Abstract

This study uses an improved surge identification method to examine composites of 29 yr of surface observations and reanalysis data alongside 10 yr of satellite precipitation data to reveal connections between flow, thermodynamic parameters, and precipitation, both within and outside of the North American monsoon (NAM) region, associated with Gulf of California (GoC) moisture surges. The North American Regional Reanalysis (NARR), examined using composites of flow during all detected moisture surges at Yuma, Arizona, and so-called wet and dry surges (those producing anomalously high and low precipitation, respectively, over Arizona and New Mexico), show markedly different flow and moisture patterns that ultimately lead to the differing observed precipitation distributions in the region. Wet surges tend to be associated with moister precursor air masses over the southwestern United States, have a larger contribution of enhanced easterly cross–Sierra Madre Occidental (SMO) moisture transport, and tend to result from a transient cyclonic disturbance tracking across northern Mexico. Dry surges tend to be associated with a more southerly tracking disturbance, are associated with less convection over the SMO, and tend to be associated with a drier presurge air mass over Arizona and New Mexico.

1. Introduction

Portions of northwestern Mexico and the southwestern United States receive in excess of 70% of their annual precipitation during July, August, and September (Douglas et al. 1993). The circulation changes associated with this boreal summer precipitation maximum in this region have been termed the North American monsoon [NAM; see Adams and Comrie (1997) for an overview of the NAM]. The large-scale circulation changes associated with the NAM modulate precipitation within and surrounding the core of the NAM region (Higgins et al. 1997, 1998; Saleeby and Cotton 2004; Diem and Brown 2009). The monsoon is responsible for at least three-quarters of the variance in total precipitation and more than half of the variance in frequency of heavy precipitation events (Diem 2005). Studies have shown that precipitation within the NAM is often highly variable; a single monsoon rain event can contribute more than a quarter of the total July–August precipitation, and severe convective storms may bring damaging winds and flash flooding (McCollum et al. 1995; Gochis et al. 2007). Because of decision makers’ need for regional to local forecasts for civil protection, water resource management, public health, and agriculture, we are motivated to understand how mechanisms that drive NAM precipitation variability operate and how they may be predicted. To date, such forecasts have proven difficult (Dunn and Horel 1994a,b; Ray et al. 2007).

As a possible cause of observed monsoon precipitation bursts in the NAM, Hales (1972) and Brenner (1974) introduced the concept of the Gulf of California (GoC) surge. Several studies have demonstrated the relationship between surges and NAM precipitation using various techniques (Berbery and Fox-Rabinovitz 2003; Higgins et al. 2004, hereafter H04; Adams and Stensrud 2007; Becker and Berbery 2008; Svoma 2010). To elucidate this relationship, H04 identified surges in surface observations at Yuma, Arizona, and used gridded rain gauge data to analyze precipitation anomalies coincident with surges. In that study, a positive precipitation anomaly appears over southern Mexico before surge arrival in Yuma, while a negative precipitation anomaly locates over northern Mexico. The positive anomaly associated with the surge then moves north along the western Mexican coastline as the surge progresses, reaching northwestern Mexico by the time the surge arrives in Yuma and the southwestern United States in the subsequent days. Despite the long-held belief that surges are associated with enhanced regional mean NAM precipitation, recent work by Ladwig and Stensrud (2009) showed that surge events might redistribute NAM precipitation within the domain rather than enhance it.

Several studies examine the structure and evolution of surges using case studies (e.g., Hales 1972; Brenner 1974; Rogers and Johnson 2007) as well as wind, temperature, pressure, and meridional moisture flux composites of several surges (e.g., Anderson et al. 2000a,b; Douglas and Leal 2003). Recent studies explored the potential to identify surges in 12-hourly Quick Scatterometer (QuikSCAT) ocean-surface winds alone (Bordoni et al. 2004; Bordoni and Stevens 2006). However, surges are most often detected using surface measurements of dewpoint, wind direction, and wind speed to exploit their high temporal resolution (e.g., Fuller and Stensrud 2000; H04; Dixon 2005). In this study, we aim to use high-resolution reanalysis data that provide a continuous, dynamically consistent dataset to examine the spatiotemporal characteristics of surges.

It is thought that the enhancement of convection along the Sierra Madre Occidental (SMO) by favorable synoptic conditions leads to the dynamic mechanism that is conducive for surge generation and propagation along the SMO. These synoptic conditions may include the passage of a tropical easterly wave (TEW), tropical cyclone, or the presence of a tropical upper-tropospheric trough near the core NAM region (Fuller and Stensrud 2000; Pytlak et al. 2005; Adams and Stensrud 2007; Douglas and Englehart 2007; Lang et al. 2007; Nesbitt et al. 2008; Bieda et al. 2009; Ladwig and Stensrud 2009). Any of these conditions may lead to enhanced moisture convergence over the SMO and often give rise to organized convection, which produces cold, relatively moist outflows that may contribute to surge air masses that bring moist air into southwestern Arizona.

