Some have claimed that the observed warming is merely an artifact of an ever expanding urban heat island effect or other issues in the temperature records. One way I thought this could be tested was by seeing whether bodies of water have exhibited the same warming trend as surrounding land-based stations. While the surface temperature of large bodies of water can be affected by a number of phenomena, they should provide a fair proxy for overall trends in the surrounding climate.

Perhaps the best record of Lake Erie water temperatures, and certainly most accessible, is that maintained by the NWS Buffalo WFO. Water temperatures have been taken continuously at the Buffalo Water Treatment Plant at the entrance to the Niagara River since 1927. The all-time highest reading is 80, which was set last summer. In addition, two all-time monthly highs have been set this year. The water temperatures reached 37 degrees in early February, and 40 degrees in late March, which were both monthly highs.

So is the warm lake water just a fluke? Or is there a trend towards increased water temperatures? This past winter was certainly unusual for its mildness and lack of ice coverage, as dicussed in this article.

Hageman, who has been running his ice guide service for 16 years, remembers winters without Lake Erie ice in 2006 and 2002. As the retired manager of Ohio State University’s Stone Laboratory on nearby Gibraltar Island, he said temperature data show Lake Erie’s winters are getting warmer.

“We’re definitely in a cycle for higher temperatures, which affect how much the lake is open,” he said. “You can’t dispute the thermometer. It’s been well-documented that it’s warmer.”

The above quote acknowledges that the lake has warmed, although it suggests the warming is cyclical in nature. I decided to take a look at this using the data from NWS BUF. Below is a graph showing the temperatures trends for the months of January, March, May, July, September, and November. For each month, I used the temperature recorded on the 15th of the month.

Figure 1. Observed lake temperatures, 1927-2012.

The graphs clearly exhibit both ups and downs over the years. To get a better idea of the trends, I graphed the linear trend line for each data set. As visible on the graphic, the trend is positive in each month, but the extent of the increase was highly variable. The largest increase occurred in May, with a trend of +0.53F/decade. The smallest increase occurred in March, with a trend of just +0.08F/decade. So what’s up? From the graph, these variations can easily be explained as a product of the decrease in seasonal ice coverage. In March, the lake is nearly always frozen over, which minimizes the variation in year-to-year temperatures. In May, however, the changes in ice coverage have exacerbated the warming effect. In the past, the east end of the lake was occasionally ice covered into the middle of May. Today, the water temperature is typically close to 50 degrees by that time of the year.

Figure 2. Observed lake temperature, 1970-2012. Note the increased upward trend present in all months.

As can be seen above, the trends have increased significantly since 1970. Again, the largest increase is in May, and the smallest increase in March. The water temperature trend in May over the period has been an impressive 1.42F/decade! The other months also demonstrate substantial warming, with a trend of +0.53F/decade in January, +0.77F/decade in July, +0.36F/decade in September, & +0.23F/decade in November.