A neural network-based scale-adaptive cloud-fraction scheme for GCMs

TL;DR

This paper introduces a neural network-based scheme for accurately simulating sub-grid cloud fraction in GCMs, improving spatial distribution and vertical structure representation with high correlation to observational data.

Contribution

The paper presents a novel neural network diagnostic scheme trained on CloudSat data that enhances cloud fraction modeling in GCMs across various resolutions and climate regimes.

Findings

Achieves correlation coefficient > 0.9 with observed cloud fraction

Significantly reduces biases in cloud vertical structure

Improves spatial distribution of cloud fraction in simulations

Abstract

Cloud fraction significantly affects the short- and long-wave radiation. Its realistic representation in general circulation models (GCMs) still poses great challenges in modeling the atmosphere. Here, we present a neural network-based diagnostic scheme that uses the grid-mean temperature, pressure, liquid and ice water mixing ratios, and relative humidity to simulate the sub-grid cloud fraction. The scheme, trained using CloudSat data with explicit consideration of grid sizes, realistically simulates the observed cloud fraction with a correlation coefficient (r) > 0.9 for liquid-, mixed-, and ice-phase clouds. The scheme also captures the observed non-monotonic relationship between cloud fraction and relative humidity and is computationally efficient, and robust for GCMs with a variety of horizontal and vertical resolutions. For illustrative purposes, we conducted comparative analyses of the 2006-2019 climatological-mean cloud fractions among CloudSat, and simulations from the new scheme and the Xu-Randall scheme (optimized the same way as the new scheme). The network-based scheme improves not only the spatial distribution of the total cloud fraction but also the cloud vertical structure (r > 0.99). For example, the biases of too-many high-level clouds over the tropics and too-many low-level clouds over regions around 60{\deg}S and 60{\deg}N in the Xu-Randall scheme are significantly reduced. These improvements are also found to be insensitive to the spatio-temporal variability of large-scale meteorology conditions, implying that the scheme can be used in different climate regimes.