1. Introduction
The Madden–Julian oscillation (MJO; Madden and Julian 1971, 1972) is the dominant planetary-scale intraseasonal mode in the equatorial Indo-Pacific warm pool, characterized by a convection-circulation complex traveling slowly eastward with a period of about 30–90 days. The MJO modulates atmospheric (e.g., tropical cyclones), oceanic (e.g., chlorophyll), and coupled ocean–atmosphere [e.g., El Niño–Southern Oscillation (ENSO)] phenomena in the tropics (e.g., Neale et al. 2008; Lau and Waliser 2011; Jin et al. 2013; Zhang 2013) by exciting a Gill-type response (Gill 1980), and its impacts also extend outside of the tropical region (Stan et al. 2017; Jiang et al. 2020). Diabatic heating related to the MJO leads to the formation of an anomalous Rossby wave source (RWS) in the subtropics and midlatitudes through upper-level MJO divergent wind anomalies impinging on regions with a strong absolute vorticity gradient near the North Pacific westerly jet (Sardeshmukh and Hoskins 1988). The excited circulation anomalies (i.e., MJO teleconnections) in the extratropics (Tseng et al. 2018) modulate weather and climate phenomena such as atmospheric rivers (ARs; a long and narrow filament of concentrated moisture in the lower atmosphere that is responsible for the majority of the poleward moisture transport across the midlatitudes) (Guan and Waliser 2015; Mundhenk et al. 2016; Zhou et al. 2021; Toride and Hakim 2022), extratropical cyclones (Deng and Jiang 2011), and precipitation extremes (e.g., Jones et al. 2004; Joseph et al. 2008; Neena et al. 2011; Guan et al. 2012; Shimizu et al. 2017), and thus could have large societal implications for water resource management on subseasonal (2–6 weeks) time scales.
Extreme precipitation events are among the most devastating natural disasters and pose substantial socioeconomic risks to human society around the world. In the United States, flooding events cost a total of $150 billion (U.S. dollars) in economic damages and caused over 2500 fatalities from 1991 to 2020 (NOAA NCEI 2022, https://www.ncdc.noaa.gov/billions/; https://www.weather.gov/hazstat/). Although flooding can occur in association with many situations, e.g., melting snowpack (Fang and Pomeroy 2016), river overflow (Bomers et al. 2019), and storm surge (Rahmstorf 2017), wet extreme precipitation events are still the dominant causes of floods over the western U.S. region (e.g., Wang et al. 2017). For example, the wettest winter (2016–17) in Northern California (hereafter referred to as “Northern CA”) in recent decades was caused by a series of long-duration (>7 days) extreme precipitation events that led to hazardous flooding (Moore et al. 2020). Given their substantial socioeconomical impacts, the scientific community has taken great efforts to understand the variability and modulation of precipitation extremes, such as their subseasonal variations (e.g., Jones et al. 2004; Singh et al. 2015; Muhammad et al. 2021), interannual modulation by ENSO (e.g., Curtis et al. 2007; Sun et al. 2015; Gore et al. 2020), and potential future changes under anthropogenic warming (e.g., Emori and Brown 2005; Sillmann et al. 2017; Paik et al. 2020). A better understanding of the variability of precipitation extremes on subseasonal time scales would be particularly helpful in improving their subseasonal forecasts to better mitigate against their potential adverse impacts. Several previous studies that were focused on MJO impacts on flooding found that during periods of active MJO days, the frequency of flooding is increased in many regions around the globe such as Indonesia, the western Pacific, Brazil, and the West Coast of North America (Jones 2000; Jones et al. 2004; Vasconcelos et al. 2018; Muhammad et al. 2021). Although there have been attempts to relate subseasonal variations of precipitation extremes to MJO activity, a clear and systematic understanding is still lacking as to how the extremes over the western United States vary on this time scale in frequency, intensity, and duration in response to MJO variability, which will be the focus of this study.
This study will examine the seasonality of MJO impacts on precipitation extremes. Previous studies have documented the pronounced seasonality of the MJO between boreal winter and summer, and have shown that MJO events are more frequent and organized during boreal winter when the mean jet stream is stronger (Zhang and Dong 2004; Adames et al. 2016). The MJO is also dominated more by an eastward propagation in boreal winter compared to being a strong northward-propagating feature in summer (Adames et al. 2016). However, differences in the MJO characteristics and their impacts on precipitation extremes within the extended winter season (October–March) have received less attention. Bond and Vecchi (2003) found substantially different MJO impacts on seasonal-mean precipitation in Oregon and Washington between late autumn/early winter [October–December (OND)] and late winter [January–March (JFM)]. Zhou et al. (2012) also showed seasonal dependence of MJO impacts on U.S. temperature and precipitation.
Inspired by these studies, the present study proposes a hypothesis that the MJO impacts on precipitation extremes may have strong seasonal variation due to changes in MJO characteristics and its associated extratropical response. First, it is expected that there will be a large difference in MJO-associated extratropical circulation patterns (i.e., MJO teleconnections), as MJO teleconnections are highly dependent on both the MJO and the basic state such as the westerly jet (e.g., Henderson et al. 2017; Wang et al. 2020) which all possess strong seasonality. The characteristic large-scale patterns that are conducive to the seasonality of MJO-related precipitation extremes will be identified. In addition to MJO teleconnections, we also examine how the moisture transport and the associated AR variations give rise to the seasonality of MJO-related precipitation extremes. ARs are the primary drivers of flooding over the western United States (e.g., Corringham et al. 2019; Lavers et al. 2020; Prince et al. 2021) and have historically caused about 88% of flood damage over this region. ARs are also found to be largely modulated by the MJO: when the enhanced MJO convection is located over the western Pacific, there is a tendency for increased AR frequency offshore of the West Coast (Ralph et al. 2011; Guan et al. 2012; Payne and Magnusdottir 2014; Guan and Waliser 2015; Mundhenk et al. 2016; Zhou et al. 2021). In this study, the seasonality of MJO modulation on AR frequency, intensity, and duration is examined to understand the AR contribution to the seasonality of MJO impacts on precipitation extremes.
The paper is organized as follows. The observational data used in this study are described in section 2, along with the definition of wet extreme intensity, frequency, and duration. The climatology of precipitation extremes and their modulations by the MJO in the boreal winter season is examined in section 3. The seasonality of MJO modulation is discussed in section 4. Physical mechanisms are examined in section 5 by discussing changes in the MJO characteristics, their teleconnections, and modulation of western U.S. AR characteristics by the MJO. The summary and discussion are provided in section 6.
2. Data and methods
a. Precipitation extremes
Observational precipitation data used in this study are obtained from the NOAA Climate Prediction Center (CPC) global unified gauge-based daily precipitation dataset (Chen et al. 2008), which has a horizontal resolution of 0.5° × 0.5°. The analysis of precipitation extremes is focused on the boreal winter season from October to March when the MJO and its extratropical impacts are most prominent (e.g., Stan et al. 2017) from 1979 to 2019. Precipitation extreme events are selected based on the two-parameter gamma frequency distributions of nonzero precipitation, constructed as follows. First, the two-parameter gamma frequency distribution is fitted to each time series of nonzero boreal winter daily precipitation via the maximum likelihood approach at each grid point (Jones et al. 2004) to estimate the precipitation frequency distribution. A fitted gamma distribution function instead of a histogram is used to estimate precipitation frequency to reduce the effect of the skewness in precipitation (Jones et al. 2004). The percentage of nonzero precipitation days is shown in Fig. 1a. The northwestern United States has a much higher (>60%) frequency of nonzero precipitation, which is equivalent to >4360 days. Overall, nonzero precipitation days account for >10% of total days (>720 days) over the entire western United States during the period we analyzed. Daily wet extreme events are defined when the observed precipitation is above the 90th percentile of the gamma distribution.
(a) Percentage of days with nonzero precipitation. (b) 90th percentile of nonzero precipitation distributions (mm day−1) over the western United States from October to March during 1979–2019. Northern, Central, and Southern CA are outlined in red, blue, and yellow, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
(a) Percentage of days with nonzero precipitation. (b) 90th percentile of nonzero precipitation distributions (mm day−1) over the western United States from October to March during 1979–2019. Northern, Central, and Southern CA are outlined in red, blue, and yellow, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
(a) Percentage of days with nonzero precipitation. (b) 90th percentile of nonzero precipitation distributions (mm day−1) over the western United States from October to March during 1979–2019. Northern, Central, and Southern CA are outlined in red, blue, and yellow, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Figure 1b shows the climatology of the observational 90th percentile (wet extreme) nonzero precipitation over the western United States. Wet extreme intensity is substantially stronger over the West Coast, with the largest intensity values exceeding 27 mm day−1 in western Washington, western Oregon, and Northern CA, consistent with Jones and Carvalho (2012). The 90th percentile precipitation values in other regions are generally around 3–12 mm day−1. The above spatial patterns and thresholds (adjusted for season and geographical location) are used for the extreme event selection. The climatology of the 90th percentile nonzero daily precipitation over the western United States derived from raw distribution (Fig. S1 in the online supplemental material) is overall similar to that derived from the fitted distribution (Fig. 1) but with some regional differences such as higher 90th percentile thresholds over CA (up to ∼3 mm day−1 higher). The main conclusions in this study are not sensitive to the usage of less stringent thresholds (i.e., 75th percentile for flooding) for selecting the extreme events (e.g., Fig. S2). Changes in characteristics of precipitation extremes related to active MJO days are examined, including 1) precipitation intensity, defined as the average daily precipitation amount of the extreme events; 2) relative frequency, defined as extreme days for a given MJO phase divided by the total days in that phase; and 3) duration, defined as maximum consecutive days that precipitation exceeds the extreme threshold within an 11-day window, which is roughly the persistence time of MJO teleconnections (Wang et al. 2020).
b. MJO
This study uses the real-time multivariate MJO (RMM) indices to characterize the phase and amplitude of the MJO (http://www.bom.gov.au/climate/mjo) that are available with no missing values after 1979. The RMM indices are derived from the combined empirical orthogonal functions (CEOF) of 15°S–15°N averaged OLR and 850- and 200-hPa zonal wind anomalies (Wheeler and Hendon 2004). This study focuses on changes in precipitation extremes in response to strong MJO days defined as those with an amplitude of the RMM index greater than one standard deviation. To increase the sample size, we combine every two MJO phases based on their similar propagation location (e.g., phases 2–3 represent enhanced MJO convection over the Indian Ocean).
c. AR objects
AR occurrence is estimated based on the global AR detection algorithm developed in Guan and Waliser (2015) and updated and validated with in situ/dropsonde data later in Guan et al. (2018). This algorithm identifies AR objects based on 1) the integrated vapor transport (IVT) magnitude thresholding (85th percentile of climatological IVT, adjusted for season and geographical location) and 2) geometric requirements (length > 2000 km and length-to-width ratio > 2). It has been applied to various reanalysis datasets and is widely used in previous literature (e.g., Mundhenk et al. 2016; Shields et al. 2018). In this study, we use the AR detection dataset based on NASA Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA2; Gelaro et al. 2017). The 6-hourly MERRA-2 AR data span from 1980 to 2020 at a horizontal resolution of 0.625° × 0.5° and provide information on the time and location of detected AR objects. Other variables from MERRA-2 include geopotential height and IVT which are used to examine the MJO teleconnections, moisture transport, and AR intensity.
d. Significance test
To assess the statistical significance of MJO impacts, a moving-blocks bootstrap test with replacement is employed in this study. This technique, unlike the traditional bootstrap which samples from individual elements from the sample pool, samples consecutive elements (i.e., blocks) to account for the data autocorrelation and thus is more suitable for MJO studies (e.g., Hamill et al. 2013; Henderson et al. 2016; Wolding et al. 2017; Baggett et al. 2018; Mundhenk et al. 2018; Martin et al. 2021; Wang et al. 2022). The bootstrap sample size M is equivalent to the number of MJO days used in the composites by joining M/5 blocks, where 5 is the block length that represents the average duration in days of an individual MJO phase (Alaka and Maloney 2012; Henderson et al. 2016). However, within the given M, only MJO days that are concurrent with extreme precipitation are included in the calculation. The effective sample size, therefore, changes regionally over time lags and is generally smaller than the total number of days in a particular MJO phase. The moving-blocks bootstrap is stricter than the traditional bootstrap given that a block length that is too small (i.e., 1 day in traditional bootstrap) can result in an overestimate of the significance if there is autocorrelation in the time series (Marchand et al. 2006). The bootstrapping procedure is repeated 1000 times to obtain a sufficiently large sample size and the 2.5th and 97.5th percentile values are used to define the 95% confidence interval. The results are significantly different from climatology if the confidence interval is entirely greater or smaller than the climatology, which indicates positive and negative anomalies, respectively.
3. MJO impacts on western U.S. precipitation extremes during boreal winter
The climatologies of wet extreme intensity, frequency, and duration independent of MJO phase are shown in Fig. 2. Only nonzero days are used in the calculation. The spatial distributions of wet extreme intensity (Fig. 2a) are consistent with the extreme thresholds (Fig. 1) which show that western Washington and western Oregon generally receive much more intense precipitation than the other regions. In California (CA), wet extreme intensity is also high, with some regions having values over 50 mm day−1. The frequency of wet extremes, which varies around 10% (Fig. 2b), shows the spatial variation as a result of the fitting of daily precipitation at each point to the gamma distribution. The climatological duration of wet extremes is calculated as the maximum consecutive days when precipitation exceeds the extreme threshold within an 11-day window independent of MJO phase and is typically around 1–2 days over the western United States, with some regions over the West Coast (e.g., Northern CA) having a duration exceeding 2 days (Fig. 2c). In summary, on average, wet extremes along the West Coast have higher intensities (up to ∼50 mm day−1) than in the other western U.S. regions and last up to 2 days during boreal winter.
Climatological (a) intensity (mm day−1), (b) frequency (percent of days), and (c) duration (number of days) of wet extremes over the western United States from October to March during the period of 1979–2019.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Climatological (a) intensity (mm day−1), (b) frequency (percent of days), and (c) duration (number of days) of wet extremes over the western United States from October to March during the period of 1979–2019.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Climatological (a) intensity (mm day−1), (b) frequency (percent of days), and (c) duration (number of days) of wet extremes over the western United States from October to March during the period of 1979–2019.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The absolute differences in wet extremes between strong MJO days and climatology are shown in Fig. 3. A 5–9-day lagged average is applied to the MJO-related precipitation extreme intensity and frequency, as it takes about 1–2 weeks for tropically forced waves to reach the midlatitudes (Hoskins and Karoly 1981). For example, for extreme intensity, we first take the composite of extreme values for events that occurred 5 days after a certain MJO phase combination. Then, another composite is taken for extremes that occurred 6 days after that MJO phase, and so on. The 5–9-day lagged average is the average of those composites from 5 to 9 days after the active MJO days. A lag window of 5–15 days is used for calculating the duration of precipitation extremes, that is, during 5–15 days after an active MJO day, we count the maximum consecutive days that precipitation exceeds the extreme thresholds and the composite is made for all MJO days. This 11-day window length is chosen to match the persistence time of MJO teleconnections (Wang et al. 2020) which varies from 10 to 18 days after active MJO days.
