1. Introduction
The study of extreme precipitation events has been an interesting topic in hydrologic research due to their socioeconomic role in communities worldwide. The effects of extreme precipitation events on society are evident and diverse. They can produce natural disasters such as landslides and floods or be a main charger for depleted water resources (Paltan et al. 2017). As such, a better understanding of extreme precipitation behavior is required to maximize benefits, sustain water resources, and improve resilience for communities. The effects of global warming have generally impacted the precipitation patterns worldwide and have a more consistent relationship with extreme precipitation events (Trenberth 2011). As a result, the occurrence of extreme precipitation events may double with each temperature degree increase if historical trends continue (Myhre et al. 2019). In addition, other local factors can cause changes in extreme precipitation events such as land-use changes, soil moisture, and atmospheric moisture transport from remote areas (Vázquez et al. 2020). For example, Li et al. (2019) showed that atmospheric circulation caused regional changes to precipitation behavior as warming ocean and land enhance evaporation and precipitated moisture transport. Further, Vázquez et al. (2020) studied the contribution of global sources of moisture transport to the mean and extreme precipitation during the peak month of precipitation and found a general reduction in the global influence of these sources. Based on the published literature, the critical role of atmospheric moisture transport in the extreme precipitation characteristics over global and regional scales motivates this work to consider the effect of moisture sources on the associated heavy precipitation events.
One of the extreme precipitation characteristics that has received attention in recent literature is seasonality or timing of the extreme precipitation events (Pryor and Schoof 2008; Pal et al. 2013; Dhakal et al. 2015; Mallakpour and Villarini 2017; Marelle et al. 2018; Henny et al. 2023; Aljoda and Dhakal 2023; Dhakal et al. 2023). Seasonality of extreme precipitation is a climatic phenomenon that has an active influence on the hydrologic processes and water resource sustainability (Pal et al. 2013). Characterization of seasonality of extreme precipitation is crucial to a wide range of engineering decision-making, including decisions related to stormwater management and water allocations for urbanization, agriculture, and ecosystem services (Pal et al. 2013; Dhakal et al. 2015; Mallakpour and Villarini 2017). Our recent study (Aljoda and Dhakal 2023) examined the timing of AMP events and determined the strong seasonality and their nonstationarity by using a comprehensive circular statistical framework for the eastern United States in 1950–2019. We concluded that 90% of the study sites experienced extreme precipitation during at least two seasons over the year. Therefore, the same framework is used here to identify the seasonality (season with strong signal) of extreme precipitation. Seasonality of extreme precipitation might be impacted by the regional pattern of the seasonal weather and atmospheric moisture pathways often determined by the large-scale, general circulation of the atmosphere (Hirschboeck 1988). To our knowledge, there is still a lack of studies that focus on understanding the moisture pathways and determining the sources origin associated with extreme precipitation seasonality for the eastern United States.
Atmospheric rivers (ARs) are narrow, filamentary regions of enhanced water vapor transport (Newell et al. 1992). Numerous studies have documented the contribution of ARs in the extreme precipitation events in local and regional scales across the globe (Eckhardt et al. 2004; Knippertz and Martin 2007; Knippertz and Wernli 2010; Neiman et al. 2013; Pfahl et al. 2014; Rutz et al. 2014; Lavers and Villarini 2013, 2015; Alexander et al. 2015; Barth et al. 2017; Aljoda 2021). Mahoney et al. (2016) indicated that based on previous studies, heavy rain events (72 h > 500 mm) are most likely to be caused by ARs in the Southeast and West Coast of the United States. While the contribution of ARs to extreme precipitation is well documented in the West Coast regions of the United States, few and limited research studies have explored heavy rainfall events associated with the atmospheric moisture transported from other water vapor sources like the Atlantic Ocean, the Gulf of Mexico, and the Caribbean Sea (Moore et al. 2012; Mahoney et al. 2016). Aljoda (2021) showed that ARs have widely contributed to floods caused by rain or rain on snow in many locations across the United States by delineating the origin of moisture sources that controlled these extremes. Thus, it is expected that ARs would also have influences on extreme precipitation across the country and they have the same origins as those causing the nationwide floods. Agel et al. (2015) defined extreme precipitation in a grid point as the top 1% wet days and found that 75%–100% of the four seasonal extremes in the Northeast United States in 1979–2008 were caused by the synoptic extratropical storms. A recent study of extreme precipitation in the mid-Atlantic and Northeast United States between 1979 and 2019 showed that there are significant changes in frequency and intensity of precipitation days associated with ARs during the four seasons (Henny et al. 2023). As noted earlier, there is no study up to the time of authoring this article that examines the seasonality of extreme precipitation associated with ARs and attributes the observed seasonality to the origin of AR moisture. Such work is crucial for weather forecasting to develop flood adaptation strategies and improve community resilience under climate change conditions.
This study focuses on characterizing the seasonality of extreme precipitation caused by ARs by delineating the constituent remote atmospheric moisture sources. To do so, we utilize daily precipitation records across the eastern United States to explore the following objectives:
-
Identify the extreme precipitation events induced by ARs and explore their regional and spatial patterns.
-
Characterize the seasonality or timing of extreme precipitation events associated with ARs and investigate the spatial patterns of significant seasonality modes.
-
Delineate the AR moisture tracks associated with extreme precipitation events and identify their sources of origin.
-
Identify the role of moisture sources in the seasonality of these extreme events.
Achieving these objectives provides a better understanding of the typical timing and frequency of ARs associated with extreme precipitation in different seasons that improves the accuracy of forecasting models and predictions. This also allows for better preparedness and timely warnings for areas at risk of flooding. As such, authorities can enhance flood monitoring systems and implement temporary flood defenses in vulnerable areas. Such knowledge helps communities in implementing resilient infrastructure, updating building codes, or enhancing drainage systems in anticipation of peak precipitation seasons.
