182
MONTHLY WEATHER REVIEW
VOLUME 141
Regional Surges of Monsoonal Moisture into the Southwestern United States
JAMES E. FAVORS
Department of Meteorology, San Jose´ State University, San Jose´, California
JOHN T. ABATZOGLOU
Department of Geography, University of Idaho, Moscow, Idaho
(Manuscript received 28 January 2012, in final form 2 July 2012)
ABSTRACT
Episodic surges of moisture into the southwestern United States are an important attribute of the North
American monsoon. Building upon prior studies that identified mesoscale gulf surges using station-based
diagnostics, regional surges in monsoonal moisture are identified using precipitable water and integrated
water vapor flux from the North American Regional Reanalysis. These regional surge diagnostics exhibit
increased skill over gulf surge diagnostics in capturing widespread significant multiday precipitation over the
state of Arizona and are associated with the northward intrusion of moisture and precipitation into the
southwestern United States. Both tropical and midlatitude circulation patterns are associated with identified
regional surge events. In the tropics, the passage of a tropical easterly wave across the Sierra Madre and
through the Gulf of California facilitates a northeastward flux of moisture toward the southwestern United
States. In midlatitudes, the breakdown and eastward shift of an upper-level ridge over the western United
States ahead of an eastward-propagating trough off the Pacific Northwest coast helps destabilize the middle
troposphere ahead of the easterly wave and provides a conduit for subtropical moisture advection into the
interior western United States.
1. Introduction
The interaction of the general circulation and terrain
of western Mexico and the interior western United
States produces diverse spatial and seasonal characteristics of the North American monsoon (Adams and
Comrie 1997; Barlow et al. 1998; Douglas et al. 1993).
The North American monsoon (NAM) exhibits variability on interannual (e.g., Gutzler 2004), intraseasonal
(e.g., Higgins and Shi 2001), and synoptic (e.g., Carleton
1986; Bordoni and Stevens 2006) time scales. Though
the focal point of the NAM is primarily centered over
western Mexico, monsoonal activity periodically infiltrates northward into the intermountain western
United States. The collective influence of this variability
across the northern fringe of the NAM results in highly
variable warm season precipitation across the southwestern United States (e.g., Higgins et al. 2004).
Corresponding author address: Dr. John T. Abatzoglou, Department of Geography, University of Idaho, P.O. Box 443021,
Moscow, ID 83844-3021.
E-mail: [email protected]
DOI: 10.1175/MWR-D-12-00037.1
Ó 2013 American Meteorological Society
A large portion of warm season precipitation across
the southwestern United States occurs with episodic
northward-propagating moisture intrusions, hereafter
referred to as monsoon surges, that foster convective
outbreaks across the southwestern United States on
a variety of spatial scales including both mesoscale, or
gulf surges (e.g., Fuller and Stensrud 2000, hereafter
FS00), and regional-scale moisture surges (e.g., Higgins
et al. 2004). Despite the influence of monsoon surges on
localized to regional-scale weather-related hazards in
the southwestern United States (e.g., McCollum et al.
1995; Ray et al. 2007; Abatzoglou and Brown 2009),
predicting the onset and evolution of monsoon surges
remains a forecasting challenge (e.g., Maddox et al.
1995). Previous studies have focused on the mesoscale
gulf surges of the NAM, characterized by the northward
flux of moist, low-level air from the Gulf of Mexico
and Gulf of California into the northern Sonoran Desert
(e.g., Adams and Comrie 1997; Bordoni and Stevens
2006). These gulf surges have been attributed to tropical
processes including the passage of a tropical easterly wave
over western Mexico (e.g., Stensrud et al. 1995, 1997;
FS00) and remnants of eastern Pacific tropical cyclones
JANUARY 2013
FAVORS AND ABATZOGLOU
(e.g., Corbosiero et al. 2009). Other studies have
stressed the importance of the midlatitude circulation in
facilitating a northward flux of moisture into the
southwestern United States and organizing widespread
vertical motion (e.g., Higgins et al. 2004; Anderson and
Roads 2002). Collectively, observational studies demonstrate that low-latitude disturbances aid moisture
advection into the southwestern United States; however, they also find that tropical disturbances are not
a sufficient condition for monsoon surges and highlight
the need to address additional hypotheses regarding the
roles played by tropical and midlatitude circulation
patterns on transient behavior of the NAM.
Whereas much prior research has focused on gulf
surges using observations from Yuma, Arizona, and
individual radiosonde sites (e.g., Stensrud et al. 1995,
1997; FS00; Dixon 2005) a conceptual and robust means
of diagnosing regional surges of the NAM that have
widespread effects across the state of Arizona has not
been fully explored. Station-based means to identify gulf
surge events were neither designed or intended to capture the regional-scale moisture surges into the southwestern United States (e.g., FS00), yet have been
examined in such context (e.g., Higgins et al. 2004).
