Abstract. The Snow, Ice, and Aerosol Radiative (SNICAR) model has been used in various capacities over the last 15 years to model the spectral albedo of snow with light-absorbing constituents (LACs). Recent studies have extended the model to include an adding-doubling two-stream solver and representations of non-spherical ice particles; carbon dioxide snow; snow algae; and new types of mineral dust, volcanic ash, and brown carbon. New options also exist for ice refractive indices and solar-zenith-angle-dependent surface spectral irradiances used to derive broadband albedo. The model spectral range was also extended deeper into the ultraviolet for studies of extraterrestrial and high-altitude cryospheric surfaces. Until now, however, these improvements and capabilities have not been merged into a unified code base. Here, we document the formulation and evaluation of the publicly available SNICAR-ADv3 source code, web-based model, and accompanying library of constituent optical properties. The use of non-spherical ice grains, which scatter less strongly into the forward direction, reduces the simulated albedo perturbations from LACs by ∼9 %–31 %, depending on which of the three available non-spherical shapes are applied. The model compares very well against measurements of snow albedo from seven studies, though key properties affecting snow albedo are not fully constrained with measurements, including ice effective grain size of the top sub-millimeter of the snowpack, mixing state of LACs with respect to ice grains, and site-specific LAC optical properties. The new default ice refractive indices produce extremely high pure snow albedo (>0.99) in the blue and ultraviolet part of the spectrum, with such values only measured in Antarctica so far. More work is needed particularly in the representation of snow algae, including experimental verification of how different pigment expressions and algal cell concentrations affect snow albedo. Representations and measurements of the influence of liquid water on spectral snow albedo are also needed.

Abstract. The Snow, Ice, and Aerosol Radiative (SNICAR) model has been used in various capacities over the last 15 years to model the spectral albedo of snow with light-absorbing constituents (LAC). Recent studies have extended the model to include an adding-doubling two-stream solver and representations of non-spherical ice particles, carbon dioxide snow, snow algae, and new types of mineral dust, volcanic ash, and brown carbon. New options also exist for ice refractive indices and solar zenith angle-dependent surface spectral irradiances used to derive broadband albedo. The model spectral range was also extended deeper into the ultraviolet for studies of extraterrestrial and high-altitude cryospheric surfaces. Until now, however, these improvements and capabilities have not been merged into a unified code base. Here, we document the formulation and evaluation of the publicly-available SNICAR-ADv3 source code, web-based model, and accompanying library of particle optical properties. The use of non-spherical ice grains, which scatter less strongly into the forward direction, reduce the simulated albedo perturbations from LAC by ~9–31 %, depending on which of the three available non-spherical shapes are applied. The model compares very well against measurements of snow albedo from seven studies, though key properties affecting snow albedo are not fully constrained with measurements, including ice effective grain size of the top sub-millimeter of the snowpack, mixing state of LAC with respect to ice grains, and site-specific LAC optical properties. The new default ice refractive indices produce extremely high pure snow albedo (> 0.99) in the blue and ultraviolet part of the spectrum, with such values measured so far only in Antarctica. More work is needed particularly in the representation of snow algae, including experimental verification of how different pigment expressions and algal cell concentrations affect snow albedo. Representations and measurements of the influence of liquid water on spectral snow albedo are also needed.

Short‐term forecasting of wind gusts, particularly those of higher intensity, is of great societal importance but is challenging due to the presence of multiple gust generation mechanisms. Wind gust observations from eight high‐passenger‐volume airports across the continental United States (CONUS) are summarized and used to develop predictive models of wind gust occurrence and magnitude. These short‐term (same hour) forecast models are built using multiple logistic and linear regression, as well as artificial neural networks (ANNs) of varying complexity. A suite of 19 upper‐air predictors drawn from the ERA5 reanalysis and an autoregressive (AR) term are used. Stepwise procedures instruct predictor selection, and resampling is used to quantify model stability. All models are developed separately for the warm (April–September) and cold (October–March) seasons. Results show that ANNs of 3–5 hidden layers (HLs) generally exhibit higher hit rates than logistic regression models and also improve skill with respect to wind gust magnitudes. However, deeper networks with more HLs increase false alarm rates in occurrence models and mean absolute error in magnitude models due to model overfitting. For model skill, inclusion of the AR term is critical while the majority of the remaining skill derives from wind speeds and lapse rates. A predictive ceiling is also clearly demonstrated, particularly for the strong and damaging gust magnitudes, which appears to be partially due to ERA5 predictor characteristics and the presence of mixed wind climates.

No abstract

No abstract