Week 2 Severe Weather Forecasts: What Should We Predict?

Sophie M. Miller, Michael E. Baldwin, Kimberly A. Hoogewind, and Harold E. Brooks

What is already known:

• Severe convective storms are a threat to life and property each year in the U.S., and they account for >50% of all $1+ billion-dollar weather disasters
• There is a growing interest in forecasting severe storms at longer lead times, including the week 2 time frame.
• Beyond day 8, no operational severe weather forecasts exist from the National Weather Service.
• Some studies have shown skill in forecasting weekly severe weather activity at the week 2–4 time period.
• There is no consensus on how severe weather frequency on a weekly basis should be quantified nor what should be predicted in a week 2 forecast.

What this study adds:

• This work explores various ways to quantify severe weather frequency during rolling 7-day periods based on the number of days per week that exceed a certain coverage threshold as defined via the practically perfect hindcast (PPH) framework.
• Monthly climatologies of weekly severe weather day frequency demonstrate that the most active weeks occur during late spring/summer.
• More frequent days of severe weather during a week tend to be higher impact, with the biggest day in the week encompassing a larger area, more frequent days with significant severe weather, and are more likely to lead to a greater number of injuries and fatalities.
• The frequency of weeks that include big tornado days peaks earlier in the year compared to more active weeks that include any type of severe weather, so hazard-specific weekly forecasts may need to be tailored to specific times within the year.
• The annual cycle informs when week 2 forecasts should be made, given a greater number of days occur during a week in the late spring/summer, while the cool season experiences much fewer events during a week.