Dams and other water-control structures can influence rivers far downstream. We calculated the Streamflow Alteration Index for 245 576 subbasins across Ontario, Canada, averaging 4.3 square kilometres. By linking structure locations with connected lake and river networks, we mapped potential downstream influence, including places without flow measurements. This province-wide dataset supports river modelling, water management, ecological studies and targeted local assessments.

The Saclay observatory, 20 km southwest of Paris, monitored greenhouse and reactive gases, as well as aerosols, over 10 years to assess the impact of urban emissions. By comparing downwind and background conditions, the study revealed significant reductions in pollutants linked to traffic: -35.6 % for CO, -52.3 % for NOₓ, -56.7 % for eBC, and -15 % for CO₂. These trends align with emission decrease estimates from AIRPARIF inventories over the same period.

This study develops a monthly near-global sea surface wind dataset (MLAWind) from 1950–2023 through an interpretable machine learning framework integrating in-situ observations and satellite data. Evaluations show that MLAWind achieves comparable performance to the widely-used reanalysis datasets at different time scales. It provides reliable historical wind data during both satellite and non-satellite periods, demonstrating broad application prospects in ocean and climate research.