It is established that the low-level thermodynamic conditions and circulation markedly change after surge arrival. At the mouth of the GoC, a low-level anticyclonic flow anomaly replaces a cyclonic flow anomaly, which has been associated with a TEW passage at the mouth of the GoC coincident with surge arrival at Yuma (Douglas and Leal 2003). Low-level wind anomalies reverse from northerly to southerly as the surge and associated thermodynamic anomalies, including decreased temperature and increased dewpoint [thus markedly increased relative humidity (RH)] progress up the GoC (e.g., Mullen et al. 1998; Anderson et al. 2000b; Douglas and Leal 2003; Bordoni et al. 2004). Previous findings also connect the surge to a GoC low-level jet (LLJ; Douglas 1995). Once the surge reaches the southwestern United States, it somewhat loses its identity because of strong sensible heat fluxes, reduced surface moisture fluxes, and the removal of confining topography (Hales 1972; Rogers and Johnson 2007). Despite strong sensible heating over the deserts of the Southwest that tries to mix surge air with the drier surrounding air, the surge air mass may still provide low-level moisture to enhance convective precipitation.

The origin of moisture for surge-associated precipitation, as well as for convective systems that ultimately contribute air to surge flows, has been a controversial topic. The role of local versus remote sources of moisture for surges must be determined to address surge sources of predictability. In the NAM, local sources, such as evaporation from coastal land and the GoC, may be important (Adams and Comrie 1997; Berbery 2001; Bosilovich 2003; Dominguez et al. 2008), while distant sources, such as the eastern Pacific Ocean, land surface moisture east of the SMO, and the Gulf of Mexico (GoM), may also provide significant quantities of moisture at certain points in surge evolution (Adams and Comrie 1997).

This study uses an improved version of H04’s identification method that more easily and objectively identifies surges, the North American Regional Reanalysis (NARR) to diagnose the moisture budget during surges, and multisatellite, gauge-bias-corrected Tropical Rainfall Measuring Mission (TRMM) precipitation data to address some of the outstanding questions regarding precipitation related to surges. These questions include: How do surges affect the distribution and timing of precipitation over the NAM region? Where does surge moisture originate? What moisture and thermodynamic characteristics differentiate the spatiotemporal evolution of high-precipitation surges and low-precipitation surges? To address these questions, surges are identified in Yuma surface observations using a more comprehensive RH-based metric that is updated to account for low presurge dewpoint and a rapid rise in moisture (Hales 1972). Composites of wet, dry, and all surges are then inspected for attendant changes in moisture and precipitation. Section 2 explains the data and methodology used herein. Section 3 examines characteristics of surges as observed at the surface in Yuma. Flow, precipitation, and moisture patterns associated with surges are discussed in sections 46. Sections 7 and 8 discuss and summarize the results, respectively.

2. Methods and data

a. Surface observations

The automated surge identification methods used herein are based conceptually on the manual detection methods used in H04 (see Table 1) to retain the high temporal resolution of surface observations [when compared to other surge identification methods, such as QuikSCAT in Bordoni and Stevens (2006)]. Surges are identified during July and August 1979–2007 using hourly surface observations from the Yuma, Arizona, Marine Corps Air Station (MCAS; 32.65°N, 114.62°W, elevation 65 m). For times without an observation on the hour, the closest observation within 10 min of the hour is used. If no suitable observation exists, then that time is not included in the analysis (nearly all of July 1987 was missing, along with a few shorter periods). A 1–2–3–2–1 weighted running mean is used to smooth high-frequency variations in wind direction. As in H04, a 25-h running mean is applied to temperature, dewpoint, RH, and wind speed to smooth diurnal variations.

Table 1.

Comparison of H04 surge identification criteria to those used in this study.

Comparison of H04 surge identification criteria to those used in this study.
Comparison of H04 surge identification criteria to those used in this study.

After smoothing, H04 required an observed dewpoint of at least 15.7°C, a minimum wind speed of 3.3 m s−1, and wind direction of roughly 150°–200°. Upon manual inspection of the time series using H04’s classification scheme, criteria were refined to better identify the beginning and end of surges at Yuma by addressing a few issues, namely, the lack of consideration for the rapid rise mentioned by H04 and Fuller and Stensrud (2000) and the inability of H04’s method to catch a surge when the background dewpoint is already >15.7°C (Bordoni and Stevens 2006).

Hales (1972) stated that relative humidity substantially increases at the surface, particularly in the deserts with surge arrival. To mark the arrival of surge air, a 12% rise in RH over 24 h is used. The RH threshold was tuned to fit visually identified surges during the period 1979–2007. In addition to detecting a rise in dewpoint (sought in previous studies), this metric corrects for the two above-mentioned shortcomings in previous methods (high background dewpoint and rapid rise in moisture) and accounts for the temperature drop associated with surge arrival. The wind direction must be 135°–203° to constrain the flow originating along the long axis of the GoC. The range is expanded eastward from H04’s to better account for the southeast–northwest tilt of the GoC. The wind speed threshold, reduced to 2 m s−1, is a weak constraint used to ensure that it is indeed surge air that reached Yuma before surface sensible heating and mixing dilute the RH signal during the surge’s first 24 h over land (Brenner 1974).

