My print forecast in the Wake Weekly made a lot of sense on Wednesday morning when I wrote it. I thought I might have been a little optimistic about Sunday’s forecast high in the mid to upper 70s. I know my personal bias is caused by my preference for warmer temperatures. However, I didn’t expect my forecast and that of the models I used for it on Wednesday morning to be so drastically different from reality.
The forecast for Saturday was for partly to mostly cloudy skies, a high in the upper 70s, and a chance for showers and thunderstorms with the possibilty that a few storms could be strong to severe. It looked to me like the cold front and its rain would approach early Saturday evening and bring those storms with it.
What happened Saturday was a bit surprising – the front stalled in Virginia, the sun shone here, and the temperature at RDU International Airport reached 87 degrees! I doubt anyone was too angry over a sunnier reality on a Saturday filled with outdoor events across the area. I know I wasn’t. Well, I was not thrilled that the forecast busted, but I was happy with the nicer weather.
The severe weather Saturday evening was mostly constrained to just north of the Virgina border where the stalled frontal boundary sat. By the time it started moving south, the atmosphere had cooled and lost its instability, which led to more rain and less thunder.
My forecast busted on the cool side yesterday. I had predicted a rainy day, but the rain and the winds from the east kept the temperature 20 degrees cooler than it looked like it would be several days before.
As I beat myself up a bit over the way my printed forecast did not verify, a friend pointed out to me, “Most people don’t expect weather people to be right. They just want some guidance on what to expect for the next few days.” He told me that he appreciated my personal need to be accurate, but he didn’t think accuracy was all that expected.
While I appreciate his sympathetic thoughs, I disagree. Accuracy should be expected. Maybe not on a weekend like this recent one — when the speed of a cold front’s movements ultimately determined the temperatures over two days and there was not a huge amount of confidence in the models three days in advance — but it should be expected.
Meteorologists’ accuracy over the long term is what gains the public’s trust. Without that trust, we can’t expect them to take us seriously when we forecast a chance for severe weather or a winter storm or a hurricane making landfall several days out. While most of the time we are actually right, it’s those times that we are not that the public seems to recall most. It’s human nature, and it is something we have to work hard to overcome — both as forecasters and as humans.
The best way for a forecaster to improve is to spend time picking apart a forecast that didn’t verify and figuring out what happened. Sometimes, there is not much to blame other than the very models we rely on to make the forecasts. Other times, we can see trends in hindsight that we should have recognized in advance. So, that is why this morning, I asked myself the question that I expected to receive from my readers. “What happened with the forecast?”