Accepted author manuscript, 3.32 MB, PDF document
Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License
Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - MODELLING EXTREMES OF SPATIAL AGGREGATES OF PRECIPITATION USING CONDITIONAL METHODS
AU - Richards, Jordan
AU - Tawn, Jonathan
AU - Brown, Simon
PY - 2022/12/31
Y1 - 2022/12/31
N2 - Inference on the extremal behaviour of spatial aggregates of precipitationis important for quantifying river flood risk. There are two classes of previousapproach, with one failing to ensure self-consistency in inference acrossdifferent regions of aggregation and the other imposing highly restrictive assumptions. To overcome these issues, we propose a model for high-resolutionprecipitation data, from which we can simulate realistic fields and explorethe behaviour of spatial aggregates. Recent developments have seen spatialextensions of the Heffernan and Tawn (2004) model for conditional multivariateextremes, which can handle a wide range of dependence structures.Our contribution is twofold: extensions and improvements of this approachand its model inference for high-dimensional data; and a novel framework forderiving aggregates addressing edge effects and sub-regions without rain.Weapply our modelling approach to gridded East-Anglia, UK precipitation data.Return-level curves for spatial aggregates over different regions of varioussizes are estimated and shown to fit very well to the data.
AB - Inference on the extremal behaviour of spatial aggregates of precipitationis important for quantifying river flood risk. There are two classes of previousapproach, with one failing to ensure self-consistency in inference acrossdifferent regions of aggregation and the other imposing highly restrictive assumptions. To overcome these issues, we propose a model for high-resolutionprecipitation data, from which we can simulate realistic fields and explorethe behaviour of spatial aggregates. Recent developments have seen spatialextensions of the Heffernan and Tawn (2004) model for conditional multivariateextremes, which can handle a wide range of dependence structures.Our contribution is twofold: extensions and improvements of this approachand its model inference for high-dimensional data; and a novel framework forderiving aggregates addressing edge effects and sub-regions without rain.Weapply our modelling approach to gridded East-Anglia, UK precipitation data.Return-level curves for spatial aggregates over different regions of varioussizes are estimated and shown to fit very well to the data.
U2 - 10.1214/22-AOAS1609
DO - 10.1214/22-AOAS1609
M3 - Journal article
VL - 16
SP - 2693
EP - 2713
JO - Annals of Applied Statistics
JF - Annals of Applied Statistics
SN - 1932-6157
IS - 4
ER -