While H04 relied on a static dewpoint threshold of 15.7°C to find the end of a surge, this study defines the end of the surge as a net decrease in RH over 12-, 24-, and 48-h periods after surge arrival. This modification was employed because background dewpoint tends to increase as the monsoon matures, which made surges appear longer in our analysis of the latter part of the monsoon. These three periods were considered to avoid prematurely ending a surge in the event of a brief decrease in relative humidity.

b. Precipitation data

To improve our understanding of surge precipitation characteristics, we use the TRMM 3B42 version 6 Multisatellite Precipitation Analysis, which provides spatially continuous, 3-hourly, 0.25° resolution data. TRMM 3B42 data combine the average precipitation rate over 3 h (centered on the time stamp of the data) from infrared and passive microwave satellite–based measurements, which are rescaled to monthly rainfall gauge totals on a 2.5° grid (Huffman et al. 2007). TRMM 3B42 data start in 1998, limiting the years available for comparison with longer-term datasets. Prior studies rely on the National Oceanic and Atmospheric Administration’s Climate Prediction Center (CPC) Unified Raingauge Dataset over the United States and Mexico. Although the gauge dataset is temporally longer, it relies on sparse operational rain gauge networks in Mexico and does not include measurements over water.

This study focuses on the mean and departures from the mean precipitation rate at times relative to surge arrival in Yuma. The nearest 3-h time following surge arrival at Yuma is used as the reference time (t = 0). A 24-h period surrounding t − 24, t = 0, and t + 48 h is averaged across all surges to account for and completely assess the effects of surges with arbitrary arrival times relative to the strong convective diurnal cycle (Nesbitt et al. 2008). Anomalies are calculated by subtracting the daily climatology for July–August 1998–2007 from the above-mentioned surge-relative means.

H04 noted that some surges over Arizona and New Mexico fail to produce significant precipitation events. To investigate the differences in precipitation pattern evolution between so-called wet and dry surges, TRMM data are composited separately for these events. Wet and dry surges are identified in the manner of H04 from 1998 to 2007. Wet surges are those that show a positive anomaly when compared to the TRMM 1998–2007 mean in the areal average precipitation over eastern Arizona and western New Mexico (32°–36°N, 112.5°–107.5°W), and dry surges consist of those that show a negative anomaly.

c. Reanalysis data

Past studies have used the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis (Kalnay et al. 1996) to study large-scale features associated with gulf surges. This study employs the NARR (Mesinger et al. 2006). The NARR boasts two major improvements over prior reanalyses: 1) it assimilates observed precipitation in addition to the usual atmospheric observations incorporated in other reanalyses and 2) it has a 32-km horizontal resolution, 36 times the horizontal resolution of the NCEP–NCAR reanalysis. NARR data, available every 3 h, are used to study the 1979–2007 monsoons.

For easy comparison, NARR fields are analyzed using the same methods as the TRMM data, except that the climatology encompasses July–August of 1979–2007. To examine synoptic-scale conditions associated with the spatiotemporal variability of surges, selected parameters are composited to reveal anomalous flow, thermodynamics, and moisture budget characteristics.

d. Statistical significance

All NARR and TRMM anomalies (which are 24-h averages) are tested at each grid point for statistical significance using the two-tailed Student’s t test (Wilks 2006) with respect to daily averages of their respective climatologies. Vector significance is calculated for zonal and meridional components separately. Unless otherwise noted, only anomalies that are statistically significant at the 90% level are discussed.

e. Moisture budgets

The moisture budget over a given area is given by (e.g., Yanai et al. 1973):

 
formula

where the integrals are evaluated over the depth of the atmosphere, g is gravitational acceleration, q is specific humidity, p is pressure, V is the vector horizontal wind, and E and P are the surface evaporation and precipitation flux, respectively. Over the specified area, the first term on the left side is the time rate of change of precipitable water (PW), while the second term is the horizontal moisture flux convergence (MFC). Moisture budget terms have been obtained from NARR, and a residual term r due to NARR’s precipitation assimilation scheme and other adjustments was estimated following Eq. (2) in Ruane (2010). NARR moisture budgets may suffer because of 1) overactive modeled convection leading to overestimates of MFC over the core monsoon region in Mexico (Becker and Berbery 2008), 2) differences between model-resolved terrain and real terrain where assimilated precipitation is observed (Ruane 2010), and 3) discontinuities in the precipitation datasets used in the assimilation scheme along the U.S.–Mexico border (Mo et al. 2005a).

Moisture budget terms have been calculated for plan views and averaged over the five regions shown as boxes in Fig. 1. Each box is seven NARR grid points (≈230 km) on a side, and the 3-hourly NARR moisture budget values were averaged spatially over the box relative to Yuma surge onset time and temporally smoothed by a nine-point running mean to remove diurnal variations. To understand the contribution of the prevailing moisture flux from the Atlantic basin (from the east) versus the Pacific basin (from the south) on surge moisture budgets, the components of the vertically integrated moisture flux F [e.g., ] normal to the eastern and southern boundaries of each box were also calculated separately.

