, so a natural stress test is to repeat the analysis with matching windows of ±1 and ±3 days and see whether the 5,000-year estimate moves.
  • Beyond the paper: the event-set generation procedure in Appendix D makes annual exceedance probabilities for any region computable in principle, but it needed a million simulated years to produce about 100 events at the four-station bankfull region, so a cheaper far-tail sampler would be needed to make flood event catalogues practical.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. This paper presents the first statistical implementation of geometric extremal graphical models, applying them to daily river flows at ten gauging stations on a river network near Preston, UK. The gauge function of the geometric limit set is parameterised through a block graph representing flow-connected and flow-unconnected stations (Eq. 7). The model is fitted via a truncated gamma model with kernel-based radial thresholds, and a correction coefficient C_corr (Eq. 3) is introduced to calibrate marginal probabilities. The authors assess fit via PP plots, limit-set projections and chi_S(u) comparisons, and use the fitted model to estimate the annual probability of simultaneous bankfull exceedance at stations (3),(4),(5),(7), reported as 1.8[0.8,2.4]*10^-4. An extended exponential-Gaussian gauge is proposed to allow joint extremes at selected pairs.

Significance. The methodological advance is real: it is the first parametric statistical implementation of geometric graphical models in a dimension (10) that is beyond previous geometric fits, and it offers a single-model alternative to conditional-extremes approaches with their conditioning-site ambiguity. The paper is careful in several respects: data and code are provided, block-bootstrap uncertainty quantification is used throughout, and the diagnostics are standard and clearly reported. The central quantitative claim, however, rests on a data-preprocessing step whose effect on dependence is acknowledged and not quantified, and on in-sample calibration. If the requested sensitivity analyses confirm stability, the method would be a useful addition to the flood-risk toolbox; as it stands, the numerical headline should be treated as conditional on the matching and threshold choices.

major comments (4)
  1. [§2.2 / App. A.1] The matched dataset is constructed by replacing each event with the maximum flow within ±p=2 days at each station, reducing 15,459 to 2,582 points. As the paper notes, this mechanically strengthens dependence (Fig. 2). Since no observation exceeds bankfull at stations (3),(4),(5),(7) (§5.2.4), the annual probability 1.8[0.8,2.4]·10^-4 is a pure extrapolation whose dependence structure comes entirely from the matching-induced correlation. The window p=2 is chosen from one 1994 event, with no sensitivity analysis over p. The reported block-bootstrap CIs condition on the matched dataset and ignore uncertainty from p, the matching procedure, and the mass-balance filter. A sensitivity analysis over p (and ideally a comparison to an un-matched analysis) is needed to support the central quantitative claim.
  2. [§5.2.2–5.2.3] The claim of 'strong extrapolation performance' (Abstract, §1) rests on the transformed PP plot (Figure 7) and χ_S(u) comparisons (Figure 9), which are in-sample diagnostics: both the empirical and model-based curves are computed from the same matched dataset and the same fitted model. These diagnostics do not validate extrapolation to the bankfull exceedance region in §5.2.4, where thresholds exceed observed maxima at all four stations. The χ_S(u) comparison is informative about dependence at observed quantile levels but not about the far-tail probability. The paper should include an out-of-sample or held-out event validation (e.g., fitting to a training period and evaluating on a test period, or a simulation-based calibration of the extrapolated probability) or temper the claim accordingly.
  3. [§3.2.3, Eq. (3)] The correction coefficient C_corr is an ad-hoc average of marginal probability ratios, justified only by 'numerical experiments' that are not shown. The reported probabilities, including 1.8·10^-4, are multiplied by this coefficient; different choices of S (which dimensions are included) and the marginal estimation used in the denominator can change the estimates. The paper should report the value of C_corr for the flood probability computation, show how the result changes without it, and provide a sensitivity analysis over S and over the exponential vs fitted marginal distribution.
  4. [§6, Table 2] The extended exponential-Gaussian model has a relatively flat likelihood and wide CIs for γ parameters (e.g., γ_{5,7}: 1.00 [0.67,1.97]) and the bivariate projection for pair (5,7) is wider than the data suggest (Fig. 10). The claim that the model can accommodate both simultaneous and non-simultaneous extremes is therefore only weakly identified by the data. This is not a fatal flaw, but the paper should explicitly discuss what information in the data identifies γ, and avoid presenting the extended model's ability to detect joint extremes as a demonstrated property of this dataset.
minor comments (5)
  1. [Data and code availability] There is a typo: 'NRF A athttps://nrfa.ceh.ac.uk/' should be 'NRF A at https://nrfa.ceh.ac.uk/'.
  2. [Figure 2] The axis labels 'GDF .9', 'GDF .5', etc. should be 'GDF_9', 'GDF_5' for readability.
  3. [§5.2.4] The bankfull values v_exp,i are given without units; please clarify that these are in units of the standard exponential margins, not m³/s.
  4. [§5.2.2] The tolerance intervals in the transformed PP plot assume independent data. The paper notes they would be wider for dependent data, but it would be helpful to quantify the impact using the block-bootstrap samples.
  5. [§3.2.3, Eq. (3)] Please state explicitly what happens when S is empty or when b_s=0 for some dimensions; currently the definition of S is informal.

