A new radiative transfer model, CARE-RTM, enables scientists to simulate the propagation of sunlight through the atmosphere and ocean with improved precision and computational speed, providing a physical tool for extracting climate and environmental information from satellite data.

The model was developed by researchers from the Aerospace Information Research Institute (AIR) of the Chinese Academy of Sciences (CAS). Relevant findings were recently published in Advances in Atmospheric Sciences.

Satellite sensors detect radiative signals that have undergone scattering and absorption as light travels through the atmosphere, clouds, and ocean. To interpret these signals, scientists rely on radiative transfer models (RTMs). However, existing models often struggle to achieve high accuracy, computational efficiency, and comprehensive functionality at the same time.

CARE stands for Cloud Remote Sensing, Atmospheric Radiation and Renewable Energy Application, reflecting the model's broad coverage of key research areas in Earth observation. The CARE-RTM model addresses these challenges by integrating and refining a suite of advanced modules.

For gas absorption calculations, the model employs a GPU‑accelerated line‑by‑line scheme that accounts for gas‑mixing effects while enhancing calculation efficiency. This GPU implementation delivers speedups exceeding 500‑fold compared to traditional CPU‑based approaches, reducing computation times from minutes to seconds for demanding tasks.

The model also incorporates a Voronoi ice crystal scattering module to represent the complex, irregular shapes of natural ice particles. This capability supports the retrieval of ice cloud properties, which influences climate modeling and weather forecasting.

In the ocean domain, CARE‑RTM features multiple bio‑optical modules and an improved coupled atmosphere‑ocean scheme that simulates underwater light fields and water‑leaving radiance. This is particularly relevant for ocean color remote sensing, where the water signal typically accounts for only about 10% of top‑of‑atmosphere observations.

Furthermore, the researchers validated CARE‑RTM against benchmark datasets and compared its performance with established models across 100 sun‑sensor geometries and four atmosphere‑ocean scenarios spanning ultraviolet to visible wavelengths.

The results confirmed the model's numerical fidelity, with simulation errors below 0.02% for total radiance and below 0.005% for the degree of linear polarization at the top of the atmosphere.

The practical utility of CARE‑RTM was also demonstrated through ocean color retrievals using real observations from the Chinese FY‑3F/MERSI satellite. Chlorophyll‑a concentrations derived using CARE‑RTM showed close agreement with official MODIS ocean color products, confirming the model's capability to capture chlorophyll‑related spatial variability in observational scenarios.

"CARE‑RTM was developed to provide the scientific community with a simulation platform that combines accuracy, efficiency, and comprehensiveness," said Dr. SHI Chong from the AIR, the lead author of the study. "We hope it will serve as a tool that links satellite signals with the physical states of the atmosphere and ocean, supporting applications such as weather forecasting, climate research, and marine ecosystem monitoring."

The team plans to extend CARE‑RTM to account for Earth's curvature and improve land surface modeling, aiming to broaden its applications in Earth observation and climate science, SHI said.