Building upon aforementioned diagnostics to identify
gulf surge events and conceptual framework of monsoonal moisture surges by Hales (1972) and Brenner
(1974), we suggest a method for diagnosing regional
surges of the NAM into the southwestern United States
using precipitable water (PW) and integrated water
vapor flux (IWVF) from the North American Regional
Reanalysis (NARR).
The motivation of this paper is twofold. First, we
evaluate the ability of the proposed regional surge
method to capture multiday periods of intense and
widespread precipitation across the state of Arizona.
We hypothesize that a threshold-based diagnostic that
includes PW and IWVF from NARR will be better
suited to characterize features associated with regional
surges than diagnostics specifically designed for gulf
surge events. Second, we examine the evolution of both
the tropical and extratropical circulation relative to the
identified regional surges to further evaluate the roles of
tropical and midlatitude circulation in initiating and
enabling such events.
2. Data and methods
Although the seasonal window of the NAM in the
southwestern United States has been referred to a variety of ways in the literature, we constrain our focus to the
core monsoonal season defined as 1 July to 15 September
(e.g., Adams and Comrie 1997) for the 28-yr period
183
(1980–2007). We employ three classes of data in our
analysis: (i) hourly surface observations for Yuma from
the Western Regional Climate Center (http://www.wrcc.
dri.edu), (ii) synoptic and mesoscale reanalysis from the
National Centers for Environmental Prediction–National
Center for Atmospheric Research (NCEP–NCAR, 4
times daily) reanalysis and the NARR (8 times daily;
Mesinger et al. 2006), respectively, and (iii) gridded
precipitation from the National Aeronautics and Space
Administration (NASA) National Land Data Assimilation System (NLDAS-2, Mitchell et al. 2004; http://ldas.
gsfc.nasa.gov/nldas/NLDAS2forcing.php, last accessed
15 June 2011). For (ii) and (iii) above, we calculated daily
means and totals ending at 0000 UTC from subdaily
and hourly data, respectively, and hereafter use only
daily time scales in our analysis.
Prior methods to diagnose moisture surges of the
NAM into the southwestern United States have utilized
both surface (e.g., Stensrud et al. 1995) and radiosonde
observations (Dixon 2005). Building off seminal work
on moisture surges by Hales (1972) and Brenner (1974),
FS00 developed an objective methodology for identifying gulf surges using surface observations from Yuma.
The criteria FS00 used to diagnose an event required
sustained maximum daily 2-m dewpoint temperature
exceeding 15.68C (608F) for several days and at least one
reported 10-m wind velocity from the south exceeding
4 m s21 on the initial day of the event. While the procedure to identify gulf surges additionally classified
events as either ‘‘strong’’ (i.e., maximum dewpoint
temperature increases during 3-day period after surge
onset) or ‘‘weak’’ (i.e., maximum dewpoint temperature
decreases during 3-day period after surge onset), we
primarily focus on all gulf surge events as identified from
FS00. Subsequent studies have argued that the use of
a single surface observation from Yuma may not be
ideal for identifying regional monsoon surges because of
diurnal and geographic biases, lack of upper-air observations, and potential false positives associated with
outflow from localized thunderstorms (e.g., Dixon
2005). Dixon (2005) provided evidence that dewpoint
temperature on mandatory pressure levels from radiosonde observations could be used to overcome some of
those biases and more accurately diagnose widespread
surges of moisture and precipitation into the southwestern United States.
Building off the work of FS00 and Dixon (2005), we
posit that the regional reanalysis provides an additional
means of diagnosing regional surges of the NAM to
overcome geographic limitations of station-based approaches. For the purpose of this study, we constrain our
focus to regional surges for Arizona and western New
Mexico, herein defined as covering 318–368N, 1088–1148W
184
MONTHLY WEATHER REVIEW
(approximated by the North American Monsoon Experiment subregion 2). Following the conceptual models
of Hales (1972) and Brenner (1974) we suggest that two
quantities efficiently encapsulate the regional northward
surge of moisture: precipitable water, or total column
integrated water vapor, and the meridional component of
the IWVF.
Precipitable water is regarded as a key factor in
summertime precipitation events across the Southwest
(Reitan 1957; Maddox et al. 1995; Jiang and Lau 2008;
Lu et al. 2009) and is often used operationally (staff at
NWS Tucson 2007, personal communications). Complimentary to PW, IWVF has been used to examine
atmospheric moisture pathways for the NAM (e.g.,
Schmitz and Mullen 1996; Castro et al. 2001; Jiang and
Lau 2008) and has generated recent interest as a forecasting tool for extreme precipitation events (e.g.,
Neiman et al. 2009). We hypothesize that IWVF is well
suited to capture the moist, southerly flow characteristic
of monsoon surge events on all scales. Given that the
bulk of water vapor in the atmosphere is below 500 hPa,
IWVF was calculated from pressure level NARR data
by vertically integrating specific humidity and wind velocity from the surface to 500 hPa (Schmitz and Mullen
1996), whereas total atmospheric column PW was acquired as direct output from NARR.
