As promised, in this blog post, we'll review the "Summer Weather Preview" and grade its accuracy :-)
Here's the post in question, and I'll bring over some of the relevant parts to here so you don't have to keep flipping back and forth. It's important to remind yourself of all the complexities of seasonal forecasting that I complained about during that last post. They still annoy me, and you'll have an even better idea why after we go through this review.
There are a number of things on which this seasonal forecast depended, so we should also look at how those panned out.
1. North Atlantic Oscillation - The forecast for this grand weather circulation feature was that it was going to turn into its positive phase, but stay below a significant value.
The projection of a weakly positive NAO cycle was incorrect, and most of June through July was spent in the negative realms, though for the most part weakly so. A negative NAO number in the summer-time is considered associated with a more zonal jetstream, given to less actual storm fronts coming through and more opportunities for heat waves (ahem). Generally speaking, the summertime weather in a negative NAO regime would be more air-mass induced with thunderstorms rather than a cold-front / warm-front thing. However, it should be noted that both positive and negative NAO regimes have the propensity to permit warmer-than-average temperature scenarios over the eastern half of the country-- just one does it through large scale weather patterns, and the other through a persistent air mass bubbling of uninterrupted summer sunshine. About the only North Atlantic Oscillation factor that could bring cooler weather to the area in summer-time would be a net-zero number (the weak stuff), though this particular circulation continues to be researched as all is not yet conclusive about it. It definitely has a greater impact on our winter weather than the summer weather.
2. El Nino - This was forecasted to become a weak El Nino event in the summer time, coming out of a period of "La Nina" conditions that persisted since mid-summer of 2011.
El Nino conditions in fact did occur, and edged slightly higher than the still weak +0.5 level that was generally anticipated. Oddly enough, both El Nino and La Nina conditions (particularly strong episodes of either) are correlated with warmer than normal temperatures in Appalachia, but have more dichotomous temperature impacts when looking at the rest of the country as a whole.
Using these two "teleconnection" patterns as guide, it almos seems like every single summer should be a forecast of higher-than-normal temperatures, but as far as precipitation goes, El Nino usually carries with it a lower precipitation expectation compared with a La Nina. [Feel free to browse my Summer Weather Preview post of old for more information on this].
3. So What Happened...
Here's what the normals were:
||Record Low||Normal Precip|
...And the forecast was for warmer and drier conditions when compared with normal. Here's the readout for Charleston and Huntington:
So you can now see some of the complexities and uselessness-es of these types of seasonal forecasts. Indeed I was accurate that the area received a slightly warmer and drier summer on average, but no one would ever agree after going through a July like we did (unless you went through an August the likes of that Charleston hadn't seen ever before). Predicting results based on a summer season that was to be marked by thunderstorm / air mass patterns over that of actual warm front / cold front type systems is quite difficult. Unless you are a farmer or a sea-farer, these types of seasonal forecasts aren't going to be of much use because they simply aren't going to capture the type of resolution you're interested (meaning, it wasn't going to detect a July set of power-grid-destroying heat waves and severe thunderstorms). In a seasonal forecast, that sort of thing is the 'noise', while the 'signal' is what gets averaged out of the whole mess.
If this sort of thing were a trend, unfortuately seasonal forecasts would just continue to be less useful to more people, because it's the variability from average that governs the daily human experience. For example, if this winter was marked with temperatures of 60-degrees and -10 degrees, who cares if the average still was around normal-- because the weather certainly wasn't normal!
The Winter Season forecast is a topic of much discussion I've seen, and I can certainly provide one on the blog... but we'll be issuing a forecast as a weather team in the upcoming issue of WSAZ's digital magazine, as well as giving it as we traditionally do on Thanksgiving day. Also, Josh I believe has his own going up on his blog... So my preference is to not muddy the waters in case we all have different winter forecasts, but I will post one if the demand is there. However, I do want you all to better understand the intricacies of seasonal forecasting, what goes into it, what it can and can't predict, and what you can do with the information.
If anyone has anything to ask about in the comments, feel free to do so, and I'll try to get at them as soon as I can :-)
+ Ashland, KY
+ Athens, OH
+ Beckley, WV
+ Charleston, WV
+ Clarksburg, WV
+ Clay, WV
+ Gallipolis, OH
+ Huntington, WV
+ Ironton, OH
+ Logan, WV
+ Morgantown, WV
+ Paintsville, KY
+ Parkersburg, WV
+ Pikeville, KY
+ Pomeroy, OH
+ Ripley, WV
+ Spencer, WV
+ Sutton, WV
+ Wayne, WV
+ Wheeling, WV
+ Williamson, WV