2. Materials and methods
a. Study area
Our study area covers the eastern United States (eUS) that is divided into five climate regions (Fig. 1) based on NOAA classification. The total number of stations used here is well and sufficiently distributed over the five selected regions (Fig. 1) as follows: Central or Ohio Valley (134 stations), East North Central or Upper Midwest (87 stations), Northeast (95 stations), South (150 stations), and Southeast (117 stations).
The eUS with 583 Historical Climatology Network precipitation stations across the study region. Different colors represent five climatically consistent regions within the eUS [Central (Ohio Valley), East North Central (Upper Midwest), Northeast, South, and Southeast] (https://www.ncdc.noaa.gov/monitoring-references/maps/us-climate-regions.php).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
b. Data
1) Daily precipitation data
We utilized 752 stations of daily precipitation data (1-day or 24-h total) for the period 1950–2021 (n = 72) from the Global Historical Climatology Network (GHCN) database (Menne et al. 2012). To maintain the quality control of the data, we used two screening criteria (Aljoda and Dhakal 2023): 1) in each year, the percentage of days with missing values should be no more than 10%, and 2) for an individual station, precipitation record length should be greater than or equal to 60 years. Based on these filtering criteria, 583 stations across the eUS were retained for analysis (Fig. 1). An extreme event of daily precipitation is defined by using the block maximum approach. For any selected location, an AMP event was described as the maximum value of 1-day (or 24-h) rainfall totals within each year.
2) AR data
For the AR analysis, two global products of reanalysis datasets are used to utilize the AR shape and intensity at each life stage. The AR database used to extract the AR shape index (https://ucla.box.com/ARcatalog) detects ARs based on integrated water vapor transport (IVT) intensity, direction, and geometry (Guan and Waliser 2015; Guan et al. 2018; Guan and Waliser 2019). The criteria to detect AR are i) IVT intensity at each grid cell must be greater than max (85th percentile, 100 kg m−1 s−1), whichever is larger; ii) mean IVT over the AR should be within 45° of AR shape orientation and with an appreciable poleward component (i.e., 50 kg m−1 s−1); iii) AR length must be greater than 2000 km and the length-width-ratio must be greater than 2; and iv) as the refinement for less well-structured ARs, requirement i is repeated for up to five times, each time with an increase of 2.5 in the IVT percentile threshold if requirements ii and iii fail. The shape index is a unique number given to each grid point within the same AR event observed around the world at a single life stage to distinguish between the available AR events at the same time step. The AR data used are based on the NCEP/NCAR reanalysis, offering 2.5° × 2.5°, 6-hourly (6-h) record over 72-yr period (1948–2021). For the magnitude of 6-h IVT, we used the IVT time series (https://www.inscc.utah.edu/∼rutz/ar_catalogs/) constructed by Jonathan Rutz based on the NCEP/NCAR reanalysis, offering 2.5° × 2.5° resolution and 6-h record over the same period.
c. Methods
1) Seasonality of extreme precipitation
2) Detection of AR-associated AMP
To detect AMP caused by AR, first we created the station effective area polygon (SP) by drawing a circle of 500-km radius centered at the site’s location (Fig. 2) to determine which grid points from the atmospheric data variable (i.e., AR shape and IVT) are included within the SP boundaries. This radius is based on visual assessment of contiguous precipitation areas and is a smaller domain comparing previous studies (Barlow 2011; Kunkel et al. 2012; Pfahl and Wernli 2012; Agel et al. 2015; Dhakal and Jain 2020) to maintain the accuracy of attributing the AMP event to an AR mechanism. However, using a radius of 250 km reduced the total AMP events associated with AR by less than 2% while the total AMP-AR events increased by 7% and 12% for using 750- and 1000-km radius. Daily time series for each atmospheric variable in each station polygon are constructed by averaging or taking the maximum of the 6-h values for the assigned grid points. We used the exact date of the AMP event to identify the association between AMP and AR (AMP-AR event) instead of time window to maximize the result accuracy. The time series of AR shape (polygon) are examined on the AMP date to detect if there is an AR event stalling over the circle (SP) or not. To this end, an AMP event is defined as caused by AR when the time series of picked grid point(s) for AR shape observed the unique number (n > 0) which is determined based on meeting the AR conditions (Guan et al. 2018) at that specified day of AMP. Each AR shape polygon has its own unique number; therefore, a single AMP event can be associated with more than one AR event. As a result, the AR shape polygons with the selected unique number are retained for the analysis of AR track and origin detection in the next section.
AMP event on 14 Aug 2014, caused by ARs for the station USC00190120 located in Amherst, MA. Black circle represents the effective area (SP) centered at the station with a radius of 500 km; the intersection of SP with AR shape polygon points (colored circles) caused AMP-AR event at the AMP date. The black line represents the track of the AR event that caused AMP from the stage of birth to its end.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
3) Detection of AR pathways and origins of moisture for AMP
Examples of the major source of moisture of AMP-AR events; (a) the PO which consists of TEP, TCP, sTP, and eTP; (b) the AO which includes TAO, sTA, and eTA; (c) the CS and the GM; and (d) local moisture that originated inside the continent. Each map in (a)–(d) has the unique (USC) ID number for the selected AMP-AR event, the state (ST) that the station is located in, and the date (year/month/day) of the AMP event. Red curves represent the track of the AR event that caused AMP from the stage of birth to its end.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
3. Results
a. Spatial pattern of AR-associated extreme precipitation
Figure 4 represents the spatial pattern of AMP-AR events in the eUS. Approximately 75% (31 438 out of 41 976) of AMP events recorded across the eUS between 1950 and 2021 occurred due to the contribution of ARs; at least 37.5% (27 AMP events) of the extreme precipitation recorded at each location are caused by ARs over the 72-yr period. Figure 4a shows the monthly boxplot summary of AMP-AR across the five climate regions. It can be noticed that ARs are the year-round dominating mechanism of AMP events across the study area except the South and Southeast regions during the months of July and August, where other well-known mechanisms are effectively contributing to AMP events such as North Atlantic tropical cyclones (Dhakal and Jain 2020). Seasonally, winter is the season with fewer AMP-AR events in the eUS over the whole record length as there are only a few stations that have a low percentage of AMP-AR-relative events over 72-yr across the regions of Central, South, and Southeast. On the other hand, most locations in the Central, East North Central, and Northeast regions have their active period of high percentage of AMP-AR during the warm part of the year in the late spring–summer–early fall (May–October). Moreover, the spring and fall (September and October) are also effective seasons for extreme precipitation induced by AR activities with locations having their highest percentages of AMP-AR in the South and Southeast regions.
(a) Monthly regional summary of location-specific AMP-AR (white) vs AMP-non-AR (gray) events. (b) Spatial pattern of AMP events induced by AR vs AMP produced by non-AR mechanisms.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
Figure 4b shows the spatial pattern of the rainfall annual maxima induced by ARs (top panel) and events caused by other different (non-AR) mechanisms (bottom panel). Generally, AR activities are very effective across all five regions. ARs have affected all the locations of the Central, East North Central, and Northeast regions, as they contribute to 60%–99% of the AMP events of each site during the 72-yr period. On the other hand, mixed population of extreme precipitation comprising AR and non-AR events are more pronounced in the South and Southeast regions. Although there are many locations having at least 60% of their annual maximum record affected by AR activities, sites within the South and Southeast regions have ARs contributing less than 40% of AMP events due to the other effective mechanisms such as the tropical cyclones (Dhakal and Jain 2020) as indicated in the non-AR event map (Fig. 4b). As such, around 60% of the sites in the South and Southeast regions have up to 25%–65% of extreme records caused by non-AR mechanisms.
b. Seasonality of AR-associated extreme precipitation
Significant seasonal mode of timing of AMP-AR events is determined by using the circular density approach, and the results are presented in Fig. 5. The bar charts in Fig. 5 represent the regional count of stations (in percentage) with number of significant seasons (from one to four) for the AMP-AR events. The bars show that approximately 50% of the locations in the East North Central and South regions have significant modes in three different seasons, while up to 40% of sites in the Central, Northeast, and Southeast regions have three different significant seasons of extreme precipitation events induced by AR activities. On the other hand, the regional distribution of stations with observed rainfall annual maxima caused by AR over two different seasons is the highest in the Northeast (65%) and Southeast (32%) regions. The highest count of sites with four significant seasons (year-round) of AMP-AR events is up to 10% in the Southeast region. However, the charts show that only a few sites have one significant season of AMP-AR event over the year in the East North Central region.
Count of significant seasons of AMP-AR events (bar charts); seasonal significant mode of AMP-AR events (maps).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
Maps in Fig. 5 display the spatial distribution of sites with significant seasonal mode of AMP-AR events. Winter’s (DJF) map shows that there are only a few stations with significant seasonal mode of AMP-AR events; there are 95 (∼17%) stations with significant AMP-AR events which are distributed as clusters over the regions of Central (24), Northeast (10), South (39), and Southeast (24). Many of these sites appeared in one large group extended over the eastern South, western Southeast, and southern part of Central regions. On the other hand, the maps of spring (MAM), summer (JJA), and fall (SON) show that many of stations have significant seasonal mode of AMP-AR events for all three seasons. The summer and fall seasons recorded the highest number of locations 347 (∼60%) and 388 (∼67%) with significant mode which appeared mainly in the Northeast and high-latitude parts of the East North Central and Southeast regions while significant locations in the Central and South regions appeared as expansion to the clusters across the regional borders as well as along the South coastline. For the spring season, around 44% (253) of the locations across the study area have significant modes which are more pronounced in low-latitude areas specifically in the Central and South regions.
Table 1 shows the regional count of stations (percentage) with a significant mode for winter, spring, summer, and fall seasons. Regionally, Fig. 5 (winter’s map) depicts that almost 25% of sites in the South and Southeast regions are significant in winter, while fall contains the highest number of stations with significant mode (Table 1). For the Central and South regions, most stations are significant during spring, summer, and fall, while summer and fall display the higher number of stations with significant mode in the regions of East North Central, Northeast, and Southeast. Last, spring had a moderate number of stations with significant mode in the East North Central region.
The regional count of stations (in percentage) with a significant mode for four different seasons.
c. The role of AR’s source of moisture in the annual and seasonal distribution of extreme precipitation
Following the approach in section 2c, we drew the major axis (pathway) of moisture tracks and identified the sources of moisture for the ARs associated with extreme precipitation events across the eUS. The AR pathway is the integrated smoothened weighted curve of all qualified grid points from a complete AR life cycle. The role of AR moisture source variation in the spatial and regional distribution of AMP-AR events is discussed based on the time horizon in the following sections.
1) Over the year contributions
The spatial distribution of fractional contribution of AMP-AR events categorized based on the four major sources of moisture is shown in Fig. 6. The map of the Pacific Ocean (PO) in Fig. 6a clearly shows the limited contribution of the PO moisture for only a few locations in the South region. The major effects of the PO moisture appeared in a small group of stations in the west part of the South region when PO moisture contributed 60%–80% of AMP-AR events of their annual extreme records. Figure 6b shows that the Atlantic Ocean (AO) moisture affected many sites in the Northeast and Southeast regions where it was associated with 20%–60% of the extreme rainfall record except in some coastal locations where the AO moisture contribution in extremes reached up to 80% of their events. The Caribbean Sea and Gulf of Mexico (CSGM) contributed to AMP events in many sites within the Central, Northeast, Southeast, and South regions (Fig. 6c). Locations with high fractions (>60%) of AMP-AR events caused by the CSGM moisture tracks are distributed in the coastal lines and lower latitude of the Southeast and South regions. Last, local ARs (terrestrial moisture) are the storms that originated over the inland sources of moisture and met the geometrical and IVT conditions of AR. These local activities are mostly effective in the East North Central region where the local moisture is associated with at least 40% of extremes record of each station (Fig. 6d). Furthermore, the impact of local moisture is expanded with having less contribution in many locations in the Central and Northeast regions.
