Abstract

Transient, narrow plumes of strong water vapor transport, referred to as atmospheric rivers (ARs), are responsible for much of the precipitation along the West Coast of the United States. The most intense precipitation events are almost always induced by an AR on the coast of Oregon and Washington and can result in detrimental impacts on society due to mudslides and flooding. To accurately predict AR events on numerical weather prediction, subseasonal, and seasonal time scales, it is important to understand the large-scale impacts on extreme AR events. Here, characteristics of ARs that result in an extreme precipitation event are compared to typical ARs on the coast of Washington State. In addition to more intense water vapor transport, notable differences in the synoptic forcing are present during extreme precipitation events that are not present during typical AR events. Subseasonal and seasonal teleconnection patterns are known to influence the weather in the Pacific Northwest and are investigated here. The Madden–Julian oscillation (MJO) plays a role in determining the strength of precipitation associated with an AR on the Washington coast. Phase 5 of the MJO (convection centered over the Maritime Continent) is the most common phase during an extreme precipitation event, while phase 2 (convection over the Indian Ocean) discourages an extreme event from occurring. Interactions between El Niño–Southern Oscillation (ENSO) and the propagation speed of the MJO result in extreme events during phase 1 of the MJO and El Niño but phase 8 during neutral ESNO conditions.

1. Introduction

Atmospheric rivers (ARs), or dynamic, narrow filaments of enhanced integrated water vapor transport (IVT) (Ralph et al. 2018), account for up to 50% of the annual precipitation along the West Coast of the United States (Neiman et al. 2011; Dettinger 2013; Lamjiri et al. 2017). Such events are important to the hydrology of the region, having been shown to relieve droughts along the West Coast, particularly in the Pacific Northwest (Dettinger 2013). ARs are an important modulator for determining whether the area has a dry or wet year (Dettinger et al. 2011; Dettinger 2013; Eldardiry et al. 2019). Precipitation from ARs provide a large fraction of the snow water equivalent throughout the region’s mountain ranges, especially in the Sierra Nevada, where snowmelt is important to the water management throughout the area (Guan et al. 2010). Despite providing beneficial precipitation at times, ARs have also been shown to be associated with extreme precipitation and flooding events along the West Coast of the United States (Ralph et al. 2006; Leung and Qian 2009; Neiman et al. 2011; Ralph et al. 2013; Dong et al. 2018). Not only can such events be detrimental to society, but landfalling ARs in western North America, as well as extreme precipitation events that are associated with ARs, are projected to increase through the twenty-first century (Gao et al. 2015; Hagos et al. 2016).

The duration, timing, and resulting precipitation from winter ARs are highly dependent on the location along the West Coast of the United States. A maximum frequency of AR landfall has been noted in the Pacific Northwest, with a decreasing landfall frequency southward such that climatologically, ARs tend to be more frequent in coastal regions of Washington and Oregon compared to California (Neiman et al. 2008; Payne and Magnusdottir 2014; Rutz et al. 2014). Oregon has been shown to have longer event conditions on average, while the shortest average duration is between the Sierra Nevada and the Rockies (Rutz et al. 2014). The topography of the West Coast is dotted with mountains and valleys that help shape precipitation regimes found throughout the region, as precipitation regimes drive from the vertical motion induced by mountain ranges on moisture-laden air. In the higher terrain of the Sierra Nevada, a small number of ARs produce a large fraction of the winter precipitation, and in the Washington coastal region, a large portion of the winter precipitation is due to a large number of ARs (Dettinger et al. 2011; Rutz et al. 2014).

A number of studies have investigated the role of teleconnections on seasonal characteristics of ARs, using a variety of regions and definitions of the cold season. Pacific Ocean teleconnection patterns have been shown to modulate landfalling AR events along the West Coast (Bao et al. 2006; Ryoo et al. 2013; Payne and Magnusdottir 2014; Mundhenk et al. 2016; Gershunov et al. 2017). Gershunov et al. (2017) noted a positive correlation between land falling ARs and the Pacific decadal oscillation (PDO); however, it was also pointed out that there is an increasing trend in land falling AR IVT associated with warming SSTs in the far western tropical Pacific. El Niño–Southern Oscillation (ENSO) has also been shown to impact the strength of the subtropical jet and the associated location of Rossby wave breaking (RWB), which modulates moisture transport and precipitation (Ryoo et al. 2013). Payne and Magnusdottir (2014) pointed out El Niño shifts the AR landfalling latitude closer to the equator, while La Niña shifts the landfalling latitude more poleward.

The Madden–Julian oscillation (MJO) is another important influence in modulating ARs (Guan and Waliser 2015; Stan et al. 2017). The peak of AR landfall along the West Coast has been found to occur during phase 6, with secondary maxima at phases 3, 7, and 8, while the minimum occurs phase 1 (Guan et al. 2012; Payne and Magnusdottir 2014). Phases 2 and 5 have the greatest influence on the average position of landfalls, where phase 2 is the southernmost location and phase 5 the northernmost (Payne and Magnusdottir 2014). The MJO has also been found to modulate the amount of average rainfall ARs produce based on the phase associated with the landfalling date, with phase 3 having the highest and phase 5 having the lowest precipitation average. Increasing latitude of the average precipitation anomaly occurs during changes from phase 1 to 8 (Payne and Magnusdottir 2014). Wet conditions occur in Washington and Oregon during phases 7 and 8 of the MJO, while it tends to be dry during phases 1, 2, and 4 for the months of October, November, and December (Bond and Vecchi 2003). This changes, however, to phase 5 being wet in January, February, and March and dry during phases 2 and 7.

In this study, we focus on linking large scale influences and extreme precipitation events induced by landfalling ARs. We differentiate ARs associated with daily extreme precipitation events that exceed the 95th percentile from ARs not producing such events impacting the coast of Washington, depicted by the white outline in Fig. 1 (and also in Fig. 3) An effort is made to characterize landfalling ARs inducing extreme precipitation events by investigating synoptic variables during the cold season. Key questions addressed in the paper are 1) Are there predominant synoptic scale features associated with AR extreme events compared to association with nonextreme events? and 2) Is there an enhancement of these events associated with teleconnection patterns? It should be noted that Warner et al. (2012) also characterized the synoptic environment associated with extreme precipitation events in Washington while Payne and Magnusdottir (2014, 2016) did so for extreme and persistent ARs within an area averaged region along the entire U.S. Pacific coast. To differentiate, here we combine extreme precipitation events with ARs for a more localized region, using a different methodology for defining an AR. We use roughly double the number of extreme events as Payne and Magnusdottir (2014), which should improve the statistics for determining the relationship to teleconnection patterns. Increased counts of ARs and extreme events are a result of less stringent filtering for ARs over the Pacific Northwest and expanding the time series through 2019. Extreme precipitation events are not necessarily less extreme than in Payne and Magnusdottir (2014); however, the IVT and other properties of typical ARs are weaker. Finally, interactions between the MJO and other teleconnections are explored in section 3b.

Fig. 1.

(a) Seasonal mean precipitation, (b) the number of ARs, and (c) the percent of 95th percentile extreme precipitation events associated with an AR during NDJ for the period of 1980–2019.

Fig. 1.

(a) Seasonal mean precipitation, (b) the number of ARs, and (c) the percent of 95th percentile extreme precipitation events associated with an AR during NDJ for the period of 1980–2019.

2. Data and methodology

a. MERRA-2

The primary data source for the detection of ARs and the resulting analysis is the Modern Era Retrospective analysis for Research and Applications, version 2 (MERRA-2; GMAO 2015a,b). MERRA-2 is a state-of-the-art global reanalysis product produced by NASA’s Global Modeling and Assimilation Office, with hourly data extending from January 1980 through roughly two weeks behind near real time at a spatial resolution of 0.625° longitude × 0.5° latitude (Gelaro et al. 2017). MERRA-2 has recently been used to detect and analyze ARs in numerous other research studies (e.g., Dettinger et al. 2018; Guan et al. 2018; Shields et al. 2018; Ralph et al. 2019; Huning et al. 2019; Rutz et al. 2019). Details on how MERRA-2 is used for AR detection here can be found in section 2c.

