11 January 2005“First snow” winterstorm

by

Richard H. Grumm

National Weather Service Office

State College PA 16803

  1. INTRODUCTION

The middle of the week of from 11 through 13 January 2005 was expected to be a warm week. However, a stronger than forecast anticyclone to the north kept the low-level cold air over the region longer than anticipated. Warm advection over the low-level frontal boundary produced rain, snow, and freezing rain over most of Pennsylvania on Tuesday, 11 January 2005. Total precipitation amounts ranged from 0.15 in the extreme southeast to around 2 inches in southwestern portions of the State.

The resulting snowfall over central sections was the first measurable and first inch of snow in many locations. State College received 2 inches of snow and fell short of the old record for the latest 1 inch snow fall set on 19 January 2000. Close, but no cigar.

This was the third in a series of marginal winter storms to affect the region in 6 days. Unlike the previous two events, there was considerably more snow associated with this event than the two previous events. The heavy rains in the southwest produced flooding in both the State College and Pittsburgh County Warning areas.

This paper will serve to document the winter storm of 11 January 2005. The focus will be on the use of Ensemble Prediction System (EPS) outputted data in forecasting the event.

2. METHODS

All data were available in real-time via the PSU-NWS data feed. All EPS output images shown here were replicated using the operationally available data and the operationally deployed software. For ice storms, determining the temperature of the boundary layer is critical to determine if the boundary will be cold enough to cause rain to freeze. For this purpose, EPS 2m temperature forecasts, including spaghetti plots and probability plots were used.

In addition to the temperature of the boundary layer, it is critical to have an estimate as to whether the precipitation will be in the form of liquid or solid upon entering the boundary layer. For this approximation, the 850 hPa 0C isotherm was used.

The SREF doe implicitly produce rain, ice pellets, snow, and freezing rain data. At this time these data are not shown.

Some fields were displayed showing the mean of the EPS forecasts for that variable verse the 30-year climatology. In these instances, the departures are shown as standardized anomalies in standard deviations (SDs) from the 30-year normal. The reproductions software used the same data and scripts used in real-time. All real-time graphics changes are based on improvements made during the case study process.

The term “winter storm” is used loosely here and in fact this was more of a winter precipitation event. There was no significant surface cyclone associated with this event.

3. RESULTS

For brevity emphasis will be on SREF forecasts from 10 January 2005. The focus being on how to anticipate the heavy rainfall and the precipitation amounts.

The SREF precipitation type forecasts valid at 1500 UTC 11 January from forecasts initialized at 2100 UTC 10 January 2005 are shown in Figure 2. These data show a high probability of snow from central Pennsylvania northward. The 2m and 850 hPa temperature forecasts valid at 1200 UTC are shown in Figures 3 and 4 respectively. These data show sub-freezing temperatures to around State College at both levels at 1200 UTC. In fact, the 2m temperatures implied the low-level cold air, east of the Alleghany

Plateau extended close to the Maryland border. This may explain the low probabilities, but a chance at least, of ice pellets and freezing rain as far south as the Maryland border in Figure 2.

Figures 5 and 6 show that the low-level cold air would rapidly erode west of the Allegheny Plateau and the 850 hPa temperatures would warm up causing mixed precipitation and rain issue across central portions of the State.

The SREF precipitation forecasts for the 24-hour period ending at 1200 UTC 12 January are shown in Figure 7. These data show that the SREFs forecast a high probability of 0.5 inches or more QPF over most of southern and central Pennsylvania. The heaviest rainfall was expected in the southwest part of the State, in good agreement with Figure 1.

Figure 8 shows the 0.50 QPF SREF QPF forecasts from 12 hours earlier (0900 UTC 10 January 2005). These data show the precipitation potential was considerably less of central and eastern Pennsylvania than the latter forecasts. The amounts were overall lighter and forecast to be farther west. There was no consensus 1.0 inch area in these forecasts.

Figures 9 and 10 show the 2m and 850 hPa temperatures from the 0900 UTC SREF runs. These can be directly compared to Figures 3 and 4. It appears that there was more clustering toward two solutions in the earlier forecasts, this is most notable in the 2m temperature forecasts. With a bimodal solution, the 2m temperatures provided conflicting guidance and the probability distribution function would have clearly been bimodal. Forecasting precipitation at this time with these data provided an interesting forecast problem with precipitation type over central and southern Pennsylvania.

4. CONCLUSIONS

A strong east-west frontal boundary led to a precipitation event over Pennsylvania on Tuesday 11 January 2005. Over the northern half of the State, most of the precipitation fell as snow, freezing rain, ice pellets and ended mainly as rains. Some pockets of freezing rain persisted until 1200 UTC 12 January along the New York border. Many locations, such as State College had their first inch of snowfall (Fig. 11).

This event was well forecast based on short-term SREF forecasts. However, the lead-time potential was limited as the models slowly converged on the solution. This can be seen in the 0.50 QPF contours between two SREF runs only 12 hours apart. This was third successive minor winter storm in less than a week where the probability and spaghetti plots from the SREFs appeared to provide valuable forecast information.

The resulting warm up is best illustrated by figure 12. Its shows the pesky nature of low-level cold air evacuating Pennsylvania east of the Allegheny Plateau. The fact that the SREFs are able to simulate this slowly retreating cold air during precipitation events is a rather remarkable feat.

5. ACKNLOWEDGEMENTS

6. REFERENCES