The major sources of moisture-induced AMP-AR events. The sources are PO, AO, CS+GM, and local (interior body moisture).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
Figure 7 and Table 2 illustrate the regional summary of fractional contribution of AR-induced AMP events categorized based on the subsources of moisture. In the Central region, Fig. 7 shows that both the Gulf of Mexico and local moisture sources are the dominant source of moisture for AMP-AR events, as their lowest contributions are around 20% of total records in a selected location, and the highest are 70% and 75%, respectively. However, other sources such as PO and AO have less contribution to AMP-AR events based on mean values of association shown in Table 2. In the East North Central region, the moisture originating over the continent (local source) was the major source of AMP-AR events as it contributed to at least 50% of AMP in each location with the highest contribution of 85%. While moisture sources located outside the continent have less impact on AMP events with the highest contribution of less than 50%, the moisture coming from the Atlantic Ocean has the major impact on the AMP events in the Northeast region with the lowest and highest contribution of 30% and 76% and mean contribution of 50%. On the other hand, the moisture of CS, GM, and local moisture has less impact on AMP with mean contribution around 30% (Table 2). Last, the Caribbean Sea and Gulf of Mexico (Fig. 7) are attributed as the highest moisture sources to the annual extreme precipitation in the South and Southeast regions with an extreme record contribution up to 90% and 95%. On the other hand, the PO and local moisture sources also contributed to AMP events within the South region with a mean of association around 40% (Table 2). The AO has a mean contribution of around 40% (Table 2) and minimum and maximum contribution of around 10% and 82% in the Southeast region.
The contribution of the subsources of moisture to the AMP-AR events.
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
The average (mean) contribution of moisture subsource regions to AMP-AR events.
2) Seasonal contributions
To explore the spatial and regional roles of ARs in the seasonality of extreme precipitation according to the variation in the sources of moisture tracks, the records of AMP-AR events are divided into four subgroups based on the four seasons of the year. Figures 8, 10, 12, and 14 display the spatial distribution of fractional AMP-AR events based on the four major sources of moisture for winter (DJF), spring (MAM), summer (JJA), and fall (SON) seasons. As such, the regional influences of moisture sources variability on AMP-AR seasonality are shown in Figs. 9, 11, 13, and 15.
In winter, there is approximately 14% (4244 out of 31 438) of total AMP-AR-related events in all locations across the five climate regions. AMP-AR events in winter are distributed over the study region with percentages of 27%, 3%, 15%, 28%, and 27% in the Central, East North Central, Northeast, South, and Southeast, respectively. The PO moisture appeared to have clear impacts on the Central and South regions (Fig. 8a). Approximately half and one-third of the locations in the Central and South regions have at least 50% of their AMP-AR extremes during the winter season caused by the PO moisture, while this source of moisture has a limited effect on the AMP-AR events in the other three regions. Figure 8b shows that the AO moisture has the least impact on the study region during winter as it contributed to at least 25% of winter extreme precipitation at only 20% of all sites. On the other hand, the CSGM moisture (Fig. 8c) appeared to have the highest effect on the winter extremes by contributing to at least half of their record in 70% of 583 stations. As such, the strongest effect of the CSGM moisture appeared in the Northeast and Southeast regions with a contribution of 81% and 92% of their locations. Another significant impact of the CSGM moisture appeared in the Central and South regions where it caused AMP-AR events in 68% and 65% of the locations, respectively. Furthermore, the CSGM source contributed to winter extremes in 120 stations and approximately 75% of them are distributed along the coastal line in the Northeast, Southeast, and South regions. However, only 25% of stations in the East North Central have at least 50% of their winter extreme events caused by the CSGM moisture. In terms of the local moisture, it affected 200 stations across the study region (Fig. 8d) with the contribution of at least 25% of local winter extremes, and 71% of these locations are distributed over the Central and South regions.
The major sources of moisture for winter (DJF) AMP-AR events. The sources are PO, AO, CS+GM, and local (interior body moisture).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
Regionally, Fig. 9 demonstrates the seasonal influences of moisture subsources on extreme precipitation during winter season. The sTP, CS, GM, and local moisture are the dominant sources of moisture associated with winter extreme precipitation events in the Central region. The moisture of the GM has the highest contribution compared to other sources such as tropical and extratropical regions of the PO and the AO basins (Fig. 9). The East North Central registered the fewest number of AMP-AR events during winter; however, Fig. 9 shows that both sTP and GM are the only active moisture sources. In the Northeast region, the GM moisture source shows a significant impact on extreme precipitation, associated with at least 50% of winter extremes in many locations. Extreme precipitation in the South region shows a similar pattern of the moisture sources from sTP, CS, GM, and local sources as the sites within the Central region (Fig. 9). On the other hand, the Southeast region mimics the pattern in the Northeast as the extremes are dominated by the GM moisture (Fig. 9). Additionally, other sources of moisture also contributed to AMP-AR events as in the Northeast.
Contribution of the subsources of moisture to the AMP-AR events during winter (DJF).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
In spring, the number of extreme precipitation events caused by ARs is double compared to winter. Approximately 29% (8952 out of 31 438) of total AMP-AR events in spring are spread over the five regions with percentages of 24%, 12%, 12%, 19%, and 26% within the Central, East North Central, Northeast, South, and Southeast, respectively. The PO moisture (Fig. 10a) contributed to at least half of extremes at 66 locations across the eUS with 76% of these locations in the East North Central and South regions. The AO moisture (Fig. 10b) contributed to at least 50% of the extremes at 45 locations. The main AO moisture impact was at 30 and 11 sites in the Northeast and along the coast in the Southeast regions, respectively. Figure 10c shows that the CSGM moisture source caused at least 50% of spring extremes in 60%–92% of the stations in the Central, Northeast, Southeast, and South regions. As a result, the source caused at least 50% of spring extremes in 65% of all stations, 80% of which are in the Central, South, and Southeast regions. As such, the significant impact of CSGM moisture appeared across the coastal sites of the South region with 100% contribution to AMP-AR events for 26 out of 33 locations. The CSGM source only affected 20% of East North Central stations causing at least half of the AMP-AR record.
