Verification:
https://www.blogger.com/u/1/blog/post/edit/2820149554058213224/3271427891031056834
Final Grade is a D+.
Here is the brief version:
Abstract
Although the 2019-2020 seasonal forecast did have some forecasting victories, such as the development of an el nino, a cold and snowy start to the season in December and the rapid warm up as the holidays approached, it ultimately left a great deal to be desired. The 1969-1970 season was utilized as a primary analog with the notable caveat that not as much seasonal blocking was anticipated in the aggregate. But rather the parallel between the two seasons was made due to the postulation that the winter of 2020 would eventually develop that same proclivity to develop high latitude blocking on the heels of another meager el nino, which the 1969-70 season did. However, this season exhibited that one month that would feature the propagating poleward heat flux necessary to induce significant blocking during the second half of the season, as opposed to the 1969-70 analog, which featured the most stout blocking in January. The very mild middle of the season was expected, but it was even more protracted and warmer than forecast. Couple this with the fact that the crucial development of blocking took several weeks longer to materialize than forecast and what had began as a very promising outlook devolved into our worst yet. And by a wide margin at that.
December-March Forecast Review
Here is the DM period:
What could be gleaned from this is that there would usually be a cold source expected to at least be nearby even in the event that sustained blocking did not materialize, which it did not, until spring.
D-M Seasonal Composite
December-March Outcome
D-M Outcome
The details for this past season are as follows
Snowfall was under forecast at none of the 15 cities.
Snowfall was over forecast at all 15 cities.
Snowfall verified at none of the cities.
City
Predicted Snowfall
Actual
Forecast Error
Boston, MA
45-55”
15.8"
+185%
New York, NY
20-30"
4.8"
+317%
Philadelphia, PA
17-27"
0.3"
+5,667%
Baltimore, MD
15-25”
1.8"
+733%
Washington, DC
15-25”
0.6"
+2,400%
Albany, NY
65-75”
49.7"
+31%
Hartford, CT
50-60”
29.1"
+72%
Providence, RI
40-50”
13.2"
+203%
Worcester, MA
70-80”
44.9
+56%
Tolland, CT
60-70"
31.4"
+91%
Methuen, MA
65-75”
44"
+48%
Hyannis, MA
20-30”
5"
+300%
Burlington, VT
80-90"
69.6"
+15%
Portland, ME
70-80”
59.6"
+17%
Concord, NH
65-75”
46.5"
+40%
Index Value
Predicted '19-'20 DM Value Range
Actual '19-'20 DM Value
Forecast Error
Pacific Decadal Oscillation (PDO)
0 to .30
-.91 (DF)
+.91
Perennial North American Pattern (PNA)
.15 to .45
-.57
+.72
ENSO
Weak Modoki El Nino (0.3 to 0.5C ONI) (NDJ)
0.6C JFM ONI
.52 Modoki
Predicted Peak two months early, and .01 too cold. Modoki value +.02 greater than forecast.
(J-M) East Pacific Oscillation (EPO)
-.15 to -.45
+.65
+.80
Arctic Oscillation (AO)
0 to -.30
+2.22
+2.22
North Atlantic Oscillation (NAO)
-.15 to .15
+1.20
+1.37
The mean forecasting error with respect to snowfall was an unfortunate 678.3%, which is by far the worst snowfall outlook dating over six seasons.
Final Grade: D+
The mean forecasting error with respect to the previous seasonal snowfall outlooks are as follows:
2018-2019: 90.4%
2017-2018: 13.2%
2016-2017: 35.3%
2015-2016: 71.4%
2014-2015: 10.9%
Six season running mean error: 149.9%
24/86 (27.9%) of snowfall outlooks have been under forecast.
46/82 (56%) of snowfall outlooks have been over forecast.
16/82 (20%) of snowfall outlooks have verified within the forecast range.
This just in: Forecasting snowfall at very extended leads is difficult.