There are two different variables for total precipitation that have been released as part of the MERRA-2 dataset, the model analyzed precipitation computed from the atmospheric general circulation model, as the variable “PRECTOT,” and the observation-corrected precipitation “PRECTOTCORR.” PRECTOTCORR influences the land surface, therefore impacting soil moisture, evaporation, and latent heat, and is used for wet deposition of aerosols. PRECTOT and PRECTOTCORR are available in the FLX file collection; however, PRECTOTCORR is duplicated in the LND collection (GMAO 2015c). Both PRECTOT and PRECTOTCORR are used here and evaluated with respect to the observations to analyze the precipitation associated with landfalling ARs. Differences surrounding these two precipitation variables and the process used to correct the model generated precipitation are documented in Reichle et al. (2017).

b. Observed precipitation and extreme events

While MERRA-2 is a reliable source for atmospheric variables, it is preferred to use observations for fields such as precipitation. Precipitation observations used here are from the Climate Prediction Center (CPC) Unified Gauge-Based Analysis of Daily Precipitation, gridded observational precipitation product at a 0.25° spatial resolution (Xie et al. 2007; Chen and Xie 2008). Prior to any calculations, the precipitation observations were regridded to match the spatial resolution of MERRA-2. Daily extreme precipitation events were determined to have occurred if the precipitation exceeded the 95th percentile of precipitation. The 95th percentile was calculated using all days within the climatology period of 1981–2010 with at least 1 mm of precipitation, following Collow et al. (2016).

c. Atmospheric river detection

Atmospheric rivers were detected in MERRA-2 using the TempestExtremes tracking algorithm (Ullrich and Zarzycki 2017). To be considered an AR, TempestExtremes required a value for IVT that exceeded 250 kg m−1 s−1 with a minimum spatial area of 120 000 km2 and a minimum Laplacian of the IVT equal to 50 000 kg m−1 s−1 radian−2 within a Laplacian size of 10 grid boxes. The Laplacian restriction is used to find local regions of enhanced IVT with respect to the surrounding area. These criteria used to detect an AR were identical to what was used for the Tier 1 group of experiments for the Atmospheric River Tracking Method Intercomparison Project (ARTMIP; Shields et al. 2018), with the exception of the input data used for the IVT. In Shields et al. (2018), IVT and integrated water vapor (IWV) for AR detection were derived from MERRA-2 3-hourly zonal and meridional winds and specific humidity. Since IVT and IWV were computed using the MERRA-2 data on pressure levels and not the native vertical resolution, values are slightly smaller than the official dataset (Shields et al. 2018, their supplemental material). We have chosen to instead use the hourly IVT data directly computed from MERRA-2.

d. Teleconnection indices

AR frequencies are evaluated in connection to common teleconnection patterns for the period of 1980–2019. Monthly values of the Niño-3.4 index (Rayner et al. 2003), PDO index (Mantua et al. 1997), and quasi-biennial oscillation (QBO) index at 30 hPa were downloaded from NOAA ESRL’s Physical Sciences Division. Daily values of the Pacific–North American (PNA) index were downloaded from NOAA’s CPC (Barnston and Livezey 1987). Daily phase data for the MJO were downloaded from the Australian Bureau of Meteorology (Wheeler and Hendon 2004). The breakdown of the frequency of each teleconnection pattern within the 40-yr time period is as follows: 35% El Niño, 34% La Niña, 31% neutral ENSO; 57% positive PDO, 43% negative PDO; 60% positive PNA, 40% negative PNA; 9.1% MJO phase 1%, 12.1% MJO phase 2%, 14.6% MJO phase 3%, 13.3% MJO phase 4%, 13.4% MJO phase 5%, 13.4% MJO phase 6%, 13.4% MJO phase 7%, 10.5% MJO phase 8%; and 48% positive QBO, 52% negative PDO.

e. Composite methodology and statistical significance

Composites of meteorological fields are used to differentiate the characteristics of ARs that result in an extreme precipitation event compared to all landfalling ARs along the Washington coast. In an effort to determine the impact of the differing sample size between extreme and all landfalling AR events, composites for all ARs were analyzed in comparison to numerous random samples from all landfalling ARs of the same size as the number of extreme precipitation events (not shown). Following this analysis, it was determined that the typical AR events are similar enough to one another that the larger sample size does not dampen out the large-scale circulation. As a result, all landfalling ARs are considered despite the difference in sample size to the extreme events. Figures in section 3 show the composited meteorological fields during AR-induced extreme precipitation events as well as the difference in the composited fields between extreme events and all landfalling ARs. A t test was used for statistical significance in the composited differences and grid points that did not meet a statistical significance of 95% have been masked out.

3. Results

a. Atmospheric river and event climatology

Along the West Coast of the United States, precipitation occurs during the cold season (Wallis et al. 2007), particularly during the months of November–January (NDJ). Enhanced precipitation falls right along the coast as well as over high terrain from the Cascade Mountains in Washington State through the Sierra Nevada in California, with the largest accumulation of 13 mm day−1 over the Olympic Mountains (Fig. 1a). Although topography is a dominant factor for precipitation, the same cannot be said for the spatial pattern of the average count of ARs within an NDJ season. The majority of ARs occur along the northern half of the Pacific coast, with a sharp gradient in the count of events in southern Oregon (Fig. 1b). As pointed out by Rutz et al. (2014) and analyzed in depth by Rutz et al. (2015), a noteworthy amount of ARs penetrate inland along the Columbia River and into northern Idaho. Orographic enhancement results in the depletion of atmospheric moisture, hence topography plays an important role in the trajectory of these ARs, as well as farther south in California where ARs are present in between the Northern Coast and Southern Coast Ranges.

As ARs transport large quantities of water vapor across from over the ocean to the land surface, moisture is provided to allow for precipitation to occur. The most extreme precipitation events, in this case, those that exceed the 95th percentile, have the largest impact on society (Ralph et al. 2013). Figure 1c shows the percentage of 95th percentile extreme precipitation events that were associated with an AR. There are characteristics of both the mean precipitation and count of ARs within the spatial map of the percentage of events induced by an AR. In general, a west to east gradient exists. Although it does not rain much in the California Central Valley, ARs are more frequent there than west of the North Coastal Range and can result in up to 90% of the 95th percentile events.

The white outline in Fig. 1 depicts the region that will be focused on for this study; the political boundary of Washington State, west of 121°W. This subregion was chosen due to the fact that it has that largest magnitude of precipitation within the region and number of ARs along the West Coast, but also because nearly every grid box within the region has over 80% of the 95th percentile extreme precipitation events associated with an AR. While it is evident that northwestern Oregon also meets these criteria, we restricted our analysis to include just Washington as the seasonality and characteristics of atmospheric rivers differs as you move south (Rutz et al. 2014). Nevertheless, northern Oregon should also be investigated in future work.

A time series of the number of days with an AR as detected from MERRA-2, 1 mm of precipitation, and an extreme precipitation event, as well as the mean precipitation in each NDJ season for the selected region using the gridded gauge observations and MERRA-2 can be found in Fig. 2. The year within the time series represents the year for January, at the end of the season. Given that MERRA-2 does not include November and December of 1979, the first year shown here is 1981. January 1980 is, however, included in the subsequent figures. The number of days within the season with an AR ranges from about 20 to 60 days through the 40-yr time series (Fig. 2a). Interannual variability is larger during the first two decades, with a more stable time series beginning in the late 1990s, as denoted by the 11-yr running standard deviation. This is likely an artifact of the assimilated observations as multiple fields in the global water budget become more stable in MERRA-2 during the same time frame as a result of the introduction of advanced microwave radiances (Bosilovich et al. 2017). Noticeably suppressed counts of ARs occur in 1985, 1988, and 2014, all years with neutral ENSO or weak La Niña conditions. There is no indication that the frequency of ARs along the Washington coast has changed over time.

Fig. 2.

Time series of (a) the number of ARs detected in MERRA-2 along the Washington coast, (b) the number of days with 1 mm of area averaged precipitation, (c) the mean area averaged precipitation in the region, and (d) the number of 95th percentile extreme precipitation events associated with an AR during NDJ for the period of 1980–2019 using the CPC gridded gauge observations, MERRA-2 model analyzed precipitation, and MERRA-2 observation corrected precipitation. Error bars in (a) represent the 11-yr running standard deviation.

Fig. 2.

Time series of (a) the number of ARs detected in MERRA-2 along the Washington coast, (b) the number of days with 1 mm of area averaged precipitation, (c) the mean area averaged precipitation in the region, and (d) the number of 95th percentile extreme precipitation events associated with an AR during NDJ for the period of 1980–2019 using the CPC gridded gauge observations, MERRA-2 model analyzed precipitation, and MERRA-2 observation corrected precipitation. Error bars in (a) represent the 11-yr running standard deviation.

Between 1980 and the early 2000s there are ~70 days in NDJ with precipitation according to the observations, though like with the ARs, there are years with a suppressed count, particularly 1988, 1994, 2003, and 2014 (Fig. 2b). Following 2004, there is a drop in the number days with precipitation that coincides with a drastic decline in the number of gauge stations (Fig. 3a). It is possible that the decrease in days with precipitation is really an artifact of the fact that there are less gauges and this should be considered when analyzing precipitation in the region. During the first half of the period, between 1980 and 1999, the gauges tend to be spread out while they are really more representative of the eastern half of the region for the last two decades (Figs. 3b,c). Despite the uncertainty surrounding the gauges, the CPC product is still the preferred source of observations as products that include radar, such as Stage IV, are not available for the entire MERRA-2 time period and radar signals are often blocked by terrain.

Fig. 3.

(a) Time series of the count of gauge stations within the region along the Washington coast, and spatial maps of the average number of gauge station within each grid box for the periods of (b) 1980–1999 and (c) 2000–2019.

Fig. 3.

(a) Time series of the count of gauge stations within the region along the Washington coast, and spatial maps of the average number of gauge station within each grid box for the periods of (b) 1980–1999 and (c) 2000–2019.

Good agreement with respect to the interannual variability can be seen when the observed number of days with 1 mm of precipitation is compared to the two precipitation products from MERRA-2 (Fig. 2b). This is particularly the case with the observation corrected precipitation (correlation of 0.96); however, that is not surprising given the gauges are used in the correction process. Although the interannual variability is highly correlated, the magnitude of the number of days with 1 mm of precipitation is lower in both the model generated and observation corrected precipitation in MERRA-2, with the model generated being farther from the observations, especially prior to 2003. Beginning around 2004, the two MERRA-2 precipitation products are nearly indistinguishable. Changes in the observing system are a concern for reanalyses. In September 2002, the Atmospheric Infrared Sounder (AIRS) was introduced in MERRA-2 and alongside an increase in advanced microwave observations, the total count of assimilated observations nearly doubles in late 2002 (McCarty et al. 2016). The atmosphere becomes much better constrained with the addition of the new observations, and therefore in better agreement with the observations. Additionally, Bosilovich et al. (2017) pointed out that the assimilation of AIRS resulted in an increase in precipitation over land in MERRA-2. The fact that the observations, model generated, and observation corrected precipitation are in such good agreement in the second half of the time series likely stems from a combination of the decrease in the number of days with precipitation seen by the observations from a reduction in gauge stations as well as a more constrained atmosphere in MERRA-2 as a result of additional satellite radiances.

Regardless of the precipitation dataset, some similarities can be seen between the seasonal mean precipitation and the count of ARs (Fig. 2c). Seasons with more ARs tend to have more precipitation; however, the seasonal accumulation of precipitation is also dependent on the strength and duration of ARs. Conversely for the number of days with 1 mm of precipitation, there is impeccable agreement between the area averaged seasonal mean precipitation within the Washington coast region throughout the entire period, around 8 mm day−1. Between 1998 and 2013, the modeled generated precipitation in MERRA-2 is exaggerated, likely due to changes in the assimilated observations (Bosilovich et al. 2015). When considered alongside the number of days with precipitation, the fact that MERRA-2 agrees so well indicates that it does not precipitate as frequently in the model; however, when it does precipitate, the quantity is higher. As a result, the value for the 95th percentile is not the same in the observations and both of the MERRA-2 precipitation variables. An extreme event detected in the observations may therefore not be flagged as extreme in MERRA-2.

Figure 2d shows a time series of the number of 95th percentile extreme precipitation events associated with an AR in NDJ for the observations and two MERRA-2 precipitation datasets. A total of 266 of these events were detected in the observations (Table 1). This is equivalent to roughly 15% of the landfalling ARs in each of the months; however, the seasonal maximum in extreme precipitation events occurs in November.

Table 1.

The total count of ARs and AR-induced 95th percentile precipitation events detected during the cool season within the region along the coast of Washington State. Values in parentheses in the total AR column indicate the percentage of days within that month that an AR was detected, while values in the parentheses in the extreme AR column indicate the percent of total ARs within the month that resulted in an extreme precipitation event.

The total count of ARs and AR-induced 95th percentile precipitation events detected during the cool season within the region along the coast of Washington State. Values in parentheses in the total AR column indicate the percentage of days within that month that an AR was detected, while values in the parentheses in the extreme AR column indicate the percent of total ARs within the month that resulted in an extreme precipitation event.
The total count of ARs and AR-induced 95th percentile precipitation events detected during the cool season within the region along the coast of Washington State. Values in parentheses in the total AR column indicate the percentage of days within that month that an AR was detected, while values in the parentheses in the extreme AR column indicate the percent of total ARs within the month that resulted in an extreme precipitation event.

1) Precipitation

Spatial maps showing precipitation composited during observed extreme precipitation events from the observations and MERRA-2 model analyzed and observation corrected precipitation are displayed in Fig. 4. The influence of the varied terrain within the region is evident. All six panels show a regional maximum south, and windward, of the Olympic Mountains and as well as a strip along the western side of the Cascades with a northern (48.5°N) and southern (46°N) secondary maxima. This agreement indicates the gauges are representative despite the fact that there are grid boxes without any gauges. The spatial distribution of precipitation is not dependent on whether the event is considered extreme, however, depending on the location, the magnitude is roughly double during an extreme precipitation event.

Fig. 4.

(top) Observed, (middle) MERRA-2 modeled, and (bottom) MERRA-2 observation corrected precipitation (left) composited on days that an AR-induced 95th percentile extreme precipitation event was observed and (right) the difference to days in which an AR was detected along the coast of Washington State during NDJ 1980–2019.

Fig. 4.

(top) Observed, (middle) MERRA-2 modeled, and (bottom) MERRA-2 observation corrected precipitation (left) composited on days that an AR-induced 95th percentile extreme precipitation event was observed and (right) the difference to days in which an AR was detected along the coast of Washington State during NDJ 1980–2019.

The observations indicate a larger amount of precipitation falls in the western portion of the region, windward of the Olympic Mountains, and less to the east (Fig. 4a). On the other hand, MERRA-2 spreads the precipitation out among the Olympic and Cascade Mountains and the model generated precipitation also shifts the northern half of the strip along the Cascade Mountains slightly to the west compared to the observations. This shift is no longer present in the observation-corrected precipitation; however, somewhat counterintuitively, the magnitude of the precipitation is lower than both the observations and model generated precipitation (Fig. 4c). As part of the process to determine the correction factor for the observation corrected precipitation, the model background precipitation is aggregated such that the temporal and spatial resolutions of the model and observations match (Reichle et al. 2017). As a result, the observation corrected precipitation ends ups being smoothed over the region and the maxima are not as large.

2) Atmospheric moisture

Figure 5 shows the magnitude of the IVT beginning three days prior to an extreme precipitation event induced by an AR, as well as for the days leading to an AR being detected within the region. Given that ARs are dynamical features, this gives an indication of the location and strength of the AR as it progresses in time and space. It is evident that three days prior to an AR on the Washington coast, an AR is present spanning the Pacific Ocean, usually oriented from southwest to northeast (Figs. 5a,b). There is not much difference in the location of the AR between the extreme precipitation and nonextreme AR events, however, as expected, ARs that resulted in an extreme precipitation event had a larger magnitude in IVT in the central west Pacific. There is considerable variability in the IVT this far ahead of an AR landfall and averaging likely smooths out any indication that the location of the AR may differ during an extreme precipitation event.

Fig. 5.

(a),(c),(e),(g) IVT composited 3 days prior to the day that an AR-induced 95th percentile extreme precipitation event was observed and (b),(d),(f),(h) the difference to days in which an AR was detected along the coast of Washington State during NDJ 1980–2019. Regions in (b), (d), (f), and (h) that are not statistically significant have been masked out.

Fig. 5.

(a),(c),(e),(g) IVT composited 3 days prior to the day that an AR-induced 95th percentile extreme precipitation event was observed and (b),(d),(f),(h) the difference to days in which an AR was detected along the coast of Washington State during NDJ 1980–2019. Regions in (b), (d), (f), and (h) that are not statistically significant have been masked out.

