Marisol Osman

Marisol Osman

Climate Researcher

Research

I study why the climate of South America varies from weeks to decades ahead, and how well we can predict it. My work combines the dynamics of the main drivers of Southern Hemisphere variability (ENSO, the Indian Ocean Dipole, the Southern Annular Mode and the stratospheric polar vortex) with multi-model forecasts, and turns that knowledge into climate services for agriculture, energy and water management.

Diagram placing each research theme on a timescale axis from days to decades

Research themes: Combined climate drivers · Seasonal prediction · Subseasonal prediction and weather regimes · Stratosphere-troposphere coupling · Circulation change and storylines · Climate services · Current directions

Combined influence of ENSO and other climate drivers on South America

Climate variability on interannual timescales in the Southern Hemisphere, particularly in South America, is influenced by several climate patterns. The El Niño-Southern Oscillation (ENSO) plays a leading role, shifting rainfall and temperature anomalies across the continent. The Southern Annular Mode (SAM), the Indian Ocean Dipole (IOD) and the stratospheric polar vortex add to the complexity of this variability by modifying the atmospheric circulation and the way ENSO signals reach the extratropics.

My research focuses on how these patterns act together, using reanalyses, large ensembles of model simulations and initialized predictions. For example, we showed that the IOD modulates the ENSO teleconnection to the Southern Hemisphere and its impacts over South America during austral spring, and that the stratospheric polar vortex and ENSO combine to shape the zonal asymmetries of the upper-tropospheric circulation. With Luciano Andrián and Nahuel Gómez we are extending this work to the role of the polar stratosphere and to the predictability of these combined effects.

Next question: can forecast systems capture how the IOD and the stratosphere modulate ENSO impacts, and does accounting for them improve seasonal forecasts for South America?

Scatter plot of ENSO and IOD indices in austral spring, 1940-2020
ENSO (ONI) and Indian Ocean Dipole (DMI) indices in austral spring, 1940-2020. Most events occur together and in the same phase, so their effects on South America have to be disentangled. From Andrian et al. (2024), CC BY 4.0.

Key papers:

Seasonal predictability and prediction in South America

The development of regional climate services requires understanding the predictability and prediction skill of climate variability on different timescales. On seasonal timescales, slowly evolving components of the Earth system, such as the ocean, soil moisture and sea ice, are the dominant source of skill.

During my PhD I studied the predictability of South American climate on seasonal timescales in multi-model ensembles. I then developed a calibrated and consolidated probabilistic seasonal forecast for South America based on Ensemble Regression, which is now used operationally by the Argentine National Weather Service (SMN) to prepare its official seasonal outlooks, under a CONICET-SMN agreement for which I am the technical lead. I have also supervised students documenting the seasonal predictability of precipitation, temperature and 10-m wind in coupled climate models.

Next question: how can seasonal forecasts of variables that users need, such as wind for the energy sector or soil water for agriculture, be calibrated and delivered operationally?

Key papers:

Subseasonal prediction and weather regimes

Weather regimes are quasi-stationary, recurrent and persistent large-scale circulation patterns that modulate surface weather on daily to weekly timescales. They are a useful predictor in a range of applications, in particular for the energy sector, and understanding their predictability on subseasonal timescales helps anticipate the evolution of the weather from days to weeks ahead.

During my postdoc at KIT I focused on the year-round North Atlantic-European weather regimes defined by Grams and co-authors, asking: 1) Can forecast systems skillfully predict weather regimes weeks ahead? 2) How does performance differ between forecast systems? 3) What are the sources of errors in weather regime forecasts? I did this research with Dr Christian Grams (KIT) and Dr Remo Beerli (AXPO Solutions), and we published the first multi-model assessment of the predictive skill of S2S models for these regimes. Back in Buenos Aires, I co-supervise PhD student Marco Micieli, who is working on a physically robust definition of weather regimes over South America and their applications. I have also studied the subseasonal predictability of heat and cold waves and of summer precipitation in South America.

Next question: can physically robust weather regimes for South America identify windows of opportunity, when subseasonal forecasts are more skillful than usual?

Maps of the seven year-round Atlantic-European weather regimes
The seven year-round Atlantic-European weather regimes and the "no regime" category (500-hPa geopotential height anomalies, gpm). From Osman et al. (2023), CC BY 4.0.

Key papers:

Stratosphere-troposphere coupling in forecast systems

The stratosphere is an important source of predictability for surface weather and climate on subseasonal to seasonal timescales, but forecast systems still struggle to represent it and its coupling with the troposphere. As part of the SNAP (Stratospheric Network for the Assessment of Predictability) community of APARC, whose coordination group I am part of, I contribute to multi-model studies that quantify stratospheric biases in subseasonal forecast systems, their sources, and how they affect the prediction of events such as sudden stratospheric warmings.

Next question: how do biases in the Southern Hemisphere stratosphere affect the prediction of surface climate in South America?

Heat map of stratospheric biases in subseasonal forecast systems
Biases in the variance of stratosphere-troposphere coupling diagnostics in week 4 (days 22-28) of Southern Hemisphere spring (SON) forecasts, for 22 forecast systems. From Garfinkel et al. (2025), CC BY 4.0.

Key papers:

Southern Hemisphere circulation change and storylines

Climate models disagree on how the Southern Hemisphere circulation and regional rainfall will respond to greenhouse gas increases and ozone recovery. Together with Julia Mindlin, Ted Shepherd and Carolina Vera, I use a storyline approach that describes this uncertainty as a small set of physically plausible futures, conditioned on the response of remote drivers such as tropical warming and the stratospheric polar vortex. This work explains and predicts the behaviour of the Southern Hemisphere eddy-driven jet and yields plausible drying and wetting scenarios for southeastern South America.

Next question: which circulation storylines are emerging in observations, and what do they imply for climate risk in South American sectors?

Schematic of the remote drivers of regional circulation change in South America
Remote drivers linking greenhouse gases and ozone to regional circulation and precipitation changes in southern South America. From Mindlin et al. (2023), CC BY 4.0.

Key papers:

Climate services and applications

Coming from a farming town, I have always been interested in how climate information can help people who depend on it. I work with anthropologists, economists, farmers and weather services to co-produce forecast products that can be used in decision making, from soil water reserve and soybean production forecasts to week-2 rainfall outlooks for small farmers. I co-coordinate the SAGE project (Subseasonal Applications in Agriculture and Environment) of the World Weather Research Programme, am a member of the WCRP Working Group on Subseasonal to Interdecadal Prediction (WGSIP) and of the WMO Study Group on Renewable Energy Transition, and contribute to rapid attribution of extreme events with ClimaMeter.

Next question: how can regime-dependent skill and uncertainty be built into the decisions of farmers, energy operators and disaster managers?

SAGE project theory of change diagram
SAGE theory of change: from understanding how subseasonal forecasts are used to making them more useful, usable and used. From Osman et al. (2026), CC BY 4.0.

Key papers: