A Climatological and Keyword-Based Analysis of National Weather Service Mesoscale Precipitation Discussions

April 17, 2026

A new paper entitled “A Climatological and Keyword-Based Analysis of National Weather Service Mesoscale Precipitation Discussions” was recently published in the National Weather Association’s Journal of Operational Meteorology. This research was led by Samuel Bartlett (CW3E), with co-authors Erica Bower (CIRES/NOAA/NWS), Jason Cordeira (CW3E), David M. Donahower (RIT), Eric Hoffman (PSU), James Nelson (NOAA/NWS), Emily Slinskey (CW3E), Matthew Steen (CW3E), and Marcus Watkins (TAMU). This study emerged from a summer research project by authors Donahower, Watkins, Cordeira, and Hoffman during the National Science Foundation (NSF) Northeast Partnership for Atmospheric and Related Sciences (NEPARS) Research Experience for Undergraduates (REU) in 2021 and was completed by authors Bartlett, Steen, Slinskey, Cordeira, Bower, and Nelson thereafter.

Flooding is one of the most hazardous weather-related phenomena in the United States, with flash flooding being a particularly dangerous form of flooding generally caused by short-duration, high-intensity precipitation. In April 2013 the National Weather Service (NWS) Weather Prediction Center introduced the Mesoscale Precipitation Discussion (MPD) product to address the forecast challenges associated with these events across the lower 48 states. MPDs are issued by NWS Weather Prediction Center forecasters at the Meteorological Watch Desk (MetWatch Desk) and are valid for areas of the lower 48 U.S. states that are at risk for flash flooding or heavy rainfall in the subsequent ~6–12 hours. These products cover an area that is approximately 200,000 km², and are valid for up to six hours apart from those related to ARs, which can be valid for up to 12 hours. MPDs contain both a graphical analysis, which highlights the area of expected impacts on a map, and a text summary which includes a discussion of the ingredients associated with the potential flash-flood producing storms.

This study evaluated the frequency, spatial distribution, and storm types associated with short-term forecasts of potential flash flood events in the U.S. from a catalog of MPD products issued by the NWS Weather Prediction Center during 2013–2023. Spatial and temporal data was accessed using the issuance information for each MPD and keyword data was extracted from the text body of each MPD. The spatial distribution of MPDs over CONUS was analyzed using a 0.25° latitude × 0.25° longitude grid and by NWS county warning areas (CWA). During this period, the highest frequency of MPDs occurred in CWAs over the Central U.S. with a secondary maximum over the Desert Southwest (Fig. 1).

The text of each MPD was queried for keywords from a variety of storm-type related keyword sets (e.g. Atmospheric River, Organized Convection, Organized Tropical, Orographic, Monsoon, Convective, Front, Cold Front, Warm Front, & Stationary Front) to determine the storm types responsible for the flash flooding conditions which necessitated the MPD issuance. The spatial climatology of “Convective” (Fig. 2a) MPDs contain the highest frequencies over the Shreveport and Jackson CWAs. The distribution of “Organized Convection” (Fig. 2b) MPDs are similar to the “Convective” MPDs (although shifted slightly to the west over Texas/Oklahoma), whereas the “Orographic” (Fig. 2c) MPDs are most frequent over the Flagstaff CWA. Secondary maxima in “Orographic” MPDs are well-aligned with the primary topographic features over the U.S. with highest frequencies over the Appalachians, Rockies, and Sierra Nevada. The distribution of “Atmospheric River” (Fig. 2d) MPDs favored U.S. West Coast locations, “Monsoon” (Fig. 2e) MPDs are favored Desert Southwest locations, and “Organized Tropical” (Fig. 2f) MPDs favored Gulf Coast locations.

There is a well-established connection between ARs and high-impact weather along the US West Coast, and a growing body of literature linking ARs to high-impact weather over the Eastern US. For the purposes of this study, two methods are chosen to identify ARs and are described below. The centroid latitude and longitude and the start and end time of each MPD that occurred between 1 January 2016 and 31 December 2023 are used to create a space-and-time window to assess the association of a given MPD with ARs defined from the two methods outlined above.

