Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Identifying Key Drivers of Heatwaves: A Novel Spatio-Temporal Framework for Extreme Event Detection

J. Pérez-Aracil, C. Peláez-Rodríguez, Ronan McAdam, Antonello Squintu, Cosmin M. Marina, Eugenio Lorente-Ramos, Niklas Luther, Veronica Torralba, Enrico Scoccimarro, Leone Cavicchia, Matteo Giuliani, Eduardo Zorita, Felicitas Hansen, David Barriopedro, Ricardo Garcia-Herrera, Pedro A. Gutiérrez, Jürg Luterbacher, Elena Xoplaki, Andrea Castelletti, S. Salcedo-Sanz

TL;DR

A novel framework (STCO-FS) is proposed to identify key immediate HW drivers by combining clustering algorithms with an ensemble evolutionary algorithm, which effectively identifies significant variables influencing HWs in this region.

Abstract

Heatwaves (HWs) are extreme atmospheric events that produce significant societal and environmental impacts. Predicting these extreme events remains challenging, as their complex interactions with large-scale atmospheric and climatic variables are difficult to capture with traditional statistical and dynamical models. This work presents a general method for driver identification in extreme climate events. A novel framework (STCO-FS) is proposed to identify key immediate (short-term) HW drivers by combining clustering algorithms with an ensemble evolutionary algorithm. The framework analyzes spatio-temporal data, reduces dimensionality by grouping similar geographical nodes for each variable, and develops driver selection in spatial and temporal domains, identifying the best time lags between predictive variables and HW occurrences. The proposed method has been applied to analyze HWs in the Adda river basin in Italy. The approach effectively identifies significant variables influencing HWs in this region. This research can potentially enhance our understanding of HW drivers and predictability.

Identifying Key Drivers of Heatwaves: A Novel Spatio-Temporal Framework for Extreme Event Detection

TL;DR

A novel framework (STCO-FS) is proposed to identify key immediate HW drivers by combining clustering algorithms with an ensemble evolutionary algorithm, which effectively identifies significant variables influencing HWs in this region.

Abstract

Heatwaves (HWs) are extreme atmospheric events that produce significant societal and environmental impacts. Predicting these extreme events remains challenging, as their complex interactions with large-scale atmospheric and climatic variables are difficult to capture with traditional statistical and dynamical models. This work presents a general method for driver identification in extreme climate events. A novel framework (STCO-FS) is proposed to identify key immediate (short-term) HW drivers by combining clustering algorithms with an ensemble evolutionary algorithm. The framework analyzes spatio-temporal data, reduces dimensionality by grouping similar geographical nodes for each variable, and develops driver selection in spatial and temporal domains, identifying the best time lags between predictive variables and HW occurrences. The proposed method has been applied to analyze HWs in the Adda river basin in Italy. The approach effectively identifies significant variables influencing HWs in this region. This research can potentially enhance our understanding of HW drivers and predictability.

Paper Structure

This paper contains 12 sections, 10 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Data description
  3. Potential drivers and target
  4. Data extraction and preparation
  5. Spatio-Temporal Cluster-Optimized Feature Selection (STCO-FS)
  6. Dimensionality Reduction through Clustering
  7. Candidate Selection: the optimization problem
  8. Machine Learning classifiers
  9. Experimental work and results
  10. The selected variables
  11. Results for different Machine Learning models
  12. Conclusions and further research

Figures (10)

  • Figure 1: Scheme of the proposed feature selection framework.
  • Figure 2: Illustration of time series: time lag, forecast horizon and sequence length parameters.
  • Figure 3: Clusters provided by the K-Means+ algorithm ($K=5$) for the first group of variables: meteorological variables.
  • Figure 4: Example of the individual solution provided by the optimization algorithm. Red boxes represent variables that have not been selected in this particular solution.
  • Figure 5: Test vs train (CV) performance of all the potential solutions tried by the optimization algorithm. The solutions represented in red are the ones selected for an in-depth study.
  • ...and 5 more figures