NDG wrote:NotSparta wrote:LarryWx wrote:In looking at EUROSIP for 2005-17, I've just figured out that its forecast put out each May for Nino 3.4 SST anomalies during the heart of the hurricane season, ASO, has a significant warm bias. I plan to post about this in detail as soon as I get a chance.
That's the problem I have with it this year. Since warm neutral is most likely for ASO, its warm bias will push it into El Niño, causing erroneously high pressures.
Absolutely, by it being very warm biased with its ENSO forecast the last couple of years is the reason for its high MSLPs forecast for the Atlantic, it could be the same reason why for this year if we end up with just a warm neutral ASO, especially if it is central base..
Actually, yes I have Eurosip strongly warm biased in 2017 though not in 2016 or 2015. However, then I have it solidly warm biased for most of the years 2014-2005. Here's what I found for how the ASO 3.4 SST anomaly forecasts made in the preceding May by Eurosip verified:
Year: Eurosip's May fcast verification for ASO 3.4
2017: 1.1 cooler (cold neutral occurred vs weak El Niño prediction)
2016: 0.1 warmer (weak La Niña correctly predicted)
2015: 0.1 warmer (strong El Niño correctly predicted)
2014: 0.5 cooler (warm neutral occurred vs weak El Niño prediction)
2013: 0.5 cooler (cold neutral occurred vs warm neutral prediction)
2012: 0.9 cooler (warm neutral occurred vs moderate El Niño prediction)
2011: 1.1 cooler (moderate La Niña occurred vs warm neutral prediction)
2010: 0.9 cooler (strong La Niña occurred vs weak La Niña prediction)
2009: 0.2 cooler (weak El Niño correctly predicted)
2008: 0.8 cooler (cold neutral occurred vs weak El Niño prediction)
2007: 0.2 cooler (moderate La Niña correctly predicted)
2006: 0.2 cooler (weak El Niño correctly predicted)
2005: 0.4 cooler (cold neutral occurred vs warm neutral prediction)
So, of the 13 May forecasts for ASO, only 5 were correctly predicted while 3 were one category too warm, 4 were two categories too warm, and 1 was three categories too warm. None were any categories too cold! The average was 0.5 too warm or one category too warm!
Out of the 7 Eurosip El Niño predictions for ASO in the preceding May, only 3 verified. In contrast, Eurosip predicted only 3 La Niñas in ASO but 4 occurred.
I'm eagerly waiting to see what the May, 2018, Eurosip prediction will be for ASO. It will almost definitely either be warm neutral or weak El Niño. Of the 3 warm neutral predictions, 2 ended up one category too warm while 1 ended up three categories too warm. Of the 5 weak El Niño predictions, 2 ended up verifying correctly while 1
was one category too warm and 2 were two categories too warm.
Out of the 13 predictions, weak El Niño was clearly the most common prediction (5) while warm neutral was 2nd most common (3). Next was weak La Niña (2). The remainder were each predicted once: strong El Niño, moderate El Niño, and moderate La Niña. There were no cold neutral or strong La Niña predictions though 4 cold neutral occurred (the most of any category) and one strong La Niña occurred.
So, of the 13 May predictions for ASO, a whopping 10 were warmer than 0.0 while only 3 were colder. It turned out that only 5 were warmer than 0.0 while 8 were colder.
The average May Eurosip ASO 3.4 prog during 2005-17 was for +0.4 vs the -0.1 actual.