From the evening of November 14, 1997 through most of the day on November 15, 1997, heavy snow blanketed extreme northeast Illinois. More than 12 inches of snow fell in some locations across Lake County, Illinois (Figure 1). A snow event of 2 to 4 inches was forecast for most of north central and northeast Illinois. This verified well with the only exception being the heavy snow in Lake County. The heavy snow was apparently induced by the development of a meso-low circulation which provided added vertical motion/convergence as it moved westward (inland) across Lake County from Lake Michigan. The meso-low then retreated back over the lake as it was being engulfed by a synoptic cyclone approaching from the southwest. The merger of the two circulations strengthened a northeast wind flow, allowing moderate to heavy snow to persist through the morning hours of November 15.
Figure 1. Total snowfall across north central and northeast Illinois from November 14-15, 1997.
Due to their small scale, meso-low circulations are rarely able to be observed in detail. However, this particular meso-low assumed a path through a local network of wind instrumentation supplied by Commonwealth Edison. The purpose of this paper is to analyze this rare event using the local wind instrumentation, in conjunction with Doppler Radar imagery and the Aeronautical Radio Incorporated Communication and Reporting System (ACARS) reports.
On the evening of November 14, a strong, closed upper low was moving eastward through south central Iowa, while a weak surface low was evident in the Davenport, Iowa area around midnight. Ten inches of snow had fallen at the National Weather Service Office in Davenport by this time.
A warm front extended eastward into northeastern Illinois from the surface low. Temperatures in the lower 30s and dew points around 30 degrees were south of the warm front, and temperatures around 30 with dew points in the upper 20s were north and west of the trough, reflecting a weak baroclinic zone.
Late in the evening of November 14, an area of convergence was evident along the western shore of Lake Michigan, near the Illinois/Wisconsin border. It was associated with the aforementioned warm front stretching into northeastern Illinois. Surface winds were northeast into the Illinois/Wisconsin border region but veered southeast further down the lake shore. WSR-88D radar imagery from National Weather Service Forecast Office at Chicago, Illinois (LOT) indicated a nearly stationary cyclonically curved echo band extending westward from Lake Michigan into Lake County. Lake Michigan water temperatures were in the middle 40s, while overlying air was in the lower 30s.
While the synoptic situation evolved as anticipated, convergent bands of lake-effect precipitation which had begun during the afternoon of November 14, became more widespread by midnight. A cooperative observer from Lake Villa in Lake County reported 1.8 inches of snowfall through midnight.
At this time, rotation in the Doppler reflectivity was not evident, but an argument for the developing and westward (inland) moving circulation can be made by looking at the10-meter wind data (Figure 2). The 10-meter winds are observed at 15 minute intervals from a series of Commonwealth Edison towers which are located 2, 5 and 15 miles inland from Lake Michigan (Figure 3). At 0700 UTC ( 0100 CST), the wind direction at the main tower shifted to the southeast, with an east wind 5 miles inland and a northeast wind at the 15-mile tower. These winds suggest a possible inverted trough just east of the 5-mile tower, from which a reasonable leap can be made toward the development of a "county-scale" circulation.
Figure 2. A 24-hour Lake County 3-tower 10-meter wind plot from midnight (0600 UTC) 15 Nov 97 to midnight (0600 UTC) 16 Nov 97.
Figure 3. Conceptual model of meso-low. Locations of towers providing 10-meter winds are shown.
Another element of the network of wind data is the Doppler Sound Detection and Ranging (SODAR), which provides a real-time, "profiler-like", hourly display of the wind direction (Figure 4). This field is displayed to 400-600 meters AGL (~1500+ ft). The SODAR site is located just south of the main tower and the 10-meter wind towers (Figure 3). Winds were east-northeast at the SODAR tower prior to midnight, but then shifted quickly to south from (midnight until 3:00 a.m.) 0600-0900 UTC (CST).
Figure 4. Plot of 10-meter winds from Lake County 3-tower network from 0600 UTC (midnight) 15 Nov 97 until 0600 UTC (midnight).
At 0900 UTC, the 0.5 degree Doppler reflectivity loop from LOT did show signs of rotation across far northeastern Illinois, with an apparent center of circulation in the middle of Lake County, IL. This center of circulation is denoted by the "L" in the 0910 UTC Doppler image (Figure 5). By this time, the circulation had already reached its furthest point inland, and was now retreating back toward Lake Michigan. This can be seen by looking at both the 10-meter and SODAR wind data again. The wind veered to the southeast up until 0845 UTC at the 15-mile tower as the meso-low was propagating inland. After 0845 UTC, the winds began to back to the northeast again. Similar backing signatures are represented at the other tower locations, as well as at the SODAR, suggesting a retreat south from each of those locations. The meso-low likely remained north of O'Hare International Airport as ASOS surface winds remained south-southeast through the night. The southerly wind can also be seen by looking at an aircraft descent sounding (ACARS) into O'Hare Airport (Figure 6), which also shows that the circulation may have extended up to 770mb (~7500 feet).
Figure 5. KLOT-WSR-88D reflectivity image (0.5 degree elevation) at 0910 UTC 15 Nov 97. Approximate location of surface meso-low is indicated by an "L".
Figure 6. 0633 UTC 15 Nov 97 aircraft descent sounding (ACARS), 26km from O'Hare International Airport.
At 1200 UTC, the observer in Lake Villa reported that 4.1 more inches of snow had fallen since midnight. At this time the winds at the SODAR became east-northeast, suggesting that the circulation may have been nearing the lake. At the same time, the synoptic circulation was moving through the southwestern suburbs of Chicago, nearing the southern tip of Lake Michigan. The larger scale flow began to engulf the smaller scale rotation. A reflection of this "merger" may be the strengthening of the northeast winds after 1200 UTC. The stronger winds off the middle 40 degree water of Lake Michigan enhanced the convergent region in Lake County and helped to destabilize the "local atmosphere". This allowed moderate to heavy snow to persist over Lake County through the morning of November 15. By 1800 UTC, Lake Villa accumulated another 6.4 inches of new snow. Storm total snow accumulations ranged from 6 to over 12 inches of snow across Lake County, with Lake Villa receiving 12.5 total inches (Figure 1).
Lake Michigan can be a major influence on local-scale weather across northeastern Illinois. In this particular case, what began as individual lake-effect snow showers in the afternoon of November 14, turned into a more intense and widespread snow after midnight over Lake County. This heavy snow event was apparently influenced by the formation of a meso-low circulation. The reason for the development of the meso-low is certainly not "concrete", however it is set against the backdrop of the synoptic convergent regime in the vicinity of the aforementioned warm front. Enriched moist air from Lake Michigan, in itself, supported higher precipitation rates in Lake and McHenry counties. The ACARS sounding also reveals a relatively more stable thermal zone in the 800mb-700mb layer. This may have acted as a weak lid to the boundary layer processes occurring beneath it, such as vorticity generation (cyclonic convergent flow) and enhanced thermal forcing due to latent heat release. These boundary layer processes may have been sufficient enough to produce the observed meso-scale circulation.
Although forecast model data was not discussed here, it should be noted that the Meso-Eta 29 (Black, 1994) forecast did an excellent job at predicting a small-scale cyclonic circulation to develop and move inland near Lake County, IL. This mesoscale model also suggested a strengthening frontogenetic zone over Lake County, which indicated where forcing would be greatest. The resulting total snowfall map lines up well this forecast. It should also be noted that the traditional ETA, NGM and AVN-models performed well with the synoptic system, and the resulting forecast of 2 to 4 inches was accurate for most of northern Illinois.
Meso-scale enhancement of snowfall can be a significant forecast and observational problem. Due to their small scale, they are not very often observed in detail. However in this case, the circulation tracked into and then retreated through a unique network of meteorological wind instrumentation in Lake County. This wind instrumentation, along with Doppler Radar imagery and ACARS data, provided an opportunity to observe the detailed structure of a meso-scale system.
Thanks to Steve Hentz, Lead Forecaster at the Milwaukee Weather Service Office, and Ken Labas, SOO at the Chicago Weather Service Forecast Office, for help in the acquisition of data for this case. Ken Labas also provided insight into the framework for this discussion. Thanks to Allan Morrison, Lead Forecaster at the Chicago Weather Service Forecast Office, for the review of the events that took place on November 15.
Black, T. L., 1994: The New NMC Mesoscale Eta Model: Description and Forecast Examples. Wea. Forecasting, 9, 265-278.