Overview
The Jülich Rapid Spectral Simulation Code (JURASSIC) is an open-source infrared radiative transfer model designed for atmospheric remote sensing applications. It provides a flexible and computationally efficient framework for simulating infrared radiances and transmittances, as well as for performing inverse modelling through an integrated optimal estimation retrieval system.
JURASSIC is intended for scientific and operational use cases that require a balance between physical realism and computational performance, such as large-scale satellite data analysis, sensitivity studies, and retrieval algorithm development.
Purpose and scope
Infrared remote sensing instruments provide key observations of atmospheric temperature, trace gases, aerosols, and clouds. While line-by-line radiative transfer models offer high spectral accuracy, their computational cost limits their applicability for large datasets, ensemble simulations, or near-real-time applications.
JURASSIC addresses this challenge by combining established spectral approximations with precomputed spectroscopic information. This allows the model to achieve high computational efficiency while maintaining the level of accuracy required for quantitative atmospheric remote sensing.
The model is not tied to a specific instrument or mission and is designed as a general-purpose tool that supports a wide range of infrared remote sensing geometries and configurations.
Radiative transfer capabilities
JURASSIC represents the atmosphere as a vertically stratified medium and performs radiative transfer calculations along curved ray paths that account for atmospheric refraction. The model supports multiple observation geometries, including:
- limb sounding
- nadir viewing
- zenith viewing
- occultation geometries
Radiative transfer calculations are based on fast spectral approximations, most notably the Emissivity Growth Approximation (EGA) and the Curtis–Godson Approximation (CGA). These methods enable accurate modelling of infrared absorption and emission without the need for computationally expensive line-by-line calculations during runtime.
Spectral absorption and emission are represented using precomputed lookup tables derived from detailed line-by-line radiative transfer models. During execution, band-averaged emissivities and transmittances are obtained through fast interpolation within these tables, preserving spectroscopic fidelity while enabling rapid simulations.
Retrieval and inverse modelling
In addition to forward modelling, JURASSIC includes an integrated optimal estimation retrieval framework. This allows atmospheric state variables to be retrieved directly from measured radiances using a consistent forward–inverse modelling approach.
Typical retrieval targets include atmospheric temperature and trace gas volume mixing ratios, but the framework is designed to be flexible and extensible. Retrieval calculations follow established optimal estimation principles and provide access to diagnostic quantities such as averaging kernels and error estimates.
Typical applications
JURASSIC has been applied in a wide range of atmospheric remote sensing studies, particularly for infrared limb and nadir measurements from satellite instruments. Applications include:
- trace gas retrievals and climatologies
- temperature retrievals for gravity wave studies
- aerosol and cloud analyses
- sensitivity studies and forward-model benchmarking
The model has been used extensively in the analysis of measurements from instruments such as the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) and the Atmospheric Infrared Sounder (AIRS), among others.
A selection of scientific publications using JURASSIC is provided in the References section.
Model assumptions and extensions
The core JURASSIC model assumes local thermodynamic equilibrium (LTE), such that molecular level populations are determined by the local atmospheric temperature. This assumption is valid for most infrared remote sensing applications in the troposphere and stratosphere.
The standard configuration of JURASSIC is designed for clear-air conditions, where scattering of infrared radiation can be neglected. Simplified parameterizations can be used to represent cloud and aerosol effects in many applications. More advanced treatments of infrared scattering have been developed in dedicated extensions and applied in specialized studies.
Over time, JURASSIC has been extended to support additional capabilities, including tomographic retrieval methods and GPU-accelerated radiative transfer. Ongoing development aims to integrate these capabilities into a unified, modular code base to improve maintainability and long-term sustainability.
Software design and performance
JURASSIC is implemented in the C programming language with a modular architecture that separates ray tracing, radiative transfer, spectroscopy, and retrieval components. The model supports hybrid MPI–OpenMP parallelization and is designed for efficient execution on multicore CPUs and high-performance computing systems.
Automated tests and example projects are provided to verify correct installation and numerical behaviour. Model outputs have been extensively validated against reference radiative transfer models and documented in intercomparison studies.
JURASSIC is distributed as open-source software under the GNU General Public License (GPL) and is openly available to support transparent, reproducible, and extensible atmospheric remote sensing research.