Circularity Check

0 steps flagged

No significant circularity: the geometric/graphical framework is cited prior work, and the flood probability is a forward extrapolation from a fitted model, not a fitted quantity.

full rationale

The paper's derivation chain does not reduce to its own inputs. The truncated gamma model (Section 3.2) and the geometric extremal graphical gauge construction (Section 4.2, eq. 4) are taken from prior theoretical/methodological work by Wadsworth & Campbell (2024) and Papastathopoulos & Wadsworth (2026). These are external, parameter-free theoretical results with stated assumptions, not quantities fitted to the Preston data; the present paper's contribution is the statistical implementation. The graphical gauge in eq. (7) is a structured parametric form, and its 12 parameters are estimated by maximum likelihood from the matched river-flow data. The headline flood probability in Section 5.2.4 is then obtained by forward simulation from the fitted model into a tail region (k=2.7, 1,000,000 simulated points) where no observations exceed bankfull. That is an extrapolation, not a re-statement of a fitted parameter or of the data. The correction coefficient C_corr (eq. 3) calibrates marginal probabilities using the ratio of exponential to fitted marginal survival probabilities; it does not encode the joint event probability. The peak-matching pre-processing is a data-construction assumption that may inflate dependence, and the paper explicitly notes the matched data show stronger dependence; however, this is a validity/bias concern about the event definition, not a circular step in which the output is defined as the input. The self-citations to Wadsworth & Campbell (2024) and Papastathopoulos & Wadsworth (2026) are load-bearing only in providing the framework, but they are independent prior results and are not derived from the present application. The model diagnostics (PP plots, χ_S(u) comparisons) are comparisons of fitted model output to empirical quantities, and are not used to define the target probability. No equation or fitted value is shown to equal the headline probability by construction. Therefore the paper exhibits no significant circularity.

Axiom & Free-Parameter Ledger

7 free parameters · 8 axioms · 1 invented entities

The model's scientific content is carried by the geometric EVT framework and the graph-structured gauge; the paper does not derive new limit theorems but implements an existing theory. All 13 estimated parameters (a, six ρ's, six θ's; plus three γ's in the extended model) are fitted to the Preston data; the tuning choices τ=0.85, p=2, k=2.7 are also data-driven. The eight axioms reflect that the central claim stands on the limit-set/truncated-gamma asymptotic model, the block-graph gauge decomposition, and the data-construction choices (peak matching, seasonality ignored, Gaussian gauge). One new statistical construct is introduced—the correction coefficient C_corr—with no external falsifiable evidence. No physical entities are invented.

