Identifying Spatio-Temporal Drivers of Extreme Events

TL;DR

The approach is trained end-to-end to predict spatio-temporally extremes and spatio-temporally drivers in the physical input variables jointly and successfully identifies drivers that are correlated with extremes.

Abstract

The spatio-temporal relations of impacts of extreme events and their drivers in climate data are not fully understood and there is a need of machine learning approaches to identify such spatio-temporal relations from data. The task, however, is very challenging since there are time delays between extremes and their drivers, and the spatial response of such drivers is inhomogeneous. In this work, we propose a first approach and benchmarks to tackle this challenge. Our approach is trained end-to-end to predict spatio-temporally extremes and spatio-temporally drivers in the physical input variables jointly. By enforcing the network to predict extremes from spatio-temporal binary masks of identified drivers, the network successfully identifies drivers that are correlated with extremes. We evaluate our approach on three newly created synthetic benchmarks, where two of them are based on remote sensing or reanalysis climate data, and on two real-world reanalysis datasets. The source code and datasets are publicly available at the project page https://hakamshams.github.io/IDE.

Figures (31)

• Figure 1: Overview of the objective of this work. We are interested in identifying spatio-temporal relations between the measurable impacts of extremes like the vegetation health index and their drivers . As drivers, we focus on anomalies in state variables of the land-atmosphere and hydrological cycle. The task is very challenging since the drivers can occur at a different region than the extreme event and earlier in time.
• Figure 2: An overview of the proposed model to identify the spatio-temporal relations between extreme agricultural droughts and their drivers. The input variables are first encoded into features. In a subsequent step, a lockup free quantization layer (LFQ) takes the extracted features and classifies the variables into a binary representation of drivers, where we consider the drivers as anomalous events in the input variables. Finally, a classifier is used to predict impacts of extreme events from the identified drivers.
• Figure 3: Qualitative results on the synthetic CERRA reanalysis from the test set at time step 2160. is the prediction, is the ground truth, and is the false positive. Albedo and relative humidity are not correlated with extremes, meaning that they do not contain drivers, but only random anomalies.
• Figure 4: F1-score with different correlation settings between the input variables and extremes.
• Figure 5: (a) Qualitative results on ERA5-Land over the EUR-11 domain. Shown are the identified drivers localized spatio-temporally $7$ weeks before the extreme agricultural drought events. (b) Temporal evolution of drivers during the extremes.