|Cloud Forecasting in the Wake of|
Clipper Systems in Toledo, Ohio
NORM VAN NESS
Seasonal transitions in and around the Great Lakes can provide forecasters with many challenges on
a daily…even hourly basis. Depending upon prevailing conditions the lakes can and do freeze over during
the late winter months in some years, while others may be marked by very little ice when warm spells
linger well into the season.
Large bodies of water respond slowly to changes in temperature, and as large Arctic surges push southward
in their quest to plunge the Continental U.S. into the throws of winter, that cold air will repeatedly
cross the 94,000 square miles of comparatively warmer open waters of the Great Lakes.
While the forecasting challenges of Lake Effect Snow have been well documented…little has been written
about the difficulties in forecasting cloud formation and coverage in the wake of storms and Arctic
surges in the region. “Miss the clouds…bust the forecast” is
never more true than in the Great Lakes.
Toledo, Ohio is situated at the very western end of Lake Erie. It is not situated in any of the
documented “Snow Belt” regions as it is situated well downstream from Lake Michigan. However, Arctic
air crossing an open and relatively “warm” Lake Michigan can, and often does, produce a persistent
bank of “downstream” stratus clouds that can linger well into the next
day after the system has moved on.
Alberta Clippers are defined in the American Meteorology Society’s Glossary of Meteorology as “A low pressure
system that is often fast-moving, has low moisture content, and originates in western
Canada (in or near Alberta province). In the wintertime, it may be associated with a
narrow but significant band of snowfall, and typically affects portions of the plains
states, Midwest, and East Coast.”1 They are most prevalent in the transitions from fall
to winter and again from winter to spring. By examining the effects of one of these systems
as it crosses the Great Lakes, specifically how it effects Toledo, Ohio, we can identify
several variables that forecasters need to examine and analyze carefully if they hope to
create and routinely verifiable forecast. Observations were monitored along the prevailing
flow from Sheboygan, WI, to Muskegon, MI, then further downstream to Toledo, OH.
On February 8th and 9th of 2007, a “typical” Alberta Clipper crossed the Great Lakes spreading light
snowfall to many areas. Lake Michigan was still open at this point, and a sampling of lake water
temperatures off of Muskegon, MI showed surface water temperatures in the range of 35F degrees2. At
8am CDT the surface temperature at Sheboygan, WI was -1F with a dew point of -12F, dew
point depression of 11F.
Observations downstream at the Muskegon Light (MKGM4) at 8am EDT showed a surface temperature of 14F and
a dew point of 8F for a depression of 6F. The cold air crossing the lake had been significantly modified
over the 70 mile journey. Surface air temperatures had risen to the tune of 15F degrees, and more
importantly, the air had picked up a significant amount of moisture swinging the dew point
upwards by a good 20F degrees.
Further downstream, Toledo had a surface temperature of 10F, and a dew point of 1F. The air
mass had “dried” significantly between Western Lower Michigan and Northwest Ohio… with the dew
point dropping 7F degrees. Toledo was under a nearly clear sky at the time. While this particular
event failed to produce any significant snow, it did produce a large deck of stratus clouds in its
wake that reached inland for nearly 100 miles creating VFR and MVFR conditions for all of SW
Michigan over the course of the next day, along with a stray snow flurry or two.
The stratus deck however did not extend far enough downstream to impact the city of Toledo, Ohio despite
it being forecast to do so by surrounding forecasters, broadcast and government alike. With low clouds
in the forecast for the preceding night, and clearing not forecast to take place until well into the next
day, the 24 hour forecast was a “bust” for both sky cover and for temperature.
This single case identifies several variables that contribute significantly to the forecast. Massive
temperature differences between the surrounding air mass and the lake waters can run at times in the
order of 30F to 40F degrees. The induced vertical motion from this difference is quite effective at
evaporating moisture off of the lake and into the passing air. At the same time, subsidence building
in behind the exiting system is working to push the air back down as vertical velocities turn
negative. The moisture in effect gets “squeezed” from below and above into a relatively small
portion of the lower atmosphere resulting in the persistent stratus deck.
Different air/water temperature depressions will produce very different results. Larger depressions will
producer more efficient evaporation, while smaller ones will produce little at all. Monitoring and
forecasting air mass qualities for temperature and moisture content are critical…as is accurate
monitoring of the temperature of the water that air mass is crossing. The presence and coverage
of ice on the lake adds yet another factor to the larger equation.
Upper air observation sites are sparsely populated through the region, with Green Bay, WI and Detroit, MI being
the closest to the site in question. While they provide some clues as to this “trapping effect” this
cloud event happens on to small of a scale for them to be generally useful. Of great help is
the BUFKIT software developed by the National Weather Service office in Buffalo, NY.3
This visual analysis product uses upper air forecast data from several computer models, (ETA, NAM,
GFS, RUC to name a few). The data can be displayed as a traditional Skew-T sounding that changes
over the time of the model run. This gives the forecaster a valuable tool to use in identifying the
conditions that produce the persistent clouds. While the typical model biases still apply, this
visualization can give the forecaster a clearer picture of how the atmosphere might react to several
different variables all at one time.
The standard suite of forecast models from the National Centers for Environmental Prediction often fail to
recognize the smaller, meso and micro-scale contributions of open lake waters in the transitions into
and out of winter. Picking up on theses contributions is critical to a correct forecast…but is really
something that must be developed over time as there is no set “rule” for
forecasting these conditions.
Jeff Logsdon, the Science Operations Officer at the National Weather Service Office in North Webster, Indiana
has been a forecaster at that office since it opened in 1997. Repeated encounters with Clippers over those
years have produced a knowledge set that can most certainly help other forecasters in the region. A
conversation on the topic of clouds behind Clippers provided some helpful clues.
Jeff lists strength and timing of the system as some of the more important factors to consider. Stronger
systems have the ability to wrap in more land surface moisture which can then be pulled in behind
the system. This additional moisture is added to the contributions from the lake and is then
more likely to produce that lingering cloud deck.
Timing is also important. A system that exits the region in the day time will use the subsequent diurnal
heating to help mix out the inversions that help to create the stratus deck in the first place…while
a “night exiting” system will have the typical night time inversions to help enhance the deck before
the sun rises, thus negating the opportunity to mix out the layer once the sun does arrive.
Jeff adds that lake contributions are lessened as the winter progresses, and the lake cools or freezes. As
spring approaches, Clippers get “warmer”, (source air mass is not as cold at initiation), and this too
lessens the cloud effects as compared to fall…but any time the lakes are open, factoring in moisture
contributions from them is key.
Forecasters at the North Webster station do indeed use the BUFKIT software to help them with their
forecasts. Not only do they use point data for local stations…they also use model grid points that
are out over the lake. By inserting the lake water temperature into the BUFKIT software, a much
improved picture of lake contributions can be visualized and analyzed which greatly
improves forecast accuracy.
Cloud forecasting technique is very much an acquired skill, and getting the clouds “right” will make you a
better, more accurate forecaster. Approaching those around you with experience in specific problem areas
can help any forecaster gain a better understanding of how to process and prioritize the many variables
that make up the problem in the first place.