A. E. Gelfand, J. Castillo-Mateo, A. C. Cebrián Guajardo, J. ASIN LAFUENTE
(Joint work with Jorge Castillo-Mateo, Ana Cebrián, Jesús Asín, Erin Schliep, María Asunción Beamonte, Jesús Abaurrea)
There is increasing evidence that global warming manifests itself in more frequent warm days and that heat waves will become more frequent. However, there is no formal definition of a heat wave in the literature. So, we consider extreme heat events which are well definedat local scales. Formally, an extreme heat event (EHE) is a run of consecutive days above a specified local temperature threshold. Here, we work with a dataset over Aragón that is rich in time (daily temperature data over 60 years) but sparse in space (18 locations).
We discuss three useful strategies for such EHE exploration. First, we offer a space-time threshold exceedance model to capture characteristics of EHEs such as incidence, duration/persistence, average exceeedance above the threshold during the EHE, and maximum exceedance above the threshold during the EHE. Then, we present an approach for developing the spatial extent of an EHE, as a generalization of a spatial cdf, with illustration over the Pyrenees and over the Ebro Valley. Finally, we examine a novel spatial quantile autoregression, with primary interest in the :95 quantile. We consider conditional quantiles to examine persistence and then offer a method to convert conditional quantiles to marginal quantiles to enable spatial interpolation. In our analyses, we focus on the spatial nature of EHEs and on the temporal change in occurrence.
Keywords:
Scheduled
STATISTICAL PLENARY SESSION - Alan E. Gelfand (Aula Magna)
June 7, 2022 10:20 AM
Auditorium