Fig. 1.

NARR topography (m) and locations of areal averages.

Fig. 1.

NARR topography (m) and locations of areal averages.

3. Surge identification

The number and timing of surges identified by the updated method closely matches those identified by H04. For example, H04 and the current method identify the same eight surges in 1986 (Fig. 2 compared to Fig. 1 in H04). Svoma (2010) found approximately three surges per year, many fewer than H04 or this study, using the 850-hPa dewpoint at Tucson, Arizona. The difference is at least in part due to location, but the identification method may also play a role. The number of surges identified using QuikSCAT winds are more in line with our findings: 51 surges in six 4-month monsoons, or approximately 5 for every 2 months (Bordoni et al. 2004). Since the commencement of an RH upswing marks a surge arrival in this study, surge arrival times differ by up to a couple of days because the threshold used by H04 may fall in the middle of the upswing in RH. Our approach to surge identification finds an average of 6.6 surges per year during the period 1979–2007 compared to 5.9 surges per year during the period 1977–2001 found by the H04 method. H04 did not quantitatively mention the length of the surges. The proportion of dry to wet surges is similar to H04. H04 showed that 54% of surges were wet and 46% were dry, whereas this study found 56% (44/79) wet and 44% (35/79) dry.

Fig. 2.

Surges identified in 1986 in the present study for comparison with Fig. 1 of H04.

Fig. 2.

Surges identified in 1986 in the present study for comparison with Fig. 1 of H04.

Average surge time series

Most surges last 24–48 h. The first 36 h show the expected sharp rise in RH, a mean temperature drop of 2°C, and a mean dewpoint rise of 6°C for all surges (Fig. 3). These are consistent with the relatively cool, moist character of the surge air mass and show that both temperature and dewpoint vary in surges. Wet surges involve higher RH and dewpoint throughout, and higher temperature presurge than dry surges. The change in RH and dewpoint is nearly the same for wet and dry surges, though the temperature decreases more during wet surges and does not rebound as quickly as postsurge. All three variables show a return toward, but not quite to, presurge values. When the all-surge lines in Fig. 2 are compared to Fig. 2a from H04, there are few disparities despite the difference in datasets and identification methods. Upon visual inspection, surges lasting longer than 96 h often show a 48-h periodicity on top of the trends associated with a single surge. The pattern is most obvious in RH, but it is also visible in temperature and dewpoint. The cause and implications of this apparent periodicity require further investigation.

Fig. 3.

Time series of temperature (°C), relative humidity (%), and dewpoint (°C) after surge arrival at the Yuma MCAS surface observation station averaged over all (solid), wet (dash–dot), and dry (dashed) surges.

Fig. 3.

Time series of temperature (°C), relative humidity (%), and dewpoint (°C) after surge arrival at the Yuma MCAS surface observation station averaged over all (solid), wet (dash–dot), and dry (dashed) surges.

4. Precipitation

Surge times contribute a disproportionately high amount of the July–August precipitation over much of the monsoon region, with some areas experiencing greater than twice the average seasonal rainfall rate. Before surge arrival, the desert Southwest and northwestern Mexico show negative precipitation anomalies that are more strongly manifest in dry surges (Figs. 4a,d,g). By t = 0 the negative anomalies are replaced by weak positive rainfall anomalies along the Arizona–Mexico border (Figs. 4b,e,h). These anomalies are entirely due to wet surges, as rainfall is still less than average over the entire southwestern United States in dry surges. The NAM core is neutral in all surges at t = 0, but wet surges show strong positive anomalies, while dry surges bring a mix of anomalies. Two days after surge arrival, positive anomalies persist over the Southwest and into the southern Rockies during wet surges and all positive anomalies are replaced by negative anomalies during dry surges (Figs. 4c,f,i). In fact, dry surges reduce precipitation over Arizona, Utah, Colorado, western New Mexico, and extreme northwestern Mexico throughout all the times examined.

Fig. 4.

Anomalies from the 24-h mean precipitation (mm day−1; 1998–2007), summed from 12 h before the indicated surge-relative time to 12 h after for t − 24, surge arrival, and t + 48 for (a)–(c) all, (d)–(f) wet, and (g)–(i) dry surges, from TRMM 3B42 data.

Fig. 4.

Anomalies from the 24-h mean precipitation (mm day−1; 1998–2007), summed from 12 h before the indicated surge-relative time to 12 h after for t − 24, surge arrival, and t + 48 for (a)–(c) all, (d)–(f) wet, and (g)–(i) dry surges, from TRMM 3B42 data.

5. Geopotential heights and flow

a. 700 hPa

The most prominent low- to midtropospheric features are the center of high geopotential height over the western United States and the transient trough of low geopotential heights that move from Mexico at t − 24 h westward into the eastern Pacific (Figs. 5a–c, 6). This meridionally extensive transient trough axis appears similar to a TEW, which has been implicated in prior studies in providing the conditions for surge initiation in the SMO. The higher heights over the United States are part of an enhanced ridge whose axis seems to lie over the Rockies.

Fig. 5.

