Data-Efficient Generative Modeling of Non-Gaussian Global Climate Fields via Scalable Composite Transformations

TL;DR

This paper introduces a data-efficient, scalable framework for modeling complex, non-Gaussian global climate fields that requires significantly fewer training samples than existing methods, enabling more effective climate variability simulations.

Contribution

It presents a novel composite transformation approach combining nonparametric Bayesian transport maps with flexible marginal models for efficient, high-fidelity climate field emulation using minimal data.

Findings

Achieves high-fidelity climate field modeling with only 10 training samples.

Outperforms state-of-the-art methods trained on 80 samples.

Computational cost scales linearly with spatial dimension.

Abstract

Quantifying uncertainty in future climate projections is hindered by the prohibitive computational cost of running physical climate models, which severely limits the availability of training data. We propose a data-efficient framework for emulating the internal variability of global climate fields, specifically designed to overcome these sample-size constraints. Inspired by copula modeling, our approach constructs a highly expressive joint distribution via a composite transformation to a multivariate standard normal space. We combine a nonparametric Bayesian transport map for spatial dependence modeling with flexible, spatially varying marginal models, essential for capturing non-Gaussian behavior and heavy-tailed extremes. These marginals are defined by a parametric model followed by a semi-parametric B-spline correction to capture complex distributional features. The marginal parameters are spatially smoothed using Gaussian-process priors with low-rank approximations, rendering the computational cost linear in the spatial dimension. When applied to global log-precipitation-rate fields at more than 50,000 grid locations, our stochastic surrogate achieves high fidelity, accurately quantifying the climate distribution's spatial dependence and marginal characteristics, including the tails. Using only 10 training samples, it outperforms a state-of-the-art competitor trained on 80 samples, effectively octupling the computational budget for climate research. We provide a Python implementation at https://github.com/jobrachem/ppptm .