Establishing a Data Assimilation Baseline for Numerical Weather Prediction in Large, Metropolitan Areas in an Arid Region

The Southwest Urban Corridor Integrated Field Laboratory (SW-IFL) is one of four Urban Integrated Field Laboratories sponsored by the US Department of Energy Office of Science. SW-IFL is enabling novel observational and modelling capabilities that improve the understanding of extreme heat as a central driver of key environmental stressors in the Arizona Sun Corridor, including availability and transport of urban air pollutants. A key component of the modelling effort is the notion of a Model of Models (MoM) approach to enable a hierarchical coupling of environmental simulations from the scale of the global atmosphere to a building and street scale in highly urbanized cores.

A key component of MoM is the use of the community Model for Prediction Across Scales Atmosphere (MPAS-A). MPAS-A uses an unstructured computational mesh to enable a smooth and consistent transition from global through regional to local scales. Its goal for MoM is to improve the fidelity of the forcing data (i.e., background fields and lateral boundaries) for, and eliminate telescoping outer nests and compensating for their artifacts with the use of community Weather Research and Forecast (WRF) model. A further goal is to reduce the computational and storage costs associated with the creation of the forcing data compared to traditional dynamical downscaling from a global model (forecast or reanalysis data). As part of the MoM effort, data assimilation of available observations at different scales will be used with both MPAS-A and WRF in this coupled framework to reduce the errors in the initial conditions for both modelling systems.

While all of the components of MoM will be properly validated, including leveraging the unique observations being collected for SW-IFL, understanding the sensitivity of each component to the assimilation of local observations has value to determine the effectiveness of this coupled framework. To that end, we apply a customized version of WRF configured for 1km resolution focused on the Sun Corridor, which can operate in forecast/hindcast mode, to establish a baseline for model fidelity. It includes the three major metropolitan areas in Arizona (Phoenix, Tucson and Flagstaff). Open conventional observations from NOAA Meteorological Assimilation Data Ingest System (MADIS) are used for data assimilation. The model has been applied to limited summer periods in recent years, since model skill in representing heat is a clear priority.

We will present an overview of the approach to the modelling as well as an assessment of the model results. Since this is work in progress, we will outline the next step in this research and some of our current applications.