Mean geopotential height (gpm) at (a)–(c) 700 and (d)–(f) 200 hPa for t − 24, t = 0, and t + 48 h, from July to August 1979–2007. Red contours are for dry surges, blue contours are for wet surges, and black contours are for all surges.

Fig. 5.

Mean geopotential height (gpm) at (a)–(c) 700 and (d)–(f) 200 hPa for t − 24, t = 0, and t + 48 h, from July to August 1979–2007. Red contours are for dry surges, blue contours are for wet surges, and black contours are for all surges.

Fig. 6.

As in Fig. 4, but for 700-hPa geopotential height (gpm; contours) and wind (m s−1) anomalies.

Fig. 6.

As in Fig. 4, but for 700-hPa geopotential height (gpm; contours) and wind (m s−1) anomalies.

Over most of the contiguous United States, the flow associated with wet and dry surges diverges from the pattern for all surges (Figs. 5a–c, 6). Wet surges are associated with a strong, broad positive geopotential height anomaly over most of the United States that slowly diminishes in intensity (Figs. 6d–f). In contrast, dry surges are associated mainly with a negative geopotential height anomaly across the eastern United States that migrates southward with time and a weaker anticyclonic anomaly in the northwestern United States prior to t = 0 (Figs. 6g–i). Both cases still show a ridge over the United States and a TEW over Mexico and the southern NAM area, but the ridge is stronger during wet surges (see Fig. 5a). The contrasting anomalies over the United States shown in Figs. 6d–i largely cancel each other in the all-surge anomalies, though the trough is still quite apparent (Figs. 6a–c). Similarly, the TEW follows different tracks and intensities in wet and dry composites (Figs. 6d–i). It takes a more northerly (southerly) track in wet (dry) surge composites as shown in Figs. 6d–f (Figs. 6g–i), providing for stronger easterly flow anomalies over most of the core monsoon region.

b. 200 hPa

At 200 hPa a positive height anomaly, the northwestward extension of the monsoon anticyclone (see Figs. 6d–f, 7d–f), sits over the desert Southwest to the southeast of the strong positive 700-hPa height anomaly before t = 0 (Fig. 7a–c; e.g., H04; Diem and Brown 2009). The enhanced ridge weakens with time at both levels. A broad negative height anomaly develops south of 20°N as the monsoon anticyclone weakens. Again, the trough is located south of a similar feature at 700 hPa and may indicate of a westward-tilted TEW or inverted trough (Kelly and Mock 1982; Bieda et al. 2009; Serra et al. 2010). Averaged across all surges, the most significant flow anomalies are easterly over northern Mexico at t − 24 h and move westward with the low-latitude trough.

Fig. 7.

As in Fig. 6, but for 200-hPa anomalies.

Fig. 7.

As in Fig. 6, but for 200-hPa anomalies.

Most upper-level features in wet and dry surges offer significant contrast as in H04. Figures 5d–f indicates a more northerly (southerly), stronger (weaker) monsoon anticyclone during wet (dry) surges. A large area of increased heights covers the United States during all wet surge times (Figs. 7d–f), while during dry surges, a low geopotential height anomaly is present over the northern United States (not statistically significant; Figs. 7g–i). Deep easterly flow anomalies are present along the southern edge of the monsoon high over the NAM core prior to an initiation of a wet surge, compared with a deep southerly flow along the western side of the high during dry surges; flow anomalies remain easterly throughout wet surges. These flow anomalies over the monsoon region are related to the locations of the trough in the tropics and the position and strength of the monsoon anticyclone.

6. Moisture and thermodynamics

A significant positive anomaly in θe, an indicator of low-level moist static energy, appears at the mouth of the GoC in these all-surge composites by t − 24 h (Fig. 8a) and begins a lockstep evolution with a positive precipitation anomaly (Figs. 4a–c). A couplet of leading positive θe and trailing negative θe anomalies is present by t = 0 (Fig. 8b), in connection with the all-surge composite TEW feature shown in Figs. 6a–c. Both anomalies strengthen as they move northward over the GoC, remaining significant until at least two days after surge arrival (Fig. 8c). The positive anomaly spreads over Arizona and New Mexico once it escapes the confines of the topography of the SMO. Conversely, the negative θe anomaly is confined between the SMO and Baja California.

Fig. 8.

As in Fig. 6, but for anomalies in equivalent potential temperature (K; contours) at 2 m and vertically integrated moisture flux (kg m−1 s; vectors).

Fig. 8.

As in Fig. 6, but for anomalies in equivalent potential temperature (K; contours) at 2 m and vertically integrated moisture flux (kg m−1 s; vectors).

During dry surges, a strong negative θe anomaly covers the southwestern United States (Figs. 8g–i). Dismantling θe into temperature (Figs. 9g–i) and specific humidity (Figs. 10g–i) reveals warmer and drier air than normal, which means the reduction in θe is entirely due to reduced moisture. This θe anomaly weakens postsurge. In the case of a wet surge, the Southwest already has anomalously high θe—the surge simply accentuates it (Figs. 8d–f). Figures 9d–f, 10d–f confirm that this anomaly indicates both anomalous warm and moist air. However, the influx of surge air clearly cools and moistens the atmosphere, so any increase in θe is entirely due to an increase in moisture.