To establish criteria for diagnosing regional surges
across the study area, we examined relationships between the primary predictor variables, PW and IWVF,
and a target variable, precipitation, by taking areal averages of all variables within the study area. Figure 1
shows a scatterplot of PW and precipitation spatially
averaged over the study area with the vector showing the
IWVF. For clarity, 3-day averages (totals for precipitation) are shown every third day over the 28-yr
period. The strong relationship between PW and accumulated precipitation (daily correlation of r 5 0.62, p ,
1 3 1026) reinforces the utility of PW as a predictor for
precipitation. Figure 1 also shows that dry periods (daily
regional precipitation less than 0.5 mm) generally occur
with low PW (,20 mm) and weak eastward to northeastward moisture flux. By contrast, a strong northward
IWVF is apparent along with high PW (.32 mm) during
wet periods.
We used a threshold-based criteria for identifying regional surges based on geographically defined percentiles (defined seasonally over the 1 July to 15 September
period), rather than absolute values, as a means of
overcoming documented biases of NARR (Mo et al.
2005) and allowing for transferability to other subregions of the NAM and other data sources. We diagnose regional surges using a simple threshold-based
approach defined as when daily averaged regional PW
VOLUME 141
FIG. 1. Scatterplot of precipitable water and precipitation accumulation spatially averaged over Arizona (318–368N, 1088–1148W,
boundary denoted by area shaded in inset map) for 1 Jul–15 Sep
1980–2007. The vector component associated with each point
corresponds to the observed integrated water vapor flux. For
clarity, observations are only plotted every third day and represent
the precipitable water and integrative water vapor flux averaged
over the previous 72 h and the 72-h precipitation amount.
and the northward IWVF both exceed their 80th percentile (30 mm and 115 kg s21 m21, respectively) for at
least two consecutive days. Sequences of days meeting
the established criteria often occurred over the course of
several days, occasionally interrupted by a day or two
not meeting criteria. To reduce redundancy, regional
surges are defined as the initial date for multiday periods
meeting the aforementioned criteria and no regional
surge event is considered within four days following the
previous detection. It is important to note that such
objective methods (i.e., using the 80th percentile) are
inherently based off subjective decisions; however, the
relationships found in Fig. 1 support the use of the criteria used here.
A universally accepted means to qualify or validate
a regional monsoon surge is lacking. In lieu of such
a means, we consider significant multiday widespread
precipitation event (SMWPE) defined by daily areaaveraged precipitation exceeding the 90th percentile for
the 1 July–15 September period (4.5 mm day21) on two
or more consecutive days. The 90th percentile threshold
was used to capture only the most extreme types of
precipitation events that occur infrequently but still
have impact. The SMWPE date is qualified as the first
of the consecutive days exceeding this threshold. Similar
to our qualification of regional surges, we discretized
SMWPE by excluding events that occur less than four
days from the previous event. A contingency matrix
is used to assess the skill of both the monsoon surge
JANUARY 2013
FAVORS AND ABATZOGLOU
185
FIG. 2. (a)–(c) Precipitable water (shaded, bold black line denoting 33-mm isoline) and integrated water vapor flux (vectors, largest
vector shown is 300 kg s21 m21). (d)–(f) Precipitation accumulation (shaded), 500-hPa geopotential height (contours with interval of 20 m,
bold line every 60 m), and 700–500-hPa averaged velocity (vectors, largest vector shown is 25 m s21) for (a),(d) 12–13 Aug 1983; (b),(e)
14–15 Aug 1983; and (c),(f) 16–17 Aug 1983. Areas with less than 1 mm of precipitation are omitted in (d)–(f). The regional surge was
diagnosed on 16 Aug 1983.
diagnostic described herein and that of FS00 to detect
SMWPE the day of and up to four days following surge
diagnosis. A contingency matrix provides a means for
quantifying forecast skill when evaluating dichotomous
events (e.g., regional surge with associated regional
precipitation). The skill of both methods is measured in
success ratio (i.e., identified surge event that did have an
accompanying SMWPE within four days divided by the
total number of surges) and false alarm ratio (i.e., identified
surge event that did not have an accompanying SMWPE
within four days divided by the total number of surges).
Finally, we use composite analysis to elucidate synoptic patterns associated with regional surge events. The
evolution of synoptic fields are examined up to six days
prior to and six days following diagnosed regional surge
dates for PW and IWVF using NARR, midtropospheric
variables 500-hPa geopotential height, and 700–500-hPa
pressure-weighted wind vectors using NCEP–NCAR reanalysis and precipitation using NLDAS-2. We also examine composite anomalies to better identify perturbations
relative to the background time-averaged state, defined
as 1 July–15 September without removal of the seasonal
cycle. The standard t test is used to examine statistical
significance for normally distributed values, whereas the
nonparametric Wilcoxon–Mann–Whitney rank-sum test
is used for precipitation. Hereafter we refer only to results significant at the 95th percentile.
