During the night of August 21, 1995, a quasi-stationary mesoscale convective system (MCS) developed over northeast Nebraska and persisted into the morning of August 22, 1995. This MCS produced extremely heavy rainfall, (an unofficial maximum of 9.50" at Tilden, NE), with some reporting stations receiving as much as three times their monthly average in a 6- to 12-hour period (Figure 1).
Figure 1. Rainfall estimates (solid contours in inches) from the Valley Nebraska (KOAX) WSR-88D for the 24-hour period ending 1733 UTC 22 August 1995).
The heavy rainfall over northeast Nebraska occurred in an area that had not been outlooked for heavy precipitation in the excessive rainfall outlooks issued at 0230 UTC and 0630 UTC 22 August 1995 (same date hereafter) by NCEP. This discussion targeted the best area for excessive precipitation over North Dakota, with no mention for the possibility of heavy rainfall over northeast Nebraska. By using the 0000 UTC Eta and NGM gridded model output along with parameters discussed by Glass et al. (1995) a favorable environment was revealed over northeast Nebraska. The profiler network was an important tool for analysis in this case.
INGREDIENTS NECESSARY FOR HEAVY RAINFALL
The ability for the atmosphere to produce heavy rainfall involves a few important parameters. The most important are persistent lift, deep moisture, and inflow of potentially unstable air. Without one of these three, the chances for heavy rainfall are greatly reduced.
Maddox et al. (1979) described three types of flash flood events common to the central U.S.; the synoptic, the frontal, and the mesohigh. Each type had a different focusing mechanism involved in initiating heavy rainfall producing systems. In all of the cases, deep moisture, usually 150 percent to 200 percent of normal precipitable water, was also needed. After convection developed a strong low-level jet (LLJ) was necessary to sustain any further development by delivering warm, unstable air to the thunderstorm complex.
In the northeast Nebraska event, two of the three important parameters were not readily apparent. With no obvious significant lifting mechanism in the area, and the best sustained moisture inflow directed well to the north, heavy rainfall was not forecast.
The 300 mb and 500 mb analyses at 0000 UTC (Figures 2a and 2b) showed a broad ridge across the northern plains. High pressure at both levels was centered near southwest Kansas with weak, anticyclonic flow evident over Nebraska. The 500 mb analyses also indicated moist mid-levels, in the form of dew point depressions less than 7°C over the Rockies and eastward into Nebraska.
The 0000 UTC 850 mb analysis (Figure 2c) indicated high pressure centered over northeast Kansas and northern Missouri with a trough over the northern Rockies. A tongue of significant low-level moisture extended from eastern New Mexico through central Nebraska and into southern North Dakota. Dew point temperatures in this area were greater than 15°C and temperatures were greater than 25°C. General flow in this warm, moist air was from the southeast at 20 to 25 knots.
A similar pattern was evident on the 700 mb 0000 UTC analyses (Figure 2d) with high pressure centered over central Kansas and a weak trough across the northern Rockies. A ridge axis extended northward from the high pressure center through eastern Nebraska and the eastern Dakotas. Warm, moist air with temperatures greater than 12°C and dew point temperatures greater than 5°C, were to the west of this ridge axis over the southern Rockies and western high plains.
Figure 2a. Plot and analysis of 300 mb data valid 0000 UTC 22 August 1995 (height contours every 120 decameters).
Figure 2b. As in Figure 2a except 500 mb height (solid contours every 60 decameters) with dew point depression <7°C (dashed) and dew point depressions <4°C (hatched area).
Figure 2c. As in Figure 2a except 850 mb height (solid contours every 30 decameters), and dew point temperatures greater than or equal to 15°C (hatched area).
Figure 2d. As in Figure 2a except 700 mb height (solid contours every 30 decameters), and dew point temperatures greater than or equal to 7°C (hatched area).
This deep moisture showed up well in soundings from both North Platte, NE (LBF) (Figure 3) and Omaha, NE (OAX) (not shown) which had precipitable water values of 1.93" and 1.45" respectfully.
Figure 3. North Platte Nebraska sounding for 0000 UTC 22 August 1995.
ANALYSIS USING GRIDDED MODEL OUTPUT AND PARAMETERS FROM GLASS et al. (1995)
The 0000 UTC 22 August Eta and NGM gridded model output was used for this case. The main focus of the analysis will be to incorporate parameters from Glass et al. (1995) to show how a favorable environment for heavy rainfall could be seen over northeast Nebraska.
Glass et al. (1995) conducted a study of eight flash flood producing events during the warm season over the mid Mississippi Valley. Some important parameters discussed in this study were:
• 850 mb equivalent potential temperature (Thetae) ridges, gradients and advection
• Weak anticyclonic and diffluent upper level flow
• The importance of the LLJ
Shi and Scofield (1987) and Juying and Scofield (1989) found that MCS initiation and propagation is favored along Thetae ridges and may also be associated with Thetae gradients and positive Thetae advection. Glass et al. (1995) found that positive Thetae advection was one of the best indicators of heavy convective rainfall potential. Equivalent potential temperature is a significant parameter since high Thetae values indicate an area of significant moisture and potential instability.
A Thetae ridge is not an area where thunderstorms will always develop. The most important aspect of a Thetae ridge is that if thunderstorms do develop in or near that ridge, they tend to evolve into a more organized area of convection, (i.e. an MCS)(Chaston 1993). Chaston (1993) also observed that a low-level Thetae ridge that is overrun by mid-level tropical moisture is an area conducive to heavy rainfall if convection develops (Figure 4). This occurs mainly because deep moisture creates better precipitation efficiency for a thunderstorm.
