El Niño is here. What will it mean for Great Lakes ice cover?

Over the weekend, temperatures finally climbed over 40ºF in Cleveland. Given the fact that the average temperature in February was all of 14.3ºF – by far the coldest February in our history – the mid-40s felt like a heat wave.

My fiancée and I decided to venture outside and headed down to Edgewater Park on Cleveland’s West Side. Edgewater, as the name suggests, sits along Lake Erie. We wanted to take an opportunity to see the lake before the ice really began to melt. Due to the frigid winter, the Great Lakes were once again covered in a thick layer of ice this year. Though we will likely remain just shy of last year’s mark, ice cover reached a peak of 88.8% on February 28. As set to continue running at or above normal, this number should continue dropping until the lakes are ice free sometime in late Spring. It has already fallen by more than 20% in the past 10 days.

We were far from the only people with this idea. While neither of us planned to actually head out onto the ice, we eventually decided to follow the pack. Someone had even decided to set up a tent on the ice a few hundred feet off shore to serve soup and coffee to passersby. At the time, I had no idea what the actual thickness of the ice we were walking on was. I flippantly estimated that it was several feet thick – a testament to my ignorance. I have since discovered, from the map below, that we were likely standing on a sheet of ice roughly 40 centimeters thick. Fortunately, that is thick enough to support a car.

lake erie ice thickness march 9, 2015

Courtesy of NOAA’s Great Lakes Environmental Research Laboratory

El Niño arrives – finally

Just as the forecast was beginning to take a turn for the better last week, NOAA made headlines by announcing that El Niño had finally arrived. Forecasters had been warning about its impending onset for more than a year, so the announcement wasn’t exactly a surprise. As I stood on the ice last weekend, I couldn’t help but wondering how this phenomenon might affect ice cover next winter.

El Niño is the warm phase of the El Niño Southern Oscillation (ENSO), during which a band of water water forms in the mid-tropic Pacific Ocean. The phenomenon is characterized by high air pressure in the western Pacific and low air pressure in the eastern reaches of the ocean. As Eric Holthaus notes at Slate,

Technically, for an official El Niño episode, NOAA requires five consecutive three-month periods of abnormal warming of the so-called Nino3.4 region of the mid-tropical Pacific, about halfway between Indonesia and Peru. It usually takes a self-reinforcing link-up between the ocean and the atmosphere to achieve this, and it finally appears the atmosphere is playing its part.

Generally speaking, El Niño brings above average temperatures to the Great Lakes region. Moreover, because the oceans have been storing vast amounts of heat over the past decade-plus, helping to limit the rate of global warming, a particularly strong El Niño could lead to a dramatic transfer of stored heat from the oceans to the surface. As a result, many observers are predicting that 2015 will be the warmest year on record.

El Niño and Great Lakes ice cover

It would be logical to assume that the onset of El Niño will limit the amount of ice that forms on the lakes. According to a 2010 NOAA study, from 1963-2008, 11 out of 16 El Niño winters saw below average ice cover. During these 16 winters, ice covered an average of 47.8% of the Great Lakes, considerably lower than the long-term annual average of 54.7%. As Raymond Assel, a scientist with NOAA’s Great Lakes Environmental Research Laboratory (GLERL) wrote in 1998 (emphasis from original):

On average, the average annual regional temperature is likely to be higher (approximately 1.2ºC and the annual regional maximum ice cover is likely to be less extensive (approximately 15%) during the winter following the onset year of a strong warm ENSO event.

But the connection between El Niño and ice cover is not quite so straightforward. In fact, three winters – 1970, 1977, and 1978 – saw above average ice cover, despite occurring during El Niño events. Ice cover during the latter two years exceeded 80%.

So what else is at play? Well, according to the literature, three factors must combine to produce a particularly mild winter for the Great Lakes region and, by extension, lead to extremely low ice cover like we saw in 1998, 2002, and 2012: the strength of the El Niño event and the modes of the Arctic and Pacific Decadal Oscillations. Let’s take a look at three these indicators to get a sense of what might be in store.

El Niño strength

Multiple studies have found that the relationship between these two factor is highly nonlinear. As this chart from Bai et al. (2010) shows, the scatter plot for ice cover and El Niño strength follows a parabolic curve. Accordingly, El Niño does tend to limit ice formation, but its effect is only significant during strong events.

Relationship between El Niño strength and Great Lakes ice cover (from Bai et al. 2010).

Relationship between El Niño strength and Great Lakes ice cover (from Bai et al. 2010).

But the current signs do not point to a strong event. As Brad Plumer explained for Vox,

Back in the spring of 2014, it really did look like a strong El Niño would emerge later in the year…

But then… things got messy. Atmospheric conditions over the Pacific Ocean didn’t shift as expected. Specifically, scientists weren’t seeing the change in atmospheric pressure over both the eastern and western Pacific that you’d expect during an El Niño.

As a result, NOAA appears to be tempering expectations about the strength and duration of this event. It is likely to be relatively weak and last through the summer, potentially limiting its impacts on the Great Lakes.

Arctic Oscillation

The Arctic Oscillation (AO) is among the most important factors that determines the severity of winter in the Great Lakes. The AO is “a climate pattern characterized by winds circulating counterclockwise around the Arctic at around 55°N latitude.” During its positive phase, strong winds around the North Pole effectively lock Arctic air in the polar region, helping to moderate winters. But in its negative phases, these westerly winds weaken, allowing Arctic air to travel further South; this is the phenomenon that caused the polar vortexes we have seen in the past two winters.

Accordingly, during the positive phase of the AO, less ice cover forms on the Great Lakes. From 1963-2008, positive AO winters have been 0.9-1.8ºC warmer than normal and seen a mean ice cover of 49.2%. The combination of an El Niño and a positive AO produced the five lowest ice cover totals during this period.

So where does the AO stand? Currently, it is in a positive phase. Unfortunately, it is difficult to determine whether this phase will persist, as the AO can fluctuate widely. But if this oscillation does remain in a positive phase next winter, it would amplify the effect of the weak El Niño.

Pacific Decadal Oscillation

Winter weather is also influenced by the Pacific Decadal Oscillation (PDO), “a long-lived El Niño-like pattern of Pacific climate variability” that helps determine sea surface temperatures in the North Pacific. Rodionov and Assel (2003) concluded that the PDO helps to modulate the impact of ENSO on the Great Lakes. Warm phases of the PDO tend to amplify the impact of El Niño and reduce ice cover.

Last year, the PDO emerged from its prolonged weak phase to reach record high levels. If it continues to remain strong, it will likely lead to warmer temperatures not just next winter, but potentially for the next 5-10 years. This would seem to suggest that the PDO will enhance the impact of the El Niño event next winter.


Overall, the picture is still a bit murky. It does not appear that the El Niño will be strong enough to produce the type of least ice cover event that we saw in 2012. Yet, at the same time, the combined effects of El Niño, a positive AO (should it remain that way), and a warm PDO (if the trend continues) will likely ensure that the Great Lakes region avoids another brutal winter, like the ones we’ve seen two years running. If this is the case, lake ice cover should regress closer to the long-term mean of approximately 50%.

But if the indicators strengthen in the next several months, the winter weather could moderate even more; this would have clear impacts for lake levels, lake-effect snow, harmful algal blooms, and local temperatures during the Spring and Summer months.

Tagged , , . Bookmark the permalink.