Absolute changes (defined as precipitation extremes in strong MJO phases minus climatology) in (left) intensity (mm day−1), (center) frequency (%), and (right) duration (day) of wet extremes averaged over 5–9 lagged days after active MJO days (shading). Dots indicate that the changes are significant relative to climatology over the 95% confidence level based on the bootstrap test. The number of days for MJO phases 1 and 8, 2–3, 4–5, and 6–7 is 747, 926, 974, and 1083, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Absolute changes (defined as precipitation extremes in strong MJO phases minus climatology) in (left) intensity (mm day−1), (center) frequency (%), and (right) duration (day) of wet extremes averaged over 5–9 lagged days after active MJO days (shading). Dots indicate that the changes are significant relative to climatology over the 95% confidence level based on the bootstrap test. The number of days for MJO phases 1 and 8, 2–3, 4–5, and 6–7 is 747, 926, 974, and 1083, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Absolute changes (defined as precipitation extremes in strong MJO phases minus climatology) in (left) intensity (mm day−1), (center) frequency (%), and (right) duration (day) of wet extremes averaged over 5–9 lagged days after active MJO days (shading). Dots indicate that the changes are significant relative to climatology over the 95% confidence level based on the bootstrap test. The number of days for MJO phases 1 and 8, 2–3, 4–5, and 6–7 is 747, 926, 974, and 1083, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Figure 3 suggests that MJO activities have a more significant and regionally consistent influence on the wet extreme frequency than intensity and duration. MJO phases 1 and 8 and 6–7 (when the enhanced MJO convection is over the Pacific) generally lead to a significant increase in wet extreme frequency in CA and Arizona while opposite changes are found in phases 2–3 and 4–5 when the enhanced MJO convection is over the Indian Ocean and Maritime Continent. Other significant changes in wet extreme frequency include a decrease in South Dakota during phases 1 and 8 and 2–3, and a decrease in Colorado and Wyoming during phases 4–5. The main change in intensity is that wet extreme intensity is significantly increased in Southern CA and Arizona during MJO phases 6–7 and decreased during MJO phases 2–3. Reductions in intensity also appear significant in Western Washington during phases 1 and 8 and a portion of Texas during phases 2–3. The only significant change in duration that is found in both phases 2–3 and 6–7 is in Oregon, such that wet extremes have a longer duration when the MJO is in phases 6–7 and a shorter duration when the MJO is in phases 2–3.
Figure 4 shows the 95th confidence intervals established from the bootstrapping samples with respect to various percentiles (median, 75th, 90th) of wet extremes averaged over Northern, Central, and Southern CA (outlined in Fig. 1b) during the 5–9-day lag time with respect to the MJO phase. The choice of boundaries for these regions is similar to Hoell et al. (2016). Only precipitation events above the 90th percentile are used to construct the percentile bars. Values along the x axis represent the ranges of wet extremes, which suggest a stronger extreme intensity in Northern CA compared to Central and Southern CA, consistent with results shown in Fig. 1. In general, significant MJO impacts are mainly found over Central and Southern CA. For example, the 90th percentile of wet extremes (i.e., the right tail of wet extremes) is statistically higher during MJO phases 4–5 over Central CA, which is consistent with stronger intensity than climatology in phases 4–5 (Fig. 3c). There is a significant positive shift in the median and the tails of the distribution of wet extremes during phases 6–7 and a negative shift during phases 2–3 over Southern CA.
Composites (dots) and the 95% confidence intervals (lines) of various percentiles (red: median; blue: 75th; green: 90th) of wet extremes for different MJO phase combinations and all extreme days (CLIM).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Composites (dots) and the 95% confidence intervals (lines) of various percentiles (red: median; blue: 75th; green: 90th) of wet extremes for different MJO phase combinations and all extreme days (CLIM).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Composites (dots) and the 95% confidence intervals (lines) of various percentiles (red: median; blue: 75th; green: 90th) of wet extremes for different MJO phase combinations and all extreme days (CLIM).
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The corresponding probability distribution functions (PDFs) of wet extremes for phases 2–3 and 6–7 are shown in Fig. S3. The distributions of climatological wet extreme events suggest a decrease in probability when the intensity of precipitation increases, which is common for precipitation distributions (e.g., Takahashi et al. 2022). For Northern CA, PDFs for wet extremes are not significantly skewed toward either lower or higher intensity during MJO days, suggesting weak MJO impacts on wet extreme intensity over this region. The largest difference in PDFs is found over Southern CA, where precipitation amounts greater than 20 mm day−1 occurred much more frequently in phases 6–7 compared to phases 2–3; in particular, phases 2–3 have very few wet extreme events greater than 30 mm day−1.
The median, 75th percentile, and 90th percentile of all nonzero precipitation events over Northern, Central, and Southern CA and events that are related to MJO activity are given in Fig. S4. Similar to MJO impacts on wet extreme intensity, the largest and most significant MJO impacts on frequency are found over the Southern CA region, given that the median and tails of nonzero precipitation distributions are all positively skewed during MJO phases 6–7 compared to climatology. The opposite impacts are found during MJO phases 4–5, when precipitation distributions are negatively skewed toward values lower than climatology in all three regions. The above results further suggest that active MJO days during phases 6–7 are associated with a significant increase in wet extreme intensity and frequency, especially over Southern CA.
We further quantify the MJO impacts on precipitation extremes in CA by calculating the area averages of the percentage changes over Northern, Central, and Southern CA over various lag times (Fig. 5). The area averages of Washington and Oregon are given in Fig. S8. The results shown in Fig. 5 are consistent with the spatial distributions in Fig. 3, which suggest that the MJO has the largest and most significant impact on extreme frequency (up to ∼40% relative to climatology for wet extremes), followed by its impacts on extreme intensity (up to ∼20%). For wet extreme frequency, there are statistically significant increases in wet extreme frequency in all three CA regions 5–14 days following an MJO in phases 6–7. A very large and statistically significant increase in wet extreme frequency is also found at a longer time lag (25–29 days) in Central and Southern CA following MJO phases 1 and 8.
Regional averages of relative changes (%) in wet extremes following active MJO days over Northern CA, Central CA, and Southern CA. Results over different time lags (x axis) are shown. Boxes with black circles are significant over the 95% confidence level.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Regional averages of relative changes (%) in wet extremes following active MJO days over Northern CA, Central CA, and Southern CA. Results over different time lags (x axis) are shown. Boxes with black circles are significant over the 95% confidence level.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
Regional averages of relative changes (%) in wet extremes following active MJO days over Northern CA, Central CA, and Southern CA. Results over different time lags (x axis) are shown. Boxes with black circles are significant over the 95% confidence level.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
We quantify the consistency of the MJO impacts from both a regional perspective (i.e., the extent to which the three different regions in CA have the same response to the MJO impacts) and a temporal perspective (i.e., how consistent the impacts are when the lag time increases; hereafter referred to as “persistence”). It is expected that more regionally consistent and persistent MJO impacts have the potential to be predicted farther in advance with higher skill. The results are summarized in Table 1.
Regional consistency and persistence of MJO impact on precipitation extremes over CA. Results of regional consistency show the total number (sum) of same-sign changes between Northern CA, Central CA, and Southern CA over all the time lags. Persistence indicates the total number (sum) of same-sign changes between different time lags over CA. Values of zero are not shown. Results are detailed in MJO phases from phases 1 and 8 (P1 and 8) to phases 6–7 (P6–7).
