Apr 13, 2012

Oshkosh Winter Prediction - Cycle 3 Performance Review

Back in late November when I pulled the first cycle data together to produce the Oshkosh Winter Forecast I banked on the cycle averaging 48 days per cycle. As we LRC followers know the cycle breathes and is never a static length. To this day the cycle in 2011-12 is averaging 46 ish days. This obviously means the dates and times of month I chose to pin point events is skewed slightly. With that said, the original forecast is still comprehendible with minimal self interpolation.

The original forecast called for the cycle 3 forecast period to fall between Feb-20 through Apr-7. In actuality it was Feb-17 through Apr-3. The OSNW3 LRC Calendar can shed more light on the cycle duration. Below I compare the OSNW3 surface observations between Feb-17 through Apr-4 against the LRC based forecast conditions for Feb-20 through Apr-7 made in late Nov and analyze how they have performed in forecast cycle 3.

Prediction
Days with Measurable Snow (0.1" or greater)
Forecast Cycle 3 = 4 days (Feb-20 thru Apr-7)
- Actual
Actual Cycle 3 (Feb-17 thru Apr-3)
- 6 days with 0.1" or greater snowfall.
- 66% accuracy

Prediction
Total Snow Accumulation
Forecast Cycle 3 = 9 inches of snow (Feb-20 thru Apr-7)
- Actual
Actual Cycle 3 (Feb-17 thru Apr-3)
- 9 inches of snow
- 100% accuracy

Prediction
Temperatures
The data trend leads me to believe that there will be enough warm spells to compensate for the majority of cold air events within each cycle leading to Above Average temperatures for the winter. The numbers tell me 1 to 3 degrees above average each cycle. Don't get me wrong, there will be some cold stretches this winter, there is no doubt about it, but with warm-ups scattered about may make the overall winter not seem so harsh temperature wise.
- Actual
- Daily Maximum temperatures were an astonishing 12 degrees above the (81-10) 30 year average for the actual cycle 3 period.

Note on temperatures... using the LRC I was convinced temps would be above average while others were calling for a cold winter. The calculated average temp from cycle 1 was 3+ degrees above average. I biased the prediction on the low side because the pros were calling for colder than average. I may have learned a lesson. The LRC data doesn't lie and it is a very powerful forecasting tool.

----

What should we expect in Cycle 4?
Sticking with the original forecast numbers made back in November for cycle 4, we should expect about 2 days with measurable snowfall accumulating around 1 inch. Temperatures should stay above average for the cycle once again. With a bump or two in the road, spring will continue to amaze and May should begin with a summer feel to it. A cool and wet middle part of May will give way to more summer like weather. While the winter predictions will end the LRC trends will continue to project our future weather. Expect a warm Memorial Day and a very warm end to June. When using the LRC forecast trend below remember to give or take a day, or two!

(Temperature & Precipitation Forecast Trend For Oshkosh)


If there are any questions or thoughts on my research and analysis of the LRC or how I presented the material just let me know in the comments section of the blog. Thanks for reading!

3 comments:

  1. Excellent blog. I have been following the LRC closely for the last couple of years. I used to work for FOX 11 in Green Bay and now work in Minneapolis for a private forecasting company. I recently did a temperature analysis at KMSP for each cycle. I found a phenomenal correlation. I would love to chat about this more. Send me an email at asthut@gmail.com

    ReplyDelete
    Replies
    1. Andrew, thanks for stopping by. I recall you using the LRC last year at FOX 11. Seems they decided to stop after you left. Surface trends are fun to watch and project using the LRC. I wish you great success in MSP! Looking forward to further communications.

      Delete
  2. Kinda fun, isn't it. I love the graphing! Nice write up and we will see what this new cycle brings to us this year.

    ReplyDelete