Latent Gaussian Models for High-Dimensional Spatial Extremes

10/06/2021
by   Arnab Hazra, et al.
0

In this chapter, we show how to efficiently model high-dimensional extreme peaks-over-threshold events over space in complex non-stationary settings, using extended latent Gaussian Models (LGMs), and how to exploit the fitted model in practice for the computation of long-term return levels. The extended LGM framework assumes that the data follow a specific parametric distribution, whose unknown parameters are transformed using a multivariate link function and are then further modeled at the latent level in terms of fixed and random effects that have a joint Gaussian distribution. In the extremal context, we here assume that the data level distribution is described in terms of a Poisson point process likelihood, motivated by asymptotic extreme-value theory, and which conveniently exploits information from all threshold exceedances. This contrasts with the more common data-wasteful approach based on block maxima, which are typically modeled with the generalized extreme-value (GEV) distribution. When conditional independence can be assumed at the data level and latent random effects have a sparse probabilistic structure, fast approximate Bayesian inference becomes possible in very high dimensions, and we here present the recently proposed inference approach called "Max-and-Smooth", which provides exceptional speed-up compared to alternative methods. The proposed methodology is illustrated by application to satellite-derived precipitation data over Saudi Arabia, obtained from the Tropical Rainfall Measuring Mission, with 2738 grid cells and about 20 million spatio-temporal observations in total. Our fitted model captures the spatial variability of extreme precipitation satisfactorily and our results show that the most intense precipitation events are expected near the south-western part of Saudi Arabia, along the Red Sea coastline.

READ FULL TEXT

page 6

page 10

page 16

page 23

page 24

page 26

page 27

research
11/09/2020

High-dimensional modeling of spatial and spatio-temporal conditional extremes using INLA and the SPDE approach

The conditional extremes framework allows for event-based stochastic mod...
research
07/10/2019

Approximate Bayesian inference for spatial flood frequency analysis

Extreme floods cause casualties, widespread property damage, and damage ...
research
07/27/2019

Max-and-Smooth: a two-step approach for approximate Bayesian inference in latent Gaussian models

With modern high-dimensional data, complex statistical models are necess...
research
05/19/2021

Modelling short-term precipitation extremes with the blended generalised extreme value distribution

The yearly maxima of short-term precipitation are modelled to produce im...
research
05/03/2022

A hierarchical Bayesian non-asymptotic extreme value model for spatial data

Spatial maps of extreme precipitation are crucial in flood protection. W...
research
12/11/2019

A low-rank semiparametric Bayesian spatial model for estimating extreme Red Sea surface temperature hotspots

In this work, we focus on estimating sea surface temperature (SST) hotsp...
research
05/25/2020

Bayesian non-asymptotic extreme value models for environmental data

Motivated by the analysis of extreme rainfall data, we introduce a gener...

Please sign up or login with your details

Forgot password? Click here to reset