free parameters (7)
  • shape multiplier a (α = a·d) = 0.33 (CI 0.33–0.61) original; 0.36 (CI 0.34–0.63) extended
    Shape parameter of truncated gamma; estimated by MLE for flexibility although theory gives α=d.
  • six ρ parameters for adjacent flow-connected pairs = ρ37=0.98, ρ47=0.95, ρ48=0.96, ρ57=0.94, ρ59=0.97, ρ67=0.89 (Table 1)
    Correlation parameters of bivariate Gaussian gauge functions; MLE.
  • six θ parameters for the 4-dimensional clique {1,2,3,10} = θ12=0.97, θ13=0.75, θ110=0.71, θ23=0.76, θ210=0.74, θ310=0.73 (Table 1)
    Off-diagonal entries of the Gaussian clique gauge; MLE.
  • γ parameters (extended model) for pairs (3,7),(4,7),(5,7) = γ37=1.06, γ47=1.27, γ57=1.00 (Table 2)
    Control possibility of joint extremes; MLE; CIs include values <1.
  • radial threshold quantile τ = 0.85
    Selected for best diagnostics in threshold sensitivity analysis (§5.1); data-dependent.
  • peak-matching window p = 2 days
    Chosen from time-lag analysis on one event (§2.2, Appendix A.1).
  • radial extrapolation multiplier k = 2.7
    Chosen as highest k with zero scaled non-exceedances in the bankfull region (§5.2.4, Campbell et al. 2026 heuristic).
axioms (8)
  • domain assumption The scaled sample cloud of exponentially-margined observations converges to a star-shaped limit set G = {x: g(x) ≤ 1} with 1-homogeneous gauge g (Section 3.1).
    Foundation of geometric EVT; assumes the river-flow data are light-tailed after transformation and lie in the domain of attraction of such a limit set.
  • domain assumption R | {W=w, R>r0(w)} follows a truncated Gamma(α, g(w)) distribution (eq 1).
    Asymptotic model from Wadsworth & Campbell (2024); the likelihood and all diagnostics depend on it.
  • standard math For block graphs, g(x) = Σ_{C∈C} g_C(x_C) − Σ_{D∈D} x_D (eq 4); for trees, g(x) = Σ_{(i,j)∈E} g_{ij}(x_i,x_j) − x_i − x_j + Σ_k x_k (eq 5).
    Theorem from Papastathopoulos & Wadsworth (2026), accepted without proof; the parsimony of the model rests on it.
  • standard math Marginal gauge functions are obtained by g_J(x_J) = min_{x_{J^c}} g(x) (eq 6).
    From Nolde & Wadsworth (2022); used to project the fitted 10-dimensional gauge to bivariate plots for fit assessment.
  • domain assumption Gaussian bivariate gauge functions with ρ∈[0,1] adequately capture the dependence of every adjacent flow-connected pair, and a 4-dim Gaussian gauge captures the unconnected clique.
    Model choice; the paper picked Gaussian after comparing with logistic/inverted-logistic/square candidates on the same data (§5.1).
  • domain assumption The dependence structure of the 10 stations factorizes according to the block graph in Figure 6: a tree on flow-connected stations {3,...,9} plus a clique {1,2,3,10} joining the unconnected stations through station 3 (eq 7).
    If the true dependence does not respect this graph separation, the gauge is misspecified; this is untested.
  • domain assumption The peak-matched dataset of 2,582 points, built by replacing each event with the maximum flow in a ±2-day window, represents the true joint flood events (Section 2.2, Appendix A.1).
    The matching increases dependence visibly (Figure 2); the whole analysis uses the matched data as if they were the real multivariate extremes.
  • domain assumption Seasonal non-stationarity can be ignored without materially biasing the extreme dependence analysis (Appendix A.2).
    The paper follows prior practice; they justify by flood-interest being winter flows and the difficulty of decoupling margins and dependence.
invented entities (1)
  • Correction coefficient C_corr (eq 3) no independent evidence
    purpose: Multiplicatively adjust model-based joint probability estimates to compensate for systematic over/under-estimation of marginal tail probabilities.
    The correction coefficient is introduced as a heuristic multiplier; its validity is supported only by in-sample numerical experiments (Section 3.2.3, Figure 4) and χ_S(u) comparisons, without a derivation or external benchmark.

reviewed 2026-08-02 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Flood risk estimation via geometric extremal graphical models." pith.science (2026). https://pith.science/paper/HXBLABTK

@misc{pith2026260715000,
  author       = {Pith},
  title        = {Pith review of: Flood risk estimation via geometric extremal graphical models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HXBLABTK}},
  note         = {Machine review of arXiv:2607.15000}
}
Share X Bluesky LinkedIn Reddit HN
abstract