Two days prior to an event, the plume of moisture progresses north and east such that the maximum in IVT is located around 160°E (Fig. 5c). The difference in IVT with respect to all landfalling ARs is stronger and a negative anomaly is present in the Gulf of Alaska (Fig. 5d). The magnitude of the IVT is somewhat muted for all ARs compared to a random subsample, indicating there is some variability at this point for all ARs. However, IVT is still greater in magnitude for the extreme events than a random subsample of 226 ARs. The day before the event, the region of enhanced IVT continues to make its way north and east toward the coast of Washington and the positive and negative differences to all ARs grow in space and magnitude, as though the AR is shifted farther south during AR-induced extreme precipitation events (Figs. 5e,f). For the extreme ARs, the mean magnitude of IVT right along the coast is already above the threshold for what would be considered an AR, and therefore some of these days must also be included in the composite for all ARs on the day of an event. The enhanced IVT associated with extreme precipitation events also suggests that these ARs could have a persisted duration (Payne and Magnusdottir 2016).

Along the coast of Washington and Oregon, the IVT exceeds 500 kg m−1 s−1 on the day of an extreme precipitation event induced by an AR (Fig. 5g), while the IVT is near the threshold of what would be classified as an AR for all events. The magnitude of the IVT actually weakens for all ARs on the day of an event from the day prior, mostly because the moisture is rained out without a source for resupply, though there is some contribution from the reduction in wind speed due to the surface roughness (not shown). An additional difference between extreme and nonextreme AR events, not obvious in Fig. 5h is the orientation of the AR along the coast. The ARs associated with extreme precipitation events maintain the southwest to northeast orientation after reaching the coast, however, for all ARs, there is a west to east orientation in the immediate area along the Washington coast, as though the subtropical high is shifted to the east during extreme events. This was a feature that was seen in numerous subsamples of typical AR events and is likely not the result of smoothing when compositing over 1700 events. It is also similar to the results shown by Payne and Magnusdottir (2016) for persistent ARs along the West Coast.

A more detailed view of the IVT in the surrounding region on the day of an extreme precipitation or AR event is shown in Fig. 6. In general, both event types have a flow of moisture from the southwest, though the magnitude is larger for the ARs that resulted in an extreme precipitation event. The area of maximum IVT is also shifted to the north and east for the extreme events, and barely extends in land for all ARs. The figure is somewhat deceptive due to the coarsening for the arrows to be visible as at least one grid box in the region needs to satisfy the AR criteria of IVT of at least 250 kg m−1 s−1. This does happen in the composite, right along the coast, but it cannot be seen in Fig. 6b. The influence of topography is apparent as the magnitude of the vertically integrated water vapor flux decreases immediately downwind of higher terrain and the enhanced atmospheric moisture travels farther in land along the Washington–Oregon border where there is a gap within the Cascades. Aside from the magnitude in the vertically integrated moisture flux, there is little difference in the directionality during an extreme event.

Fig. 6.

Vertically integrated water vapor fluxes composited on days with (a) an AR-induced 95th percentile extreme precipitation event and (b) the difference to days with a landfalling AR on the Washington coast during NDJ 1980–2019. Gray contours indicate topography in MERRA-2.

Fig. 6.

Vertically integrated water vapor fluxes composited on days with (a) an AR-induced 95th percentile extreme precipitation event and (b) the difference to days with a landfalling AR on the Washington coast during NDJ 1980–2019. Gray contours indicate topography in MERRA-2.

3) Dynamics

Synoptic scale patterns are a key factor for an AR to reach the coast of Washington, particularly through the position and strength of the Aleutian low and North Pacific High, which are both apparent in the composites of sea level pressure during extreme and all ARs (Figs. 7a,b). Climatologically, the location of the Aleutian low is more indicative of the mean for earlier in the cool season, as the low is generally shifted to the west during December and January. Aside from a slightly larger number of ARs in November, landfalling ARs are equally spread throughout the season. As a result, the eastward position of the Aleutian low in the Gulf of Alaska is more favorable for ARs making landfall on the Washington coast. Furthermore, the Aleutian low is deeper than what climatology would suggest for all ARs. Aside from being deeper on days with an extreme precipitation event, there is also a negative tilt, or a northwest to southeast orientation, in the low indicating a mature low pressure system (Fig. 7a). On the other hand, the Aleutian low for all ARs is more zonally oriented (not shown). The gradient between the Aleutian low and the North Pacific High is stronger on days with an extreme event as evident by the difference in sea level pressure between extreme and all ARs (Fig. 7b). Much of this comes from a deeper low; however, there is also a stronger high that extends slightly to the northeast.

Fig. 7.

Sea level pressure composited on days with (a) an AR-induced 95th percentile extreme precipitation event and (b) the difference to days with a landfalling AR, 500-hPa height zonal anomaly composited on days with (c) an AR-induced 95th percentile extreme precipitation event and (d) the difference to days with a landfalling AR, and 250-hPa winds composited on days with (e) an AR-induced 95th percentile extreme precipitation event and (f) the difference to days with a landfalling AR on the Washington coast during NDJ 1980–2019. Regions that are not statistically significant are masked out in (b).

Fig. 7.

Sea level pressure composited on days with (a) an AR-induced 95th percentile extreme precipitation event and (b) the difference to days with a landfalling AR, 500-hPa height zonal anomaly composited on days with (c) an AR-induced 95th percentile extreme precipitation event and (d) the difference to days with a landfalling AR, and 250-hPa winds composited on days with (e) an AR-induced 95th percentile extreme precipitation event and (f) the difference to days with a landfalling AR on the Washington coast during NDJ 1980–2019. Regions that are not statistically significant are masked out in (b).

For a different perspective the anomalies with respect to the zonal mean are shown for 500-hPa heights (Figs. 7c,d). There is a clear distinction on days with an extreme precipitation event in the low within the middle troposphere. During extreme precipitation events, a clearly defined low is present in the Gulf of Alaska for a vertically stacked low pressure system (Fig. 7c); however, this is not the case for all ARs in general (Fig. 7d). A region of high heights is present over the western United States, though stronger and located south and west on days with an extreme precipitation event. Overall, zonal anomalies at 500 hPa agree with the sea level pressure in the sense that the gradient between the high and low is tighter for extreme precipitation events. Aside from this, the longwave pattern at 500 hPa is quite similar across the rest of the Northern Hemisphere (not shown). A deeper low is present for both event types farther west, centered at 140°E. It is generally in the same location regardless of whether there was an extreme precipitation event, although perhaps shifted slightly north for all ARs.

4) Upper-level winds

A 250-hPa jet streak is present over the region during extreme precipitation events, with mean speeds nearing 50 m s−1 (Fig. 7e). The direction of the flow is indicative of an upper-level ridge peaking just east of the region. While this ridge is somewhat present for all ARs, the flow is a bit more zonal within the jet stream (Fig. 7f). Additionally, the maximum wind speeds are located well to the southwest. Winds are generally weaker for all ARs, consistent with the reduced pressure gradient, though some of this could be related to the larger sample size. Nevertheless, the jet stream still passes over Washington State when an AR is present. Payne and Magnusdottir (2014) noted that more extreme ARs are associated with anticyclonic Rossby wave breaking (RWB) in the eastern Pacific that is less extensive for weaker ARs, while Payne and Magnusdottir (2016) noted an inland shift in the location of RWB for persistent ARs. This is consistent with our composites for the 250-hPa winds; however, given the in-depth discussions in Ryoo et al. (2013), Payne and Magnusdottir (2014), Payne and Magnusdottir (2016), and Hu et al. (2017), potential vorticity and RWB are not elaborated upon here.

b. Relationship to teleconnection patterns

The influence of teleconnection patterns on landfalling ARs has garnered attention recently in order to make predictions on seasonal and subseasonal time scales. ARs were analyzed for teleconnection patterns that have been shown to influence the weather in the Pacific Northwest: the PDO on a decadal time scale (Mantua and Hare 2002), ENSO and the PNA on an interannual time scale (Blackmon et al. 1984), and on a subseasonal time scale, the MJO (Madden and Julian 1994; Wheeler and Hendon 2004). Indices representing these patterns allow for simultaneous analyses of these teleconnection patterns regardless of the varying time scales. To avoid overlapping statistics,

ARs that result in an extreme precipitation are not included in the sample of typical ARs, termed nonextreme ARs. Figures 8 and 9 show the normalized phase frequency for the four teleconnection patterns for nonextreme ARs and extreme precipitation events such that the value given represents the percentage of days with a given teleconnection pattern that an AR made landfall in Washington. This methodology removes the impact of a given phase of a teleconnection generally being present more frequently than other phases. Statistical significance for the preference of a given phase of a teleconnection pattern was assessed using Monte Carlo simulations with 10 000 random iterations.

Fig. 8.