3. Results
A total of 49 regional surges of the NAM were identified over the 28-yr period (1980–2007). An example of
a regional surge event detected on 16 August 1983 is
shown in Fig. 2. Large-scale ridging at 500 hPa across
much of the United States was observed in the days prior
to the event with widespread precipitation confined to
northern Mexico and the remnants of Hurricane Ismael
off the coast of Baja California (Figs. 2a,d). In subsequent days, the ridge breaks down and progresses
eastward as a weak trough approaches the west coast of
186
MONTHLY WEATHER REVIEW
California and Oregon eventually becoming cut off from
the jet stream and forming a closed low at 500 hPa, several hundred kilometers off the central California coast
(Figs. 2b,e). The configuration of Rossby wave patterns
with a closed low upstream of Arizona and eastward
displacement of the subtropical ridge allows for the advection of a prolonged fetch of moisture emanating from
the Gulf of California and the subtropical eastern Pacific
with the remnants of Ismael into the southwestern United
States (Fig. 2b). The strong northward moisture flux into
the southwestern United States and elevated PW coincide
with the regional surge diagnostics and produce widespread precipitation across the Sonoran, Mojave, and
Great Basin deserts (Fig. 2c,f).
By contrast, approximately 150 gulf surges were
identified using the methods of FS00 during the period
of 1980–2007, 14% of which were observed within five
days following regional surges as diagnosed using the
methods of the current study. To evaluate the characteristics associated with the regional surges of the present study, and contextualize this approach relative to the
gulf surge identification methods of FS00, we examined
a lead–lag composite of precipitation, PW, and meridional IWVF averaged over the study area. A strong increase in PW, northward IWVF, and precipitation were
observed for regional surges diagnosed in this study,
whereas modest increases in PW and precipitation were
observed for gulf surges diagnosed using FS00 (Fig. 3). A
subsequent analysis of only strong gulf surges, per the
taxonomy of FS00, found that strong gulf surges resulted
in a more pronounced increase in PW and precipitation
than weak gulf surges; however, the regional increase in
PW and precipitation fell far below levels observed for
regional surges. This difference is likely tied to the discrepancy in scale between the two surge diagnostics and
phenomena of interest. Similar to the results of Higgins
et al. (2004), observed precipitation for the FS00 events
tended to peak 2–3 days following the initial detection
date, whereas the regional schema of the present work
showed regional precipitation synchronized with the
detection date. This difference in timing of precipitation
occurrence is likely due to FS00 identifying the leading
edge of the surge moving out of the Gulf of California as
detected in Yuma, whereas the current study relies on
diagnostics geographically averaged over the region
that are strongly coupled to concurrent regional precipitation (i.e., Fig. 1). We note that some of the observed
differences could also be due to the number of monsoon
surges diagnosed in the present study being approximately one-third of that diagnosed using FS00 and likely
a function of scale difference between the surge features
identified (regional vs local) and the stricter thresholds of
the diagnostics employed in the current work.
VOLUME 141
FIG. 3. Composite of daily (a) precipitation, (b) precipitable
water, and (c) southerly IWVF averaged over the study area (318–
368N, 1088–1148W) relative to regional surge dates diagnosed by
this study (solid) and gulf surges following the method of FS00
(dashed).
An additional comparison between diagnostic methods
was conducted to examine how well methods were a
proxy for the 28 SMWPE found during the 28-yr period.
More than half of these events (15) occurred on the day
of or within two days following regional surges whereas
10 were found to occur within four days following gulf
surges. Additionally, regional surges were significantly
more accurate predictors of SMWPE as seen through
the contingency matrix statistics. During the 1980–2007
study period gulf surges had a higher false-alarm ratio at
approximately 93% compared to the regional method
that resulted in a false alarm for 69% of the identified
regional surges. The regional surge diagnostic also had a
higher success ratio of 31% versus the 6.7% success ratio
of the gulf surge diagnostic. Collectively, these results
suggest that our extension of diagnosing gulf surges
by FS00 to regional surges provided added skill in detecting high-impact rainfall events on the regional scale.
JANUARY 2013
FAVORS AND ABATZOGLOU
187
FIG. 4. Composite plots for the 49 regional surge events of (a)–(c) precipitable water (shaded) and IWVF (vectors, largest vector shown
is 160 kg s21 m21), and (d)–(f) precipitation accumulation (shaded), 500-hPa geopotential height (contours), and 700–500-hPa averaged
wind velocity (vectors, largest vector shown is 15 m s21). The composite evolution of these fields is shown (a),(d) 5–6 days prior to; (b),(e)
2–3 days prior to; and (c),(f) the day of and following the date the regional monsoon surge was detected.
