Evaluating the Impact of Non-Conventional Observations on High-Resolution Analyses and Forecasts Using Observing System Experiments

A key recommendation of a report published in 2009 by the National Research Council (NRC) was to create a “network of networks” by integrating existing mesoscale networks with newly created ones.  This recommendation originated in response to deficiencies in U.S. mesoscale observations, particularly the lack of high spatiotemporal resolution moisture, temperature, and wind observations within the lower troposphere.  The report further recommended that research testbeds be used to assess the potential forecast impact of such a network.  One such testbed has been established in north Texas, namely the Center for Collaborative Adaptive Sensing of the Atmosphere (CASA) Dallas-Fort Worth Urban Testbed.

The purpose of this research is to evaluate the impact of non-conventional observations in the aforementioned research testbed on high-resolution analyses and forecasts of convection.  Some examples of non-conventional observations that are being evaluated include Mobile Platform Environmental Data (MoPED) from Global Science and Technology (GST), surface observations from Understory Weather, and several CASA X-band radars covering the testbed.  The Advanced Regional Prediction System (ARPS) model is used to perform observing systems experiments that are designed to assess the impact of non-conventional observing systems, with a particular focus on surface observations.  The ARPS three-dimensional variational (3DVAR) analysis and its associated cloud and hydrometeor analysis are used to produce analysis increments every ten minutes, which are then applied to the model forecast using incremental analysis updating (IAU) during a 30-minute assimilation window.  Experiments are performed using high-impact severe hail events from the spring of 2016, including a quasi-linear convective system (QLCS) on 23-24 March 2016 and a supercell on 11 April 2016.  Quantitative forecast verification metrics, including root-mean-square error (RMSE) and threat scores for precipitation, are used to assess how forecasts of meteorological fields such as wind, temperature, and moisture compare to observations.