The Severe Clear Air Turbulence Colliding with Aircraft Traffic (SCATCAT) experiment was conducted as part of the 2001 Winter Storms Reconnaissance Program to test the performance of Rapid Update Cycle (RUC) model predictors of turbulence and to improve understanding of turbulence generation mechanisms. We have analyzed in-flight and dropsonde data that were collected by the NOAA Gulfstream-IV (G-IV) aircraft in a region extending from the core of an intense upper-level jet to its left exit region on 17-18 February 2001. The mission consisted of making in-flight observations at several altitudes and launching dropsondes at ~40-km intervals from the 12.5 km level. A stack of four constant-altitude legs was made nearly perpendicular to the jet streak at altitudes of 12.5, 11.3, 10.7, and 10.1 km. In addition, the RUC model was run on a 20-km grid with 50 hybrid isentropic-sigma levels in a manner closely mirroring the operational RUC model, except that the domain was shifted to the data-sparse central north Pacific and boundary conditions were supplied by the AVN model instead of the Eta model.

Cross-section analyses computed from the dropsonde data showed vertically propagating mesoscale gravity waves in the region of strong vertical wind shear extending from the jet core (and its cyclonic side) into the lower stratosphere. Coherent streaks of moderate or greater turbulence were indicated here as well by a diagnostic turbulent kinetic energy (TKE) parameter applied to the dropsonde data. The G-IV did, in fact, encounter turbulence in this same region on the three lower legs of the stack. Diagnosed TKE in this same region appeared in the RUC simulation fields as mesoscale bands oriented parallel to the flow. Detailed comparisons made between flight-level observations and the RUC fields showed that the general patterns seen in the aircraft data were well represented by RUC, but that variations occurring at scales smaller than ~150 km (e.g., gravity waves) were absent in the model.

Rapid fluctuations in ozone measured by the G-IV correlated nicely with fluctuations in the potential temperature data at the 10.1 km altitude, as the aircraft penetrated a pronounced gravity wave within an upper-tropospheric frontal zone. The ozone variations also correlated well in a more general sense to potential vorticity variations in the RUC model. Spectral analyses of the aircraft 25-Hz data revealed that turbulence occurred in conjunction with this gravity wave. Phase spectrum analyses showed that potential temperature and the jet-normal wind component exhibited a strong in-phase relationship in the frequency range of 0.021 – 0.049 Hz, as did the potential temperature and ozone data – suggesting the presence of either deep propagating gravity waves or decaying (evanescent) waves. Synthesis of these analyses suggests the following hypothesis: turbulence related to energetic fluctuations in the inertial subrange (horizontal wavelength < 400 m) occurred within a packet of gravity waves (with wavelengths of 6 – 80 km) shed within an upper-level front on the cyclonic shear side of the jet core. The challenge now is to attempt to model and understand these features using a cloud-resolving model.