Dr. Rob Carver's WunderBlog

Snow for the Appalachians and Rain for the East Coast

By: rcarver, 7:33 AM GMT on October 15, 2009

It's weekends like this that make me glad I'm not in Happy Valley (State College, PA). It's only October, but the NWS has issued a Winter Winter Advisory covering most of central PA from Happy Valley to the Allegheny Plateau from Thurs. night into Saturday morning. What's going on? Well, it's a bit complicated.

A low forming off the Carolina coastline is expected to move NE while deepening and producing a wide shield of rain. As the low moves past the Chesapeake Bay region, the rain will turn into snow in the higher elevations of the Appalachians of VA, MD, and PA. Also, the circulation around the low will pull cold air down from Canada, making snow possible in the lower elevations of central PA east of the Allegheny Ridge. State College will be right at the edge of the rain-snow line.

From personal experience, I can tell you what it's like. First comes the rain, not rain showers but a sustained, heavy drizzle with temperatures in the low 40's/high 30's Then as the cold air moves in, snow starts mixing in with the drizzle with minimal accumulations. As the temperature drops, the precipitation changes over to wet, heavy snow that can fall in big heavy clumps. No matter how it falls, the snow will start sticking to trees, while it might melt on the roads. Since the leaves haven't fallen yet, there's more for the snow to stick to and tree branches start snapping as the wet snow accumulates. Power outages are frequent with this type of event. For you Nittany Lion fans (and misguided Gopher fans) who are going to be at Beaver Stadium Saturday afternoon, the current game-time forecast is for rain and 38 deg F, but I highly recommend using our Game Day Weather page for the current forecast.

This weekend's weather event isn't limited to snow. The rain associated with the coastal low is expected to produce over 2 inches of rain over the Baltimore-Philadelphia region according to the Hydrometeorological Prediction Center precipitation forecast for Oct. 15-17 in Figure 1 with reduced precipitation along the NY/New England coast.

Fig. 1 72 hour Quantitative Precipitation Forecast (QPF) for October 15-17. Image courtesy of HPC

After the low moves out, another low will affect the NE coast. The timing is still a bit uncertain, but an upper-level trough in the Midwest/OH river valley will trigger surface low formation east of the Delmarva peninsula on Saturday. The current HPC forecast holds that rain from this event on Oct 18-19, Figure 2, will not be as heavy as what occurred before. Still, it ought to be enough to keep everybody indoors on a Sunday.

Fig. 2 48 hour QPF for October 18-20. Image courtesy of HPC

Stay warm and dry out there.

Updated: 9:34 AM GMT on October 15, 2009


Storm update from Wunderground's SF office

By: rcarver, 11:35 PM GMT on October 13, 2009

As of 4pm PDT, it looks like the worst of the rain will be ending soon for the SF Bay region. It's a bit early to tabulate the results across the region, but I will note that the Personal Weather Station located on top of our building had recorded a total of pi (3.14) inches of precipitation so far today.

From a meteorological perspective, the winds with this storm haven't been anything unusual for a Pacific storm. At least for the Bay Region, rain seems to be the main trouble maker. However, as the storm moves towards the Sierras, I expect there will be some rather impressive sustained wind speeds and wind gust reports.

Now that the storm has made landfall, we can use actual observations to check how accurate the models' representation of the storms were. For example, Figure 1 shows the weather balloon sounding recorded over Oakland, CA this morning.

Fig. 1 Radiosonde report from Oakland CA at 12Z Oct 13. Image provided courtesy of the Storm Prediction Center.

It's apparent that the deep layer of moisture I discussed in my last blog entry was a real feature. Looking at the precipitable water index, the Oakland sounding had 320 mm, while the forecast value was 322 mm. So, at first glance, the GFS model and it's data assimilation system did an excellent job of representing the storm's characteristics in a data-sparse region.

As for my commute today, let's just say that my rain gear wasn't waterproof enough.


California is going to get a little wet

By: rcarver, 7:26 AM GMT on October 12, 2009

Here's a tip, when the local NWS office helpfully lists the one day and two day precipitation records in their forecast discussion, get out your rain gear.

