Observing System Experiments in the Dallas-Fort Worth Testbed

In 2009, the National Research Council (NRC) published a report that described deficiencies in U.S. mesoscale observations, especially the lack of high spatial and temporal resolution observations of moisture, temperature, and winds in the lower troposphere. The report recommended that existing mesoscale networks be integrated together with new networks to develop a “network of networks”. Testbeds were also recommended as a means of testing the forecast impact of the “network of networks”. One such testbed was established by the Center for Collaborative Adaptive Sensing of the Atmosphere (CASA) in the Dallas-Fort Worth metroplex. One purpose of this project is to assess the impact of assimilating non-conventional weather observations on high-resolution model analyses and forecasts using two separate approaches. Two new data sources tested in this presentation include Mobile Platform Environmental Data (MoPED) from Global Science and Technology (GST) and CASA X-band radars.

The first approach uses the Advanced Regional Prediction System (ARPS). Incremental analysis updating (IAU) is used to introduce the analysis increments produced every five minutes by the ARPS three-dimensional variational (3DVAR) analysis and associated cloud and hydrometeor analysis. Data denial experiments are conducted to evaluate the impact of each non-conventional dataset on the accuracy of high-resolution analyses and forecasts of convective weather. The second approach is to conduct additional data denial experiments examining the impact of the different data sources using a GSI (Global Statistical Interpolation)-based EnKF (Ensemble Kalman Filter) data assimilation system coupled with the WRF-ARW (Advanced Research Weather Research and Forecasting) model. The structure, intensity, and timing of the meteorological features of interest seen in the model fields are compared with independent observations to determine the accuracy of individual forecasts. The effects of the increased observation density from these additional datasets on the high resolution analyses and forecasts produced by this system are then compared with that of the ARPS/3DVAR system. Using case studies from spring, 2015, the impact of the non-conventional observations on both specific severe weather events and over a month-long period are determined.