Methods to identify gulf surge events were not intended
to capture regional-scale phenomena (e.g., Stensrud
et al. 1997) and key in surges different spatial scales;
however, given the widespread usage of gulf surges to
broader geographic scales, this comparison is warranted.
Lead–lag composite analysis was performed relative
to the 49 regional surge events to better understand associations with tropical and midlatitude circulation
patterns. Five to six days prior to regional surges,
a strong westward flux of moisture is noted across
northern Mexico from the Gulf of Mexico and is corroborated by above-normal precipitation amounts across
Mexico (Figs. 4a,d), while an amplified ridge dominates
much of the interior western United States (Fig. 4d).
Strong radiative heating coincident with the ridge enhances low-level heating and primes the region for instability ahead of the moisture plume. Two to three days
prior to the regional surge there is a strong northward
flux of subtropical moisture from the Sierra Madre
through the Gulf of California and enhanced precipitation across the Sierra Madre Occidental, Baja
California, and the Sonoran deserts of California and
Arizona (Figs. 4b,e). Coincidentally, an eastward shift in
the midlatitude longwave pattern, synchronizes the lowlevel moisture flux (IWVF) and midtropospheric flow
(700–500-hPa steering flow), providing a conduit for
subtropical moisture into the southwestern United
States. Increased PW and strong northward IWVF into
and widespread precipitation across the southwestern
United States are seen concurrent to the detection of the
regional surge (Figs. 4c,f). The breakdown of the midlatitude ridge enhances lower-to-midtropospheric instability and convective potential due to diffluent flow
aloft ahead of the trough and the advection of cooler air
into the midlevels (700–400 hPa) following multiple
days of enhanced surface heating (e.g., Carleton 1986).
Similar analysis for both all and strong gulf surges found
that while such events were associated with a relative
increase in PW and precipitation following such events,
the magnitude of PW and northward flux over the region
was well below that seen concurrent with regional surges
(not shown).
188
MONTHLY WEATHER REVIEW
VOLUME 141
FIG. 5. Composite PW anomaly (shaded), IWVF standardized anomaly (vectors, largest vector shown is two standard deviations) and
500-hPa geopotential height standardized anomaly (contours, starting from 0.2s, contour interval 0.05s with solid red isolines and dashed
blue isolines representing positive and negative anomalies, respectively). The composite evolution of these fields is shown (a) 5–6 days
prior to, (b) 2–3 days prior to, and (c) the day of and following the date the regional monsoon surge was detected. Only statistically
significant values (p , 0.05) are shown.
Composites of PW, IWVF, and 500-hPa geopotential
height anomaly fields were examined to better facilitate the recognition of statistically significant subtropical and midlatitude patterns associated with
regional surges. A well-defined westward-moving subtropical wave propagating over the Sierra Madre and
into the Gulf of California is apparent in the subtropical IWVF fields (Fig. 5a). The strong westward
flux of water vapor results in a significant increase in
PW across much of northern Mexico and southwestern
United States two to three days prior to the surge, while
the enhanced ridge over the western half of the United
States shifts northeastward ahead of below-normal height
fields off the coast of the Pacific Northwest (Fig. 5b). This
same northeastward shift of the midlatitude ridge and
weak trough off the west coast of the United States was
shown to be a common feature in ‘‘wet’’ and strong
subclasses of gulf surges in Higgins et al. (2004) and FS00,
respectively, and a feature observed in intraseasonal to
interannual variability in monsoonal precipitation over
Arizona and New Mexico in Kiladis and Hall-McKim
(2004). The commonality of patterns in midlatitude
Rossby waves across these studies suggests that the
midlatitudes are not a passive contributor to northward
intrusions of moisture along the northern fringe of the
NAM. As the tropical wave migrates west of the Baja
Peninsula, enhanced northward IWVF can be seen from
the tip of the Baja peninsula, through the interior western
United States and extending well into the upper Midwestern states (Fig. 5c). A composite analysis done specifically on the SMWPE exhibits a signature similar to
that for regional surge events (not shown); however,
SMWPE not associated with regional surges failed to
include a coherent propagating tropical easterly wave and
sustained northward moisture transport into the southwestern United States as seen in the regional surge
events.
The characteristics associated with the regional surges
identified in this current study align well with the synoptic signatures of wet gulf surges in Higgins et al.
(2004). Those wet events were marked by sharp increases in precipitation two to four days after surge
detection identified at Yuma, although the patterns
shown here are oriented slightly farther west. Likewise,
composite 500-hPa geopotential height fields concurrent
with regional surges (Fig. 5c) are very similar to those
referred to in wet surges of Higgins et al. (2004). These
circulation patterns are in stark contrast to those associated with ‘‘dry’’ events where the midlatitude ridge
remains fixed over the western United States and
dampen the enhanced midlevel southerly flow over the
northern extent of the Gulf of California.
