Many previous observational studies of inertia-gravity waves (IGWs) have focused on identifying and describing the wave behavior and structure (e.g., Bosart and Seimon 1988; Koch and Golus 1988; Schneider 1990; Seimon et al. 1995). Recently, other studies have begun to investigate the synoptic-scale environments in which IGWs evolve. For example, Uccellini and Koch (1987) showed that for a small sample (13) of published case studies, IGWs occur primarily poleward of a surface frontal boundary beneath the inflection point between an upstream trough and downstream ridge. Uccellini and Koch (1987) also observe a jet streak moving through the trough into the downstream ridge and suggest that geostrophic adjustment may be an important IGW initiation mechanism. A more extensive climatology of large- amplitude IGW events (defined on the basis of hourly surface pressure changes > 4.5 hPa) has recently been compiled by Koppel (1995). This is the first climatology that identifies the spatial, seasonal and diurnal distribution of IGWs in the United States.

Knowledge of the relationship between the synoptic-scale environment and the life cycle and evolution of IGWs is critical to an increased understanding of wave genesis mechanisms, an important unresolved scientific issue, and to making short-term forecasts of the significant weather events often associated with wave passage. Therefore, the goal of this research is to identify the evolution and characteristic vertical structures of environments conducive to the formation and presence of large- amplitude IGWs. The vertical structure of IGW environments is examined by constructing seasonal and regional composites of soundings nearest to the IGW occurrences identified by Koppel (1995). Since our research is tied closely to Koppel (1995), a brief review of her methodology and results is presented in section 2. Section 3 will describe the data and the objective sounding composite methodology. A discussion of our results is presented in section 4.

2. IGW CLIMATOLOGY

Koppel (1995) identified large-amplitude IGW occurrences (hereafter, IGW will refer to large-amplitude IGWs) in the United States through subjective examination of hourly station pressure changes greater than 4.5 hPa from a network of 150-200 National Weather Service observing stations. The hourly surface reports were available from archives at the National Center for Atmospheric Research (NCAR) for two periods: 1949-1963 and 1984-1993. She further subjectively partitioned IGW occurrences into two categories: 1) IGWs associated with other meteorological events (i.e., convection or cyclone passage) and 2) distinct IGWs. Koppel (1995) then constructed maps of the regional, seasonal, and diurnal distribution and frequency of IGWs, an example of which appears in Fig. 1.

Her primary results indicate that IGWs occur almost exclusively east of the Rocky Mountains (see Fig. 1). The maximum IGW frequency occurs along a generally north-south axis extending from the upper Midwest into Arkansas and Oklahoma. Secondary axes extend eastward from the primary axis across the Great Lakes and the Southeast. IGW activity is virtually absent over the mountainous western United States and is uncommon over portions of the central Appalachians and Florida. This pattern is consistent with the relatively infrequent occurrence of low- level stable layers (wave ducts) associated with warm fronts in these regions.

The monthly frequency distribution (not shown) indicates a maximum in the late winter/early spring (Feb., Mar., Apr.) and a minimum in late summer/early fall (Aug., Sept., Oct.). The diurnal distribution (not shown) shows two peaks: one near 0300 LST and a second, higher maximum near 1200 LST. Koppel (1995) also showed composite 500 hPa and surface structures for IGW events that correspond closely to the synoptic signatures outlined by Uccellini and Koch (1987).

3. IGW SOUNDING COMPOSITES

3.1 Data and Methodology

Construction of regional and seasonal composite soundings associated with IGWs is accomplished using Koppel's (1995) results and the North America sounding data (1946-1992) archived on compact disc (CD). An objective methodology has been used to identify the individual soundings of the composite. Our composite is restricted to soundings corresponding only to those IGW events that Koppel (1995) categorized as distinct (see section 2). The three closest sounding stations in time and in space to each IGW occurrence are identified excluding those soundings stations located far to the southeast of the IGW (see Fig. 2). Exclusion of soundings in this sector is based on results from Uccellini and Koch (1987) and Koppel (1995) indicating that IGWs occur primarily poleward of warm fronts. Soundings in the excluded area are more likely to be located in the warm sector of a cyclone, equatorward of the warm front. We search the sounding CD data and include the closest sounding of the three for which data are available. Individual soundings corresponding to multiple IGW occurrences are only added to the composite once. After the appropriate sounding has been identified, the temperature, dewpoint, height, and u and v wind components are linearly interpolated with respect to the log of pressure to 10 hPa increments from 1000 to 100 hPa. Finally, because climatological temperatures in the troposphere vary seasonally and spatially, each sounding is identified with one of four regions (see Fig. 3): 1) Midwest, 2) Northeast, 3) Southern Plains and 4) Southeast. Each variable is composited by season and region at the interpolated levels.

Figure 2. Diagram showing the 10 x 10 degree area in which the three closest soundings to the IGW event are identified.

Figure 3 is currently unavailable for this WWW version. The Midwest and Northeast (Southern Plains and Southeast) regions are bounded on the south (north) by 36.5 N. The Midwest and Northeast regions are west and east of 83.0 W, respectively. The Southern Plains and Southeast regions are west and east of 89.0 W, respectively.

