FuXi-RTM: A Physics-Guided Prediction Framework with Radiative Transfer Modeling

TL;DR

This paper tackles the challenge of physically inconsistent outputs in DL-based weather forecasts by introducing FuXi-RTM, a hybrid framework that couples a trainable FuXi forecaster with a fixed, DL-based radiative-transfer surrogate (DLRTM) to enforce radiative physics during training. The radiative surrogate is frozen, while a physics-regularized loss guides the FuXi predictions, enabling accurate forecasts with improved radiative flux consistency and energy conservation on ERA5 data. Across 3320 variable/lead-time combinations, FuXi-RTM achieves an $88.51\%$ improvement over FuXi-base, with near-perfect gains for radiative flux outputs and substantial improvements in ISSRD, demonstrating enhanced physical fidelity and forecast reliability. The architecture promises scalable extension to additional physical processes (e.g., convection, PBL, cloud microphysics), offering a path toward operationally trustworthy, physically constrained, next-generation weather prediction systems.

Abstract

Similar to conventional video generation, current deep learning-based weather prediction frameworks often lack explicit physical constraints, leading to unphysical outputs that limit their reliability for operational forecasting. Among various physical processes requiring proper representation, radiation plays a fundamental role as it drives Earth's weather and climate systems. However, accurate simulation of radiative transfer processes remains challenging for traditional numerical weather prediction (NWP) models due to their inherent complexity and high computational costs. Here, we propose FuXi-RTM, a hybrid physics-guided deep learning framework designed to enhance weather forecast accuracy while enforcing physical consistency. FuXi-RTM integrates a primary forecasting model (FuXi) with a fixed deep learning-based radiative transfer model (DLRTM) surrogate that efficiently replaces conventional radiation parameterization schemes. This represents the first deep learning-based weather forecasting framework to explicitly incorporate physical process modeling. Evaluated over a comprehensive 5-year dataset, FuXi-RTM outperforms its unconstrained counterpart in 88.51% of 3320 variable and lead time combinations, with improvements in radiative flux predictions. By incorporating additional physical processes, FuXi-RTM paves the way for next-generation weather forecasting systems that are both accurate and physically consistent.

Figures (6)

• Figure 1: The overall structure of our method. (a) Schematic of FuXi-RTM. (b) Architecture of FuXi-base. (c) Schematic of the deep learning-based radiative transfer model (DLRTM) utilizing a bidirectional long short-term memory (Bi-LSTM) architecture.
• Figure 2: Visualized sampling strategy. Left: Global random sampling. Right: The proposed SRC sampling. The dark regions (0 values) indicating areas without direct sunlight, which are excluded as potential SRC center points. Red points or box represents the sampling locations.
• Figure 3: Scorecard of nRMSE differences in globally-averaged, latitude-weighted RMSE between FuXi-base and FuXi-RTM. Each subplot corresponds to one of the FuXi direct output variable: 18 surface variables and 5 upper-air variables. For upper-air variables, the rows of each heatmap represent 13 pressure levels. The columns correspond to 40 forecast lead times at 6-hour intervals, spanning from 6 hours to 10 days. The color of each cell indicates the nRMSE differences, with blue denoting negative values (FuXi-RTM outperforms FuXi-base) and red indicating positive values (FuXi-base outperforms FuXi-RTM). The nRMSE difference ranges between -2 and 2, with numeric values overlaid on cells that fall outside this range.
• Figure 4: Scorecard of nRMSE differences in globally-averaged, latitude-weighted RTM RMSE between FuXi-base and FuXi-RTM. Each subplot corresponds to one of the DLRTM output variable: ISSRD, SWDFLX and SWUFLX at the surface level and 50 hPa. Blue denoting negative values (FuXi-RTM outperforms FuXi-base).
• Figure 5: Snapshot examples of ISSRD. From left to right: GT (ground truth), FuXi-base (model predictions), FuXi-RTM (model predictions), FuXi-base (diff) (difference between FuXi-base and GT), and FuXi-RTM (diff) (difference between FuXi-RTM and GT). The forecasts are initialized at four different times: 06 UTC on September 27, 2018; 00 UTC on September 28, 2018; 00 UTC on January 1, 2022; and 18 UTC on January 10, 2022. The corresponding forecast horizons are 6, 6, 120, and 240 hours, respectively.