Ask Ellen: What happened with the ice storm?

GRAND RAPIDS, Mich. (WOOD) — With another storm on the horizon this weekend, we wanted to take a moment to look at our ice storm this past Saturday. It had the makings of being a monster.

The warnings in place were valid. But the impacts here were lower than expected (thankfully.) Less than 300 miles away, more than 1″ of ice loaded down trees in Hopeville and Damascus, Ontario.

Courtesy of @isabel_ONwx
Courtesy of @isabel_ONwx

Here’s where we ended up.

Courtesy of NWS Grand Rapids

Now, to be fair. Much of this storm unfolded the way we expected. The massively heavy rain for this time of year arrived first and on schedule. The transition to cold as an area of high pressure wedged in a shallow cold pool from north to south, a wide swath of freezing rain and sleet for more than 10 hours and a final burst of snow as the storm exited.

But why were some impacts and totals so low? Well weather nerds of West Michigan, here are a few of the most likely reasons why…


A matter of a degree Fahrenheit can chance a lot in the science realm. This makes sense. A pot of water is just hot at 211°. It boils at 212°. That’s a big chance in impact for a small change in temperature.

The same can be true near freezing.

While we saw roughly 10 hours of freezing rain across West Michigan on Saturday, the ice had a hard time accreting. For the best freezing rain accretion, temperatures need to be below freezing instead of hovering AT freezing. This graph is technically used for ships and icing. But, it does show how ice accretion is much lower near 30° and increases exponentially as the temperature drops.

Courtesy: The Swedish Meteorological and Hydrological Institute

Please note this graph also includes the wind impact. That is significant because wind will often bring in lower dew point air, which can also help with icing. But what I’m saying here isn’t too surprising. Things freeze better just below freezing rather than at freezing.


To be fair, we accounted for this. But, I’m not sure we accounted enough. A warm ground will melt ice. That’s a no-brainer. But while we sliced our expected totals considerably due to the warm ground, it definitely helped keep the roads driveable until after nightfall.


We saw this in the above massive ice totals from Ontario. Days before the event, we used our longer range forecast models to predict where the storm would go. It was carrying a massive amount of moisture with it and all eyes were on the storm.

On Friday, the storm began producing volatile weather to the south. High winds, hail and tornadoes.

Long range forecast models start chugging away fresh model data every six hours based on current conditions at the time they are run. Before the severe weather, many forecast models had the upper-level portion of the storm negatively tilted at 500 mb, which would have given us the worst of the impacts once it traveled downstream a day later.

However, as one of my meterological friends puts it “models really struggle dealing with convection and its effects on mass/momentum.” Thunderstorms are such a compact package of energy, that it is highly possible the storminess down south changed the parent low that produced our storm in Michigan.

Here is the side by side of the 500mb pressure level of the parent storm. I drew a line along through axis of the storm. Notice how the first model run has the storm negatively tilted, and the second one doesn’t? It is possible the heavy convection (rain and storms) the low began producing shifted it, and weakened it.

Courtesy of Tropical Tidbits

A weaker storm means less freezing rain & snow for us. It also means less wind.

A change in the tilt of the parent storm changes how the water vapor gets wrapped in on the back side of the storm, and also likely weakened the cold air transport on the north side of the freezing line. (Perhaps why we sat at 32° instead of 30 ° as expected on Saturday.)

Courtesy of Tropical Tidbits

In the picture above the blue is cold air transport at ~5000 feet. The red is warm air transport. Notice how minimal our cold air transport is over the mitten? Meanwhile downstream, conditions aligned better for our friends in Canada later in the event.


A weaker storm means lighter wind. Wind is generated as air rushes from high pressure to low pressure. The stronger the area of low pressure, the more wind that’s produced.

Because the storm weakened slightly, it produced slightly less wind. Here’s the kicker: The force exerted by wind is not linear, it is exponential. That means if you double the wind speed you will actually see four times the force!

Courtesy of The Engineering Toolbox

Recall we saw 0.1″-0.3″ of ice accretion on lines and trees. If the winds would have gusted to 40 mph instead of the 30 mph range, we would have had significantly more power outages.

As it stood Saturday, we saw 30,000 without power.

If the storm would have been slightly stronger, as expected, the increased ice and increased wind would have likely caused outages for days. That is why so many power crews were stationed here in West Michigan to do just that.


After the icing event, we knew all things would turn to snow. However, when that actually happened here on Saturday night, the snow totals were drastically low.

Howard City Snow by Brenda Paul

The first thing to cut into this total was the lazy freezing line (as previously mentioned, that kept things a touch to warm — likely because of the weaker cold air transport by the parent low.)

The second was the fact the snow-part of the storm that arrived Saturday night had about half the water than expected.

A little bit of water can make a decent amount of snow. Usually 1″ of water makes 10″ of snow (but not always.) So, even a small change in the water can have a huge impact on the snow totals.

A meteorologist friend of mine is credited once again for pointing out the storm tilt theory. A tilted system means less water vapor is transported to the back half of the storm.


No one wants to be right about a forecast more than the meteorologist making it. All of us at Storm Team 8 were working on this one in conjunction with meteorologists across the state. We do not take your trust lightly, and are already working on new ways to communicate our certainty in a storm better.

Ice storms will always hinge on the tip of a sword. If any of the above factors would have tipped the other direction, we would have been inundated.

In the end, we are glad so many erred on the safe side. The fact so few were on the roads likely spared us some accidents.