The major sources of moisture for spring (MAM) AMP-AR events. The sources are PO, AO, CS+GM, and local (interior body moisture).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
The regional contribution of moisture subsources on AMP-AR events for spring season is shown in Fig. 11. The extremes in the Central and South regions are mainly dominated by the moisture of GM and local sources (Fig. 11). As such, Fig. 11 highlights the association of sTP, sTA, and CS with spring extremes in both regions with sTP and CS having stronger effects than sTA on the Central and South sites. On the other hand, Fig. 11 shows that extreme precipitation events in the Northeast and Southeast regions are mainly controlled by the GM moisture source, while sTA is the secondary source with local moisture in the Northeast and with CS in the Southeast region. Finally, the local source is the dominant moisture source for AMP-AR events in the East North Central (Fig. 11) with a minimum and maximum contribution of 25% and 95%, respectively. Additionally, the sTP and GM are secondary sources of moisture inducing precipitation extremes in the region.
Contribution of the subsources of moisture to the AMP-AR events during spring (MAM).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
In summer, there is approximately 46% (14 533 out of 31 438) of total AMP-AR events in all locations across the five climate regions with percentages of 24%, 22%, 17%, 20%, and 17% in the Central, East North Central, Northeast, South, and Southeast regions, respectively. Figure 12a shows that the PO moisture has a modest contribution to the AMP-AR events in the eUS contributing to at least 25% of extremes in 125 locations, 85% of these sites are in the East North Central and Southeast regions, while 12% of the sites are in the Central region. Figure 12b demonstrates the wide impact of the AO moisture on summer AMP-AR events. Mainly, the AO source is associated with at least 50% of summer extremes at 245 locations across the study area, especially in the Northeast and Southeast regions. As such, the AO moisture source contributed to AMP at 80% and 85% of the total number of stations in both regions. Similarly, Fig. 12c displays the strong summer impact of the CSGM moisture on the annual maxima rainfall across all five regions. Most of the CSGM sources impacted the summer extremes in the South and Southeast regions where the moisture caused at least 50% of AMP records for 91% of 160 locations. On the other hand, the local moisture source (Fig. 12d) impacted the annual extremes in 252 locations, and they were mainly (about 70%) distributed over the Central and East North Central regions while the South and Northeast regions have 20% and 10%, sites, respectively. As a result, the local moisture effect showed a clear spatial pattern as it controlled the birth of AMP-AR events in 65% and 98% of the locations in the Central and East North Central regions, respectively.
The major sources of moisture for summer (JJA) AMP-AR events. The sources are PO, AO, CS+GM, and local (interior body moisture).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
Regionally, many sites in the Central region are influenced by the local moisture source as it is associated with up to 90% of summer extreme events (Fig. 13). In addition, the moisture of sTA and GM sources contributed to AMP-AR events in the region. In the East North Central region, the local moisture source significantly contributed to AMP-AR events causing 50%–100% of extreme records (Fig. 13). For this region, Fig. 13 also highlights modest contribution of other sources such sTP and GM. In the Northeast, sTA and local moisture are the most effective sources as they are associated with extreme precipitation events at majority locations (Fig. 13). However, there are secondary sources that may influence the summer extremes in the region such as eTA and GM. In the South, Fig. 13 depicts the effectiveness of both local and GM moisture sources in creating extreme precipitation events while less contribution from other subsources. Last, Fig. 13 shows that sTA is the main source of moisture associated with AMP-AR events for Southeast region while CS, GM, and local moisture sources make modest contribution to the summer extremes in the area.
Contribution of the subsources of moisture to the AMP-AR events during summer (JJA).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
In fall, there is approximately 33% (10 429 out of 31 438) of total AMP-AR-related events at all locations across the five climate regions which are distributed over the study regions with percentages of 21%, 12%, 21%, 25%, and 21% in the Central, East North Central, Northeast, South, and Southeast, respectively. Figure 14a depicts the vast effect of the PO moisture on AMP-AR events over the five regions. The PO source impact is mainly observed on the regions closer to the PO: Central, East North Central, and South. As a result, the PO moisture contributed to at least 50% of fall extremes at 90 locations 61% of which are in the South region, while the rest of them are divided between the Central and East North Central with a percentage of 16 and 23, respectively. The AO moisture (Fig. 14b) produced at least 50% of AMP-AR events at 125 stations during the fall season which are distributed in the Northeast, Southeast, and Central regions. Figure 14c shows a similar pattern for the CSGM moisture impacting extremes across the five regions but with a stronger spatial pattern than the previous sources. The CSGM source contributed to at least 50% of fall extremes at 240 (41%) locations. Many of them (70%) are evenly distributed over the South and Southeast regions, while the rest are mostly in the Central region. As such, the CSGM moisture contributed to AMP-AR events in approximately 60% and 70% of the sites in the South and Southeast regions, respectively. Similarly, Fig. 14d demonstrates the ability of local moisture to affect the annual maximum rainfall across the five regions. The local moisture causes at least half of the fall extremes at 159 stations which are mainly distributed in the East North Central, South, and Central regions with 41%, 28%, and 26% of all selected sites, respectively. As a result, the local moisture impact shows a strong spatial pattern across these three regions contributing to the AMP-AR extremes in 70%, 30%, and 30% of their stations, respectively.
