Analysis of Warm Season Low-Level Jets Along the Southern New England Coast Using IMPOWR Field Data

The IMPOWR (Improving the Mapping and Prediction of Offshore Wind Resources) field study was conducted in the southern New England coastal region during the spring and summer months of 2013–2014. The field program collected surface, tower, and aircraft observations over the coastal waters in order to help our understanding diurnal coastal flows and improve models for offshore wind power. Mesoscale models are known to have large biases in wind speeds and temperatures at lower levels over coastal waters.

Using data from surface buoys, National Weather Service radiosondes, and a Long-EZ aircraft, the structure of warm season low-level jets (LLJs) are analyzed as well as the performance of the various PBL schemes in the Weather Research Forecasting (WRF-ARW version 3.6.1). Three LLJ cases (16 May 2013, 21 June in 2013 and 12 May in 2014) and analyzed and simulated using WRF nested down to 1.33-km grid spacing (12- and 4-km outer domains). A small ensemble of initial and boundary conditions came from the Rapid Refresh (RAP), the Global Forecasting System (GFS), and the North American Model (NAM) analyses as well as the YSU, ACM2, and MYNN2 PBL schemes. A relatively large inner 1.33 km nest covering much of the Northeast U.S. was used, since the strength of the LLJ is sensitive to the nest size. The flight-level data was used to generate cross sections, soundings, and horizontal plots at a constant height which were then compared with the WRF

In all three cases the low-level jet (LLJ) is relatively shallow (centered around 150 m), with peak winds of 15-20 m s^-1. , The WRF is generally too weak (by ~3 m s^-1) and is too slow to develop the jet by a few hours. The WRF is also too shallow and cool with the marine boundary layer in all the PBLs. These errors are more significant further offshore, with more accurate simulations of these jets occurring nearer the coasts. As a result, there is little change in the PBL depth in the WRF offshore during the afternoon, which is reflected in a weaker coastal pressure gradient at the surface than observed. There is more variation using different initial conditions than different PBL physics, and there is relatively little LLJ sensitivity to using a different (higher resolution) sea surface temperature analysis. The LLJ is highly transient in all three events as reflected in the flight legs and time series of short-tower and buoys near the coast. The WRF also has some variations but less amplified. These small wind (and temperature) perturbations move rapidly along the coast in WRF, suggestive of weak gravity waves along marine layer, which may be produced given the shear instability on the top of the marine layer.