The values of regional consistency are the total numbers of boxes of each column in Fig. 5 that have the same sign of the response for all three regions of CA for a given MJO phase combination summed across all lag times and then divided by three. For example, the value of regional consistency for wet extreme intensity is 16. This is derived as the sum of 6 boxes of decreased wet extreme intensity in all three regions of CA for phases 2–3 and 4–5 and 3 boxes of increased intensity for phases 6–7 during the lag days from day 0 to day 4, 3 boxes of decreased intensity for phases 1–8 and 6 boxes of increased intensity for phases 4–5 and 6–7 during lag days 5–9, and so on for the following time lags, leading to a total of 48 boxes divided by 3. The larger the value, the larger the regional consistency of MJO impacts on CA precipitation extremes.
The values of persistence in Table 1 represent the numbers of rows in Fig. 5 for which precipitation extremes over a given region have the same response across all time lags for a specific MJO phase combination, summed over the three regions of CA. For example, there is only one row (Northern CA, phases 1 and 8) for the wet extreme intensity that has the above feature (Fig. 5a), while there are four rows (Central and Southern CA, phases 1 and 8; Northern and Central CA, phases 6–7) for the wet extreme frequency that meet the above criteria. A larger metric value suggests a longer persistence of MJO impacts.
The results listed in Table 1 show that impacts from MJO phases 1 and 8 and 2–3 are generally more persistent and have larger regional consistency for CA precipitation extremes. Impacts from MJO phases 6–7, however, show large regional variation during the extended boreal winter. Among the three extreme characteristics we analyzed, changes in frequency in response to the MJO may have higher predictability potential, as these changes have the largest regional consistency and longest persistence.
4. Seasonality (OND versus JFM) of MJO impacts on western U.S. precipitation extremes
In this section, the seasonal dependence of MJO impacts on precipitation extremes is examined. We focus mainly on MJO phases 2–3 and 6–7 (results for phases 1 and 8 and 4–5 are shown in Figs. S5 and S6 in the supplemental material). Changes in wet extreme events in response to the MJO during early winter (OND) and late winter (JFM) relative to their corresponding climatology are compared in Fig. 6. Stronger seasonality is found for extreme intensity and frequency compared to duration. For MJO phases 2–3: in OND, wet extreme events occur more frequently than the climatological average in Central-Southern CA, and less frequently in Washington, Oregon, eastern Montana, North Dakota, and South Dakota; in JFM, the less frequent wet extremes are significant in a more southward location over CA, indicating a shift in the preferred location of MJO impacts along the West Coast, while eastern Montana and North Dakota generally experience an increased extreme frequency. Some significant differences are also found in extreme intensity, such as more intense extremes over western Washington and less intense extremes over Northern and Southern CA in OND, and opposite or insignificant changes in JFM. For MJO phases 6–7: in OND, there is an increase in wet extreme intensity and frequency near the West Coast except for Central CA, while in JFM, the MJO impact can be the opposite and less significant. Significant opposite response of wet extreme frequency between OND and JFM is also found in some inland regions such as New Mexico and Texas. The destructive interference between opposing changes in frequency (and to a lesser extent, intensity) during OND and JFM leads to weaker and less significant signals examined over the entire winter season (Fig. 6 versus Fig. 3). These findings suggest value for future studies to consider the seasonal dependence of the MJO–precipitation extreme relationship.
As in Fig. 3, but for a comparison between early winter (OND) and late winter (JFM) during MJO phases 2–3 and 6–7. The number of days for MJO phases 2–3 and 6–7 in OND (JFM) are 434 (481) and 440 (628), respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
As in Fig. 3, but for a comparison between early winter (OND) and late winter (JFM) during MJO phases 2–3 and 6–7. The number of days for MJO phases 2–3 and 6–7 in OND (JFM) are 434 (481) and 440 (628), respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
As in Fig. 3, but for a comparison between early winter (OND) and late winter (JFM) during MJO phases 2–3 and 6–7. The number of days for MJO phases 2–3 and 6–7 in OND (JFM) are 434 (481) and 440 (628), respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
A comparison of median, 75th percentile, and 90th percentile of wet extremes between OND and JFM is given in Fig. 7 for MJO phases 2–3 and 6–7 and the corresponding PDFs are shown in Fig. S7. Again, only precipitation events above the 90th percentile are used to construct these confidence intervals. Climatologically, larger seasonality in precipitation extremes exists over Central–Southern CA compared to Northern CA, such that wet extremes are generally more intense in JFM compared to OND. In Fig. 6, we found that the magnitude of the decrease in wet extreme intensity is stronger in OND compared to JFM for phases 2–3 over Northern CA. This could be due to the stronger negative shift of the median wet extreme distributions (Fig. 7a). Opposite responses in wet extremes between OND and JFM are found during MJO phases 6–7 over Central CA (Fig. 7b), which shows a negative shift of the median and 75th percentile wet extremes during OND and a positive shift during JFM. The wet extreme intensity in response to MJO phases 6–7 is thus overall weaker than climatology in OND and stronger than climatology during JFM in Central CA (Figs. 6c,d). In southern CA, MJO impacts are more significant during OND than in JFM, when wet extreme intensity is weaker than climatology during phases 2–3 and stronger than climatology during phases 6–7 (Fig. 7c).
As in Fig. 4, but for a comparison between OND and JFM.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
As in Fig. 4, but for a comparison between OND and JFM.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
As in Fig. 4, but for a comparison between OND and JFM.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The seasonality of MJO impacts is further quantified in Fig. 8. Similar to Fig. 5, we calculated the area averages of the percentage changes over Northern, Central, and Southern CA and compared them between OND and JFM. Results for Washington and Oregon are given in Fig. S9. Significant differences in MJO impacts on wet extremes over CA are discussed below: 1) There is a robust and statistically significant increase over most time lags in wet extreme frequency during OND for phases 2–3 and 6–7, and a robust and statistically significant decrease during JFM for phases 2–3, especially over Central and Southern CA. The magnitude of the increase is generally greater during OND than the decrease during JFM; 2). The wet extreme intensity is more likely to decrease by ∼10% during OND and increase by ∼7% during JFM over Central and Southern CA in phases 2–3. 3) There are large, robust, and persistent decreases in wet extreme intensity in OND over Central CA for MJO phases 6–7 compared to the significant increases at lag times of up to 2 weeks during JFM; 4) The MJO impacts on wet extreme duration are generally more pronounced during JFM compared to OND, e.g., a robust and persistent decrease during MJO phases 2–3.
As in Fig. 5, but for a comparison between OND and JFM.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
As in Fig. 5, but for a comparison between OND and JFM.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
As in Fig. 5, but for a comparison between OND and JFM.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
We now discuss changes in the regional and temporal consistency (Table 2). The results in Table 2 are calculated using the same method as in section 3 and are shown as relative changes between OND and JFM (i.e., values in OND minus values in JFM, normalized by values in JFM). Positive differences suggest stronger regional consistency or longer persistence in OND than in JFM, and vice versa for negative differences. The MJO in OND, on average, leads to more regionally consistent impacts on wet extremes compared to JFM during phases 6–7, which is especially significant for intensity and frequency. In addition, the MJO in OND is associated with a significantly more persistent response in wet extremes. The above results further confirm that the MJO in OND will have a more pronounced impact on precipitation extremes over CA compared to JFM, especially in phases 6–7, and therefore may be a more important factor to consider during early boreal winter for subseasonal prediction of precipitation over CA.
As in Table 1, but for a comparison between OND and JFM. Results are shown as relative changes (%), i.e., values in OND minus values in JFM and then divided by values in JFM to facilitate comparison. If the results are positive, it indicates a stronger regional consistency or persistence of MJO impacts in OND, and vice versa if the results are negative. Significant differences greater than the 95% confidence level are in bold. Results are not shown if the metric values are the same between OND and JFM.
