ReliefWeb

Technical summary. Details on the attribution system in Christidis (2021).
Nikolaos Christidis, May 2022.

A real-time attribution study was conducted during the heatwave that affected large parts of North India and Pakistan in April-May 2022. The analysis was produced with a system developed in Hadley Centre for the attribution of extremes in near-real time (Christidis, 2021). It employs an unconditional attribution framing, which estimates the changing risk of critical temperature threshold crossings under any possible conditions. The system also adopts the well-established and peer-reviewed risk-based methodology that infers probabilities of extreme events with and without the effect of human influence from large multi-model ensembles of climate model simulations. The new analysis employed coupled model data drawn from large multi-model ensembles of the Coupled Model Intercomparison Project phase 6 (CMIP6, Eyring et al., 2016). In total, the study uses data from 14 models that provide the necessary experiments for event attribution, namely historical simulations extended to the end of the 21st century with the medium emissions scenario SSP2-4.5 (ALL; Riahi et al., 2017) and simulations with natural forcings only (NAT) to year 2020. Each model provides several simulations for each experiment and there are in total 82 ALL and 71 NAT model runs for temperature, and similar numbers for rainfall (87 ALL and 72 NAT) as listed in Table 1 in the PDF.