Skip to main
Atmospheric science
George Mason University
Events

16 October (CLIM) Compo, Niño and MJO from Data

Oct 16, 2024, 1:30 - 2:30 PM

Planetary Hall 224

Gil Compo, University of Colorado, Boulder 

Title: Learning the Dynamics of El Nino and the Madden-Julian Oscillation from Data   

October 16 2024, 1:30pm 

Planetary Hall 224 and via Zoom (for link, email lortizur@gmu.edu) 

Host: Tim Delsole 

 

We have performed a diagnosis of the tropical Indo-Pacific climate system using an empirical-dynamical Linear Inverse Model (LIM) of weekly anomalies derived from atmospheric and oceanic reanalysis data for the 1979-2018 period. The model captures the essential features of ENSO and the MJO including their patterns and spectra, and by construction has no mean biases. The relative roles of various feedbacks in the system are diagnosed by performing a systematic set of feedback denial experiments with the model. This diagnosis suggests that the negative surface shortwave flux (SW) feedback on the SSTs (the so-called cloud shielding effect) is a dominant negative feedback on ENSO and is too weak in most climate models. The weakness of this feedback is likely behind their tendency to extend ENSO too far west into the western tropical Pacific. This compromises seasonal and longer-term predictions around the globe through spurious teleconnections. A weak SW feedback over the maritime continent is also consistent with the mean cold tongue and easterly trade wind biases of many climate models over the western tropical Pacific. Further diagnosis suggests that the magnitude of this feedback depends on the sensitivity of deep atmospheric convection and cloudiness to SST forcing, which is importantly influenced by a rectified effect (a “noise-induced drift”) of rapidly varying atmospheric diabatic and boundary layer processes. The rectification reduces the effective damping of low-level wind convergence and thence of tropospheric vertical velocities. To the extent that this rectification is too weak in models, the vertical velocities are also too weak, which contributes not only to their mean climate and ENSO biases but also to weakening the MJO in many models.