We have created a 14-year hourly air quality dataset at 12 km resolution by combining satellite observations of atmospheric composition with air quality models over the contiguous United States (CONUS) . The dataset has been found to reproduce key observed features of air quality over the CONUS. To enable easy visualization and interpretation of county level air quality measures and trends by stakeholders, an ArcGIS air quality dashboard has also been developed.

We develop two machine-learning-based approaches to temporally extrapolate uncertainty in hub-height wind speed modeled by a numerical weather prediction model. We test our approaches in the California Outer Continental Shelf, where a significant offshore wind energy development is currently being planned, and we find that both provide accurate results.

The authors demonstrate how the analog ensemble (AnEn) can efficiently generate deterministic and probabilistic forecasts of air quality. The method avoids the complexity and real-time computational expense of dynamical (i.e., model-based) ensembles. AnEn deterministic predictions have lower errors and are better correlated with observations. Probabilistic forecasts from AnEn are statistically consistent, reliable, and sharp, and they quantify the uncertainty of the underlying prediction.