Bar chart depicting the normalized frequency of nonextreme ARs during the different phases of (a) ENSO, (b) PNA, (c) PDO, and (d) MJO for NDJ 1980–2019. Deeper shaded bars indicate statistical significance of at least 0.1 confidence.

Fig. 8.

Bar chart depicting the normalized frequency of nonextreme ARs during the different phases of (a) ENSO, (b) PNA, (c) PDO, and (d) MJO for NDJ 1980–2019. Deeper shaded bars indicate statistical significance of at least 0.1 confidence.

Fig. 9.

Bar chart depicting the normalized frequency of AR-induced 95th percentile extreme precipitation events on the Washington coast during the different phases of (a) ENSO, (b) PNA, (c) PDO, and (d) MJO for NDJ 1980–2019. Deeper shaded bars indicate statistical significance of at least 0.1 confidence.

Fig. 9.

Bar chart depicting the normalized frequency of AR-induced 95th percentile extreme precipitation events on the Washington coast during the different phases of (a) ENSO, (b) PNA, (c) PDO, and (d) MJO for NDJ 1980–2019. Deeper shaded bars indicate statistical significance of at least 0.1 confidence.

Payne and Magnusdottir (2014) showed that landfalling ARs along the U.S. West Coast occur most frequently during El Niño, but the average landfall latitude during La Niña tended to be along the Washington coast. In a broad sense, our results are in agreement. For AR events impacting coastal Washington, El Niño and La Niña were found to be the most common phases of Niño-3.4, regardless of whether the AR resulted in an extreme precipitation event (103 and 85 days for extreme events and 518 and 514 days for nonextreme events). ARs occur less frequently when the Niño-3.4 region is neutral (78 extreme events and 418 nonextreme AR events), and this discouragement of ARs is statistically significant at a 0.05 confidence interval.

The PDO has been shown to influence water vapor transport through an equatorward shift of the North Pacific AR belt when in the positive phase (Liu et al. 2016) as well as impacting the Aleutian low (Newman et al. 2016), however, there is little distinction in the phase of the PDO for nonextreme ARs compared to those that result in an extreme precipitation event.

On a subseasonal scale, the PNA is correlated to temperature and precipitation in the Pacific Northwest (Leathers et al. 1991) as the positive phase for PNA has a breakdown of the normally prevalent midlevel anticyclone of the Pacific Northwest. During the positive phase of the PNA, there were 956 nonextreme AR events and 495 during the negative phase. The positive phase is more common and statistically significant at 0.01 confidence, but this does not outweigh the fact that ARs and AR-induced extreme precipitation events occur more frequently when the PNA is in the positive phase. Perhaps the lack of deviation between extreme and all ARs associated with the PNA stems from the necessity of the southeasterly flow associated, which occurs with the positive phase of the PNA, in order for an AR to make landfall in Washington.

Results for the MJO are more interesting as a distinction between all ARs and those that resulted in an extreme precipitation event becomes apparent. Little contrast between the phases of the MJO can be seen for nonextreme ARs, with the exception of a preference for phase 4 (convection over the Maritime Continent) at a 0.1 confidence level (Fig. 8d). Extreme event days were more common during phases 1, 5, and 8 of the MJO, with phase 5 (convection over the Maritime Continent, north of Australia) statistically significant at 0.1 confidence, and the discouragement of extreme events during phase 2 is statistically significant at 0.05 confidence (Fig. 9d). The relationship between phase 5 of the MJO and extreme precipitation events associated with AR is consistent with Payne and Magnusdottir (2014) from the perspective that they saw a more poleward mean landfalling latitude during phase 5. This, however, contradicts their finding that increased ARs also occur during phases 6 and 7, and that phase 7 has the largest anomaly in the moisture flux. Guan et al. (2012) also found phase 6 to be the most frequent phase for extreme snowfall in the California Sierra Nevada, while Mundhenk et al. (2016) showed a higher frequency of ARs during phases 7 and 8 for the entire West Coast. These differences are likely the result of the more northern landfalling latitude of ARs investigated here compared to the previous studies.

Teleconnection patterns can interact with one another and given the distinction between nonextreme ARs and those that result in an extreme precipitation event associated with the MJO, the MJO is further analyzed with respect to ENSO, PNA, and QBO (Fig. 10). Given the relationship of the MJO to the QBO discussed in the literature, the QBO has been added to the analysis (Yoo and Son 2016; Zhang and Zhang 2018). This type of analysis was also completed by Mundhenk et al. (2016); however, the authors did so for all ARs on an annual time scale with the entire West Coast of the United States considered as one region. While much of what can be seen in Fig. 10 duplicates the results already discussed there are a few additional things that can be learned when considering teleconnections alongside individual MJO phases. During phase 5 of the MJO, convection is centered to the southeast of the Philippines and north of central Australia. Acting through the excitement of Rossby waves, convection to the east of the Philippines has been shown to force a positive PNA-like pattern (Stan et al. 2017). The positive phase of the PNA has a statistically significant increased normalized frequency of AR landfalls due to cyclonic wave breaking in the northeastern Pacific (Franzke et al. 2011; Payne and Magnusdottir 2014). In addition, El Niño conditions result in an increased frequency of anomalous low pressure patterns in the region, which can be considered as a positive-like PNA pattern excited through ENSO (Horel and Wallace 1981; Johnson and Feldstein 2010). The enhanced frequency of extreme precipitation events associated with an AR during phase 5 of the MJO is likely to be a result of the combination of these teleconnection patterns.

Fig. 10.

Histogram of the (a),(b) count and (c),(d) frequency of (left) AR-induced 95th percentile extreme precipitation events and (right) all ARs on the Washington coast for teleconnection patterns during each phase of the MJO during NDJ 1980–2019. Red asterisks in (c) and (d) indicate statistical significance of at least 0.1 confidence.

Fig. 10.

Histogram of the (a),(b) count and (c),(d) frequency of (left) AR-induced 95th percentile extreme precipitation events and (right) all ARs on the Washington coast for teleconnection patterns during each phase of the MJO during NDJ 1980–2019. Red asterisks in (c) and (d) indicate statistical significance of at least 0.1 confidence.

Numerous studies have shown that the MJO behaves differently depending on the phase of ENSO, with the MJO propagating faster during El Niño (e.g., Moon et al. 2011; Chen et al. 2016; Wu and Song 2018). The synoptic conditions during phase 8 of the MJO counteract the unfavorable conditions present during neutral ENSO conditions and the weakening of the Aleutian low during the negative phase of the PDO (Mantua and Hare 2002). However, ARs resulting in extreme precipitation events occur most frequently during El Niño conditions when the MJO is in phase 1. It is hypothesized that this is actually a lagged response of the Rossby wave forcing associated with the MJO. Due to the faster propagation of the MJO during El Niño, extreme precipitation events occur during phase 1 of the MJO during an El Niño yet phase 8 under neutral ENSO conditions. During phase 3 of the MJO and La Niña conditions, an anticyclonic anomaly is present over the North Pacific Ocean, thus encouraging dry conditions along the West Coast of the United States (Moon et al. 2011). The occurrence of extreme precipitation events is therefore suppressed, with larger counts of events occurring during positive or neutral ENSO conditions under phase 3 of the MJO. There tends to be more frequent, and stronger, MJO events during the negative, or easterly phase, of the QBO (Densmore et al. 2019; Klotzbach et al. 2019), which explains the increased AR frequency during phase 4 of the MJO.

4. Summary and conclusions

In this work, we link large scale circulations influencing ARs in coastal Washington to better characterize the precipitation, synoptic regimes, and teleconnection patterns associated with AR-induced extreme precipitation events. ARs were detected using the TempestExtremes tracking algorithm with MERRA-2 integrated water vapor transport, while extreme precipitation events were determined using gridded gauge observations. Among the 1716 detected AR events landfalling along Washington’s coast, 266 extreme precipitation events were induced by a landfalling AR during the months of November, December, and January. The monthly distribution of events follows the trend noticed by Payne and Magnusdottir (2014) and Neiman et al. (2008), where the frequency of landfalling ARs decreases with each consecutive month.

Precipitation associated with ARs on the Washington coast is focused on the windward side of the Olympic Mountains as well as the Cascades; however, the magnitude is not as large to the east. MERRA-2 well represents the spatial pattern except the model precipitation accumulation is equivalent between the Olympic and Cascade Mountains instead of producing more precipitation windward of the Olympic Mountains.1

When analyzing the spatial pattern and time series of precipitation, consideration must be given to uncertainties in the dataset. Within the region, a decrease in the number of reporting gauge stations contributes to a decline in the frequency of precipitation. On the other hand, MERRA-2 becomes better constrained with the addition of new satellites in the early 2000s and likely has a related improvement in prognostic variables such as precipitation. Observation corrected precipitation in MERRA-2 has its own set of biases due to the methodology used in the correction process, smoothing out the precipitation across the entire region. As a result, it is recommended that multiple precipitation datasets be compared when evaluating the precipitation associated with an AR. Furthermore, precipitation is only one contributor to the impacts on the land surface and society as factors such as the preceding soil moisture and the duration of an event are also influential.