3.2 Results

For the 347 occurrences of distinct IGWs ('49-'63 and '84-'93) identified by Koppel (1995), a total of 299 soundings are available for the composite. Several features common to the composite soundings in all regions for the winter and spring can be seen in the winter Midwest (Fig. 4a) and spring Northeast (Fig. 4b) composites. Most prominently, a low-level inversion extends from the surface to at least 850 hPa (Fig. 4b) and sometimes to above 700 hPa (Fig 4a). The winds in this inversion veer from easterly or southeasterly tosouthwesterly at the top of the inversion, indicative of warm advection. In the layer between the surface inversion and the tropopause, the lapse rate is significantly steeper and in some cases approaches moist neutrality. Winds in this layer increase in magnitude to a maximum at jet level of 35-50 m s-1 with little directional shear.

The summer and fall composites (not shown) generally have a similar wind structure (i.e., a low-level veering wind capped by a layer with primarily speed shear). However, the low-level temperature inversion is either shallower (i.e., only up to 900 hPa) or absent altogether. Some of the summer and fall composite soundings possess positive convective available potential energy (CAPE) values. This observation suggests that perhaps IGWs in the summer and fall identified as distinct by Koppel (1995) may be embedded in a convective environment.

Figure 4. Skew-T Log-P diagram of IGW composite soundings. Temperture (C, solid), dewpoint (C, dashed), wind barbs (kts): (a) winter, Midwest; (b) spring, Northeast.

4. DISCUSSION

The overall structure of the winter and spring composites is consistent with the Lindzen and Tung (1976) theoretical model of IGW maintenance. The depth of the low-level inversion is similar to those reported by Ralph et al. (1993) and Ramamurthy et al. (1993). The wind speeds in the upper layer of 35-50 m s-1 suggest that all but the fastest propagating waves would have a critical level. Although the unidirectional wind shear observed in the composites is not specified by Lindzen and Tung (1976), a similar shear profile in the upper layer also has been observed by Ralph et al. (1993).

These composite soundings also conform to the synoptic pattern identified by Uccellini and Koch (1987). The southwesterly winds in the upper troposphere, together with the relatively high tropopause (see Fig. 4), indicate that IGWs occur in the southwesterly flow downstream (upstream) of a synoptic-scale trough (ridge). The location or movement of jet streaks can not be determined from sounding composites alone. We plan in the near future to composite European Centre for Medium-Range Weather Forecasts (ECMWF) gridded analyses for the 1985-1993 IGW events in order to address this issue and to develop a fully three- dimensional composite picture of the IGW environment.

Another important unresolved issue with forecasting implications is the difference between the composite IGW soundings and composite soundings poleward of a warm front in the absence of IGWs. Clearly, the low-level inversion with veering winds is common to both. Short-term forecasting of IGWs will depend crucially upon the ability to differentiate between an IGW environment and a "typical" warm front. Therefore, we need to identify what, if any, are these differences. Furthermore, if differences exist, are they significant? If not, what factors make the IGW environment unique? In order to address these questions, we also will be developing a composite sounding for "typical" warm fronts in the near future.

Acknowledgments. This research has been supported by the United States Air Force Office of Scientific Research through Grants F496209310002 and F496209510492.

Bosart, L. F., and A. Seimon, 1988: A case study of an unusually intense atmospheric gravity wave. Mon. Wea. Rev., 116, 1857-1886.

Koch, S. E., and R. E. Golus, 1988: A mesoscale gravity wave event observed during CCOPE. Part I: Multiscale statistical analysis of wave characteristics. Mon. Wea. Rev., 116, 2527-2544.

Koppel, L. L., 1995: Climatology of large-amplitude inertia- gravity waves in the conterminous United States. Masters Thesis. University at Albany, State University of New York, 191 pp.

Lindzen, R. S., and K. K. Tung, 1976: Banded convective activity and ducted gravity waves. Mon. Wea. Rev., 104, 1602-1617.

Ralph, F. M., M. Crochet, and S. V. Venkateswaran, 1993: Observations of a mesoscale ducted gravity wave. J. Atmos. Sci., 50, 3277-3291.

Ramamurthy, M. K., R. M. Rauber, B. P. Collins, and N. K. Malhotra, 1993: A comparative study of large-amplitude gravity-wave events. Mon. Wea. Rev., 121, 2951-2974.

Schneider, R. S., 1990: Large-amplitude mesoscale wave disturbances within the intense Midwest extratropical cyclone of 15 December 1987. Wea. Forecasting, 5, 533-558.

Seimon, A., L. F. Bosart, W. E. Bracken, and W. Snyder, 1995: Large-amplitude inertia-gravity waves. Part II: Structure of an extreme gravity wave event over New England on 4 January 1994 revealed by WSR-88D radar and mesoanalysis. Preprints, 14th Conference on Weather Forecasting and Analysis, Dallas, TX, Amer. Meteor. Soc., 434-437.

Uccellini, L. W., and S. E. Koch, 1987: The synoptic setting and possible energy sources for mesoscale wave disturbances. Mon. Wea. Rev., 115, 721-729.