CNRM-CM5_for_ENSO_predictability

Published: 06-08-2020| Version 1 | DOI: 10.17632/bsnd8md962.1
Contributor:
Yann Planton

Description

Control experiment (piControl like) with CNRM-CM5.1 and ensemble experiments designed to test the influence of ocean heat content on ENSO predictability

Files

Steps to reproduce

Control experiment: 1007-years simulation (after a 200-year spin-up) with the CNRM-CM5.1 global climate model (Voldoire et al. 2013), with greenhouse gas concentrations and solar irradiance fixed to their observed value for 1850, as in CMIP5 pre-industrial control (piControl) simulations (Taylor et al. 2012). Ensemble experiments: 70-member experiments starting on November 1st, and lasting for 17 months (i.e., until the end of March, two years later), designed to test whether the long-lead memory of the ocean can lead to asymmetrical predictability of El Niño (EN) and La Niña (LN), building on Planton et al. (2018). Ensembles are generated by applying random, 0.1°C amplitude white noise perturbations on the initial SST field, as in Puy et al. (2019). In each file, the first member is the corresponding period from the control experiment (January first Y0 to March 31st Y0+2). In each file, the first 10 months (January first to October 31st) correspond to the period from the control experiment preceeding the start of the ensemble experiment. EXTnR1: starts from an extreme LN with a moderate recharge (defined during 2329-10) MODnR1: starts from a moderate LN a moderate recharge (defined during 2339-10) MODnR2: starts from a moderate LN a moderate recharge (defined during 3098-10) MODnR3: starts from a moderate LN with a moderate recharge (defined during 2278-10) NEUtR1: starts from a neutral ENSO with a moderate recharge (defined during 2411-10) MODnN1: starts from a moderate LN with a neutral preconditioning (defined during 2783-10) MODnNR: starts from a moderate LN with a preconditioning between neutral and moderate recharge (defined during 2351-10) NEUtN1: starts from a neutral ENSO with a neutral preconditioning (defined during 2515-10) NEUtNR: starts from a neutral ENSO with a preconditioning between neutral and moderate recharge (defined during 2499-10) MODpN1: starts from a moderate EN with a neutral preconditioning (defined during 2390-10) NEUtD1: starts from a neutral ENSO with a moderate discharge (defined during 2209-10) MODpD1: starts from a moderate EN with a moderate discharge (defined during 3141-10) MODpD2: starts from a moderate EN with a moderate discharge (defined during 2566-10) EXTpD1: starts from a moderate EN with an extreme discharge (defined during 2474-10) Westerly wind event: Westerly wind events (WWEs) are detected using Puy et al.’s (2016) methodology. Daily zonal wind stress (Taux) time series are intraseasonally filtered (5-90 days) and averaged between 2°S-2°N, then WWEs are defined when Taux above 0.04 N/m2 cover a zonal extent of at least 10° during at least 5 days. WWEs separated by less than 3° and 3 days and are grouped to be a single event. Weighted average (by the intraseasonally filtered Taux) are used to define the location and date of WWEs. The intensity of each WWE is defined as the space–time integration of the intraseasonally filtered Taux over the wind event patch.