Surface ozone is an air pollutant that contributes to hundreds of thousands of premature deaths annually. Accurate short-term ozone forecasts may allow improved policy actions to reduce the risk to human health. However, forecasting surface ozone is a difficult problem as its concentrations are controlled by a number of physical and chemical processes that act on varying timescales. We implement a state-of-the-art transformer-based model, the temporal fusion transformer, trained on observational data from three European countries. In four-day forecasts of daily maximum 8-hour ozone (DMA8), our novel approach is highly skillful (MAE = 4.9 ppb, coefficient of determination $ {\mathrm{R}}^2=0.81 $) and generalizes well to data from 13 other European countries unseen during training (MAE = 5.0 ppb, $ {\mathrm{R}}^2=0.78 $). The model outperforms other machine learning models on our data (ridge regression, random forests, and long short-term memory networks) and compares favorably to the performance of other published deep learning architectures tested on different data. Furthermore, we illustrate that the model pays attention to physical variables known to control ozone concentrations and that the attention mechanism allows the model to use the most relevant days of past ozone concentrations to make accurate forecasts on test data. The skillful performance of the model, particularly in generalizing to unseen European countries, suggests that machine learning methods may provide a computationally cheap approach for accurate air quality forecasting across Europe.

A survey of the Milky Way disk and the Magellanic System at the wavelengths of the 21-cm atomic hydrogen (H i) line and three 18-cm lines of the OH molecule will be carried out with the Australian Square Kilometre Array Pathfinder telescope. The survey will study the distribution of H i emission and absorption with unprecedented angular and velocity resolution, as well as molecular line thermal emission, absorption, and maser lines. The area to be covered includes the Galactic plane (|b| < 10°) at all declinations south of δ = +40°, spanning longitudes 167° through 360°to 79° at b = 0°, plus the entire area of the Magellanic Stream and Clouds, a total of 13 020 deg2. The brightness temperature sensitivity will be very good, typically σT≃ 1 K at resolution 30 arcsec and 1 km s−1. The survey has a wide spectrum of scientific goals, from studies of galaxy evolution to star formation, with particular contributions to understanding stellar wind kinematics, the thermal phases of the interstellar medium, the interaction between gas in the disk and halo, and the dynamical and thermal states of gas at various positions along the Magellanic Stream.

An electromigration study has determined the lifetime characteristics and failure mode of dual-damascene Cu/oxide interconnects at temperatures ranging between 200 and 325 °C at a current density of 1.0 MA/cm2. A novel test structure design is used which incorporates a repeated chain of “Blech-type” line elements. The large interconnect ensemble permits a statistical approach to addressing interconnect reliability issues using typical failure analysis tools such as focused ion beam imaging. The larger sample size of the test structure thus enables efficient identification of “early failure” or extrinsic modes of interconnect failure associated with process development. The analysis so far indicates that two major damage modes are observable: (1) via-voiding and (2) voiding within the damascene trench.