The major sources of moisture for fall (SON) AMP-AR events. The sources are PO, AO, CS+GM, and local (interior body moisture).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
Regionally, Fig. 15 shows that the local and GM moisture are the main subsources of creating AMP-AR events in the Central and South regions. It also shows the role of other secondary subsources like sTP in creating extreme events in both regions. Similar subsources (local and GM) are actively associated with extreme precipitation events in the East North Central region with the strongest impact belonging to the local moisture source (Fig. 15). On the other hand, the results reported in Fig. 15 describe the critical role of moisture from sTA, GM, and local sources in producing AMP-AR events in the eastern regions of the study area. As such, the moisture of sTA is the major influential factor of extreme precipitation in the Northeast region, while the moisture of GM, sTA, and local moisture are mainly associated with extremes in the Southeast region.
Contribution of the subsources of moisture to the AMP-AR events during fall (SON).
Citation: Journal of Hydrometeorology 25, 11; 10.1175/JHM-D-23-0210.1
4. Summary and conclusions
The objective of this study is to evaluate the contribution of ARs and their moisture sources to the distribution and seasonality of extreme precipitation across the eUS during 1950–2021. Using the station records of daily precipitation data from the GHCN database, and the NCEP/NCAR reanalysis-based AR catalog, we assessed the percentages of AMP associated with ARs and their moisture sources. The results suggest that AR is a dominant factor of AMP in the eUS as they contributed to 75% (31 438 out of 41 976) of total AMP events recorded over the 72-yr period, and each station had at least 27 events associated with ARs. Furthermore, a monthly analysis of AMP-AR events shows that ARs are the year-round dominant mechanisms of producing extreme rainfall events at most locations in the eUS. We used a circular density approach to evaluate the seasonality of AMP-AR events. The findings indicate that many of sites in the Central, East North Central, and South regions experienced three significant seasons of AMP-AR events, while locations in the Northeast and Southeast experienced mixture of two and three significant seasons. Seasonally, only 17% of all locations observed significant mode of AMP-AR events during the winter season which are mainly clustered in the Central, South, and Southeast regions. The other three seasons show a strong signal of seasonality as spring is strongest in the South region, while summer and fall display a wide picture of strong seasons across the study area, especially the East North Central and Northeast regions.
Using the AR-detection algorithm, we drew the pathway of moisture tracks to identify the origins of the ARs associated with extreme precipitation events in the eUS from the four major sources: (i) the Pacific Ocean (PO), (ii) the Atlantic Ocean (AO), (iii) the combined source of Caribbean Sea (CS) and Gulf of Mexico (GM), and (iv) the local source. The spatial patterns of AMP events associated with the four major sources of moisture indicate that the PO has a marked contribution to AMP-AR events in western part of the South region, the AO effectively causes up to 80% of AMP-AR events in the Northeast and Southeast regions, CSGM is associated with up to 100% of extremes along the coast and lower latitude areas of the Southeast and South regions, and local moisture controls at least 40% of AMP-AR events at all stations within the East North Central region in addition to effective contribution to annual maxima in the Central and South regions. The findings of the moisture subsource’s effects on the regional pattern of extreme precipitation seasonality show that the GM moisture impacted majority of winter precipitation extremes across the five regions. The GM and local moisture appeared to be the dominant sources associated with spring AMP-AR events in the study area. In addition, the secondary sources are active during spring such as subtropical Atlantic Ocean (sTA) in the Northeast and Southeast regions and subtropical Pacific Ocean (sTP) in the Central, East North Central, and South regions. Local moisture plays a key role causing summer extreme precipitation across the five regions beside other oceanic sources like sTA in the Northeast and Southeast regions as well as GM in the South region. Last, the results of fall season show the vital role of ARs originated over the GM and the land in producing extreme precipitation events across the five regions in addition to the significant impact of sTA in the Northeast and Southeast regions as well as the contribution of sTP in the other three regions.
Results herein reinforce the key role ARs play in transporting water vapor to the eastern United States from the surrounding water bodies as well as the inland originated moisture. In addition, it augments knowledge on AR-associated extreme precipitation seasonality which governs the hydrologic cycle of this region. This work improves the quantitative understanding of ARs role as causative factor to the seasonality and distribution of extreme precipitation in the eUS. Future work will focus on characterizing the AR events associated with extremes in terms of IVT magnitude, frequency, and timing to enhance extremes alerts by developing more accurate forecasting system; hence, investigations of the hydrological response to ARs across the study area will be beneficial.
Acknowledgments.
This work is supported by the National Science Foundation (NSF) under Award HBCU-EiR-1901426.
Data availability statement.
The datasets used in this study were derived from the following public domain resources: Precipitation: The Global Historical Climatology Network database (https://www.ncei.noaa.gov/pub/data/ghcn/daily/), AR shape index: the Global Atmospheric Rivers Database, Version 3 (https://dataverse.ucla.edu/dataset.xhtml?persistentId=doi:10.25346/S6/YO15ON), and IVT magnitude: AR Catalog by Jonathan Rutz (https://www.inscc.utah.edu/∼rutz/ar_catalogs/).
REFERENCES
Agel, L., M. Barlow, J.-H. Qian, F. Colby, E. Douglas, and T. Eichler, 2015: Climatology of daily precipitation and extreme precipitation events in the Northeast United States. J. Hydrometeor., 16, 2537–2557, https://doi.org/10.1175/JHM-D-14-0147.1.
Alexander, M. A., J. D. Scott, D. Swales, M. R. Abel, K. Mahoney, and C. A. Smith, 2015: Moisture pathways into the U.S. Intermountain West associated with heavy winter precipitation events. J. Hydrometeor., 16, 1184–1206, https://doi.org/10.1175/JHM-D-14-0139.1.
Aljoda, A., 2021: Understanding hydrologic risk within the context of climate variability and atmospheric moisture pathways. Ph.D. dissertation, The University of Maine, 115 pp.
Aljoda, A., and N. Dhakal, 2023: Examining seasonality based on probabilistic properties of extreme precipitation timing in the eastern United States. Atmosphere, 14, 366, https://doi.org/10.3390/atmos14020366.
