Development of novel techniques in data-driven methods has motivated the application of these methods to space weather. For operational space weather in particular, it is vital to clarify regimes of validity for model fits, and ensure model generalizability. This avoids compromising model performance, especially in operational scenarios, where generalizability is especially important. We show an example of the importance of this class of techniques for the problem of F10.7 nowcasting.

The F10.7 solar radio flux is a critical quantity for operational space weather nowcasting and forecasting, where it is routinely used as a driver for coupled atmospheric models to estimate a variety of important quantities such as the neutral atmospheric density. Although there have been several successful developments in the way of parametric modeling to ensure F10.7 coverage during outages (often using the sunspot number or radio flux observations at neighboring wavelengths), these developments refrained from employing comprehensive cross-validation schemes to ensure model generalizability, and can stand to benefit from recently-developed techniques for modeling nonlinear phenomena. We present an approach that uses Feature Ordering by Conditional Independence (FOCI) to identify favorable surrogates for the F10.7 index through a series of distance computations, and combines this with modeling of F10.7 with linear models and Generalized Additive Models (GAM). We find that this approach offers notable improvements in reconstructing F10.7 over gaps of various lengths, with GAMs yielding mean error of ∼2.7%, compared to polynomial methods that yield mean errors of ∼3.1%. We additionally demonstrate the effect of reconstruction error on neutral densities modeled by the NRLMSISE2.0 thermosphere model.

The need for reliable estimates of solar extreme ultraviolet (EUV) for operational space weather nowcasting and forecasting has motivated the development of new model of solar irradiance variability in 59 wavelength bins between 1 and 1750 Angstroms. The model employs a nonlinear parameterization of rebinned daily FISM2 irradiances and is termed NEUVAC: Nonlinear Extreme Ultraviolet Irradiance Model for Aeronomic Calculations. The model is parameterized in terms of F10.7 and 54-day averaged F10.7 in a backwards-looking window. The model takes advantage of the high accuracy and resolution of FISM2, and the consistent availability of F10.7 data in order to provide the space weather community with an empirical irradiance model that requires only two inputs, and is applicable for solar EUV forecasting. Additionally, the model is capable of running ensembles and providing estimates of quantified uncertainty for usage by downstream ionosphere-thermosphere models. The model is evaluated against native FISM2 outputs in the 37 bins used by the EUVAC model, as well as the bins specified by Solomon and Qian. Its accuracy and precision is also compared to that of EUVAC, HEUVAC, and the HFG models.

A document by Daniel Brandt. Click on the document to view its contents.

Characterization of the thermospheric horizontal wind is an important challenge in atmospheric modeling, due to its vital role in the transport of densities and energy, associations with the diurnal tide, and interplay with vertical winds that drive changes in the thermosphere neutral composition. The mechanisms and drivers that underlie the physics of thermospheric horizontal winds remain under investigation and, to date, no comprehensive statistical study between thermospheric winds generated by a physics-based atmospheric model and those retrieved from satellite measurements has been performed. Comparisons between cross-track horizontal winds from a 10-month run of GITM and those derived from the Gravity field and steady-state Ocean Circulation Explorer (GOCE) satellite showed that GITM’s modeled horizontal winds best in the polar zone and overestimated them at midlatitudes in the equatorial ionization anomaly region. GITM’s wind response to AE was best at polar noon and worst in the midnight auroral zone, its ability to capture seasonality was best in the northern high latitudes and worst in the southern high latitudes, and GITM displayed less wind variability as a function of F10.7 than GOCE, matching it best for F10.7~150. Discrepancies in GITM’s performance may be explained by inaccurate modeling of ion drift, ion drag, and electron densities.