SUMMARYThe correlation between dimensionless shear φ m and dimensionless height z/L, where L is the Obukhov length, for stable conditions is strongly influenced by self-correlation for the present datasets. This effect is quite large for stronger stability but still significant for near-neutral conditions. A conditional analysis of nocturnal stable boundary-layer data, where 'non-turbulent' parts of the record are removed, reduces the impact of nonstationarity and therefore reduces the scatter. The conditional analysis also reduces the relative importance of self-correlation. Difficulties with estimating self-correlation are also discussed.

Boundary layer wind data observed by a Doppler lidar and sonic anemometers during the mornings of three intensive observational periods (IOP2, IOP3, and IOP7) of the Joint Urban 2003 (JU2003) field experiment are analyzed to extract the mean and turbulent characteristics of airflow over Oklahoma City, Oklahoma. A strong nocturnal low-level jet (LLJ) dominated the flow in the boundary layer over the measurement domain from midnight to the morning hours. Lidar scans through the LLJ taken after sunrise indicate that the LLJ elevation shows a gradual increase of 25-100 m over the urban area relative to that over the upstream suburban area. The mean wind speed beneath the jet over the urban area is about 10%-15% slower than that over the suburban area. Sonic anemometer observations combined with Doppler lidar observations in the urban and suburban areas are also analyzed to investigate the boundary layer turbulence production in the LLJ-dominated atmospheric boundary layer. The turbulence kinetic energy was higher over the urban domain mainly because of the shear production of building surfaces and building wakes. Direct transport of turbulent momentum flux from the LLJ to the urban street level was very small because of the relatively high elevation of the jet. However, since the LLJ dominated the mean wind in the boundary layer, the turbulence kinetic energy in the urban domain is correlated directly with the LLJ maximum speed and inversely with its height. The results indicate that the jet Richardson number is a reasonably good indicator for turbulent kinetic energy over the urban domain in the LLJ-dominated atmospheric boundary layer.

Length scales determined by maximum turbulent kinetic energy (TKE), the integral scale, and two length scales based on Reynolds stress-tensor anisotropy are compared to the often stated outer length scales of boundary-layer depth and distance from the earth's surface, z. The scales are calculated using sonic anemometer data from two elevations, 5 and 50 m above the ground at the main tower site of the CASES-99 field campaign. In general, none of these scales agrees with the other, although the scale of maximum TKE is often similar to the boundary-layer depth during daytime hours, and the length scales derived from anisotropy characteristics are sometimes similar to κz, z, and 2z depending on scale definition and thermal stability. Except for the scale with the strictest isotropy threshold, the turbulence is anisotropic for each of the various candidates for the outer scale. Length scales for maximum buoyancy flux and temperature variance are evaluated and the turbulence characteristics at these scales are almost always found to be anisotropic.

To eliminate the need to correct for instrument tilt, a process that can be problematic in complex terrain, a new way to calculate the turbulent friction velocity is derived based on invariants of the Reynolds stress tensor. In utilizing Reynolds stress tensor invariants, this new method eliminates the need for tilt correction. The friction velocity is calculated without any reference to the wall normal or other terrain features making this method a candidate for future use with data from complex environments. Since this new method is derived from a different theoretical basis than the well-established methods, it is evaluated using data from flat terrain to compare the new method to the standard calculation method, treated here as a baseline truth. For neutral thermal stratification the values calculated using the new method nearly identically match the control values calculated using the standard method. Although for nonneutral stratification the values calculated using the new method do not closely match the values calculated using the standard method, the new friction velocity produces the same dimensionless shear versus dimensionless height Monin–Obukhov scaling relationship over the full range of stabilities as does the standard friction velocity.

A variety of atmospheric boundary layer parameters are examined as a function of wind direction in both urban and suburban settings in Oklahoma City, Oklahoma, derived from measurements during the Joint Urban 2003 field campaign. Heterogeneous surface characteristics result in significant differences in upwind fetch and, therefore, statistically significant differences in measured values, even for small changes in wind direction. Taller upwind obstructions yield larger measured values of drag coefficient and turbulence intensity than do shorter upwind obstructions regardless of whether the obstruction is a building or a tree. The fraction of turbulent kinetic energy going into streamwise, cross-stream, and vertical variances differs depending on the upwind fetch, and reduced cross-stream values may indicate locations of persistent wind stream convergence. In addition, a quadrant analysis of burst/sweep behavior near the surface is examined as a function of wind direction in urban and suburban environments.