Our work explores the use of extended reality (XR) to improve scientific discovery with numerical weather/climate models that inform Earth science digital twins, specifically the NASA Goddard Earth Observing System (GEOS) global atmospheric model. The overall project is named the Vi-sualization And Lagrangian dynamics Immersive eXtended Reality Toolkit (VALIXR), which has two main areas of focus: (1) enhancing the understanding of and interaction with model output data through advanced visualizations in the XR environment, and (2) the integration of Lagrangian dynamics into the GEOS model, which allows a natural, feature-specific analysis of Earth science phenomena as opposed to traditional, fixed-point Eulerian dynamics. Here, we report initial work on these focus areas.

Our work explores the use of extended reality (XR) to improve scientific discovery with numerical weather/climate models that inform Earth science digital twins, specifically the NASA Goddard Earth Observing System (GEOS) global atmospheric model. The overall project is named the Vi-sualization And Lagrangian dynamics Immersive eXtended Reality Toolkit (VALIXR), which has two main areas of focus: (1) enhancing the understanding of and interaction with model output data through advanced visualizations in the XR environment, and (2) the integration of Lagrangian dynamics into the GEOS model, which allows a natural, feature-specific analysis of Earth science phenomena as opposed to traditional, fixed-point Eulerian dynamics. Here, we report initial work on these focus areas.

New radar remote sensing measurements of the turbulent hurricane boundary layer (HBL) are examined through analysis of airborne (Imaging Wind and Rain Airborne Profiler; IWRAP) and spaceborne (synthetic aperture radar; SAR) data from Hurricanes Dorian (2019) and Rita (2005). These two systems provide a wide range of storm intensities and intensity trends to examine the turbulent HBL. The central objective of the work is to document the characteristics of coherent turbulent structures (CTSs) found in the eyewall region of the HBL. Examination of the IWRAP data in Dorian shows that the peak, localized wind speeds are found inside the CTSs near the eye-eyewall interface. The peak winds are typically located at lower levels (0.15 - 0.50 km), but sometimes are found at higher levels (1.0 - 1.5 km) when the CTSs are stretched vertically. A SAR overpass of Dorian’s eyewall showed ocean surface backscatter perturbations at the eye-eyewall interface that have connections to the CTSs identified in IWRAP data. Wavelet analysis, including detailed significance testing, was performed on the IWRAP and SAR data to study the CTS wavelengths and power characteristics. Both datasets showed a multi-scale structure in the wavelet power spectrum with peaks at ~ 10 km (eyewall), ~ 4 - 5 km (merger of small-scale eddies) and ~ 2 km (native scale of the CTSs). The ~ 2 km native scale of the CTSs is robust across intensity trends (rapid intensification, weakening and steady-state), storm cases and region of the storm. This information is useful for turbulence parameterization schemes used in numerical models that require the specification of a turbulent length scale.

The computational fluid dynamics of hurricane rapid intensification (RI) is examined through idealized simulations using two codes: a community‐based, finite‐difference/split‐explicit model (WRF) and a spectral‐element/semi‐implicit model (NUMA). The focus of the analysis is on the effects of implicit numerical dissipation (IND) in the energetics of the vortex response to heating, which embodies the fundamental dynamics in the hurricane RI process. The heating considered here is derived from observations: four‐dimensional, fully nonlinear, latent heating/cooling rates calculated from airborne Doppler radar measurements collected in a hurricane undergoing RI. The results continue to show significant IND in WRF relative to NUMA with a reduction in various intensity metrics: (a) time‐integrated, mean kinetic energy values in WRF are ∼20% lower than NUMA and (b) peak, localized wind speeds in WRF are ∼12 m/s lower than NUMA. Values of the eddy diffusivity in WRF need to be reduced by ∼50% from those in NUMA to produce a similar intensity time series. Kinetic energy budgets demonstrate that the pressure contribution is the main factor in the model differences with WRF producing smaller energy input to the vortex by ∼23%, on average. The low‐order spatial discretization of the pressure gradient in WRF is implicated in the IND. In addition, the eddy transport term is found to have a largely positive impact on the vortex intensification with a mean contribution of ∼20%. Overall, these results have important implications for the research and operational forecasting communities that use WRF and WRF‐like numerical models.

Key Points: • The WRF dynamic core dissipates ∼ 20% more kinetic energy than NUMA for a dry vortex forced by four-dimensional latent heating observations. • Values of the eddy diffusivity in WRF need to be reduced by ∼ 50% from those in NUMA in order to produce a similar intensity time series. • Budgets and sensitivity tests indicate that the low-order approximation of the pressure gradient is the source of the dissipation in WRF.Abstract The computational fluid dynamics of hurricane rapid intensification (RI) is examined through idealized simulations using two codes: a community-based, finite-difference/split-explicit model (WRF) and a spectral-element/semi-implicit model (NUMA). The focus of the analysis is on the effects of implicit numerical dissipation (IND) in the energetics of the vortex response to heating, which embodies the fundamental dynamics in the hurricane RI process. The heating considered here is derived from observations: four-dimensional, fully nonlinear, latent heating/cooling rates calculated from airborne Doppler radar measurements collected in a hurricane undergoing RI. The results continue to show significant IND in WRF relative to NUMA with a reduction in various intensity metrics: (1) time-integrated, mean kinetic energy values in WRF are ∼20% lower than NUMA and (2) peak, localized wind speeds in WRF are ∼12m/s lower than NUMA. Values of the eddy diffusivity in WRF need to be reduced by ∼50% from those in NUMA to produce a similar intensity time series. Kinetic energy budgets demonstrate that the pressure contribution is the main factor in the model differences with WRF producing smaller energy input to the vortex by ∼23%, on average. The low-order spatial discretization of the pressure gradient in WRF is implicated in the IND. In addition, the eddy transport term is found to have a largely positive impact on the vortex intensification with a mean contribution of ∼20%. Overall, these results have important implications for the research and operational forecasting communities that use WRF and WRF-like numerical models. Plain Language Summary The intensity of a hurricane is primarily a balance between energy production and dissipation from various physical processes. Numerical models calculate this energy balance by solving complicated equations that attempt to capture these physical processes. Previous research has shown that the methods used to solve these equations can introduce additional dissipation into the system that affects the prediction of the storm intensity. In this paper, we examine this "numerical dissipation" idea more closely by conducting carefully designed comparisons between the community numerical model (WRF) and an advanced, research model (NUMA). Using observational estimates of heating in clouds, which feed the production of energy, we find that the WRF model produces significantly more numerical dissipation relative to NUMA that results in a reduced intensity of the storm. Our analysis indicates that the reason for the anomalous numerical dissipation in WRF is due to how the pressure gradient is computed. These results can have potentially important consequences for operational forecasts, especially the rapid intensification process. For example, the under-prediction or low bias of rapid intensi-fication forecasts may be partly due to excessive numerical dissipation.