Figure 4. Six-hour forecast from 0000 UTC 22 August 1995 NGM run of 850 mb Thetae (thick solid contours every 5 K), 700 mb relative humidity (dashed contours greater than or equal to 70% every 10%), and 500 mb relative humidity (thin solid contours greater than or equal to 70% every 10%) valid 0600 UTC 22 August.
The 6-hour NGM 850 mb Thetae forecast valid for 0600 UTC (Figure 5) showed a ridge over western and central Nebraska with a tight gradient over eastern Nebraska. The 850-700 mb layer Thetae advection showed a maximum over north central Nebraska. The MCS that developed over northeast Nebraska initiated on the east side of the Thetae ridge within the area of highest positive Thetae advection and generally followed the Thetae contours southeastward. Glass et al. (1995) created a conceptual model of the location of heavy rainfall with respect to positive Thetae advection (Figure 6). In this model the location of the heaviest rainfall was to the east-southeast of the maximum positive Thetae advection, which correlated well in this case.
Figure 5. Six-hour forecast from 0000 UTC 22 August 1995 NGM run of 850 mb Thetae (dashed contours every 5 K) and 850-700 mb layer positive Thetae advection (solid contours deg K day-1) valid 0600 UTC 22 August.
Figure 6. Conceptual model from Glass (1995) of location of heavy rainfall (shaded area) with respect to fronts, thickness contours (dashed), and 850 mb positive Thetae advection maximum.
A flow pattern that is often correlated with heavy rain producing MCS's is anticyclonic and diffluent flow (Glass et al. 1995). The 0000 UTC 300 mb and 500 mb analyses (Figures 2a and 2b) showed this relatively weak, anticyclonic flow with wind speeds ranging from 10 to 25 knots. The 6-hour NGM 300 mb wind and divergence forecast valid at 0600 UTC (Figure 7) indicated continuation of this weak anticyclonic and diffluent flow over north central and northeast Nebraska. This would imply that any thunderstorms that developed in the area would tend to move very slowly.
Figure 7. Six-hour forecast from 0000 UTC 22 August 1995 NGM run of 300 mb wind (knots) and divergence (solid positive contours in 10-6*s-1).
One parameter that has historically been difficult to analyze, until recent years, was the LLJ. With upper air observations taken only twice daily at 0000 UTC and 1200 UTC, an accurate analysis of what occurred above the surface during the other hours of the day was not possible. Mitchell et al. (1995) found that the peak diurnal occurrence of the LLJ was from 0600 UTC to 0900 UTC with a noticeable decrease from 0900 UTC to 1200 UTC. The peak period occurs directly between upper air observation times.
In this case, analysis using only model output and sounding data would have lead one to believe that the LLJ over southwest Nebraska and northwest Kansas from 0000 UTC through 1200 UTC would be fairly weak (20 to 25 knots) and be from the south to southeast. Analysis using the hourly profiler data in comparison with the model output became an important tool in evaluating flash flood potential. The profiler network showed a large discrepancy between the gridded model output 6-hour and 12-hour LLJ forecast, valid at 0600 UTC and 1200 UTC respectfully, and what actually occurred.
The time section display from the McCook, NE (RWD) profiler (Figure 8) indicated veering and increasing low-level winds from 0300 UTC to around 1200 UTC. A speed of approximately 45 knots from the southwest can be seen in the second gate at 0800 UTC. This was 25 to 30 knots stronger and about 20 degrees southwest of what the models forecasted (Figures 9 and 10). This advected the high Thetae air over central Nebraska toward northeast Nebraska rather than into the Dakotas as the models had indicated.
Figure 8. Time-height series of wind (knots) from McCook Nebraska profiler (RWD) for 0300 to 1800 UTC 22 August 1995.
Figure 9. Time-height section of forecast wind (knots) from 0000 UTC 22 August 1995 NGM run.
Figure 10. As in Figure 9 except from Eta run.
SUMMARY AND CONCLUSIONS
At first glance, ingredients necessary for the formation of a flash flood producing MCS were not readily apparent in this case. However, information from profiler data combined with the Glass et al. (1995) conceptual model indicated an environment more conducive to the production of heavy precipitation by 0600 UTC over northeast Nebraska. A Thetae analysis at 850-mb, 700-mb, 500-mb and 300-mb showed that high Thetae air was present at all levels. A quasi-stationary baroclinic zone over eastern Nebraska combined with moist, unstable air being lifted above this zone by a strong, southwesterly LLJ provided the necessary ingredients for extremely heavy rainfall. This case illustrates the necessity of examining datasets and parameters beyond those upon which the classical conceptual models were developed.
Chaston, P., 1993: Graphical Guidance. DOC/NOAA/NWS, Training Center; Kansas City,MO, 95 pp.
Glass, F. H., D.L. Ferry, and S.M. Nolan, 1995: Characteristics of heavy convective rainfall events across the mid-Mississippi Valley during the warm season: Meteorological conditions and a conceptual model. Preprints, 14th Conference on Weather Analysis and Forecasting, AMS (Boston), 34-41.
Juying, X., and R.A. Scofield, 1989: Satellite-derived rainfall estimates and propagation characteristics associated with Mesoscale Convective Systems (MCSs). NOAA Tech. Memo. NESDIS 25, 49 pp.
Maddox, R.A., C.F. Chappell, and L.R. Hoxit, 1979: Synoptic and meso-alpha scale aspects of flash flood events. Bull. Amer. Meteor. Soc., 60, 115-122.
Mitchell, M.J., R.W. Arritt, and K. Labas, 1995: A climatology of the warm season great plains low-level jet using wind profiler observations. Wea. and Forecasting, 10, 576-591.
Shi, J., and R.A. Scofield, 1987: Satellite observed Mesoscale Convective Systems (MCS) propagation characteristics an a 3-12 hour heavy precipitation forecast index. NOAA Tech. Memo. NESDIS 20, 43 pp.