Aside from more intense ARs with enhanced water vapor transport, subtle differences in the synoptic environment are present that distinguish an extreme precipitation event from the typical precipitation associated with an AR. The gradient between the Aleutian low and the North Pacific high is tighter, the wind speed is increased, and the water vapor transport is intensified. Not only is the low pressure in the Gulf of Alaska deeper, but on average, and during numerous individual extreme precipitation events, there is a negative tilt indicating a mature cyclonic system. During extreme precipitation events, the anomalously low pressure in the Gulf of Alaska extends vertically into the middle troposphere, which is not necessarily the case for all ARs. A jet streak at 250 hPa is also located directly over the region during extreme precipitation events, as opposed to southwest, over the Pacific Ocean. Despite not being discussed here, Rossby wave breaking is an important mechanism for AR-induced extreme precipitation events (Ryoo et al. 2013; Payne and Magnusdottir 2014; Hu et al. 2017). Some of these results are not surprising, but rather natural.

Influences of teleconnection patterns on AR-induced extreme precipitation events were evaluated through an analysis of the count and frequency of events within the different phases of decadal, interannual, and intraseasonal teleconnection patterns. Little influence can be seen from ENSO, PDO, or PNA in distinguishing an AR that resulted in an extreme precipitation event from all other ARs. The phase of the MJO, on the other hand, can impact whether there is an extreme precipitation event. Phase 5 is the most common for an extreme precipitation event while there is a statistically significant distinction with phase 4 for a typical landfalling AR. When considered alongside other teleconnection patterns, phase 1 is preferential during El Niño, phases 3 and 8 during neutral ENSO conditions, and phases 5 and 6 during La Niña. Although mirror phases of the MJO, phases 1 and 5 can both have a deepening of the Aleutian low due to the tropical forced Rossby wave train the in extratropics, thereby influencing extreme precipitation events along the Washington coast. Some of this interaction is likely to be dependent on the timing within the cool season as Bond and Vecchi (2003) demonstrated an increase in precipitation in the region during phase 8 during October–December, but phase 5 during January–March.

Some limitations to this study are worth noting and expanding upon in future work. The region of focus was along the coast of Washington; however, ARs impact the entire Pacific coast, spanning British Columbia to Southern California. As previous studies have demonstrated, it cannot be expected that these results would be indicative of ARs impacting elsewhere along the coast, or farther inland. Furthermore, using a reanalysis such as MERRA-2 restricts the sample size of the dataset to a 40-yr period. By nature, extreme precipitation events are a rare occurrence, and therefore limits the statistics that can performed. One possible way to expand the sample size and ensure these results hold is to conduct a similar analysis with the version of the GEOS atmosphere–ocean general circulation model used for the subseasonal-to-seasonal (S2S) forecasting system, which is capable of producing a realistic MJO (Molod et al. 2020). This will then allow for an assessment of the subseasonal forecast skill of the S2S system in producing ARs and extreme events induced by an AR associated with the MJO together with ENSO and QBO conditions. The connection between ENSO and the propagation speed of the MJO also warrants additional study of a potential lag relationship between teleconnection patterns and extreme precipitation events. It is also possible that the upstream transport and maintenance of enhanced water vapor necessary for an extreme precipitation event is influenced by the MJO’s feedback on the extratropics.

Acknowledgments

The authors thank Randy Koster for assisting with Monte Carlo significance testing on the teleconnection patterns, Paul Ullrich and Beth McClenny for providing guidance and the ideal settings for the tempestExtremes atmospheric river detection, and Siegfried Schubert and Amin Dezfuli for their useful discussions. This work was supported by NASA’s Earth Science Research Program. H. M. was additionally supported by USRA through Goddard Earth Sciences Technology and Research (GESTAR).

