Understanding Perturbed Parameter Ensemble Sensitivities Using A Contrastive Learning Approach
Abstract
Perturbed parameter ensembles (PPEs) reveal how physics parameters affect climate simulations, but interpreting parameter sensitivities across multivariate, spatially structured outputs remains challenging, particularly when calibrating models against observations. We develop an explainable contrastive learning model that maps 5 monthly cloud and radiation fields into a shared representation space. We train the model on the fields of two 100-member Community Atmosphere Model version 6 (CAM6) PPEs, spanning 34 parameters, that only differ in the warm rain microphysics scheme: KK2000, the default bulk microphysics scheme, and TAU-ML, a neural network emulator of a bin microphysics scheme. The learned representations separates two PPEs with over 94\% linear classification accuracy while preserving the seasonal variability and ensemble spread due to parameter perturbations. In the shared representation space, the representations of satellite observations occupy the same low-dimensional manifold as the PPEs but are displaced from them most strongly during boreal spring and autumn. TAU-ML PPE has a lower distance to observations compared to KK2000 in the representation space. Integrated Gradients attributions highlights the contributions in subtropical low-cloud regions, Northern and Southern Hemisphere storm track regions, and tropical convection regions to differences between PPEs and observations. Regional attributions correlate most strongly with parameters associated with cloud microphysics, boundary layer turbulence, and deep convection. These results demonstrate that explainable representations of climate fields can attribute model differences to specific variables, regions, seasons, and physical parameters.
Related
- Quantifying key parameter sensitivities for water table depth in hydrological schemes of CoLM-PSUADE at a mountain forest site
- What’s in flower? Phenological temperature sensitivities in a restricted alpine flora
- Hydrological Constraints on Temperature Sensitivities of Precipitation Frequency and Intensity
- Sensitivities of a simulated Arctic stratocumulus to aerosol and microphysical parameters
- CLIVeg: Contrastive Image-Vegetation Alignment for Botany-Aware Earth Observation Representations
BibTeX
@misc{fan2026understanding,
title = {{Understanding Perturbed Parameter Ensemble Sensitivities Using A Contrastive Learning Approach}},
author = {Da Fan and David John Gagne II and Gregory S Elsaesser and Brian Medeiros and Addisu G Semie and Qingyuan Yang and Akila Sampath and Subashree Venkatasubramanian},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.30420},
eprint = {2609.30420},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.30420},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Understanding Perturbed Parameter Ensemble Sensitivities Using A Contrastive Learning Approach
AU - Da Fan
AU - David John Gagne II
AU - Gregory S Elsaesser
AU - Brian Medeiros
AU - Addisu G Semie
AU - Qingyuan Yang
AU - Akila Sampath
AU - Subashree Venkatasubramanian
PY - 2026
DA - 2026/09/24
PB - arXiv
DO - 10.48550/arxiv.2609.30420
UR - https://arxiv.org/abs/2609.30420
AB - Perturbed parameter ensembles (PPEs) reveal how physics parameters affect climate simulations, but interpreting parameter sensitivities across multivariate, spatially structured outputs remains challenging, particularly when calibrating models against observations. We develop an explainable contrastive learning model that maps 5 monthly cloud and radiation fields into a shared representation space. We train the model on the fields of two 100-member Community Atmosphere Model version 6 (CAM6) PPEs, spanning 34 parameters, that only differ in the warm rain microphysics scheme: KK2000, the default bulk microphysics scheme, and TAU-ML, a neural network emulator of a bin microphysics scheme. The learned representations separates two PPEs with over 94\% linear classification accuracy while preserving the seasonal variability and ensemble spread due to parameter perturbations. In the shared representation space, the representations of satellite observations occupy the same low-dimensional manifold as the PPEs but are displaced from them most strongly during boreal spring and autumn. TAU-ML PPE has a lower distance to observations compared to KK2000 in the representation space. Integrated Gradients attributions highlights the contributions in subtropical low-cloud regions, Northern and Southern Hemisphere storm track regions, and tropical convection regions to differences between PPEs and observations. Regional attributions correlate most strongly with parameters associated with cloud microphysics, boundary layer turbulence, and deep convection. These results demonstrate that explainable representations of climate fields can attribute model differences to specific variables, regions, seasons, and physical parameters.
ER -