Hovmöller plots of 500-hPa geopotential height
anomalies latitudinally averaged between 358–47.58N
and subtropical IWVF averaged between 208–258N
help illustrate potential synchronization of eastwardpropagating midlatitude Rossby waves and westwardpropagating tropical easterly waves concurrent with
regional surges (Fig. 6). The configuration of Rossby
wave patterns over the United States with the ridge axis
shifted to the east of the region and trough axis located
west of the region may facilitate the advection of subtropical moisture northward into the southwestern
United States and mirror the conceptual model for gulf
JANUARY 2013
FAVORS AND ABATZOGLOU
189
FIG. 6. Hovmöller plot of (a) standardized anomalies of 500-hPa geopotential height latitudinally averaged between 37.58–47.58N, (b) standardized anomalies of meridional IWVF latitudinally averaged between 208–258N, and
(c) standardized anomalies of PW and IWVF longitudinally averaged between 1058–1158W (unit vector of one
standard anomaly shown for reference) relative to the 49 regional surges. Statistically significant values in (a) and (b)
are denoted by 3 and IWVF west of 1258W are not included in (b). Vectors of IWVF that are not statistically
significant were excluded from (c).
surge onset of FS00 and analysis of Higgins et al. (2004).
A complimentary Hovmöller plot of PW anomalies
and IVWF anomalies averaged along a south–north
transect between 1058–1158W further clarify the signal
of northward-surging moisture into the interior western
United States over a 5–7-day period associated with an
initial anomalous westward flux of moisture that pivots
to a northward flux coincident with the surge date.
Our results suggest that synchronized propagation of
subtropical waves westward and midlatitude waves
eastward catalyze a northward surge of regional-scale
moisture into the southwestern United States, reinforcing
the results of previous studies on local to regional scales
(e.g., FS00; Higgins et al. 2004). While tropical easterly
waves appear prominent in the composite fields, only
about 17% of tropical easterly waves, identified following
the methods of Ladwig and Stensrud (2009), were followed by a regional surge. This is in agreement with prior
studies that show tropical easterly waves as important but
not solely sufficient features in driving all NAM surge
events (e.g., FS00). Additionally, just over half of regional
surges were preceded by a tropical easterly wave within
5 days prior to regional surge identification. Likewise,
we did not observe a significant link between east Pacific
tropical cyclones (from Corbosiero et al. 2009) and regional surges; however, when our analysis is extended
to include the latter half of September and October, landfalling tropical cyclones whose remnants tracked through
the study area exhibited a character akin to regional
surges and associated widespread precipitation (i.e.,
Corbosiero et al. 2009; Ritchie et al. 2011).
Finally, while regional surges diagnosed by this
study were infrequent, daily precipitation totals spanning one day prior, through four days following regional surges, accounted for between 20%–40% of the
1 July–15 September climatological precipitation (1980–
2007) across the interior southwestern United States
(Fig. 7a). In addition, we found a statistically significant
interannual correlation (Spearman’s rank correlation)
between the number of regional surges and the total
1 July–15 September precipitation across much of the
Sonoran, Mojave, and Great Basin deserts (Fig. 7b).
This highlights the potential importance of extreme events
on interannual monsoonal precipitation variability. Either
regional surges are preconditioned by low-frequency
variability through coupled ocean–atmosphere–terrestrial
processes (e.g., antecedent soil moisture, large-scale
modes of climate variability), or regional surges are
independent of low-frequency climate variability and
thus limit our ability to more fully realize potential predictability of the NAM in the southwestern United
States.
190
MONTHLY WEATHER REVIEW
VOLUME 141
FIG. 7. (a) Percentage of 1 Jul–15 Sep that occurs within three days following regional monsoon surge events,
values less than twice climatological normal, or approximately 15%, are omitted. (b) Interannual Spearman’s rank
correlation between the number of regional monsoon surge events and 1 Jul–15 Sep precipitation for 1980–2007.
Only statistically significant values (p , 0.05) are shown.
4. Discussion and conclusions
We demonstrate a regional approach to diagnosing
monsoon surges for the southwestern United States that
proves a more successful predictor of regional precipitation events compared to methods designed to diagnose gulf surges. The methods described herein
on gridded fields allow for transferability to short- to
medium-range operational forecasting and research
activities using reanalyses and regional climate models.
The premise of the methods employed here for Arizona
may also be applicable identifying regional surge events
in other subregions of the NAM. Likewise, further efforts to
understand low-frequency variability of monsoonal precipitation over the southwestern United States may benefit
from the consideration of regional monsoon surges given
their contribution both to climatology and interannual
variability of summer precipitation (i.e., Fig. 6).
The synoptic characteristics associated with regional
surges reiterate previous findings and suggest that the
passage of a tropical easterly wave across Mexico and
into the eastern Pacific provides the subtropical circulation pattern for moisture advection into the southwestern United States (e.g., FS00; Higgins et al. 2004).