In summary, the seasonality of MJO impacts is examined and quantified in this section. The results suggest that active MJO days that occur in OND are associated with more significant changes in CA precipitation extremes with larger regional consistency and longer persistence compared to JFM. An opposite or asymmetric response may exist between the seasons. For example, MJO phases 2–3 are associated with a ∼51% increase in wet extreme frequency in CA during OND when averaging over all the time lags compared to a ∼38% decrease during JFM. MJO phases 6–7 are associated with a ∼47% increase in wet extreme frequency during OND, but their impacts are much less significant (only ∼4% decrease) during JFM.
5. Seasonality of MJO teleconnections and AR activities
To understand the physical mechanisms that contribute to the seasonality of the MJO– extreme precipitation relationship discussed in section 4, we examine the 5–9-day lagged average of MJO teleconnections (represented by 25–90-day filtered 500-hPa geopotential height anomalies; Z500a) and moisture transport (represented by filtered 1000–300-hPa IVT anomalies; IVTa) after the MJO activity at day 0 during phases 2–3 and 6–7. The comparison between the seasons is given in Fig. 9, along with the 250-hPa mean zonal wind to indicate the differences in the basic state westerly jet. The results show that the MJO and the jet in OND are overall weaker and situated more westward and northward compared to their magnitude and position during JFM. This can lead to different patterns and amplitudes in MJO teleconnections (Wang et al. 2020) such as the weaker and northwestward shifted Z500a in OND than in JFM as shown in Fig. 9. The anomalous moisture transport is also influenced by the MJO and its teleconnections, which tend to be more zonally oriented in OND than in JFM.
The 5–9-day averaged lagged response of filtered Z500a (shading; m) and IVTa (vectors; kg m−1 s−1) to day 0 MJO phases 2–3 and 6–7 [OLR contour: green (brown) represents enhanced (suppressed) convection, interval: 5 W m−2] in (left) OND and (right) JFM. The dotted areas represent significant Z500a exceeding the 95% confidence level according to the two-tailed Student’s t test. Vectors that are shown are significant IVTa. Black contours (interval: 10 m s−1; values smaller than 30 m s−1 are omitted) indicate the 250-hPa mean zonal wind in OND and JFM, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The 5–9-day averaged lagged response of filtered Z500a (shading; m) and IVTa (vectors; kg m−1 s−1) to day 0 MJO phases 2–3 and 6–7 [OLR contour: green (brown) represents enhanced (suppressed) convection, interval: 5 W m−2] in (left) OND and (right) JFM. The dotted areas represent significant Z500a exceeding the 95% confidence level according to the two-tailed Student’s t test. Vectors that are shown are significant IVTa. Black contours (interval: 10 m s−1; values smaller than 30 m s−1 are omitted) indicate the 250-hPa mean zonal wind in OND and JFM, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The 5–9-day averaged lagged response of filtered Z500a (shading; m) and IVTa (vectors; kg m−1 s−1) to day 0 MJO phases 2–3 and 6–7 [OLR contour: green (brown) represents enhanced (suppressed) convection, interval: 5 W m−2] in (left) OND and (right) JFM. The dotted areas represent significant Z500a exceeding the 95% confidence level according to the two-tailed Student’s t test. Vectors that are shown are significant IVTa. Black contours (interval: 10 m s−1; values smaller than 30 m s−1 are omitted) indicate the 250-hPa mean zonal wind in OND and JFM, respectively.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
During MJO phases 2–3 in OND (Fig. 9a), anomalous ridges located over the northwestern Pacific and North America give rise to significant zonal IVTa from the northwestern United States to the Pacific Ocean, which is associated with a significant decrease in extreme frequency over the Pacific Northwest (Fig. 6e). In JFM (Fig. 9c), strong troughs are present over North America, leading to significant southward IVTa around the strong Pacific ridge that extends eastward near the West Coast. This anomalous moisture transport does not penetrate the western United States, especially the southwestern portion. As a result, the southwestern United States is left with a significant decrease in extreme frequency (Fig. 6f). During MJO phases 6–7, there is a weak trough over the western United States in OND instead of a ridge in JFM. The IVTa associated with this trough is transporting more moisture from the Pacific Ocean to the southwestern United States, providing a moisturized environment in this region that favors an increase in extreme frequency (Fig. 6g). On the other hand, the northward IVTa near the West Coast in JFM associated with the ridge transports the moisture toward higher latitudes and leaves the U.S. West Coast with relatively dry conditions compared to OND (Fig. 6h).
The above differences in the MJO, its teleconnections, and associated moisture transport all suggest different responses in western U.S. AR activity between the seasons. The OND and JFM AR climatologies are shown in Fig. S10 for AR intensity (defined as the mean IVT averaged over AR events), AR frequency (defined as AR occurrence divided by the number of active MJO days, where AR occurrence is expressed as the number of 6-hourly time steps divided by four), and AR duration (similar definition with precipitation extremes with the unit in hours). It is shown that the AR characteristics have strong seasonality (Payne and Magnusdottir 2014; Rutz et al. 2014; Guan and Waliser 2015). ARs generally have stronger intensity, lower frequency, and shorter duration in OND than in JFM. There is also a southward shift in the high AR occurrence region in JFM compared to OND. The seasonality of MJO impacts on AR intensity, frequency, and duration are shown in Fig. 10. Generally, the MJO has a more significant influence on AR intensity in OND than in JFM over the western United States. In OND, a significant decrease in AR intensity is observed in Oregon, Nevada, and Idaho 5–9 days after MJO phases 2–3 (Fig. 10a) and a significant increase in AR intensity over Northern CA, Oregon, and Washington is observed after MJO phases 6–7 (Fig. 10c). The response in AR frequency is consistent with Fig. 9 that changes are more zonally oriented in OND and meridionally oriented in JFM along the U.S. West Coast. When the MJO is in phases 2–3, an elongated band of decreasing AR frequency is found south of 40°N in OND (Fig. 10e) including the Northern–Central CA and Oregon along the south side of ridges where westward moisture transport is found (Fig. 9a). In JFM, the significant decrease in AR frequency is generally found at a more southward location over the Central–Southern CA and Baja California (Fig. 10f). When the MJO is in phases 6–7, the opposite response is found over North America between the two seasons; a significant increase in AR frequency is expected over CA in OND and over British Columbia in JFM. MJO impacts on AR duration are more significant during phases 2–3. ARs tend to have a shorter duration in OND and a longer duration in JFM in CA (Figs. 10i,j).
The 5–9-day averaged lagged response of (left) AR intensity, (center) AR frequency, and (right) AR duration in OND and JFM after MJO phases 2–3 and 6–7. The dotted areas represent significant AR changes exceeding the 95% confidence level according to the moving-blocks bootstrapping test.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The 5–9-day averaged lagged response of (left) AR intensity, (center) AR frequency, and (right) AR duration in OND and JFM after MJO phases 2–3 and 6–7. The dotted areas represent significant AR changes exceeding the 95% confidence level according to the moving-blocks bootstrapping test.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The 5–9-day averaged lagged response of (left) AR intensity, (center) AR frequency, and (right) AR duration in OND and JFM after MJO phases 2–3 and 6–7. The dotted areas represent significant AR changes exceeding the 95% confidence level according to the moving-blocks bootstrapping test.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
In summary, the seasonality of MJO impacts on precipitation extremes discussed in section 4 could be due to the different patterns and amplitude of the MJO and basic state between the seasons which then lead to different responses in MJO teleconnections, moisture transport, and AR activities. When the MJO is over the Indian Ocean (phases 2–3), the decrease in AR frequency over the U.S. West Coast is located more northward in OND than that in JFM, leading to a decrease in wet extreme frequency that is located in Oregon and Washington in OND and CA in JFM. When the MJO is over the western Pacific (phases 6–7), MJO teleconnections in OND are more favorable for the inland moisture transport from the Pacific Ocean toward the western United States, which gives rise to an increase in AR intensity and frequency over this region and leads to an increase in wet extreme intensity and frequency, while weak AR and wet extreme response are found in JFM. Note that although most of the seasonality of MJO–wet extreme relationships could be explained by changes in MJO teleconnections and the associated variations in ARs, some changes in wet extremes are not consistent with AR changes. For example, a shorter duration of CA wet extremes is expected during JFM after MJO phases 2–3 (Fig. 6j) although the AR activity generally lasts longer over this region (Fig. 10j). More investigation is thus needed in a future study to further understand the contributions of ARs to the MJO-related precipitation extremes.