Fig. 9.

As in Fig. 6, but for anomalies in 2-m temperature (K; contours) and 10-m wind (m s−1; vectors).

Fig. 9.

As in Fig. 6, but for anomalies in 2-m temperature (K; contours) and 10-m wind (m s−1; vectors).

Fig. 10.

As in Fig. 6, but for anomalies in 2-m specific humidity (g kg−1; contours) and 10-m wind (m s−1; vectors).

Fig. 10.

As in Fig. 6, but for anomalies in 2-m specific humidity (g kg−1; contours) and 10-m wind (m s−1; vectors).

Anomalous moisture transport during surge events originates from both sides of the SMO, as shown in a vertical cross section of zonal and vertical moisture flux vectors and θe anomalies constructed from the Pacific Ocean to the GoM (Fig. 11) along the line shown in Fig. 1. Anomalous east-to-west moisture transport and elevated θe from the GoM extends across the SMO the day before surge arrival in Yuma through t = 0, while anomalous west-to-east moisture transport west of the SMO leads to moisture convergence and elevated low-level θe along the SMO slopes (Figs. 11a,b). The accompanying low θe region following the TEW axis at and after t − 24 h (Fig. 8), which also partially originates east of the SMO and from drying following active convection (Fig. 4), leads to negative low-level θe anomalies along the western slopes of the SMO after surge arrival (Fig. 11c). All of these flux anomalies, save the up-gulf advection of moist air during dry surges, are almost entirely due to wet surges (Figs. 8d–i, 11d–i). Dry surges generally show much a weaker moisture flux that contributes little to the all-surge average.

Fig. 11.

As in Fig. 6, but for vertical cross section (see dashed line in Fig. 3) of mean equivalent potential temperature (K; contours), moisture flux (kg m−1 s; vector horizontal component), and vertical velocity (×100 Pa s−1; vector vertical component).

Fig. 11.

As in Fig. 6, but for vertical cross section (see dashed line in Fig. 3) of mean equivalent potential temperature (K; contours), moisture flux (kg m−1 s; vector horizontal component), and vertical velocity (×100 Pa s−1; vector vertical component).

Moisture budgets and moisture fluxes relative to surge onset

Based on the above, large differences in the bulk moisture budgets and flows would be expected among all-surge averages, wet surges, and dry surges. Herein the spatiotemporal variability of the remaining quantities of the moisture budget are examined: PW, MFC, and E, as well as r, the residual term due to NARR’s assimilation process.

In Fig. 12, plan views of PW composites are shown along with vertically integrated moisture flux to examine the time evolution of columnar moisture. In the all-surge composites (Figs. 12a–c), anomalously low PW is shown over the Colorado basin into the tropical east Pacific at t − 24 h, replaced by a westward-moving region of anomalously high PW. In wet surges, strong positive PW anomalies exist over the desert Southwest prior to surge arrival, accompanied by zonally significant easterly moisture transport anomalies over most of the NAM region (including up the axis of the Rio Grande river valley into Arizona and New Mexico). Over and downstream of the SMO, weak (and locally statistically significant) positive PW anomalies exist at t − 24 h (Fig. 12d). At this time, a slight negative PW anomaly exists over north-central Mexico; despite this, moisture flux is significantly greater than normal upstream of the SMO. By t = 0 (Fig. 12e), the PW pattern remains similar to t − 24 h, but the anomalous moisture transport becomes more southerly by t + 48 h (Fig. 12f), while conditions over the SMO now have anomalously low PW. In contrast, statistically significant low PW sits over the desert Southwest in dry surges for most of the period. At t = 0 (Fig. 12h) statistically significant southeasterly anomalous moisture transports, originating over the Pacific, are associated with increased PW and positive PW anomalies (although not statistically significant) over the slopes of the SMO.

Fig. 12.

As in Fig. 6, but for anomalies in PW (kg m−2; contours) and vertically integrated moisture flux (kg m−1 s; vectors).

Fig. 12.

As in Fig. 6, but for anomalies in PW (kg m−2; contours) and vertically integrated moisture flux (kg m−1 s; vectors).

In terms of MFC (Fig. 13), in all surge classifications at t − 24 h (Figs. 13a,d,g) positive anomalies exist over the SMO, with anomalous moisture flux divergence over Arizona and New Mexico, although in comparing wet and dry surges the maximum is located more over the SMO and northeastern Mexico in wet surges (Fig. 13d) compared with dry surges (Fig. 13g). By t = 0, wet surge maxima in MFC are more focused in the SMO, perhaps provided for by convective systems there in the half day prior to surge arrival, whereas in dry surges there is a center of strong positive MFC anomalies over the northern GoC and near Yuma. By t + 48 h, positive (negative) MFC is located over most of northern Arizona and New Mexico (GoC and the SMO) in wet surges (Fig. 13f), while negative anomalies exist over most of the northern SMO, Arizona, and New Mexico by this time in dry surges (Fig. 13i).