Barlow, M., 2011: Influence of hurricane-related activity on North American extreme precipitation. Geophys. Res. Lett., 38, L04705, https://doi.org/10.1029/2010GL046258.
Barth, N. A., G. Villarini, M. A. Nayak, and K. White, 2017: Mixed populations and annual flood frequency estimates in the western United States: The role of atmospheric rivers. Water Resour. Res., 53, 257–269, https://doi.org/10.1002/2016WR019064.
Dhakal, N., and S. Jain, 2020: Nonstationary influence of the North Atlantic tropical cyclones on the spatio-temporal variability of the eastern United States precipitation extremes. Int. J. Climatol., 40, 3486–3499, https://doi.org/10.1002/joc.6409.
Dhakal, N., S. Jain, A. Gray, M. Dandy, and E. Stancioff, 2015: Nonstationarity in seasonality of extreme precipitation: A nonparametric circular statistical approach and its application. Water Resour. Res., 51, 4499–4515, https://doi.org/10.1002/2014WR016399.
Dhakal, N., B. Tharu, and A. Aljoda, 2023: Changing seasonality of daily and monthly precipitation extremes for the contiguous USA and possible connections with large-scale climate patterns. Int. J. Climatol., 43, 2647–2666, https://doi.org/10.1002/joc.7994.
Dirmeyer, P. A., and J. L. Kinter III, 2010: Floods over the U.S. Midwest: A regional water cycle perspective. J. Hydrometeor., 11, 1172–1181, https://doi.org/10.1175/2010JHM1196.1.
Eckhardt, S., A. Stohl, H. Wernli, P. James, C. Forster, and N. Spichtinger, 2004: A 15-year climatology of warm conveyor belts. J. Climate, 17, 218–237, https://doi.org/10.1175/1520-0442(2004)017<0218:AYCOWC>2.0.CO;2.
Erlingis, J. M., J. J. Gourley, and J. B. Basara, 2019: Diagnosing moisture sources for flash floods in the United States. Part II: Terrestrial and oceanic sources of moisture. J. Hydrometeor., 20, 1511–1531, https://doi.org/10.1175/JHM-D-18-0120.1.
Gimeno, L., R. Nieto, and R. Sori, 2020: The growing importance of oceanic moisture sources for continental precipitation. npj Climate Atmos. Sci., 3, 27, https://doi.org/10.1038/s41612-020-00133-y.
Guan, B., and D. E. Waliser, 2015: Detection of atmospheric rivers: Evaluation and application of an algorithm for global studies. J. Geophys. Res. Atmos., 120, 12 514–12 535, https://doi.org/10.1002/2015JD024257.
Guan, B., and D. E. Waliser, 2019: Tracking atmospheric rivers globally: Spatial distributions and temporal evolution of life cycle characteristics. J. Geophys. Res. Atmos., 124, 12 523–12 552, https://doi.org/10.1029/2019JD031205.
Guan, B., D. E. Waliser, and F. M. Ralph, 2018: An intercomparison between reanalysis and dropsonde observations of the total water vapor transport in individual atmospheric rivers. J. Hydrometeor., 19, 321–337, https://doi.org/10.1175/JHM-D-17-0114.1.
Henny, L., C. D. Thorncroft, and L. F. Bosart, 2023: Changes in seasonal large-scale extreme precipitation in the mid-Atlantic and Northeast United States, 1979–2019. J. Climate, 36, 1017–1042, https://doi.org/10.1175/JCLI-D-22-0088.1.
Hirschboeck, K. K., 1988: Flood hydroclimatology. Flood Geomorphology, V. R. Baker, R. C. Kochel, and P. C. Patton, Eds., Wiley, 27–49.
Insua-Costa, D., G. Miguez-Macho, and M. C. Llasat, 2019: Local and remote moisture sources for extreme precipitation: A study of the two catastrophic 1982 western Mediterranean episodes. Hydrol. Earth Syst. Sci., 23, 3885–3900, https://doi.org/10.5194/hess-23-3885-2019.
Jammalamadaka, S. R., and A. Sengupta, 2001: Topics in Circular Statistics. Series on Multivariate Analysis, Vol. 5, World Scientific, 336 pp.
Knippertz, P., and J. E. Martin, 2007: A Pacific moisture conveyor belt and its relationship to a significant precipitation event in the semiarid southwestern United States. Wea. Forecasting, 22, 125–144, https://doi.org/10.1175/WAF963.1.
Knippertz, P., and H. Wernli, 2010: A Lagrangian climatology of tropical moisture exports to the Northern Hemispheric extratropics. J. Climate, 23, 987–1003, https://doi.org/10.1175/2009JCLI3333.1.
Kunkel, K. E., D. R. Easterling, D. A. R. Kristovich, B. Gleason, L. Stoecker, and R. Smith, 2012: Meteorological causes of the secular variations in observed extreme precipitation events for the conterminous United States. J. Hydrometeor., 13, 1131–1141, https://doi.org/10.1175/JHM-D-11-0108.1.
Lavers, D. A., and G. Villarini, 2013: Atmospheric rivers and flooding over the central United States. J. Climate, 26, 7829–7836, https://doi.org/10.1175/JCLI-D-13-00212.1.
Lavers, D. A., and G. Villarini, 2015: The contribution of atmospheric rivers to precipitation in Europe and the United States. J. Hydrol., 522, 382–390, https://doi.org/10.1016/j.jhydrol.2014.12.010.
Li, C., and Coauthors, 2019: Larger increases in more extreme local precipitation events as climate warms. Geophys. Res. Lett., 46, 6885–6891, https://doi.org/10.1029/2019GL082908.
Lu, M., and U. Lall, 2016: Tropical moisture exports, extreme precipitation and floods in Northeast US. Hydrol. Earth Syst. Sci. Discuss., https://doi.org/hess-2016-403.
Mahoney, K., and Coauthors, 2016: Understanding the role of atmospheric rivers in heavy precipitation in the southeast United States. Mon. Wea. Rev., 144, 1617–1632, https://doi.org/10.1175/MWR-D-15-0279.1.