We exploit the new framework of multivariate geometric extreme value theory for the statistical analysis of river flow extremes at multiple locations on a river network. Current methodologies within the geometric framework are limited to a relatively low number of dimensions. This is insufficient for the purposes of flood risk estimation, since the number of gauging stations on a river network is often of the order $10-20+$. In order to create a parsimonious model in higher dimensions, we translate recent theoretical work on geometric extremal graphical models into statistical practice. We define the gauge function, a key object in geometric extremes, in a structured way using block graphs, which are a natural way of expressing the river network. We introduce both simple models, and more complex ones that can accommodate both simultaneous and non-simultaneous flows, and apply them to extreme flows at 10 locations on a river network around Preston, in north-west England. The models are shown to fit well and indicate strong extrapolation performance. We also introduce a correction coefficient for the geometric framework to address potential over- or under-estimation of marginal probabilities. The overall utility of our approach is illustrated through calculation of probabilities of simultaneous flooding at four locations on the network.

Figures

Figures reproduced from arXiv: 2607.15000 by Jennifer L. Wadsworth, Kristina Grolmusova, Thordis L. Thorarinsdottir.

Figure 1
Figure 1. Figure 1: Partial map of the UK showing the rivers as blue lines with gauging stations as [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Left: Plot of the GDF at gauging stations (9), (5) and (3) with an extreme event [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Illustration of the convergence of scaled sample cloud [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Left: Plot of the data (black) and simulated points (orange) with the region of [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: An example of a block graph. A graph G is defined as G = (V, E), where V = {1, . . . , d} is a set of vertices and E ⊂ V × V is a set of edges between pairs of distinct vertices. We focus on an undi￾rected graph, meaning that for two vertices i, j ∈ V, (i, j) ∈ E implies (j, i) ∈ E. This means that there is not a distinct edge when going from i to j and when go￾ing from j to i. A special class of graphs, t… view at source ↗
Figure 6
Figure 6. Figure 6: Left: River structure of the gauging stations of the Preston river network. The [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Transformed PP plot for the truncated gamma model with the graphical gauge [PITH_FULL_IMAGE:figures/full_fig_p013_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Plot of matched data (black) and simulated points (orange) for flow-unconnected [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Plot of empirical (solid black) and model-based [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Unit level sets of the bivariate projections of the graphical gauge function with [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Plot of empirical (solid black) and model-based [PITH_FULL_IMAGE:figures/full_fig_p019_11.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

38 extracted references · 5 linked inside Pith

  1. [1]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume = 86, number = 5, pages =

    Statistical inference for multivariate extremes via a geometric approach , author =. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume = 86, number = 5, pages =

  2. [2]

    arXiv preprint arXiv:2601.00239 , year =

    Geometric extremal graphical models and coefficients of extremal dependence on block graphs , author =. arXiv preprint arXiv:2601.00239 , year =

    arXiv

  3. [3]

    Extremes on river networks , author =

  4. [4]

    2025 , howpublished =

    2025

  5. [5]

    , year = 2020, month = sep, journal =

    Engelke, Sebastian and Hitz, Adrien S. , year = 2020, month = sep, journal =. Graphical

    2020

  6. [6]

    and Tawn, Jonathan A

    Coles, Stuart G. and Tawn, Jonathan A. , year = 1991, journal =. Modelling

    1991

  7. [7]

    arXiv preprint arXiv:2412.05195 , year =

    Piecewise-linear modeling of multivariate geometric extremes , author =. arXiv preprint arXiv:2412.05195 , year =

    Pith/arXiv arXiv

  8. [8]

    arXiv preprint arXiv:2411.17013 , year =

    Conditional extremes with graphical models , author =. arXiv preprint arXiv:2411.17013 , year =

    Pith/arXiv arXiv

  9. [9]

    2022 , month = sep, journal =

    Linking representations for multivariate extremes via a limit set , author =. 2022 , month = sep, journal =

    2022

  10. [10]

    arXiv preprint arXiv:2512.24392 , year =

    Geometric criteria for identifying extremal dependence and flexible modeling via additive mixtures , author =. arXiv preprint arXiv:2512.24392 , year =

    arXiv

  11. [11]

    2023 , howpublished =

    2023

  12. [12]

    and Griffin, A

    Vesuviano, G. and Griffin, A. , title =. 2025 , howpublished =

    2025

  13. [13]

    Flood Estimation Handbook , year =

  14. [14]

