ESRL Global Systems Division
Doppler Lidar Observations Improve Regional Forecasts
Working with other research institutions, FSL has completed a series of regional observing system simulation experiments (OSSEs) designed to test the potential impact of a space-based Doppler lidar wind profiling system on regional-scale numerical weather predictions using the Rapid Update Cycle (RUC) assimilation and forecast system. A numerical prediction model is used to mimic a typical atmospheric evolution in OSSEs, which present a cost-effective method for evaluating the forecast improvement from potential new observing systems. The regional experiments were run in coordination with a series of global OSSE runs conducted by the National Centers for Environmental Prediction (NCEP). Lateral boundary conditions used in the regional OSSEs were obtained from matched global lidar OSSEs. The simulated observations were obtained from the MM5 (version 5 of the Penn State/NCAR Mesoscale Model) regional nature run, completed by FSL's Local Analysis and Prediction Branch. Conventional observations (rawinsondes, profilers, velocity azimuth displays, automated aircraft reports, meteorological aviation reports, and buoys) were produced directly from FSL's regional nature run, and the Environmental Technology Laboratory (ETL) created the lidar observations from MM5 fields. Calibration of the MM5 regional nature run against the European Center global nature run, and coding of the regional verification software was completed by the National Center for Atmospheric Research (NCAR).
As expected, the regional forecast improvements shown in this study can be attributed to two sources: direct assimilation of the lidar observations on the regional domain and improved lateral boundary conditions resulting from the assimilation of lidar observations within the global model. Improvement from the direct assimilation of lidar observations is greatest at the analysis time, and then slowly diminishes with forecast projection, while improvements from the lateral boundary conditions increase with forecast projection. Initial OSSE runs focused on a best-case scenario, in which no lidar observations were lost due to cloud attenuation and no errors were added to the lidar observations. These experiments show that lidar wind observations produce modest short-range (0 to 12-h) forecast improvements. Upper-level wind forecasts benefit the most, with 6-h forecast errors reductions of up to 10% resulting from the assimilation of lidar observations. Other experiments, which have accounted for the loss of lidar observations due to clouds and the existence of small, random errors in the lidar observations, show only a slight reduction in the forecast improvement compared to the best-case experiments. As part of this project, the formulation of the RUC three-dimensional variational (3DVAR) analysis scheme was modified to facilitate the direct assimilation of lidar line-of-sight wind observations. This modification also allows for the direct assimilation of Doppler radar radial velocity observations, in accordance with planned RUC analysis enhancements.
Presentations of these results are planned in the near future.
Name: Steve Weygandt