What's going on? Well, the remnants of Super typhoon Melor are about to make landfall along the northern CA coastline Monday night. For Northern CA, the NWS is forecasting 2-3 inches of rain along the Bay, 4-6 inches of rain for the Santa Cruz and Sierra foothills, and several feet of snow in the Sierra Nevada proper (snow level starting at 7500 ft, heaviest snow about 9000 ft). Oh, and did I mention the wind? For the SF Bay on Tuesday, the sustained winds are forecast to be 20-40 MPH along the coast and in the Bay area hills with gusts to 60 MPH possible. The Central Valley can also expect sustained winds of 20-30 MPH, with speeds increasing to 45 MPH in the foothills. However, the Sierra ridges near Truckee are the winner, with speeds "in excess of 120 MPH".

Why? As previously noted, the remnants of Super typhoon Melor are associated with this storm. One of the common hallmarks of tropical air is high moisture, and this is no exception. Figure 1 shows a skew-T from our Wundermap model data layer showing the forecast conditions of a point just SW of San Francisco on Tuesday morning.

Fig. 1 48 hr skew-T from the 2009-10-11 18Z GFS run

At a glance, anybody who has survived their junior-level meteorological thermodynamics course can see the problem. The atmosphere has a lot of water vapor (Precipitable water of 322 mm to be exact). For everybody else, here's a really short summary of thermo: the RH for a level in the atmosphere is inversely proportional to the distance between the left trace (dew point temp.) and right trace (air temperature). Also, as temperature increases, the maximum amount of water vapor the atmosphere can have before condensation becomes likely also goes up. So looking at the traces, what do we see? The dew-point and air temperature are very similar from the surface to a pressure level of 300 mb (~9 km or 30,000 ft). This means the RH is very high through a deep layer in the atmosphere. But just how moist? The relatively warm temperature profile and high RH indicate that this deep layer is VERY moist. In fact, using a technique known as forecast anomalies, such moisture levels only occur 1 out of 1,744,208 times in the month of October (precipitable water anomaly is estimated at +5 for the mets out there). So this is an unusually wet event.

As for the wind, that can be explained by the upper-level winds associated with the storm as shown by Figure 2.

Fig. 2 48h forecast of 300 mb heights and winds from the 2009-10-11 18Z GFS run

Again, this is for Tuesday morning, and the subtropical jet is just to the south of the upper-level low. At this time, the peak wind speeds are a few hours from making landfall, so the peak winds for SF will be late Tuesday night/early Wednesday morning. The strength of the winds is due to the strong horizontal difference in temperatures around the jet. The area north of the jet is much cooler than the area just south of the jet, and the wind speed is proportional to such temperature differences.

As for me, my commute to Wunderground's office is usually a nice walk, but I'm thinking a kayak may be more appropriate this week.

Important Update Shaun Tanner, our senior meteorologist in the SF office (and Bay area native), has his thoughts here. Shaun points out something I forgot about in the above entry, forecasts tend to be overly "optimistic" about the first storm of the season. It doesn't matter if it's Pacific storms, Atlantic hurricanes, or the first forecast severe weather outbreak over the southern Plains (something I learned at OU). That said, I still think this will be a major event for an October storm. The amount of moisture present and the upper-level winds are rather impressive. Also to corroborate a point Shaun made about the models, forecasts from the NAM now cover the time period of the storm and look
a lot like the GFS forecasts with some minor differences (the orientation of the upper-level winds change to take a slightly glancing hit on the Santa Cruz mountains vs the straight-on hit in the GFS).

However, this is not surprising from a modeling point of view. Right now, the storm is over what we call a "data void", an area with no surface obs or direct upper-level observations (traditional weather balloons). Again drastically simplifying, the only data available to the NAM for the Pacific ocean is forecasts from the GFS, so there isn't much to cause forecast differences.


About rcarver

Rob is the Research and Development Scientist for Weather Underground. He has a Ph.D. in meteorology from Penn State University.

rcarver's Recent Photos

Sent from my iPhone

Sent from my iPhone
Cell Phone Image 33
Cell Phone Image 32