Mallakpour, I., and G. Villarini, 2017: Analysis of changes in the magnitude, frequency, and seasonality of heavy precipitation over the contiguous USA. Theor. Appl. Climatol., 130, 345–363, https://doi.org/10.1007/s00704-016-1881-z.
Marelle, L., G. Myhre, Ø. Hodnebrog, J. Sillmann, and B. H. Samset, 2018: The changing seasonality of extreme daily precipitation. Geophys. Res. Lett., 45, 11 352–11 360, https://doi.org/10.1029/2018GL079567.
Menne, M. J., and Coauthors, 2012: Global Historical Climatology Network - Daily (GHCN-Daily), version 3. NOAA National Climatic Data Center, accessed 1 October 2023, https://doi.org/10.7289/V5D21VHZ.
Moore, B. J., P. J. Neiman, F. M. Ralph, and F. E. Barthold, 2012: Physical processes associated with heavy flooding rainfall in Nashville, Tennessee, and vicinity during 1–2 May 2010: The role of an atmospheric river and mesoscale convective systems. Mon. Wea. Rev., 140, 358–378, https://doi.org/10.1175/MWR-D-11-00126.1.
Myhre, G., and Coauthors, 2019: Frequency of extreme precipitation increases extensively with event rareness under global warming. Sci. Rep., 9, 16063, https://doi.org/10.1038/s41598-019-52277-4.
Nakamura, J., U. Lall, Y. Kushnir, A. W. Robertson, and R. Seager, 2013: Dynamical structure of extreme floods in the U.S. Midwest and the United Kingdom. J. Hydrometeor., 14, 485–504, https://doi.org/10.1175/JHM-D-12-059.1.
Neiman, P. J., F. M. Ralph, B. J. Moore, M. R. Abel, K. M. Mahoney, J. M. Cordeira, and M. D. Dettinger, 2013: The landfall and inland penetration of a flood-producing atmospheric river in Arizona. Part I: Observed synoptic-scale, orographic, and hydrometeorological characteristics. J. Hydrometeor., 14, 460–484, https://doi.org/10.1175/JHM-D-12-0101.1.
Newell, R. E., N. E. Newell, Y. Zhu, and C. Scott, 1992: Tropospheric rivers? – A pilot study. Geophys. Res. Lett., 19, 2401–2404, https://doi.org/10.1029/92GL02916.
Oliveira, M., R. M. Crujeiras, and A. Rodriguez-Casal, 2013: Nonparametric circular methods for exploring environmental data. Environ. Ecol. Stat., 20 (1), 1–17, https://doi.org/10.1007/s10651-012-0203-6.
Pal, I., B. T. Anderson, G. D. Salvucci, and D. J. Gianotti, 2013: Shifting seasonality and increasing frequency of precipitation in wet and dry seasons across the U.S. Geophys. Res. Lett., 40, 4030–4035, https://doi.org/10.1002/grl.50760.
Paltan, H., D. Waliser, W. H. Lim, B. Guan, D. Yamazaki, R. Pant, and S. Dadson, 2017: Global floods and water availability driven by atmospheric rivers. Geophys. Res. Lett., 44, 10 387–10 395, https://doi.org/10.1002/2017GL074882.
Pfahl, S., and H. Wernli, 2012: Quantifying the relevance of cyclones for precipitation extremes. J. Climate, 25, 6770–6780, https://doi.org/10.1175/JCLI-D-11-00705.1.
Pfahl, S., E. Madonna, M. Boettcher, H. Joos, and H. Wernli, 2014: Warm conveyor belts in the ERA-Interim dataset (1979–2010). Part II: Moisture origin and relevance for precipitation. J. Climate, 27, 27–40, https://doi.org/10.1175/JCLI-D-13-00223.1.
Pryor, S. C., and J. T. Schoof, 2008: Changes in the seasonality of precipitation over the contiguous USA. J. Geophys. Res. Atmos., 113, D21108, https://doi.org/10.1029/2008JD010251.
Ralph, F. M., and M. D. Dettinger, 2012: Historical and national perspectives on extreme West Coast precipitation associated with atmospheric rivers during December 2010. Bull. Amer. Meteor. Soc., 93, 783–790, https://doi.org/10.1175/BAMS-D-11-00188.1.
Ralph, F. M., P. J. Neiman, G. A. Wick, S. I. Gutman, M. D. Dettinger, D. R. Cayan, and A. B. White, 2006: Flooding on California’s Russian River: Role of atmospheric rivers. Geophys. Res. Lett., 33, L13801, https://doi.org/10.1029/2006GL026689.
Rutz, J. J., W. J. Steenburgh, and F. M. Ralph, 2014: Climatological characteristics of atmospheric rivers and their inland penetration over the western United States. Mon. Wea. Rev., 142, 905–921, https://doi.org/10.1175/MWR-D-13-00168.1.
Slinskey, E. A., P. C. Loikith, D. E. Waliser, B. Guan, and A. Martin, 2020: A climatology of atmospheric rivers and associated precipitation for the seven U.S. National Climate Assessment regions. J. Hydrometeor., 21, 2439–2456, https://doi.org/10.1175/JHM-D-20-0039.1.
Tan, X., T. Y. Gan, and Y. D. Chen, 2018: Moisture sources and pathways associated with the spatial variability of seasonal extreme precipitation over Canada. Climate Dyn., 50, 629–640, https://doi.org/10.1007/s00382-017-3630-0.
Trenberth, K. E., 2011: Changes in precipitation with climate change. Climate Res., 47, 123–138, https://doi.org/10.3354/cr00953.
Vázquez, M., R. Nieto, M. L. R. Liberato, and L. Gimeno, 2020: Atmospheric moisture sources associated with extreme precipitation during the peak precipitation month. Wea. Climate Extremes, 30, 100289, https://doi.org/10.1016/j.wace.2020.100289.