A cohort of 1,547,244 Medicaid-enrolled mother-child pairs linked prenatal PM2.5 exposure with higher autism risk, while ADHD and postnatal findings varied by timing and race or ethnicity.
A commentary on the AMOC–cryosphere system argues that stabilizing feedbacks deserve a place in tipping-risk assessments, while leaving their net effect unresolved.
A preprint using Sentinel-3 reflectance and Norwegian monitoring data found large accuracy differences across feature-engineering choices for chlorophyll-a and Secchi disk depth, with the best setup varying by model and target.