Improving Accuracy of a Microscale Fast-Response Wind Modeling Platform (QES-Winds) By Downscaling Wind Data from a Mesoscale Numerical Weather Prediction Model (HRRR)

Quick Environmental Simulation (QES) is a microclimate simulation platform for computing 3D environmental scalars in urban areas and over complex terrain. Several applications, such as modeling wildfires, pollution propagation and dispersion in cities, transport of pathogens through vineyards, and effects of wildfire smoke on grape quality require high-resolution wind field modeling faster than real-time. QES-Winds, a fast-response 3D diagnostic wind model written in C++, solves a mass-conservation equation for the wind field rather than slower yet more physics-based solvers that include conservation of momentum. An initial wind field is specified in QES-Winds utilizing empirical parameterizations. QES-Winds uses a variational analysis technique to minimize the differences between the initial guess field and the final wind field. Using data from mesoscale numerical weather prediction models like Weather Research and Forecasting (WRF) or High-Resolution Rapid Refresh (HRRR) to create the initial wind field, can improve accuracy of the final velocity field.

Each hour NOAA conducts HRRR model runs that have an 18-hour prediction horizon and daily they produce 48-hour future forecasts. Importantly, model products from HRRR and HRRR-Smoke are archived by NOAA and provided for public use through readily available cloud-based services (e.g., Amazon Web Services, AWS). The HRRR model has 3km grid resolution over the entire continental U.S. providing excellent coverage for weather-related applications. HRRR near surface wind output is used to drive QES-Winds simulations. QES-Winds downscales the 3km HRRR wind field to 100m resolution. The coarse QES grid resolves topographic and land use changes associated with vineyard blocks and the immediate surrounding area that are subgrid to HRRR. Comparison between QES-Winds’ results with 100m horizontal resolution and the measured data showed better performance than HRRR’s model with 3km horizontal grid size. Therefore, QES-Winds simulations driven by HRRR wind data is more capable of providing high-resolution wind fields for modeling wildfires, pollution propagation and dispersion in cities, transport of pathogens through vineyards, and effects of wildfire, agricultural burn, and smoke on grape quality. In the next step, HRRR near-surface smoke field will be downscaled to QES in order to develop a fast-response tool to simulate smoke exposure on grapes.