“Snowman” by torange.biz is licensed under CC BY 4.0
It was a wild week -- perhaps the weather itself couldn't stand the pressure and simply folded at the last minute. The general conditions were favorable for a period of snow sometime Friday morning into Friday afternoon, to be followed sleet and rain. It seemed like it would snow for the first time this year.
However, our forecast model consistently indicated that precipitation would not arrive until after the cold air had exited the region. Precipitation would arrive, but only at night, and it would be rain. (Some folks did see some snow in Jerusalem at night, and there were light snow showers here in Efrat during the day, so these prior forecasts were very accurate.) Then, disaster struck. As part of a forecast I made to predict short term weather changes, I looked at what would happen just several hours later. Much to my surprise, the forecast model indicated a snow accumulation. The precipitation as shown in the forecast would arrive at the peak of the time of coldest air aloft, which would lead to "convective" snow events. Unfortunately, it was not to be. I went back after the fact and reran another forecast with an improved representation of melting and cloud processes. This is the forecast model I usually use for longer forecasts. It still showed snow, but no accumulation. Still, it too brought the precipitation earlier than actually occurred (based on an incorrect Global Forecast System (GFS) forecast of the storm position). It looks like two or three things went wrong Friday morning: a) the best forecast version of the model was not used; b) the GFS incorrectly forecast the position of the storm closer to us than it actually was (it was further west and south); and c) I was rushed and did not properly review all the data the model was producing. We already spoke about the forecast model. The incorrectly forecast track of the storm meant that the storm took longer to arrive. This was an issue we spoke about last week, so that the precipitation was "delayed." The fact that I was very rushed (it was Friday morning and there was a client who needed personal service) and the surprise of seeing snow in the forecast meant that I issued my "winter storm warning" without completely reviewing the data. Had I taken more time, I might have noticed that the temperatures were just marginal for snow (and its possible accumulation), meaning at freezing or even above. I might then have tempered my forecast (enthusiasm). You can imagine that after all this I and the weather are both quite tired. In fact, there is very little weather this coming week. In fact it should be increasingly mild (and sunny -- besides a few possible showers on Monday). The next storm system may not arrive until early next week.
Image credit: The Jerusalem Herald
When I was in sixth grade, there was a forecast for rain. It snowed. The next day, there was a forecast for snow -- again. This rarely happened in the 1970s. We'd have snow one week, rain the next, maybe snow the next. So, this was pretty exciting. After sleeping with difficulty, I woke up in the morning to to see that we'd had just rain, which had almost washed away all of our snow. I was so disappointed that my teacher sent me to the library to learn about the weather. I've been learning since (with a break to receive a liberal arts degree from Oberlin College). Unfortunately, despite my best efforts, I still make mistakes, and if I don't, then the forecast models do. So, even though I have a high success rate predicting snow, its the last forecast people remember; I can understand those that are disappointed and those who are just happy it didn't snow. It's a lot of pressure, and I am not sure I want to continue making such difficult forecasts. I don't sleep the night of the storm, and if I bit my nails they'd be down to the bone. Maybe it’s time for some other sixth grader to build the next snowman?
Dr. Lynn is a lecturer at The Hebrew University of Jerusalem, Earth Sciences Department. He is also CEO of Weather It Is, LTD, a company that specializes in reducing weather risk.