Fig. 13.

As in Fig. 6, but for anomalies in moisture convergence (kg m−2 s; contours) and vertically integrated moisture flux (kg m−1 s; vectors).

Fig. 13.

As in Fig. 6, but for anomalies in moisture convergence (kg m−2 s; contours) and vertically integrated moisture flux (kg m−1 s; vectors).

Finally, E anomaly patterns (Fig. 14), although significantly less in magnitude than MFC anomalies (Fig. 13), show that the GoC provides the most dynamic moisture source for surges, with anomalously low E prior to t = 0 over most of the GoC (Figs. 14a,d,g). Contrasting wet (Figs. 14d,e,f) and dry (Figs. 14g,h,i) surges through the period examined, it appears that 1) positive E anomalies are stronger and extend farther northward over the length of the GoC in wet surges; and 2) near-zero to small positive E anomalies exist over the high terrain in Arizona and New Mexico in wet surges, whereas statistically significant negative E anomalies exist in these regions in dry surges.

Fig. 14.

As in Fig. 6, but for anomalies in evaporation (kg m−2 s; contours) and vertically integrated moisture flux (kg m−1 s; vectors).

Fig. 14.

As in Fig. 6, but for anomalies in evaporation (kg m−2 s; contours) and vertically integrated moisture flux (kg m−1 s; vectors).

To examine the source regions for atmospheric water vapor for precipitation associated with wet versus dry surges, terms of the atmospheric water vapor budget relative to Yuma surge onset time are presented for five key regions in Fig. 15 (see section 2e for details on how these budgets were constructed and Fig. 1 for the location of the regions). In southwestern Arizona there is a strong increase in PW with surge arrival; however, this is balanced quickly by strong moisture flux divergence as the moisture is carried into the interior of the southwestern United States. Precipitation increases negligibly in this region in NARR, which may be due to the aforementioned errors in the assimilation scheme near the U.S.–Mexico border. Further inland over the Mogollon Rim (Fig. 15b), PW tendencies are strongly negative from t − 72 to t − 36 h prior to wet surges (and near zero for dry surges) and positive in the 36 h following surge arrival (higher tendencies are present in dry surges, but PW is already high prior to wet surge arrival; see Fig. 12). Moisture flux convergence in wet surges is positive at t − 24 h but negative in dry surges. By t + 24 h, convergence during wet surges increases to twice that in dry surges, which is temporally associated with increased precipitation in wet surges that is more than twice that during dry surges. Over central New Mexico (Fig. 15c), precipitation variations and water budget components are more muted than over Arizona, but precipitation, MFC, and change in PW are on the upswing up to a day prior to surge arrival at Yuma.

Fig. 15.

Time series of NARR moisture budget terms (kg m−2 s) from t − 72 to t + 72 h, smoothed using a nine-point running mean, averaged over geographic boxes shown in Fig. 3: (a) southwestern Arizona, (b) Mogollon Rim, (c) central New Mexico, (d) northern portion of the NAM core, and (e) southern portion of the NAM core.

Fig. 15.

Time series of NARR moisture budget terms (kg m−2 s) from t − 72 to t + 72 h, smoothed using a nine-point running mean, averaged over geographic boxes shown in Fig. 3: (a) southwestern Arizona, (b) Mogollon Rim, (c) central New Mexico, (d) northern portion of the NAM core, and (e) southern portion of the NAM core.

Over the core monsoon region where surges initiate, the water budget varies little between the northern and southern SMO (Figs. 15d,e). With nearly constant E, increases in PW (particularly for dry surges) are associated with increases in MFC and precipitation (stronger and earlier in wet versus dry surges), reaching peak values near t − 24 h. Under these conditions precipitation systems develop that help initiate and propagate surges up the GoC. Note that the residual value r is particularly negative over the SMO, indicating that the model produces too little precipitation or too much MFC and/or E. However, it is likely that positive model precipitation errors are offset by too much MFC and E in the core monsoon region (Becker and Berbery 2008; Ruane 2010).

To examine the temporal evolution of moisture sources in these five regions during wet and dry surges, Fig. 16 shows the vertically integrated moisture flux in prevailing flow directions normal to the east and south boundaries of the regional boxes relative to surge arrival at Yuma. Over southwestern Arizona (Fig. 16a), increased moisture transport coincides with surge arrival. Moisture transport from the south is larger than from the east in both wet and dry surges, with higher peak values in wet surges. Farther north over the Mogollon Rim (Fig. 16b), however, moisture transport from the south and east are of comparable magnitude but out of phase, with transport from the east increasing well before surge arrival and transport from the south peaking 1–2 days after surge arrival at Yuma. Large differences in moisture transport from the east prior to wet versus dry surges may help explain why conditions are moist (dry) prior to wet (dry) surge arrival in Arizona. In central New Mexico (Fig. 16c) moisture transport from the south is more important in the mean, but the increase from the east prior to wet surge arrival (not seen in dry surges) is of equal magnitude to transport from the south the day prior to surge arrival.

Fig. 16.

