What's in the latent space? Exploring coupled tropical Pacific variability within a Multi-branch $β$-Variational Autoencoder

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wisinski, Emily F., Molina, Maria J., Hall, Kyle J. C., Bao, Hannah, Mahajan, Salil, Rosenbloom, Nan, Fasullo, John
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911583834210304
author Wisinski, Emily F.
Molina, Maria J.
Hall, Kyle J. C.
Bao, Hannah
Mahajan, Salil
Rosenbloom, Nan
Fasullo, John
author_facet Wisinski, Emily F.
Molina, Maria J.
Hall, Kyle J. C.
Bao, Hannah
Mahajan, Salil
Rosenbloom, Nan
Fasullo, John
contents What is encoded in the latent space of a multi-branch $β$-variational autoencoder ($β$-VAE) trained on coupled tropical Pacific climate fields? To answer this question, we assess the reconstruction skill and physical interpretability of the latent space of a multi-branch $β$-VAE trained on sea surface temperature, ocean heat content, and outgoing longwave radiation across the tropical Pacific from a 500-year preindustrial control simulation. The model generalizes well, with only modest degradation from training to test performance, and preserves the dominant basin-scale structure of all three fields. Latent-space diagnostics show that variability is organized unevenly across dimensions: sea surface temperature is concentrated in a smaller subset of latent dimensions, whereas ocean heat content and outgoing longwave radiation are more broadly distributed across multiple dimensions. Comparisons with conventional tropical Pacific diagnostics further show that several latent dimensions align with known El Niño and La Niña variability, while others capture related coupled ocean-atmosphere variability on decadal or longer timescales. Sensitivity experiments and latent traversals identify dimensions associated with eastern-Pacific-like, central-Pacific-like, coastal, subsurface-dominant, and atmosphere-dominant variability. Together, these results show that the multi-branch $β$-variational autoencoder yields a skillful and physically informative reduced representation of coupled tropical Pacific variability.
format Preprint
id arxiv_https___arxiv_org_abs_2604_07137
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle What's in the latent space? Exploring coupled tropical Pacific variability within a Multi-branch $β$-Variational Autoencoder
Wisinski, Emily F.
Molina, Maria J.
Hall, Kyle J. C.
Bao, Hannah
Mahajan, Salil
Rosenbloom, Nan
Fasullo, John
Atmospheric and Oceanic Physics
What is encoded in the latent space of a multi-branch $β$-variational autoencoder ($β$-VAE) trained on coupled tropical Pacific climate fields? To answer this question, we assess the reconstruction skill and physical interpretability of the latent space of a multi-branch $β$-VAE trained on sea surface temperature, ocean heat content, and outgoing longwave radiation across the tropical Pacific from a 500-year preindustrial control simulation. The model generalizes well, with only modest degradation from training to test performance, and preserves the dominant basin-scale structure of all three fields. Latent-space diagnostics show that variability is organized unevenly across dimensions: sea surface temperature is concentrated in a smaller subset of latent dimensions, whereas ocean heat content and outgoing longwave radiation are more broadly distributed across multiple dimensions. Comparisons with conventional tropical Pacific diagnostics further show that several latent dimensions align with known El Niño and La Niña variability, while others capture related coupled ocean-atmosphere variability on decadal or longer timescales. Sensitivity experiments and latent traversals identify dimensions associated with eastern-Pacific-like, central-Pacific-like, coastal, subsurface-dominant, and atmosphere-dominant variability. Together, these results show that the multi-branch $β$-variational autoencoder yields a skillful and physically informative reduced representation of coupled tropical Pacific variability.
title What's in the latent space? Exploring coupled tropical Pacific variability within a Multi-branch $β$-Variational Autoencoder
topic Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2604.07137