Similar to previous work, we find that while the likelihood of a regional surge is enhanced by the passage of
a tropical easterly wave, such waves are not a sufficient
condition for surge occurrence. Additionally, we provide further evidence of the influence of the midlatitude
circulation on transient activity of the NAM (e.g.,
Carleton 1986; FS00; Higgins et al. 2004; Corbosiero
et al. 2009). We suggest that the configuration of the
midlatitude Rossby wave pattern can be important in
facilitating regional surges and widespread precipitation
events across the southwestern United States. Regional
surges preferentially occurred following the amplification
and subsequent breakdown and eastward shift in the
ridge over the western United States. Though previous
research has shown that the placement of a midlatitude
trough upstream adjacent to the NAM core causes
a shutdown of surges (e.g., Stensrud et al. 1997; FS00), this
analysis suggest that the steering influence of the upstream
trough assists the transport of water vapor northward from
the core of the NAM into the southwestern United States
and enhances large-scale convective midtropospheric instability (e.g., Lang et al. 2007; Anderson and Roads 2002;
Carleton 1986). Such findings are consistent with Higgins
et al. (2004) who showed that the northeast shift in the
monsoonal ridge allows for widespread convective outbreaks over the southwestern United States due to
enhanced northward moisture transport into the southwestern United States.
Thus, we hypothesize that the eastward shift of the
monsoonal anticyclone and upstream trough provide a
conduit for the northward extension of subtropical
moisture organized by a tropical disturbance (e.g., easterly wave, tropical cyclone) into the interior southwestern
United States. The regional surges that affect the southwestern United States tend to lack the strong midlatitude
dynamics of tropical moisture plumes, or atmospheric
rivers, that impact the central and western United States
(e.g., Dirmeyer and Kinter 2009; Higgins et al. 2011;
Ralph et al. 2006). However, we suggest that regional
surges associated with the NAM adhere to a subtle, albeit
important, coupling of tropical–extratropical dynamics.
Acknowledgments. The authors appreciate the use
NLDAS-2 data from acquired as part of the mission
of NASA’s Earth Science Division and archived and
distributed by the Goddard Earth Sciences (GES) Data
and Information Services Center (DISC). They also
JANUARY 2013
FAVORS AND ABATZOGLOU
acknowledge Jim Ashby from the Western Regional
Climate Center for providing station data and helpful
reviews from Eugene Cordero, Robert Bornstein, and
two anonymous reviewers. Funding for this work was
provided from NSF Grant ATM-0801474.
REFERENCES
Abatzoglou, J. T., and T. J. Brown, 2009: Influence of the Madden–
Julian oscillation on summertime cloud-to-ground lightning
activity over the continental United States. Mon. Wea. Rev.,
137, 3596–3601.
Adams, D. K., and A. C. Comrie, 1997: The North American
monsoon. Bull. Amer. Meteor. Soc., 78, 2197–2213.
Anderson, B. T., and J. O. Roads, 2002: Regional simulation of
summertime precipitation over the southwestern United
States. J. Climate, 15, 3321–3342.
Barlow, M., S. Nigam, and E. Berbery, 1998: Evolution of the
North American monsoon system. J. Climate, 11, 2238–2257.
Bordoni, S., and B. Stevens, 2006: Principal component analysis of
the summertime winds over the Gulf of California: A gulf
surge index. Mon. Wea. Rev., 134, 3395–3414.
Brenner, I. S., 1974: A surge of maritime tropical air—Gulf of
California to the southwestern United States. Mon. Wea. Rev.,
102, 375–389.
Carleton, A. M., 1986: Synoptic-dynamic character of ‘bursts’ and
‘breaks’ in the South-West U.S. summer precipitation singularity. J. Climatol., 6, 605–623.
Castro, C. L., T. B. McKee, and R. A. Pielke, 2001: The relationship
of the North American monsoon to tropical and North Pacific
sea surface temperatures as revealed by observation analyses.
J. Climate, 14, 4449–4473.
Corbosiero, K. L., M. J. Dickinson, and L. F. Bosart, 2009: The
contribution of eastern North Pacific tropical cyclones to the
rainfall climatology of the Southwest United States. Mon.
Wea. Rev., 137, 2415–2435.
Dirmeyer, P. A., and J. L. Kinter III, 2009: The ‘‘Maya Express’’:
Floods in the U.S. Midwest. Eos, Trans. Amer. Geophys.
Union, 90, 101, doi:10.1029/2009EO120001.
Dixon, P. G., 2005: Using sounding data to detect gulf surges during
the North American monsoon. Mon. Wea. Rev., 133, 3047–3052.
Douglas, M. W., R. A. Maddox, K. Howard, and S. Reyes, 1993:
The Mexican monsoon. J. Climate, 6, 1665–1677.
Fuller, R. D., and D. J. Stensrud, 2000: The relationship between
tropical easterly waves and surges over the Gulf of California
during the North American monsoon. Mon. Wea. Rev., 128,
2983–2989.