    Kjeldsen, T. R. and Jones, D. A. and Bayliss, A. C. , title =. 2008 , pages =

    2008

  15. [15]

    and Tawn, Jonathan A

    Heffernan, Janet E. and Tawn, Jonathan A. , year = 2004, month = aug, journal =. A

    2004

  16. [16]

    and Reich, Brian J

    Majumder, Reetam and Shaby, Benjamin A. and Reich, Brian J. and Cooley, Daniel S. , year=. Semiparametric estimation of the shape of the limiting bivariate point cloud , volume=. Bayesian Analysis , publisher=

  17. [17]

    Simpson and Jonathan A

    Emma S. Simpson and Jonathan A. Tawn , title =. Electronic Journal of Statistics , number =. 2024 , howpublished =

    2024

  18. [18]

    arXiv preprint arXiv:2406.19936 , archivePrefix =

    Deep learning of multivariate extremes via a geometric representation , author =. arXiv preprint arXiv:2406.19936 , archivePrefix =

    Pith/arXiv arXiv

  19. [19]

    arXiv preprint arXiv:2505.02957 , year =

    Generative modelling of multivariate geometric extremes using normalising flows , author =. arXiv preprint arXiv:2505.02957 , year =

    Pith/arXiv arXiv

  20. [20]

    arXiv preprint arXiv:2310.06130 , year =

    Statistical inference for radial generalized Pareto distributions and return sets in geometric extremes , author =. arXiv preprint arXiv:2310.06130 , year =

    Pith/arXiv arXiv

  21. [21]

    arXiv preprint arXiv:2603.18149 , year =

    Analysing extreme rainfall via a geometric framework , author =. arXiv preprint arXiv:2603.18149 , year =

    arXiv

  22. [22]

    Environmetrics , volume = 24, number = 1, pages =

    Estimating the probability of widespread flood events , author =. Environmetrics , volume = 24, number = 1, pages =

  23. [23]

    Journal of Multivariate Analysis , author =

    Estimation of the conditional distribution of a multivariate variable given that one of its components is large:. Journal of Multivariate Analysis , author =

  24. [24]

    arXiv preprint arXiv:2511.08192 , year =

    Geometric modelling of spatial extremes , author =. arXiv preprint arXiv:2511.08192 , year =

    arXiv

  25. [25]

    and Wadsworth, Jennifer L

    Huser, Raphaël G. and Wadsworth, Jennifer L. , year =. Modeling spatial processes with unknown extremal dependence class , volume =

  26. [26]

    2021 , date =

    Spatial Statistics , author =. 2021 , date =

    2021

  27. [27]

    Spatial Statistics , author =

    Higher-dimensional spatial extremes via single-site conditioning , volume =. Spatial Statistics , author =. 2022 , date =

    2022

  28. [28]

    , booktitle =

    Lauritzen, Steffen L. , booktitle =. Graphical models , year =

  29. [29]

    Environmetrics , author =

    Model‐based inference of conditional extreme value distributions with hydrological applications , volume =. Environmetrics , author =. 2019 , pages =

    2019

  30. [30]

    Journal of Hydrology , author =

    Spatial dependence in extreme river flows and precipitation for. Journal of Hydrology , author =. 2009 , pages =

    2009

  31. [31]

    Journal of the Royal Statistical Society Series C: Applied Statistics , author =

    Spatial risk assessment for extreme river flows , volume =. Journal of the Royal Statistical Society Series C: Applied Statistics , author =

  32. [32]

    Nat Commun , author =

    Flood exposure and poverty in 188 countries , volume =. Nat Commun , author =

  33. [33]

    Hydrology , author =

    Generation of spatially heterogeneous flood events in an. Hydrology , author =

  34. [34]

    Water Resources Research , author =

    Developing a fluvial and pluvial stochastic flood model of. Water Resources Research , author =

  35. [35]

    Water Resources Research , author =

    The spatial dependence of flood hazard and risk in the. Water Resources Research , author =

  36. [36]

    Water Resources Research , author =

    Use of hydrological models in global stochastic flood modeling , volume =. Water Resources Research , author =

  37. [37]

    Water Resources Research , author =

    Toward global stochastic river flood modeling , volume =. Water Resources Research , author =

  38. [38]

    Stat , author =

    Spatial extremal modelling:. Stat , author =

This paper was first reviewed by deepseek-v4-flash on August 2, 2026.