REFERENCES

REFERENCES
Bao
,
J.-W.
,
S. A.
Michelson
,
P. J.
Neiman
,
F. M.
Ralph
, and
J. M.
Wilczak
,
2006
:
Interpretation of enhanced integrated water vapor bands associated with extratropical cyclones: Their formation and connection to tropical moisture
.
Mon. Wea. Rev.
,
134
,
1063
1080
, https://doi.org/10.1175/MWR3123.1.
Barnston
,
A. G.
, and
R. E.
Livezey
,
1987
:
Classification, seasonality, and persistence of low-frequency atmospheric circulation patterns
.
Mon. Wea. Rev.
,
115
,
1083
1126
, https://doi.org/10.1175/1520-0493(1987)115<1083:CSAPOL>2.0.CO;2.
Blackmon
,
M. L.
,
Y.
Lee
, and
J. M.
Wallace
,
1984
:
Horizontal structure of 500 mb height fluctuations with long, intermediate and short time scales
.
J. Atmos. Sci.
,
41
,
961
980
, https://doi.org/10.1175/1520-0469(1984)041<0961:HSOMHF>2.0.CO;2.
Bond
,
N. A.
, and
G. A.
Vecchi
,
2003
:
The influence of the Madden–Julian oscillation on precipitation in Oregon and Washington
.
Wea. Forecasting
,
18
,
600
613
, https://doi.org/10.1175/1520-0434(2003)018<0600:TIOTMO>2.0.CO;2.
Bosilovich
,
M. G.
, and et al
,
2015
:
MERRA-2: Initial evaluation of the climate. NASA Tech. Memo. NASA/TM-2015-104606/Vol. 43, 145 pp.
, https://gmao.gsfc.nasa.gov/pubs/docs/Bosilovich803.pdf.
Bosilovich
,
M. G.
,
F. R.
Robertson
,
L.
Takacs
,
A.
Molod
, and
D.
Mocko
,
2017
:
Atmospheric water balance and variability in the MERRA-2 reanalysis
.
J. Climate
,
30
,
1177
1196
, https://doi.org/10.1175/JCLI-D-16-0338.1.
Chen
,
M.
, and
P.
Xie
,
2008
:
CPC unified gauge-based analysis of global daily precipitation
.
Western Pacific Geophysics Meeting
,
Cairns, QLD, Australia, Amer. Geophys. Union, Abstract A24A-05
.
Chen
,
X.
,
J.
Ling
, and
C.
Li
,
2016
:
Evolution of the Madden–Julian oscillation in two types of El Niño
.
J. Climate
,
29
,
1919
1934
, https://doi.org/10.1175/JCLI-D-15-0486.1.
Collow
,
A. B. M.
,
M. G.
Bosilovich
, and
R. D.
Koster
,
2016
:
Large-scale influences on summertime extreme precipitation in the northeastern United States
.
J. Hydrometeor.
,
17
,
3045
3061
, https://doi.org/10.1175/JHM-D-16-0091.1.
Densmore
,
C. R.
,
E. R.
Sanabia
, and
B. S.
Barrett
,
2019
:
QBO influence on MJO amplitude over the maritime continent: Physical mechanisms and seasonality
.
Mon. Wea. Rev.
,
147
,
389
406
, https://doi.org/10.1175/MWR-D-18-0158.1.
Dettinger
,
M. D.
,
2013
:
Atmospheric rivers as drought busters on the U.S. West Coast
.
J. Hydrometeor.
,
14
,
1721
1732
, https://doi.org/10.1175/JHM-D-13-02.1.
Dettinger
,
M. D.
,
F. M.
Ralph
,
T.
Das
,
P. J.
Neiman
, and
D. R.
Cayan
,
2011
:
Atmospheric rivers, floods, and the water resources of California
.
Water
,
3
,
445
478
, https://doi.org/10.3390/w3020445.
Dettinger
,
M. D.
,
F. M.
Ralph
, and
J. J.
Rutz
,
2018
:
Empirical return periods of the most intense vapor transports during historical atmospheric river landfalls on the U.S. West Coast
.
J. Hydrometeor.
,
19
,
1363
1377
, https://doi.org/10.1175/JHM-D-17-0247.1.
Dong
,
L.
,
L. R.
Leung
,
F.
Song
, and
J.
Lu
,
2018
:
Roles of SST versus internal atmospheric variability in winter extreme precipitation variability along the U.S. West Coast
.
J. Climate
,
31
,
8039
8058
, https://doi.org/10.1175/JCLI-D-18-0062.1.
Eldardiry
,
H.
,
A.
Mahmood
,
X.
Chen
,
F.
Hossain
,
B.
Nijssen
, and
D. P.
Lettenmaier
,
2019
:
Atmospheric river–induced precipitation and snowpack during the western United States cold season
.
J. Hydrometeor.
,
20
,
613
630
, https://doi.org/10.1175/JHM-D-18-0228.1.
Franzke
,
C.
,
S. B.
Feldstein
, and
S.
Lee
,
2011
:
Synoptic analysis of the Pacific–North American teleconnection pattern
.
Quart. J. Roy. Meteor. Soc.
,
137
,
329
346
, https://doi.org/10.1002/qj.768.
Gao
,
Y.
,
J.
Lu
,
L. R.
Leung
,
Q.
Yang
,
S.
Hagos
, and
Y.
Qian
,
2015
:
Dynamical and thermodynamical modulations on future changes of landfalling atmospheric rivers over western North America
.
Geophys. Res. Lett.
,
42
,
7179
7186
, https://doi.org/10.1002/2015GL065435.
Gelaro
,
R.
, and et al
,
2017
:
The Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2)
.
J. Climate
,
30
,
5419
5454
, https://doi.org/10.1175/JCLI-D-16-0758.1.
Gershunov
,
A.
,
T.
Shulgina
,
F. M.
Ralph
,
D. A.
Lavers
, and
J. J.
Rutz
,
2017
:
Assessing the climate-scale variability of atmospheric rivers affecting western North America
.
Geophys. Res. Lett.
,
44
,
7900
7908
, https://doi.org/10.1002/2017GL074175.
GMAO
,
2015a
:
MERRA-2 tavg1_2d_slv_Nx: 2d, 1-hourly, time-averaged, single-level, assimilation, single-level diagnostics V5.12.4. Goddard Earth Sciences Data and Information Services Center, accessed May 2019
, https://doi.org/10.5067/VJAFPLI1CSIV.
GMAO
,
2015b
:
MERRA-2 tavg1_2d_int_Nx: 2d,1-Hourly, time-averaged, single-level, assimilation, vertically integrated diagnostics V5.12.4. Goddard Earth Sciences Data and Information Services Center, accessed May 2019
, https://doi.org/10.5067/Q5GVUVUIVGO7.
GMAO
,
2015c
:
MERRA-2 tavg1_2d_flx_Nx: 2d,1-Hourly, time-averaged, single-level, assimilation, surface flux diagnostics V5.12.4. Goddard Earth Sciences Data and Information Services Center, accessed May 2019
, https://doi.org/10.5067/7MCPBJ41Y0K6.
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.
,
N. P.
Molotch
,
D. E.
Waliser
,
E. J.
Fetzer
, and
P. J.
Neiman
,
2010
:
Extreme snowfall events linked to atmospheric rivers and surface air temperature via satellite measurements
.
Geophys. Res. Lett.
,
37
,
L20401
, https://doi.org/10.1029/2010GL044696.
Guan
,
B.
,
D. E.
Waliser
,
N. P.
Molotch
,
E. J.
Fetzer
, and
P. J.
Neiman
,
2012
:
Does the Madden–Julian oscillation influence wintertime atmospheric rivers and snowpack in the Sierra Nevada?
Mon. Wea. Rev.
,
140
,
325
342
, https://doi.org/10.1175/MWR-D-11-00087.1.
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.
Hagos
,
S. M.
,
L. R.
Leung
,
J.-H.
Yoon
,
J.
Lu
, and
Y.
Gao
,
2016
:
A projection of changes in landfalling atmospheric river frequency and extreme precipitation over western North America from the Large Ensemble CESM simulations
.
Geophys. Res. Lett.
,
43
,
1357
1363
, https://doi.org/10.1002/2015GL067392.
Horel
,
J. D.
, and
J. M.
Wallace
,
1981
:
Planetary-scale atmospheric phenomena associated with the southern oscillation
.
Mon. Wea. Rev.
,
109
,
813
829
, https://doi.org/10.1175/1520-0493(1981)109<0813:PSAPAW>2.0.CO;2.
Hu
,
H.
,
F.
Dominguez
,
Z.
Wang
,
D. A.
Lavers
,
G.
Zhang
,
F. M.
Ralph
,
2017
:
Linking atmospheric river hydrological impacts on the U.S. West Coast to Rossby wave breaking
.
J. Climate
,
30
,
3381
3399
, https://doi.org/10.1175/JCLI-D-16-0386.1.
Huning
,
L. S.
,
B.
Guan
,
D. E.
Waliser
, and
D. P.
Lettenmaier
,
2019
:
Sensitivity of seasonal snowfall attribution to atmospheric rivers and their reanalysis-based detection
.
Geophys. Res. Lett.
,
46
,
794
803
, https://doi.org/10.1029/2018GL080783.
Johnson
,
N. C.
, and
S. B.
Feldstein
,
2010
:
The continuum of North Pacific sea level pressure patterns: Intraseasonal, interannual, and interdecadal variability
.
J. Climate
,
23
,
851
867
, https://doi.org/10.1175/2009JCLI3099.1.
Klotzbach
,
P.
,
S.
Abhik
,
H. H.
Hendon
,
M.
Bell
,
C.
Lucas
,
A. G.
Marshall
, and
E. C. J.
Oliver
,
2019
:
On the emerging relationship between the stratospheric Quasi-Biennial oscillation and the Madden-Julian oscillation
.
Sci. Rep.
,
9
,
2981
, https://doi.org/10.1038/s41598-019-40034-6.
Lamjiri
,
M. A.
,
M. D.
Dettinger
,
F. M.
Ralph
, and
B.
Guan
,
2017
:
Hourly storm characteristics along the U.S. West Coast: Role of atmospheric rivers in extreme precipitation
.
Geophys. Res. Lett.
,
44
,
7020
7028
, https://doi.org/10.1002/2017GL074193.
Leathers
,
D. J.
,
B.
Yarnal
, and
M. A.
Palecki
,
1991
:
The Pacific/North American teleconnection pattern and United States climate. Part I: Regional temperature and precipitation associations
.
J. Climate
,
4
,
517
528
, https://doi.org/10.1175/1520-0442(1991)004<0517:TPATPA>2.0.CO;2.
Leung
,
L. R.
, and
Y.
Qian
,
2009
:
