Meteorological summer begins Wednesday, and after a spring that turned out warmer, drier and snowier than average, it’s time to look at what we can expect this season.

I think summer is the most difficult season to predict. The reason: The jet stream is at its weakest and, overall, June through August features warm temperatures. Smaller-scale summer patterns are harder to forecast.

Even a cool summer won’t bring temperatures in the 50s and 60s during the day. In winter, we have the potential for bigger fluctuations and the larger-scale features can be easier, though not easy, to forecast.

A weak jet stream also doesn’t bring a lot of weather, usually just scattered showers and thunderstorms. The question for forecasters is whether the season will be warmer than usual or will there be more outbreaks of Canadian air, keeping the extreme heat at bay?

This summer, most forecasters and models agree, with be hotter than average.

This would mean more days when the temperature tops 90 or reaches at least well into the 80s, and more humidity.

The last two years have brought periods of comparable coolness in the middle of the summer. Most notable were the first two weeks of August 2014, which were remarkably cool, and many nights last summer when temperatures fell to the 40s and 50s, making for a great stretch of sleeping weather.

Hot summers, like those of 2010 and 2013, have more 90-degree days than average, and not just one or two more, but many.  During those summers there were 10 days reaching or exceeding 90 degrees at the Portland International Jetport.  It’s safe to guess that, unless you lived on the water, you experienced even more.

The chart above shows the total number of 90-degree days and is sorted by those with the most. On average, Portland will experience about four such days, with many more inland over the south, but fewer up north and along the immediate coast.

The National Weather Service is predicting the greatest odds are for a hot summer in the Northeast.  This is, of course, an odd forecast, meaning that, although there is a greater than 50 percent chance we are going to have a hot summer, there is still a 20 to 30 percent chance of a summer with average heat.

It’s very important to keep in mind that this is a three-month forecast.  July could end up below average with June and August above, making the summer forecast of warmth valid.

One set of computer models I have found to be quite reliable is the Canadian one.  This model is predicting a somewhat warmer summer, with June forecast to have the highest anomaly or greatest variance above average.

Since June is going to start with seasonable or even below seasonable temperatures, this means either the forecast is wrong or the second half of the month will turn quite hot.

The models are forecasting adequate rainfall or even more than we typically see in a summer.  Before you cancel your week at the beach, consider that this is likely a result of the belief that there will be more heat, because with more heat comes more humidity and a greater chance of afternoon showers and thunderstorms, something that was lacking during the last two years.