Making Sense of Multiple Boundary Layer Meteorological Observations during the Jack Rabbit II Chlorine Field Experiment

When can you have too many meteorological observations? One answer is: when you need to choose optimum meteorological inputs for an operational dispersion model, and the scenario is a comprehensive field experiment with over 100 near-surface meteorological sensors and several vertical profilers on a 10 to 20 km domain. The 2015 and 2016 Jack Rabbit II field experiment at Dugway Proving Ground (DPG), Utah, involved short duration releases of large quantities of pressurized liquefied chlorine, and made use of extensive sets of surface meteorological towers, spread over the 10 to 20 km source and sampling domain. The near surface winds were observed by aerovanes and sonic anemometers. Temperature and relative humidity were observed, as well as components of the surface energy balance. 2 m towers were used over most of the wind network domain, and three 32 m towers, with five levels of observation, were used near the source and at downwind distances of 1 and 2 km. Winds and temperatures in the boundary layer above the surface towers were observed by a minisodar, radiosondes, and 924 and 449 MHz profilers.

Most operational dispersion models being applied to the JR II chorine releases require, as a minimum, inputs of wind speed and direction, temperature, and stability class near the source location. The surface roughness length, z_o, is also needed. Some models also desire to have inputs of friction velocity u*, Monin-Obukhov length L, convective velocity scale w*, and perhaps turbulence components σ_u, σ_v, and σ_w. Some models allow input of vertical profiles of these variables. A few models that make use of a diagnostic mass-consistent wind model can use the wind fields from the mesoscale networks.

We present the rationale for selecting JR II meteorological observations from a subgroup of wind, turbulence, and energy flux instruments near the source, and recommending meteorological inputs, separated into surface and upper air. Several difficulties are discussed. One is related to the fact that all releases were in early morning, during a period of boundary layer transition from stable to unstable. This time variability also affects the cloud as it traverses to 11 km across the sampling network. Another difficulty is related to the fact that the chlorine release took place in an environment that could be classified as a “drainage wind”, with large wind direction shears above 100 or 200 m. Another problem is caused by the mesoscale meandering or sloshing that is taking place across the sampling domain.