Gutzler, D. S., 2004: An index of interannual precipitation variability in the core of the North American monsoon region.
J. Climate, 17, 4473–4480.
Hales, J. E., 1972: Surges of maritime tropical air northward over
the Gulf of California. Mon. Wea. Rev., 100, 298–306.
Higgins, R. W., and W. Shi, 2001: Intercomparison of the principal
modes of interannual and intraseasonal variability of the
North American Monsoon System. J. Climate, 14, 403–417.
——, ——, and C. Hain, 2004: Relationships between Gulf of
California moisture surges and precipitation in the southwestern United States. J. Climate, 17, 2983–2997.
——, V. E. Kousky, and P. Xie, 2011: Extreme precipitation events
in the south-central United States during May and June 2010:
Historical perspective, role of ENSO, and trends. J. Hydrometeor., 12, 1056–1070.
191
Jiang, X., and N.-C. Lau, 2008: Intraseasonal teleconnection between the North American and western North Pacific monsoons with a 20-day time scale. J. Climate, 21, 2664–2679.
Kiladis, G. N., and E. A. Hall-McKim, 2004: Intraseasonal modulation of precipitation over the North American monsoon
region. Preprints, 15th Symp. on Global Change and Climate
Variations, Seattle, WA, Amer. Meteor. Soc., 11.4. [Available
online at http://ams.confex.com/ams/pdfpapers/72428.pdf.]
Ladwig, W. C., and D. J. Stensrud, 2009: Relationship between
tropical easterly waves and precipitation during the North
American monsoon. J. Climate, 22, 258–271.
Lang, T. J., D. A. Ahijevych, S. W. Nesbitt, R. E. Carbone, S. A.
Rutledge, and R. Cifelli, 2007: Radar-observed characteristics
of precipitating systems during NAME 2004. J. Climate, 20,
1713–1733.
Lu, E., X. Zeng, Z. Jiang, Y. Wang, and Q. Zhang, 2009: Precipitation and precipitable water: Their temporal-spatial behaviors and use in determining monsoon onset/retreat and
monsoon regions. J. Geophys. Res., 114, D23105, doi:10.1029/
2009JD012146.
Maddox, R. A., D. McCollum, and K. Howard, 1995: Large-scale
patterns associated with severe summertime thunderstorms
over central Arizona. Wea. Forecasting, 10, 763–778.
McCollum, D. M., R. A. Maddox, and K. W. Howard, 1995: Case
study of a severe mesoscale convective system in central
Arizona. Wea. Forecasting, 10, 643–665.
Mesinger, F., and Coauthors, 2006: North American Regional
Reanalysis. Bull. Amer. Meteor. Soc., 87, 343–360.
Mitchell, K. E., and Coauthors, 2004: The multi-institution North
American Land Data Assimilation System (NLDAS) project:
Utilizing multiple GCIP products and partners in a continental distributed hydrological modeling system. J. Geophys. Res.,
109, D07S90, doi:10.1029/2003JD003823.
Mo, K. C., M. Chelliah, M. L. Carrera, R. W. Higgins, and
W. Ebisuzaki, 2005: Atmospheric moisture transport over the
United States and Mexico as evaluated in the NCEP regional
reanalysis. J. Hydrometeor., 6, 710–728.
Neiman, P. J., A. B. White, F. M. Ralph, D. Gottas, and S. I.
Gutman, 2009: A water vapor flux tool for precipitation
forecasting. Water Manage., 162 (2), 83–94.
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, doi:10.1029/2006GL026689.
Ray, A. J., G. M. Garfin, M. Wilder, M. Vásquez-León, M. Lenart,
and A. C. Comrie, 2007: Applications of monsoon research:
Opportunities to inform decision making and reduce regional
vulnerability. J. Climate, 20, 1608–1627.
Reitan, C. H., 1957: The role of precipitable water vapor in Arizona’s summer rains. Tech. Rep. on the Meteorology and
Climatology of Arid Regions 2, Institute of Atmospheric
Physics, The University of Arizona, Tucson, AZ, 19 pp.
Ritchie, E. A., K. M. Wood, D. S. Gutzler, and S. R. White, 2011:
The influence of eastern Pacific tropical cyclone remnants on
the southwestern United States. Mon. Wea. Rev., 139, 192–210.
Schmitz, T. J., and S. L. Mullen, 1996: Water vapor transport associated with the summertime North American monsoon as
depicted by ECMWF analyses. J. Climate, 9, 1621–1634.
Stensrud, D. J., R. L. Gall, S. L. Mullen, and K. W. Howard, 1995: Model
climatology of the Mexican monsoon. J. Climate, 8, 1775–1794.
——, ——, and M. K. Nordquist, 1997: Surges over the Gulf of
California during the Mexican monsoon. Mon. Wea. Rev., 125,
417–437.