The representation of vegetative sub‐canopy wind is critical in numerical weather prediction (NWP) models for the determination of the air‐surface exchange processes of heat, momentum, and trace gases. Because of the relationship between wind speed and fire behaviors, the influence of the canopy on near‐surface wind speed is critical for prognostic fire spread models used in regional NWP models. In practice, the wind speed at the midflame point of fires (midflame wind speed) is used to determine the rate of fire spread. However, the wind speeds from most in situ measurements and NWP models are taken at some reference height above the canopy and fire flames. Hence, this study develops a modular and computationally‐efficient one‐dimensional model set composed of a canopy wind model and a wind adjustment factor (WAF) model for NWP applications across scales. The model set uses prescribed foliage shape functions to represent the vertical vegetation profile and its impacts on the three‐dimensional structure of horizontal wind speeds. Results from the canopy wind model well agree with ground‐based observations with average mean absolute bias, root mean square error and determination coefficients around 0.18 m s⁻¹, 0.40 m s⁻¹and 0.90, respectively. The WAF model provides midflame wind speeds by estimating the WAF based on canopy, fire and flame characteristics. Various user‐definable options provide flexibility to adapt to variations in canopy characteristics and additional complexities associated with wildfires. The model set is expected to improve NWP models by providing an improved representation of the sub‐grid wind flows at any spatial scale.

In New York State (NYS), episodic high fine particulate matter (PM_2.5) concentrations associated with aerosols originated from the Midwest, Mid-Atlantic, and Pacific Northwest states have been reported. In this study, machine learning techniques, including multiple linear regression (MLR) and artificial neural network (ANN), were used to estimate surface PM_2.5 mass concentrations at air quality monitoring sites in NYS during the summers of 2016–2019. Various predictors were considered, including meteorological, aerosol, and geographic predictors. Vertical predictors, designed as the indicators of vertical mixing and aloft aerosols, were also applied. Overall, the ANN models performed better than the MLR models, and the application of vertical predictors generally improved the accuracy of PM_2.5 estimation of the ANN models. The leave-one-out cross-validation results showed significant cross-site variations and were able to present the different predictor-PM_2.5 correlations at the sites with different PM_2.5 characteristics. In addition, a joint analysis of regression coefficients from the MLR model and variable importance from the ANN model provided insights into the contributions of selected predictors to PM_2.5 concentrations. The improvements in model performance due to aloft aerosols were relatively minor, probably due to the limited cases of aloft aerosols in current datasets.

Fire activities introduce hazardous impacts on the environment and public health by emitting various chemical species into the atmosphere. Most operational air quality forecast (AQF) models estimate smoke emissions based on the latest available satellite fire products, which may not represent real‐time fire behaviors without considering fire spread. Hence, a novel machine learning (ML) based fire spread forecast model, the Fire Intensity and spRead forecAst (FIRA), is developed for AQF model applications. FIRA aims to improve the performance of AQF models by providing realistic, dynamic fire characteristics including the spatial distribution and intensity of fire radiative power (FRP). In this study, data sets in 2020 over the continental United States (CONUS) and a historical California fire in 2024 are used for model training and evaluation. For application assessment, FIRA FRP predictions are applied to the Unified Forecast System coupled with smoke (UFS‐Smoke) model as inputs, focusing on a California fire case in September 2020. Results show that FIRA captures fire spread with R‐squared (R²) near 0.7 and good spatial similarity (∼95%). The comparison between UFS‐Smoke simulations using near‐real‐time fire products and FIRA FRP predictions show good agreements, indicating that FIRA can accurately represent future fire activities. Although FIRA generally underestimates fire intensity, the uncertainties can be mitigated by applying scaling factors to FRP values. Use of the scaled FIRA largely outperforms the experimental UFS‐Smoke model in predicting aerosol optical depth and the three‐dimensional smoke contents, while also demonstrating the ability to improve surface fine particulate matter (PM_2.5) concentrations affected by fires.

The presence of dense forest canopies significantly alters the near-field dynamical, physical, and chemical environment, with implications for atmospheric composition and air quality variables such as boundary layer ozone (O3). Observations show profound vertical gradients in O3 concentration beneath forest canopies; however, most chemical transport models (CTMs) used in the operational and research community, such as the Community Multiscale Air Quality (CMAQ) model, cannot account for such effects due to inadequate canopy representation and lack of sub-canopy processes. To address this knowledge gap, we implemented detailed forest canopy processes – including in-canopy photolysis attenuation and turbulence – into the CMAQv5.3.1 model, driven by the Global Forecast System and enhanced with high-resolution vegetation datasets. Simulations were conducted for August 2019 over the contiguous US. The canopy-aware model shows substantial improvement, with mean O3 bias reduced from +0.70 ppb (Base) to −0.10 ppb (Canopy), and fractional bias from +9.71 % to +6.37 %. Monthly mean O3 in the lowest model layer (∼ 0–40 m) decreased by up to 9 ppb in dense forests, especially in the East. Process analysis reveals a 75.2 % drop in first-layer O3 chemical production, with daily surface production declining from 673 to 167 ppb d−1, driven by suppressed photolysis and vertical mixing. This enhances NOx titration and reduces O3 formation under darker, stable conditions. The results highlight the critical role of canopy processes in atmospheric chemistry and demonstrate the importance of incorporating realistic vegetation-atmosphere interactions in CTMs to improve air quality forecasts and health-relevant exposure assessments.