VLES and LES Forecasting Experiments for US Army Aberdeen Proving Ground Using NCAR RTFDDA-LES

The NCAR-ATEC (National Center for Atmospheric Research and Army test and Evaluation Command) RTFDDA-LES (real-time four-dimensional data assimilation and LES simulation) model was employed to study multi-scale flow circulations at US Army Aberdeen Proving Ground (APG), MD. Military test activities strongly rely on the fine scale flows and other weather variables over the test range which extends over a 10x15 km2 peninsula and surrounding areas in the upstream of Chesapeake Bay. The area is featured with very complex coastlines, hills and heavily man-modified landscapes. In this paper, the LES version of the NCAR real-time four-dimensional data assimilation and forecasting system (RTFDDA-LES) was employed to study the multiple-scale flow interaction of synoptic, local land-surface forcing, and thermally driven land-sea flows. Six nested-grid domains, with grid sizes of 8100, 2700, 900, 300, 100, and 33m, respectively, were configured and a 48h simulation was carried out simultaneously on the six nested grids for a two-day period during summer 2013. The data assimilation of RTFDDA was turned on for the mesoscale domains (1, 2 and 3), while the LES domains (4, 5 and 6) were run with “free forecasting”. The mesoscale data assimilation on the coarse meshes provide realistic mesoscale forcing for the LES simulation, so that the model outputs of the LES domains can be reasonably verified using high-resolution (every 1 – 5 minutes) measurements of APG surface weather stations and the remote-sensing systems. The model results are analyzed to evaluate the feature, value, and feasibility of LES-scale precision weather forecasts for applications over complex coastal regions. Meanwhile, by inter-comparing the mesoscale, VLES and LES scale model output, we also assess validity and values of running VLES (very-large-eddy simulation; DX=100s meters) NWP for real-time forecasting with the currently available computing capability.