Atmospheric river related keywords were seldom used in the text of MPDs, with the highest occurrence in county warning areas in the Western US. This distribution is mostly by design, as the Weather Prediction Center guidance on using the term “Atmospheric River” and its related keywords is primarily reserved for events along the U.S. West Coast between September and April. However, many additional MPDs across the U.S. occur with a combination of ingredients that may also be described by the synoptic scale environment of an AR. Overall, between 1 January 2016 and 31 December 2023 the “Atmospheric River” keyword set was used in only 2.9% of MPDs, whereas the tARget and the AR Scale ARDTs identified 43.7% and 60.6% of these MPDs as AR-related, respectively (Fig. 3). These results suggest that AR-related forecast diagnostics and the environments they describe may be useful for ingredient-based operational forecasting of flash flood conditions for various storm event types in locations beyond the U.S. West Coast.

This research supports the “Atmospheric Rivers and Extreme Precipitation Research, Prediction, and Applications” priority area of the CW3E’s 2025-2029 Strategic Plan. In particular, it fulfills the key objectives to 1) expand research and forecast verification efforts on ARs and the different storm types responsible for extreme precipitation and 2) Work with strategic partners in atmospheric sciences to improve the understanding of atmospheric rivers and their associated impacts. Additionally, these activities are well aligned with the objectives of NOAA’s Precipitation Prediction Grand Challenge NOAA’s Precipitation Prediction Grand Challenge. Among other objectives, the evaluation of the MPD product, its verification, and potential future improvements specifically serve to enhance and sustain user engagement (Objective 1), to improve precipitation prediction products and applications (Objective 2), and to understand precipitation predictability, its sources, and barriers (Objective 6.2) in the NOAA PPGC.

Figure 1. Frequency of MPDs overlapping each 0.25° grid cell during (a) all months, (b) winter, (c) spring, (d) summer, and (e) fall. The frequency is shaded according to scale. Figure 4 from Bartlett et al. 2026.

Figure 2. Frequency of MPDs containing at least one of the keywords or keyword sets in Table 1 summarized by CWA for (a) “Convective”, (b) “Organized Convection”, (c) “Orographic”, (d) “Atmospheric River”, (e) “Monsoon” and (f) “Organized Tropical”. The frequency is shaded according to scale which varies among panels. Figure 5 from Bartlett et al. 2026.

Figure 3. Frequency of MPDs associated with ARs between 1 January 2016 and 31 December 2023, summarized by CWA using (a) the MPD keyword methodology, (b) the Guan and Waliser tARget ARDT, and (c) the AR Scale ARDT, and (d) the relative frequency of MPDs in a given month associated with ARs using the aforementioned methods. Figure 8 from Bartlett et al. 2026.

Acknowledgements:

This study emerged from a project funded by the National Science Foundation (AGS-1757009) as part of the Northeast Partnership for Atmospheric and Related Studies (NEPARS) led by authors DD, MW, JC, and EH. The project was completed by authors JC, SB, and MS with funding at the Center for Western Weather and Water Extremes (CW3E) at the University of California, San Diego Scripps Institution of Oceanography from the National Oceanic & Atmospheric Administration (NOAA) awarded to the Cooperative Institute for Research to Operations in Hydrology (CIROH) through the NOAA Cooperative Agreement with The University of Alabama (NA22NWS4320003). The statements, findings, conclusions, and recommendations are those of the author(s) and do not necessarily reflect the opinions of NOAA.

Citation:

Bartlett, S. M., Bower, E., Cordeira, J., Donahower, D. M., Hoffman, E., Nelson, J., Slinskey, E., Steen, M., & Watkins, M. (2026). A Climatological and Keyword-Based Analysis of National Weather Service Mesoscale Precipitation Discussions. Journal of Operational Meteorology, 14(7), 99–109. https://doi.org/10.15191/nwajom.2026.1407