Atmospheric composition of WASP-77 b revealed by new high-speed computational model

A new high-performance software tool called BeAR allows researchers to analyze exoplanet atmospheres at unprecedented speeds by utilizing graphics processing units. By applying this system to the hot Jupiter WASP-77 b, the team confirmed specific chemical signatures, including sub-solar metallicity and a near-solar carbon-to-oxygen ratio, demonstrating the model's high accuracy compared to previous methods.

Star system: WASP-77 A | Distance from the Sun: 344 light years

Close

Atmospheric composition of WASP-77 b revealed by new high-speed computational model

Star system: WASP-77 A | Distance from the Sun: 344 light years

Researchers have developed a powerful new tool known as BeAR, or the Bern Atmospheric Retrieval code, designed to analyze the compositions of distant worlds with remarkable speed and precision. This software acts as a successor to earlier atmospheric modeling tools and specializes in combining data from both high-resolution and low-resolution telescopes. By processing all calculations entirely on powerful graphics processing units, the framework can compute millions of atmospheric models every minute, significantly accelerating the pace of exoplanet research.

To test the effectiveness of this new system, the team performed retrievals on data collected from the hot Jupiter WASP-77 b. These retrievals involve comparing observed light patterns from a planet to mathematical models to determine what chemicals exist in its atmosphere. The analysis successfully reproduced previously published findings regarding the planet's chemical abundances and thermal structure with high consistency. When combining different sets of observational data, the system confirmed that WASP-77 b possesses a sub-solar metallicity, meaning it has fewer heavy elements than the Sun, and a carbon-to-oxygen ratio that is close to the solar value.

While the system offers significant advantages in speed, the research team also highlighted important considerations for future use. Testing showed that the current statistical methods may slightly underestimate the true uncertainty when calculating the ratios of different chemical elements. Despite this, the framework represents a major step forward in creating efficient, combined models that can handle massive amounts of data from modern space and ground-based telescopes in a fraction of the time required by previous standards.

Key findings
The BeAR framework utilizes graphics processing units to complete atmospheric forward models in approximately one millisecond.
Analysis of the hot Jupiter WASP-77 b confirmed a sub-solar metallicity and a near-solar carbon-to-oxygen ratio.
The model consistently reproduced previously published abundances and thermal structures within high degrees of statistical accuracy.
Joint retrieval capabilities allow for the simultaneous use of high-resolution and low-resolution observational data.
Performance testing indicates that complex retrievals with 17 free parameters can be completed in under an hour on standard high-end hardware.

Scientific study

BeAR: The Bern Atmospheric Retrieval code - A GPU-accelerated framework for low-resolution, high-resolution, and joint retrievals of exoplanet atmospheres