Activity 2 Updates – Prediction of Rainfall Extremes at Subseasonal to Seasonal Periods

Research Activity 2 Updates

July 2020

Hello, my name is Melanie Schroers and I am a second year Master’s student in the School of Meteorology. I am currently using former grad student Gregory Jennrich’s extreme-event classification system to look into the impacts and large-scale drivers of those extreme precipitation events that he found. Extreme rainfall events were characterized by their size, number of rainfall days, and rainfall totals in seven regions across the United States. Regions were defined by a latitude/longitude box around areas with similar rainfall regimes (Figure 1). The seven regions are: the Northeast, Southeast, Great Lakes, Southern Plains, Northern Plains, Mountain West, and West Coast. From 1981 to 2019, we found that the Great Lakes, Southern Plains, Southeast, and West Coast experienced one event per year or more, while the Mountain West, North Plains and Northeast experienced less than one event per year.

**Figure 1:** Extreme precipitation geographic regions overlaid with S2S extreme precipitation event threshold values

IMPACTS

To better understand the impacts of these precipitation events, we examined the National Oceanic and Atmospheric Administration storm reports. This database has reports entered by National Weather Service offices all across the nation. Because reporting became more consistent after 1996, we examined events from 1996 to 2018. In Figure 2, we see that these events are associated with higher-than-normal flash flood, flood and heavy rain reports when compared to random 14-day periods. In every region, reports of floods significantly increased during extreme events when compared to random 14-day periods. This result suggests extreme precipitation events can have serious societal impacts related to flooding.

The Mountain West has the lowest frequency of reports, as seen in Figure 2, which could be due to fewer events or the lower amount of rain that is needed to produce an extreme event within this region. Impacts will continue to be assessed as the project progresses.

**Figure 2:** Total storm reports for Flash Flood (left pair), Flood (center pair) and Heavy Rainfall (right pair) in seven regions for our extreme rainfall events (dark blue) compared to a random selection of non-extreme events (light blue). When the dark blue box is taller than the light blue box, the extreme rainfall events had more impacts, as reported by the National Weather Service, than the non-extreme events. (Statistically significant differences (p< 0.1) are indicated with an asterisk (*) next to the storm report labels along the x-axis.) Note: each y-axis has a different scale due to differences in report occurrences.

SEASONALITY

We also examined the seasonality of precipitation events. We found the dry and wet season for each region and compared them to the occurrence of extreme precipitation events. Figure 3 shows the number of events per month from 1981-2018, with the wet seasons in blue and the dry seasons in gray. Most extreme events happen within the respective region’s wet season, with only a few events occurring in the dry season. There is an exception in the Southeast, where there are two distinct wet seasons, which is another pitfall found of using latitude and longitude boundaries to form precipitation regions.

**Figure 3:** Number of extreme events per month in seven regions from 1981-2018. Wet seasons are seen in blue, while dry seasons are seen in grey.

INTO THE FUTURE

We will continue examining the weather patterns of these events and focus on finding common atmospheric environments associated with major S2S precipitation events. Once these patterns are found, days with similar environments will be examined. In addition, finding events with similar patterns and no extreme rainfall, will help lead the way to better predictability and forecasts.