Atmospheric rivers induced heavy precipitation and flooding in the western U.S. simulated by the WRF regional climate model
.
Geophys. Res. Lett.
,
36
,
L03820
, https://doi.org/10.1029/2008GL036445.
Liu
,
X.
,
X.
Ren
, and
X.
Yang
,
2016
:
Decadal changes in multiscale water vapor transport and atmospheric river associated with the Pacific decadal oscillation and the North Pacific Gyre oscillation
.
J. Hydrometeor.
,
17
,
273
285
, https://doi.org/10.1175/JHM-D-14-0195.1.
Madden
,
R.
, and
P.
Julian
,
1994
:
Observations of the 40–50 day tropical oscillation—A review
.
Mon. Wea. Rev.
,
122
,
814
837
, https://doi.org/10.1175/1520-0493(1994)122<0814:OOTDTO>2.0.CO;2.
Mantua
,
N. J.
, and
S. R.
Hare
,
2002
:
The Pacific decadal oscillation
.
J. Oceanogr.
,
58
,
35
44
, https://doi.org/10.1023/A:1015820616384.
Mantua
,
N. J.
,
S. R.
Hare
,
Y.
Zhang
,
J. M.
Wallace
, and
R. C.
Francis
,
1997
:
A Pacific interdecadal climate oscillation with impacts on salmon production
.
Bull. Amer. Meteor. Soc.
,
78
,
1069
1079
, https://doi.org/10.1175/1520-0477(1997)078<1069:APICOW>2.0.CO;2.
McCarty
,
W.
,
L.
Coy
,
R.
Gelaro
,
A.
Huang
,
D.
Merkova
,
E. B.
Smith
,
M.
Seinkiewicz
, and
K.
Wargan
,
2016
:
MERRA-2 input observations: Summary and assessment. NASA Tech. Rep. NASA/TM-2016-104606/Vol. 46, 61 pp.
, https://gmao.gsfc.nasa.gov/pubs/docs/McCarty885.pdf.
Molod
,
A.
, and et al
,
2020
:
GEOS-S2S Version 2: The GMAO high-resolution coupled model and assimilation system for seasonal prediction
.
J. Geophys. Res. Atmos.
,
125
, e2019JD031767, https://doi.org/10.1029/2019JD031767.
Moon
,
J.-Y.
,
B.
Wang
, and
K.-J.
Ha
,
2011
:
ENSO regulation of MJO teleconnection
.
Climate Dyn.
,
37
,
1133
1149
, https://doi.org/10.1007/s00382-010-0902-3.
Mundhenk
,
B. D.
,
E.
Barnes
, and
E.
Maloney
,
2016
:
All-season climatology and variability of atmospheric river frequencies over the North Pacific
.
J. Climate
,
29
,
4885
4903
, https://doi.org/10.1175/JCLI-D-15-0655.1.
Newman
,
M.
, and et al
,
2016
:
The Pacific decadal oscillation, revisited
.
J. Climate
,
29
,
4399
4427
, https://doi.org/10.1175/JCLI-D-15-0508.1.
Neiman
,
P. J.
, and et al
,
2008
:
Meteorological characteristics and overland precipitation impacts of atmospheric rivers affecting the West Coast of North America based on eight years of SSM/I satellite observations
.
J. Hydrometeor.
,
9
,
22
47
, https://doi.org/10.1175/2007JHM855.1.
Neiman
,
P. J.
,
L. J.
Schick
,
F. M.
Ralph
,
M.
Hughes
, and
G. A.
Wick
,
2011
:
Flooding in western Washington: The connection to atmospheric rivers
.
J. Hydrometeor.
,
12
,
1337
1358
, https://doi.org/10.1175/2011JHM1358.1.
Payne
,
A. E.
, and
G.
Magnusdottir
,
2014
:
Dynamics of landfalling atmospheric rivers over the North Pacific in 30 years of MERRA reanalysis
.
J. Climate
,
27
,
7133
7150
, https://doi.org/10.1175/JCLI-D-14-00034.1.
Payne
,
A. E.
, and
G.
Magnusdottir
,
2016
:
Persistent landfalling atmospheric rivers over the west coast of North America
.
J. Geophys. Res. Atmos.
,
121
,
13 287
13 300
, https://doi.org/10.1002/2016JD025549.
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.
Ralph
,
F. M.
,
T.
Coleman
,
P. J.
Neiman
,
R. J.
Zamora
, and
M. D.
Dettinger
,
2013
:
Observed impacts of duration and seasonality of atmospheric-river landfalls on soil moisture and runoff in coastal Northern California
.
J. Hydrometeor.
,
14
,
443
459
, https://doi.org/10.1175/JHM-D-12-076.1.
Ralph
,
F. M.
,
M. D.
Dettinger
,
M. M.
Cairns
,
T. J.
Galarneau
, and
J.
Eylander
,
2018
:
Defining “atmospheric river”: How the Glossary of Meteorology helped resolve a debate
.
Bull. Amer. Meteor. Soc.
,
99
,
837
839
, https://doi.org/10.1175/BAMS-D-17-0157.1.
Ralph
,
F. M.
, and et al
,
2019
:
ARTMIP-early start comparison of atmospheric river detection tools: How many atmospheric rivers hit northern California’s Russian River watershed?
Climate Dyn.
,
52
,
4973
4994
, https://doi.org/10.1007/s00382-018-4427-5.
Rayner
,
N. A.
,
D. E.
Parker
,
E. B.
Horton
,
C. K.
Folland
,
L. V.
Alexander
,
D. P.
Rowell
,
E. C.
Kent
, and
A.
Kaplan
,
2003
:
Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century
.
J. Geophys. Res.
,
108
,
4407
, https://doi.org/10.1029/2002JD002670.
Reichle
,
R. H.
,
Q.
Liu
,
R. D.
Koster
,
C. S.
Draper
,
S. P.
Mahanama
, and
G. S.
Partyka
,
2017
:
Land surface precipitation in MERRA-2
.
J. Climate
,
30
,
1643
1664
, https://doi.org/10.1175/JCLI-D-16-0570.1.
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.
Rutz
,
J. J.
,
W. J.
Steenburgh
, and
F. M.
Ralph
,
2015
:
The inland penetration of atmospheric rivers over western North America: A Lagrangian analysis
.
Mon. Wea. Rev.
,
143
,
1924
1944
, https://doi.org/10.1175/MWR-D-14-00288.1.
Rutz
,
J. J.
, and et al
,
2019
:
The Atmospheric River Tracking Method Intercomparison Project (ARTMIP): Quantifying uncertainties in atmospheric river climatology
.
J. Geophys. Res. Atmos.
,
124
,
13 777
13 802
, https://doi.org/10.1029/2019JD030936.
Ryoo
,
J. M.
,
Y.
Kaspi
,
D. W.
Waugh
,
G. N.
Kiladis
,
D. E.
Waliser
,
E. J.
Fetzer
, and
J.
Kim
,
2013
:
Impact of Rossby wave breaking on U.S. West Coast winter precipitation during ENSO events
.
J. Climate
,
26
,
6360
6382
, https://doi.org/10.1175/JCLI-D-12-00297.1.
Shields
,
C. A.
, and et al
,
2018
:
Atmospheric River Tracking Method Intercomparison Project (ARTMIP): Project goals and experimental design
.
Geosci. Model Dev.
,
11
,
2455
2474
, https://doi.org/10.5194/gmd-11-2455-2018.
Stan
,
C.
,
D. M.
Straus
,
J. S.
Frederiksen
,
H.
Lin
,
E. D.
Maloney
, and
C.
Schumacher
,
2017
:
Review of tropical-extratropical teleconnections on intraseasonal time scales
.
Rev. Geophys.
,
55
,
902
937
, https://doi.org/10.1002/2016RG000538.
Ullrich
,
P. A.
, and
C. M.
Zarzycki
,
2017
:
TempestExtremes: A framework for scale-insensitive pointwise feature tracking on unstructured grids
.
Geosci. Model Dev.
,
10
,
1069
1090
, https://doi.org/10.5194/gmd-10-1069-2017.
Wallis
,
J. R.
,
M. G.
Schaefer
,
B. L.
Barker
, and
G. H.
Taylor
,
2007
:
Regional precipitation-frequency analysis and spatial mapping for 24-hour and 2-hour durations for Washington State
.
Hydrol. Earth Syst. Sci.
,
11
,
415
442
, https://doi.org/10.5194/hess-11-415-2007.
Warner
,
M. D.
,
C. F.
Mass
, and
E. P.
Salathè
,
2012
:
Wintertime extreme precipitation events along the Pacific Northwest coast: Climatology and synoptic evolution
.
Mon. Wea. Rev.
,
140
,
2021
2043
, https://doi.org/10.1175/MWR-D-11-00197.1.
Wheeler
,
M. C.
, and
H. H.
Hendon
,
2004
:
An all-season real-time multivariate MJO index: Development of an index for monitoring and prediction
.
Mon. Wea. Rev.
,
132
,
1917
1932
, https://doi.org/10.1175/1520-0493(2004)132<1917:AARMMI>2.0.CO;2.
Wu
,
R.
, and
L.
Song
,
2018
:
Spatiotemporal change of intraseasonal oscillation intensity over the tropical Indo-Pacific Ocean associated with El Niño and La Niña events
.
Climate Dyn.
,
50
,
1221
1242
, https://doi.org/10.1007/s00382-017-3675-0.
Xie
,
P.
,
A.
Yatagai
,
M.
Chen
,
T.
Hayasaka
,
Y.
Fukushima
,
C.
Liu
, and
S.
Yang
,
2007
:
A gauge-based analysis of daily precipitation over East Asia
.
J. Hydrometeor.
,
8
,
607
626
, https://doi.org/10.1175/JHM583.1.
Yoo
,
C.
, and
S.-W.
Son
,
2016
:
Modulation of the boreal wintertime Madden-Julian oscillation by the stratospheric quasi-biennial oscillation
.
Geophys. Res. Lett.
,
43
,
1392
1398
, https://doi.org/10.1002/2016GL067762.
Zhang
,
C.
, and
B.
Zhang
,
2018
:
QBO-MJO connection
.
J. Geophys. Res. Atmos.
,
123
,
2957
2967
, https://doi.org/10.1002/2017JD028171.

Footnotes

For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).
1

The underestimation of precipitation over the Olympic mountains and exaggeration over the Cascades has since been reduced in GEOS through the use of the U.S. Geological Survey’s Global Multiresolution Terrain Elevation Dataset for the topography in version 5.17.0 of GEOS, though there is still room for improvement (not shown).