Seasonal forecasting is one of the more tedious and intricate tasks for meteorologists. Not only is it difficult to predict by nature (no pun intended) — but forecast model accuracy skill is greatly reduced at that range. Often, meteorologists find themselves looking back towards past events, and analogs, for help with predicting the months ahead. In our case, interest was piqued when we noted what seemed to be higher than normal frequency of high latitude blocking this year. Hurricane Sandy, the nor’easter which immediately followed, and many events this winter featured patterns that were driven by blocking patterns over Canada, Greenland, and the higher latitudes towards the pole. The blocking pattern was not overly anomalous at least on a per-year basis, but it certainly seemed to be more frequent than the calendar year which preceded it.
As we look forward to May, forecast models are in agreement on the continuation of higher than normal height anomalies at 500mb over Central and Eastern Canada as well as farther north towards Greenland. The continuation of blocking in the higher latitudes, relative to our location, can offer some interesting insights into the summer temperature forecast as we move forward. Is there a common theme amongst the historically warm and historically cool summers in the NYC Area — and can we relate it to Springtime blocking patterns? The answer, may surprisingly to some, is yes. For the sake of example, take the summer of 2009 and the summer of 2010. Both of these years featured wildly different patterns, with 2009 remaining very cool throughout the summer while 2010 was warm. In 2009, there was very little blocking observed from March through May. 2010, on the other hand, featured periodic blocking in the high latitudes from February through May. Interested yet?
The data becomes even more compelling once you take a look at the 15 warmest and 15 coolest summer in New York City over the last 30 years. Using the 15 warmest summers in New York City, we can back track to the May which preceded that summer and analyze the data at 500mb. The height anomalies show a bit of a striking picture — high latitude blocking, and plenty of it. Also, notice the ridge north of Hawaii which typically teleconnects with warmer summers not only in our area but in much of the Eastern United States. The exact temperature departure from the analog years in New York City was +1.5 degrees from normal. The image below shows the aforementioned information, with the 500mb height anomalies of the May preceding the 15 warmest summers of the last 30 years on the left, and the composite temperature anomaly of the summers that followed on the right.
The analog works both ways, as well, with the 15 coolest summers in New York City thrown into the data pile. When one analyzes the data from the 15 coolest summers in NYC, a temperature departure of exactly -1.3 degrees is observed. However, a closer look at the May 500mb pattern which preceded those summers reveals some interesting information once again. High latitude blocking is completely absent. Instead, lower than normal height anomalies are spread throughout much of the higher latitudes. In addition, the ridge north of Hawaii, which was present during the hotter summers in New York City over the last 30 years, is not there — it is replaced by a mean trough and negative height anomalies. The image below shows the 15 coolest summers in New York City and composite temperature anomalies across the United States (right), and the 500mb height anomalies and pattern during the May which preceded those summers (left).
So, what does this tell us? That the amount of blocking over the high latitudes during the month of May, over the last 30 years, has proven to be (on average) a relatively accurate predictor of summer temperatures in New York City. May’s which feature high latitude blocking average a summer temperature departure around a degree above average, while May’s which feature little or no high latitude blocking averaged a summer temperature departure of about a degree below average.
Although we’ll have to wait a week or two to see exactly where the average height departures for May are heading, we can at least take an amateur glance at long range forecast models to get a “first look” at what May looks like during the initial 5 to 7 days. The GFS and Euro models are in decent agreement that, on average over the next 8-10 days, 500mb heights over the high latitudes will average below normal. Also, as a point of reference from the above data, check out the area north of Hawaii. The 8-10 day forecast on both the GFS and Euro matches the “cooler” summer composite for May fairly well.
That being said, the 8-10 day forecasts are mostly trivial at this point, at least for our purpose. We’ll have to check back in a few weeks to see where the average amount of blocking in May is actually heading this year. Nevertheless, given the data above, what we see at that point should give us a decent idea where we’re heading in the summer months to come.