6. Summary and discussion
In this work, we investigate the impact of the MJO on the intensity, frequency, and duration of boreal winter precipitation extremes over the western United States and their differences between late autumn/early winter and late winter (OND versus JFM). During the extended boreal winter from October to March, active MJO days in phases 6–7 are generally associated with more intense, more frequent, and longer-lasting wet extremes over portions of the western United States relative to climatology; active MJO days in phases 2–3 are generally associated with the opposite response, which is especially significant in CA. The above MJO impacts are more pronounced in OND and can be opposite in JFM, especially for extreme frequency. More regionally consistent and persistent MJO impacts are seen in OND compared to JFM.
The seasonality of MJO impacts on large-scale circulations and AR activities are examined to further understand the underlying mechanisms. The key findings are summarized in Fig. 11 for an example of MJO phases 6–7, which are associated with the strongest seasonality in precipitation extremes among the MJO phases. The MJO has a weaker amplitude and is located more northwestward in OND than in JFM (i.e., the enhanced convection is located near the Maritime Continent in OND and over the central Pacific in JFM). The jet core is weaker and positioned further north in OND than in JFM. The above MJO and basic state differences together contribute to the weaker MJO teleconnections that are located more northwestward in OND than in JFM according to Wang et al. (2020). As a result, the western United States is dominated by a weak trough in OND as compared to a ridge in JFM. The moisture transport around the trough in OND then brings moisture from the Pacific Ocean toward the western United States, leading to an increase in AR intensity and frequency along the south side of the trough and a decrease in AR activity to the north. The increased moisture supply over the western United States favors an increase in extreme precipitation intensity over Oregon and Washington and extreme precipitation frequency over CA in OND. In comparison, the eastward extended Pacific trough and southward extended ridge in JFM lead to a more poleward diverted moisture transport toward Alaska, leaving the western United States with relatively dry conditions and hence a decrease in AR activity and precipitation extremes over this region.
The schematic diagram of extratropical response (precipitation extremes, Z500a, anomalous moisture transport, and AR activity) to MJO phases 6–7 in (top) OND and (bottom) JFM. Results are derived from the 5–9-day lagged average after the active MJO day at day 0. The green triangle indicates the center longitude of enhanced MJO convection at day 0. The jet (mean 250-hPa zonal wind as the basic state) interval is 20 m s−1 starting at 30 m s−1. The relative magnitude of each component is indicated by the difference in the thickness and length, and the variables shown may not be entirely precise in location and pattern.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The schematic diagram of extratropical response (precipitation extremes, Z500a, anomalous moisture transport, and AR activity) to MJO phases 6–7 in (top) OND and (bottom) JFM. Results are derived from the 5–9-day lagged average after the active MJO day at day 0. The green triangle indicates the center longitude of enhanced MJO convection at day 0. The jet (mean 250-hPa zonal wind as the basic state) interval is 20 m s−1 starting at 30 m s−1. The relative magnitude of each component is indicated by the difference in the thickness and length, and the variables shown may not be entirely precise in location and pattern.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
The schematic diagram of extratropical response (precipitation extremes, Z500a, anomalous moisture transport, and AR activity) to MJO phases 6–7 in (top) OND and (bottom) JFM. Results are derived from the 5–9-day lagged average after the active MJO day at day 0. The green triangle indicates the center longitude of enhanced MJO convection at day 0. The jet (mean 250-hPa zonal wind as the basic state) interval is 20 m s−1 starting at 30 m s−1. The relative magnitude of each component is indicated by the difference in the thickness and length, and the variables shown may not be entirely precise in location and pattern.
Citation: Journal of Hydrometeorology 24, 1; 10.1175/JHM-D-22-0089.1
This study examined the MJO impacts on precipitation extremes over the western United States and their possible changes between seasons and the underlying mechanisms. In addition to the seasonality, the MJO impacts may also be modulated by different atmospheric background states. For example, studies have shown a strong QBO modulation of MJO amplitude and propagation (e.g., Yoo and Son 2016; Toms et al. 2020) and the associated MJO teleconnections (e.g., Wang et al. 2018). ENSO is also shown to be influential to the propagation speed of the MJO, and therefore ARs (Collow et al. 2020). How the MJO impacts are sensitive to different background states merits investigation in future studies. The MJO impacts examined in this study are conditioned upon the individual MJO phases (i.e., no propagation characteristics are considered). More analysis needs to be done in future studies to further analyze how precipitation extremes change in response to different MJO events with different amplitude and propagation characteristics, e.g., strong MJO amplitude events that propagate from the Indian Ocean to the western Pacific versus weak MJO amplitude events that decay over the Maritime Continent. The sensitivity of the MJO–AR relationship found in this study warrants further analysis in future studies, such as how the relationship changes with different reanalysis datasets and different AR detection algorithms, which have been found to potentially influence the detection results of AR objects (Collow et al. 2022), and how the relationship changes with different AR scales (Ralph et al. 2019). To that end, it is worth noting that the AR detection algorithm used in the current study is among the ones that show minimal sensitivity to the input reanalysis dataset (Collow et al. 2022).
Acknowledgments.
Wang, DeFlorio, and Castellano were supported by the California Department of Water Resources AR Program (Grant 4600013361). Guan was supported in part by the California Department of Water Resources AR Program via University of California, San Diego, and by NASA Grant 80NSSC22K0926.
Data availability statement.
The ERA5 data were obtained from the Copernicus Climate Change Service Climate Data Store (CDS) via https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-single-levels?tab=form (Hersbach et al. 2020). The CPC data were downloaded from the NOAA/Physical Sciences Laboratory https://psl.noaa.gov/data/gridded/data.cpc.globalprecip.html (Chen et al. 2008). The MERRA2 data were downloaded from NASA Earthdata https://disc.gsfc.nasa.gov/datasets?page=1&keywords=MERRA-2. The MERRA-2 AR dataset was obtained at https://ucla.box.com/ARcatalog.
REFERENCES
Adames, Á. F., J. M. Wallace, and J. M. Monteiro, 2016: Seasonality of the structure and propagation characteristics of the MJO. J. Atmos. Sci., 73, 3511–3526, https://doi.org/10.1175/JAS-D-15-0232.1.
Alaka, G. J., Jr., and E. D. Maloney, 2012: The influence of the MJO on upstream precursors to African easterly waves. J. Climate, 25, 3219–3236, https://doi.org/10.1175/JCLI-D-11-00232.1.
Baggett, C. F., K. M. Nardi, S. J. Childs, S. N. Zito, E. A. Barnes, and E. D. Maloney, 2018: Skillful subseasonal forecasts of weekly tornado and hail activity using the Madden–Julian Oscillation. J. Geophys. Res. Atmos., 123, 12 661–12 675, https://doi.org/10.1029/2018JD029059.
Bomers, A., R. M. Schielen, and S. J. Hulscher, 2019: Consequences of dike breaches and dike overflow in a bifurcating river system. Nat. Hazards, 97, 309–334, https://doi.org/10.1007/s11069-019-03643-y.
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.
