The role of inertial oscillations in the generation of nocturnal LLJs during CASES99

In October 1999, the Cooperative Atmosphere Surface Exchange Study (CASES99) field program took place in the Walnut River Watershed of southeastern Kansas. Numerous in-situ and remote-sensing instruments were deployed to probe the evolution of the nocturnal stable boundary layer. One of the stated goals of CASES99 was to assess the role of the evening transition period in the initiation of inertial oscillations and the enhancement of nocturnal low-level jets.

As turbulent stresses in the atmospheric boundary layer decay near sunset, an elevated residual layer decouples from a relatively shallow, very stable surface layer that contains the principal turbulent stresses. According to Blackadar's 1957 theory, winds aloft, in the residual layer, accelerate in the form of ageostrophic inertial oscillation (IOs), inertia-gravity waves with a frequency equal to the Coriolis parameter. Theoretically, the IOs generated by this decoupling then enhance winds to create the nocturnal low-level jet profile. Though some observational and modeling case studies (Zhong et at 1996, among others) have confirmed the importance of inertial oscillations to the development of the low-level jet, other studies have shown that IOs are present within a nearly neutral layer near the surface (Ostdiek and Blumen 1997), and that the frictional decoupling described by Blackadar is not required for their development. By combining data from the CASES99 instrumental array and numerical models, this study seeks to resolve the role of developing stability in the generation of IOs, and in turn, their role in the development of the LLJ.

Data from radar boundary-layer wind profilers, sodars, and the 55-m CASES99 anchor tower are examined for inertial oscillations occurring on nights with strong LLJs. Data from three boundary-layer profilers show that out of 29 operational nights, 20 nights exhibit a jet profile developing in the lowest 1 km after sunset and before 600 LST. The same data reveal only one night with inertial oscillations in the time series of winds from all three profilers. Our discussion of the results of our analysis will focus on the timescale for the decoupling of the residual layer, the time- and height-dependence of the development of IOs, and the resulting evolution of the LLJ profile.

The CASES99 dataset is supplemented with numerical models to explore the initiation of IOs. We initialize a one-dimensional K-closure model (McNider et al 1988) with CASES99 data to simulate the evolution of IOs and the LLJ. A case study is also performed with the Kosovic and Curry (1997) large-eddy-simulation model; we discuss that model's ability to capture the details of the decay of convective turbulence.