Sources of Predictability for Subseasonal Precipitation in South America

Prof. Kathy Pegion

University of Oklahoma

Monday November 10, 2025, 2 PM ET

 

Abstract:

Most global land regions have little to no significant average subseasonal skill for  precipitation. However, parts of South America, particularly Brazil, have modest but significant  skill in most subseasonal prediction models. The skill indicates that predictability exists and the  source of the skill represents a source predictability. We investigate where this skill is coming from  to better understand subseasonal precipitation predictability in South America. 

Using subseasonal re-forecasts from the NCAR-CESM2 model, we demonstrate that significant  skill persists even when interannual variability is removed. The highest skill occurs during austral  summer and spring. To isolate sources of predictability, we analyze a novel set of re-forecast  experiments initialized with climatological atmosphere, land and ocean states. Results indicate  that atmospheric initial conditions are essential for achieving skill, while ocean initializations  contribute to skill containing interannual variability and land initializations contribute minimally to  skill. 

Canonical correlation analysis is used to identify the most skillful spatial patterns and associated  time series during the high-skill seasons. The most skillful patterns in December–February contain  additional skill beyond the South American dipole – the dominant pattern of precipitation  variability. This additional skill is not associated with propagating tropical convection or sea  surface temperature anomalies indicating that the Madden–Julian Oscillation (MJO), El Niño– Southern Oscillation (ENSO), or South Atlantic SST variability are not the sources of predictability. 

Instead, a wave-train-like structure extends across the South Pacific, resembling a Rossby wave  response. Idealized experiments using a simplified atmospheric general circulation model show  that this pattern can be reproduced by stationary tropical heating over the Maritime Continent,  suggesting a dynamical link between tropical heating and South American precipitation variability  on subseasonal timescales independent of ENSO and the MJO.

 

Biosketch:

Dr. Kathy Pegion is an Associate Professor and Williams Chair in Earth System Prediction at the University of Oklahoma. She earned her Ph.D. in Climate Dynamics from George Mason University. She leads the Earth System Prediction Lab (ESPLab), which focuses on subseasonal-to-seasonal (S2S) predictability—identifying sources and limits of forecast skill on weeks-to-months timescales and  applying predictions to benefit society. Her team uses physical- dynamical models, explainable AI, and advanced statistics. As lead of  the Subseasonal Consortium (SubC), she collaborates across agencies and institutions to improve extended-range forecasting. Current ESPLab projects address food and water security, transportation, and extremes.

 

Webinar:

Event site: https://go.umd.edu/pegion

Zoom Webinar: https://go.umd.edu/essicseminarwebinars

Zoom Meeting ID: 918 7733 3086
Zoom password: essic

US Toll: +13017158592
Global call-in numbers: https://umd.zoom.us/u/aMElEpvNu

Add to Google Calendar

 

For IT assistance:
Cazzy Medley: cazzy@umd.edu


Resources:

Seminar schedule & archive: https://go.umd.edu/essicseminar

Seminar Google calendar: https://go.umd.edu/essicseminarcalendar

Seminar recordings on Youtube: https://www.youtube.com/user/ESSICUMD

Date

Nov 10 2025

Time

2:00 am - 3:00 pm

Location

Remote via Zoom

Category

Organizer

John Xun Yang
Email
jxyang@umd.edu