Chen, M., W. Shi, P. Xie, V. B. Silva, V. E. Kousky, R. Wayne Higgins, and J. E. Janowiak, 2008: Assessing objective techniques for gauge‐based analyses of global daily precipitation. J. Geophys. Res., 113, D04110, https://doi.org/10.1029/2007JD009132.
Collow, A. B. M., H. Mersiovsky, and M. G. Bosilovich, 2020: Large-scale influences on atmospheric river–induced extreme precipitation events along the coast of Washington State. J. Hydrometeor., 21, 2139–2156, https://doi.org/10.1175/JHM-D-19-0272.1.
Collow, A. B. M., and Coauthors, 2022: An overview of ARTMIP’s Tier 2 Reanalysis Intercomparison: Uncertainty in the detection of atmospheric rivers and their associated precipitation. J. Geophys. Res. Atmos., 127, e2021JD036155, https://doi.org/10.1029/2021JD036155.
Corringham, T. W., F. M. Ralph, A. Gershunov, D. R. Cayan, and C. A. Talbot, 2019: Atmospheric Rivers drive flood damages in the western United States. Sci. Adv., 5, eaax4631, https://doi.org/10.1126/sciadv.aax4631.
Curtis, S., A. Salahuddin, R. F. Adler, G. J. Huffman, G. Gu, and Y. Hong, 2007: Precipitation extremes estimated by GPCP and TRMM: ENSO relationships. J. Hydrometeor., 8, 678–689, https://doi.org/10.1175/JHM601.1.
Deng, Y., and T. Jiang, 2011: Intraseasonal modulation of the North Pacific storm track by tropical convection in boreal winter. J. Climate, 24, 1122–1137, https://doi.org/10.1175/2010JCLI3676.1.
Emori, S., and S. Brown, 2005: Dynamic and thermodynamic changes in mean and extreme precipitation under changed climate. Geophys. Res. Lett., 32, L17706, https://doi.org/10.1029/2005GL023272.
Fang, X., and J. W. Pomeroy, 2016: Impact of antecedent conditions on simulations of a flood in a mountain headwater basin. Hydrol. Processes, 30, 2754–2772, https://doi.org/10.1002/hyp.10910.
Gelaro, R., and Coauthors, 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.
Gill, A. E., 1980: Some simple solutions for heat‐induced tropical circulation. Quart. J. Roy. Meteor. Soc., 106, 447–462, https://doi.org/10.1002/qj.49710644905.
Gore, M., B. J. Abiodun, and F. Kucharski, 2020: Understanding the influence of ENSO patterns on drought over southern Africa using SPEEDY. Climate Dyn., 54, 307–327, https://doi.org/10.1007/s00382-019-05002-w.
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., 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.
Hamill, T. M., G. T. Bates, J. S. Whitaker, D. R. Murray, M. Fiorino, T. J. Galarneau, Y. Zhu, and W. Lapenta, 2013: NOAA’s second-generation global medium-range ensemble reforecast dataset. Bull. Amer. Meteor. Soc., 94, 1553–1565, https://doi.org/10.1175/BAMS-D-12-00014.1.
Henderson, S. A., E. D. Maloney, and E. A. Barnes, 2016: The influence of the Madden–Julian oscillation on Northern Hemisphere winter blocking. J. Climate, 29, 4597–4616, https://doi.org/10.1175/JCLI-D-15-0502.1.
Henderson, S. A., E. D. Maloney, and S.-W. Son, 2017: Madden–Julian oscillation Pacific teleconnections: The impact of the basic state and MJO representation in general circulation models. J. Climate, 30, 4567–4587, https://doi.org/10.1175/JCLI-D-16-0789.1.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hoell, A., M. Hoerling, J. Eischeid, K. Wolter, R. Dole, J. Perlwitz, T. Xu, and L. Cheng, 2016: Does El Niño intensity matter for California precipitation? Geophys. Res. Lett., 43, 819–825, https://doi.org/10.1002/2015GL067102.
Hoskins, B. J., and D. J. Karoly, 1981: The steady linear response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38, 1179–1196, https://doi.org/10.1175/1520-0469(1981)038<1179:TSLROA>2.0.CO;2.
Jiang, X., and Coauthors, 2020: Fifty years of research on the Madden–Julian oscillation: Recent progress, challenges, and perspectives. J. Geophys. Res. Atmos., 125, e2019JD030911, https://doi.org/10.1029/2019JD030911.
Jin, D., D. E. Waliser, C. Jones, and R. Murtugudde, 2013: Modulation of tropical ocean surface chlorophyll by the Madden–Julian oscillation. Climate Dyn., 40, 39–58, https://doi.org/10.1007/s00382-012-1321-4.
Jones, C., 2000: Occurrence of extreme precipitation events in California and relationships with the Madden–Julian oscillation. J. Climate, 13, 3576–3587, https://doi.org/10.1175/1520-0442(2000)013<3576:OOEPEI>2.0.CO;2.
Jones, C., and L. M. Carvalho, 2012: Spatial–intensity variations in extreme precipitation in the contiguous United States and the Madden–Julian oscillation. J. Climate, 25, 4898–4913, https://doi.org/10.1175/JCLI-D-11-00278.1.
Jones, C., D. E. Waliser, K. Lau, and W. Stern, 2004: Global occurrences of extreme precipitation and the Madden–Julian oscillation: Observations and predictability. J. Climate, 17, 4575–4589, https://doi.org/10.1175/3238.1.
Joseph, S., A. Sahai, and B. N. Goswami, 2008: Eastward propagating MJO during boreal summer and Indian monsoon droughts. Climate Dyn., 32, 1139–1153, https://doi.org/10.1007/s00382-008-0412-8.
Lau, W. K.-M., and D. E. Waliser, 2011: Intraseasonal Variability in the Atmosphere-Ocean Climate System. 2nd ed. Springer, 613 pp.
Lavers, D. A., F. M. Ralph, D. S. Richardson, and F. Pappenberger, 2020: Improved forecasts of atmospheric rivers through systematic reconnaissance, better modelling, and insights on conversion of rain to flooding. Commun. Earth Environ., 1, 39, https://doi.org/10.1038/s43247-020-00042-1.
Madden, R. A., and P. Julian, 1971: Detection of a 40–50 day oscillation in the zonal wind in the tropical Pacific. J. Atmos. Sci., 28, 702–708, https://doi.org/10.1175/1520-0469(1971)028<0702:DOADOI>2.0.CO;2.
Madden, R. A., and P. Julian, 1972: Description of a global-scale circulation cells in the tropics with a 40–50 day period. J. Atmos. Sci., 29, 1109–1123, https://doi.org/10.1175/1520-0469(1972)029<1109:DOGSCC>2.0.CO;2.
Marchand, R., N. Beagley, S. E. Thompson, T. P. Ackerman, and D. M. Schultz, 2006: A bootstrap technique for testing the relationship between local-scale radar observations of cloud occurrence and large-scale atmospheric fields. J. Atmos. Sci., 63, 2813–2830, https://doi.org/10.1175/JAS3772.1.
Martin, Z., C. Orbe, S. Wang, and A. Sobel, 2021: The MJO–QBO relationship in a GCM with stratospheric nudging. J. Climate, 34, 4603–4624, https://doi.org/10.1175/JCLI-D-20-0636.1.
Moore, B. J., A. B. White, D. J. Gottas, and P. J. Neiman, 2020: Extreme precipitation events in Northern California during winter 2016–17: Multiscale analysis and climatological perspective. Mon. Wea. Rev., 148, 1049–1074, https://doi.org/10.1175/MWR-D-19-0242.1.
Muhammad, F. R., S. W. Lubis, and S. Setiawan, 2021: Impacts of the Madden–Julian oscillation on precipitation extremes in Indonesia. Int. J. Climatol., 41, 1970–1984, https://doi.org/10.1002/joc.6941.
