APRIL 2020 FORECAST - How did we do?
By Justin Poublon
May 5, 2020 - 10:01 AM CST | 36 0
Only two months in. March went very well (see review). April was not the same level of success however I've learned a few things.
My main goal is not to be totally wrong; for example avoid precipitation A++ when it's B--. As I develop this method I may use near average if the signals are not strong enough or I'm uncertain. I wasn't certain enough to draw in a below average precip zone considering wetter tendencies on the last few years.
The temperature forecast across Wisconsin was pretty good. I didn't expect the colder temps across the Midwest which may have something to do with a lack of southwest flow/eastern US ridging. I thought +AMO would have a stronger influence which it did not. But at least I was right across Wisconsin.
The precipitation forecast could have been better. Happy that I hit the above avg forecast in northern Michigan! Disappointing to miss that drier than average zone in western WI. Initially I had this precipitation forecast nailed but I backed off thinking I was too biased towards the AO. This was my FEB 14th prediction...
"FEB 14 - If we take +AO Feb trend into Apr we see some interesting anomalies in the models. A tendency towards below average temperatures in the Great Lakes region. Below average precip in western/northern WI..."
Here is what I predicted on MAR 1...
"The maps are based on top analog years of 1989-90 which were colder, dry(er) than average. Coming out of March which will be wetter, warmer than average. Perhaps well below average precipitation in April? Latest MJO forecast is heading from 8 to 3-ish range which is on the colder side of things."
Predicted the snow event in mid-April on MAR 31...
"It's not out of the question to have a snow event somewhere in here if the AO goes negative...mid-month maybe?"
I can do better being more consistent carrying initial predictions/trends through the entire forecast and being clearer on my final prediction.
Today I'm testing ways to add more detail to the monthly forecasts by comparing top analog years on a daily/weekly basis. From this I predict we can say what the weather could be like week to week within that month. I feel like I'm on the right path.