AI Predicts Solar Eruptions Hours Early

☀️ AI predicts solar eruptions early

🔭 Brown dwarf orbits massive A-star

🌌 Dark photon search expands

🕷️ NASA crafts colorful nebula portrait

🕳️ Black holes may birth giant planets

NJIT researchers developed EarlyDetect, a machine learning model trained on NASA's SDO (Solar Dynamics Observatory) data to identify precursor signals of solar active regions before they become visible on the Sun's surface.

EarlyDetect identified emergence signatures averaging 9.24 hours before visibility by detecting acoustic power drops in Doppler-velocity measurements, outperforming standard Transformer models and previous benchmark approaches in testing on unseen active regions.

The team released SolARED, the first public dataset for solar active region emergence, which NJIT's Mengjia Xu says provides shared resources for both machine learning and heliophysics communities to advance space weather prediction.

NASA's TESS (Transiting Exoplanet Survey Satellite) mission team identified a rare brown dwarf companion orbiting a massive A-type star, adding to the sparse population of known brown dwarfs in close stellar orbits.

The brown dwarf detection utilized TESS photometric transit observations combined with radial velocity follow-up measurements, revealing orbital parameters and mass constraints for this substellar object around its luminous host star.

According to the research team, this discovery helps address the "brown dwarf desert" phenomenon, where few brown dwarfs exist in close orbits, providing constraints on formation mechanisms for substellar companions.

Perimeter Institute researchers Junwu Huang and Mohamad Shalaby, collaborating with University of Maryland physicist Anson Hook, reanalyzed how hypothetical dark photons would interact with early universe plasma using new computer simulations.

Simulations revealed plasma becomes violently nonlinear during energy conversion, shutting off heating before significant transfer occurs, reopening parameter space across ten orders of magnitude from 10⁻¹⁵ to 10⁻⁶ electron volts.

Hook states previous exclusions claimed dark matter strength had to be 10⁸ weaker than actually possible, meaning this paper opens substantial new experimental possibilities for detecting dark matter particles.

Ohio State University researcher Jennifer Rodriguez led a team combining observations from NASA's Chandra X-ray Observatory, James Webb Space Telescope, and Hubble Space Telescope to study the Tarantula Nebula's energy dynamics.

Chandra detected X-rays from gas heated to millions of degrees, while Webb captured infrared data revealing thousands of young stars and cool dust within the nebula located 160,000 light-years away in the Large Magellanic Cloud.

The Astrophysical Journal paper concludes the nebula loses energy through three channels: hot gas leaking through shell walls, mixing between hot and cold gas, and thermal conduction with cooler material.

New Mexico State University astronomer Wladimir Lyra and collaborator Bhupendra Mishra simulated conditions in accretion disks around supermassive black holes, discovering potential planet-forming environments in their outer regions resembling protoplanetary disks.

Computer models showed dust condensing into clumps over millions of years, with resulting bodies potentially exceeding Jupiter's mass approaching brown dwarf scales, representing a bottom-up formation mechanism distinct from gravitational collapse.

Lyra suggests the upcoming Nancy Grace Roman Space Telescope, launching late August with its 300-megapixel infrared camera, could detect these objects through microlensing events in active galactic nuclei observations.