Mundhenk, B. D., E. A. Barnes, and E. D. 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.
Mundhenk, B. D., E. A. Barnes, E. D. Maloney, and C. F. Baggett, 2018: Skillful empirical subseasonal prediction of landfalling atmospheric river activity using the Madden–Julian oscillation and quasi-biennial oscillation. npj Climate Atmos., 1, 20117, https://doi.org/10.1038/s41612-017-0008-2.
Neale, R. B., J. H. Richter, and M. Jochum, 2008: The impact of convection on ENSO: From a delayed oscillator to a series of events. J. Climate, 21, 5904–5924, https://doi.org/10.1175/2008JCLI2244.1.
Neena, J., E. Suhas, and B. Goswami, 2011: Leading role of internal dynamics in the 2009 Indian summer monsoon drought. J. Geophys. Res., 116, D13103, https://doi.org/10.1029/2010JD015328.
Paik, S., S. K. Min, X. Zhang, M. G. Donat, A. D. King, and Q. Sun, 2020: Determining the anthropogenic greenhouse gas contribution to the observed intensification of extreme precipitation. Geophys. Res. Lett., 47, e2019GL086875, https://doi.org/10.1029/2019GL086875.
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.
Prince, H. D., P. B. Gibson, M. J. DeFlorio, T. W. Corringham, A. Cobb, B. Guan, F. M. Ralph, and D. E. Waliser, 2021: Genesis locations of the costliest atmospheric rivers impacting the Western United States. Geophys. Res. Lett., 48, e2021GL093947, https://doi.org/10.1029/2021GL093947.
Rahmstorf, S., 2017: Rising hazard of storm-surge flooding. Proc. Natl. Acad. Sci. USA, 114, 11 806–11 808, https://doi.org/10.1073/pnas.1715895114.
Ralph, F. M., P. J. Neiman, G. N. Kiladis, K. Weickmann, and D. W. Reynolds, 2011: A Multiscale observational case study of a Pacific atmospheric river exhibiting tropical–extratropical connections and a mesoscale frontal wave. Mon. Wea. Rev., 139, 1169–1189, https://doi.org/10.1175/2010MWR3596.1.
Ralph, F. M., J. J. Rutz, J. M. Cordeira, M. Dettinger, M. Anderson, D. Reynolds, L. J. Schick, and C. Smallcomb, 2019: A scale to characterize the strength and impacts of atmospheric rivers. Bull. Amer. Meteor. Soc., 100, 269–289, https://doi.org/10.1175/BAMS-D-18-0023.1.
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.
Sardeshmukh, P. D., and B. J. Hoskins, 1988: The generation of global rotational flow by steady idealized tropical divergence. J. Atmos. Sci., 45, 1228–1251, https://doi.org/10.1175/1520-0469(1988)045<1228:TGOGRF>2.0.CO;2.
Shields, C. A., and Coauthors, 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.
Shimizu, M. H., T. Ambrizzi, and B. Liebmann, 2017: Extreme precipitation events and their relationship with ENSO and MJO phases over northern South America. Int. J. Climatol., 37, 2977–2989, https://doi.org/10.1002/joc.4893.
Sillmann, J., and Coauthors, 2017: Understanding, modeling and predicting weather and climate extremes: Challenges and opportunities. Wea. Climate Extremes, 18, 65–74, https://doi.org/10.1016/j.wace.2017.10.003.
Singh, P., V. Vasudevan, J. Chowdary, and C. Gnanaseelan, 2015: Subseasonal variations of Indian summer monsoon with special emphasis on drought and excess rainfall years. Int. J. Climatol., 35, 570–582, https://doi.org/10.1002/joc.4004.
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.
Sun, X., B. Renard, M. Thyer, S. Westra, and M. Lang, 2015: A global analysis of the asymmetric effect of ENSO on extreme precipitation. J. Hydrol., 530, 51–65, https://doi.org/10.1016/j.jhydrol.2015.09.016.
Takahashi, C., Y. Imada, and M. Watanabe, 2022: Influence of the Madden–Julian oscillation on wintertime extreme snowfall and precipitation in Japan. J. Meteor. Soc. Japan, 100, 257–283, https://doi.org/10.2151/jmsj.2022-014.
Toms, B. A., E. A. Barnes, E. D. Maloney, and S. C. van den Heever, 2020: The global teleconnection signature of the Madden–Julian oscillation and its modulation by the quasi‐biennial oscillation. J. Geophys. Res. Atmos., 125, e2020JD032653, https://doi.org/10.1029/2020JD032653.
Toride, K., and G. J. Hakim, 2022: What distinguishes MJO events associated with atmospheric rivers? J. Climate, 35, 6135–6149, https://doi.org/10.1175/JCLI-D-21-0493.1.
Tseng, K.-C., E. A. Barnes, and E. D. Maloney, 2018: Prediction of the midlatitude response to strong Madden–Julian oscillation events on S2S time scales. Geophys. Res. Lett., 45, 463–470, https://doi.org/10.1002/2017GL075734.
Vasconcelos, F. C., Jr., C. Jones, and A. W. Gandu, 2018: Interannual and intraseasonal variations of the onset and demise of the pre-wet season and the wet season in the northern Northeast Brazil. Rev. Bras. Meteor., 33, 472–484, https://doi.org/10.1590/0102-7786333007.
Wang, J., H. M. Kim, E. K. Chang, and S. W. Son, 2018: Modulation of the MJO and North Pacific storm track relationship by the QBO. J. Geophys. Res. Atmos., 123, 3976–3992, https://doi.org/10.1029/2017JD027977.
Wang, J., H. M. Kim, D. Kim, S. A. Henderson, C. Stan, and E. D. Maloney, 2020: MJO teleconnections over the PNA region in climate models. Part II: Impacts of the MJO and basic state. J. Climate, 33, 5081–5101, https://doi.org/10.1175/JCLI-D-19-0865.1.
Wang, J., H. M. Kim, and M. J. DeFlorio, 2022: Future changes of PNA-like MJO teleconnections in CMIP6 models: Underlying mechanisms and uncertainty. J. Climate, 35, 3459–3478, https://doi.org/10.1175/JCLI-D-21-0445.1.
Wang, S.-Y. S., J.-H. Yoon, E. Becker, and R. Gillies, 2017: California from drought to deluge. Nat. Climate Change, 7, 465–468, https://doi.org/10.1038/nclimate3330.
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.
Wolding, B. O., E. D. Maloney, S. Henderson, and M. Branson, 2017: Climate change and the Madden–Julian oscillation: A vertically resolved weak temperature gradient analysis. J. Adv. Model. Earth Syst., 9, 307–331, https://doi.org/10.1002/2016MS000843.
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., 2013: Madden–Julian oscillation: Bridging weather and climate. Bull. Amer. Meteor. Soc., 94, 1849–1870, https://doi.org/10.1175/BAMS-D-12-00026.1.
Zhang, C., and M. Dong, 2004: Seasonality in the Madden–Julian oscillation. J. Climate, 17, 3169–3180, https://doi.org/10.1175/1520-0442(2004)017<3169:SITMO>2.0.CO;2.
Zhou, S., M. L’Heureux, S. Weaver, and A. Kumar, 2012: A composite study of the MJO influence on the surface air temperature and precipitation over the continental United States. Climate Dyn., 38, 1459–1471, https://doi.org/10.1007/s00382-011-1001-9.
Zhou, Y., H. Kim, and D. E. Waliser, 2021: Atmospheric river lifecycle responses to the Madden–Julian oscillation. Geophys. Res. Lett., 48, e2020GL090983, https://doi.org/10.1029/2020GL090983.