As in Fig. 15, but for the component of moisture flux normal to the southern and eastern borders of each box. Positive is defined as flow into the box.

Fig. 16.

As in Fig. 15, but for the component of moisture flux normal to the southern and eastern borders of each box. Positive is defined as flow into the box.

Little variation in behavior is seen between the two regions over the complex terrain of the monsoon core (Figs. 16d,e). Note that easterly moisture transports are significant throughout the surge cycle, peak nearly a day prior to the increase in transport from the south, and are stronger in both wet and dry surges at all times compared to transport from the south. Moisture transport from the east begins to increase a day before surge arrival at Yuma, roughly 12 h earlier over the southern SMO than over the northern SMO. Northerly moisture transport peaks near or shortly after surge arrival, then decreases, and finally turns southerly, riding the surge into the southwestern United States. From this analysis, it is clear that moisture transport from the Atlantic basin prior to surge arrival is important for surge-initiating precipitation systems over the SMO and may be an important precursor to wet surges in the desert Southwest of the United States.

7. Discussion

Centered on the time of surge arrival in Yuma, composites of TRMM 3B42 0.25° precipitation and NARR fields reveal several characteristics of surges at temporal and spatial resolutions that have not been shown previously. Variations in the strength of the upper-level anticyclone and the track of the cyclonic flow anomaly at lower latitudes appear key in providing for moisture and thermodynamic anomalies associated with air masses in the region prior to surge arrival, conditions in the monsoon core region for surge initiation, and pathways for moisture into the southwestern United States for postsurge air masses that ultimately produce precipitation in Arizona, New Mexico, and the surrounding states. The results presented herein suggest that there are fundamental differences in causes and implications of surges that do and do not produce anomalously high precipitation over the southwestern United States.

Associated with enhanced cross-barrier moisture flux, MFC, and high θe air over the SMO, conditions fuel convective systems as shown by TRMM precipitation data (Lang et al. 2007), with enhanced rainfall starting over the SMO the day prior to surge arrival (Figs. 4a–c). Precipitation in the SMO is enhanced and more cohesive for wet surges, during which a more geographically extensive portion of the western slopes are a focus of heavy precipitation compared with dry surges (Figs. 4d–i). Surges and associated precipitation seem to depend greatly on the track and strength of upper-level transient cyclonic flow anomalies, such as TEWs and upper-level troughs (Fig. 6; Adams and Stensrud 2007; Ladwig and Stensrud 2009; Bieda et al. 2009). This study shows several of the flow patterns shown in H04’s conceptual model (their Fig. 13); however, this study emphasizes that the more northerly track of the TEW in wet surges is key to enhancing east-to-west moisture flux presurge, as well as south-to-north moisture flux postsurge (Ladwig and Stensrud 2009). Resolution, model, and data assimilation improvements in NARR over the NCEP reanalysis may help in elucidating these differences in this study. Because of the synoptic-scale flows at mid- and upper levels (including a more northerly tracking cyclonic flow anomaly), stronger and deeper cross-barrier flow, enhanced moisture flux and MFC, and high θe air are all more predominant and meridionally extensive over the SMO compared with dry surges (Figs. 7, 8; Douglas and Leal 2003; Bordoni and Stevens 2006; Johnson et al. 2007; Adams and Stensrud 2007; Ladwig and Stensrud 2009; Bieda et al. 2009; Finch and Johnson 2010).

In addition to varying conditions associated with surge generation, the state of the presurge air mass over Arizona and New Mexico is implicated in providing for enhanced (reduced) precipitation in wet (dry) surges. Low-level warm, moist, high θe air and high PW values exist prior to wet surges, along with enhanced easterly moisture flux, MFC, and to a lesser extent, anomalous E all contributing to high PW values over the southwestern United States, contrasting with conditions discussed above yielding significantly drier surge presurge air masses. Along with less favorable conditions for surge initiation and propagation into the region, the warmer, drier conditions in the southwestern United States during dry surges may promote destruction of the surge air mass through mixing as well as increased entrainment, higher cloud-base heights, and increased subcloud evaporation experienced by convective systems that do develop (Nesbitt et al. 2008).

Composites of wet and dry surges show a trough–ridge–trough pattern across the United States, but the amplitude of this wave train appears greater during wet surges (H04 does not differentiate amplitude). The seesaw pattern between the GoC LLJ and the Great Plains LLJ (Mo and Berbery 2004; Mo et al. 2005a) is evident in wet versus dry surges. The southern Great Plains experiences reduced precipitation during wet surges, while western Mexico and the southwestern U.S. experiences increased precipitation (Figs. 4d–f). Positive precipitation anomalies exist in the central United States during dry surges, while it remains dry over most of Arizona and New Mexico (Figs. 4g–i). Surge propagation and enhanced moisture fluxes may periodically enhance the GoC LLJ and modify the typical evening rainfall maximum observed along the SMO (Dunn and Horel 1994a,b; Douglas 1995; Anderson et al. 2001; Bordoni et al. 2004), especially during wet surges. All of these teleconnections reinforce the idea of a precipitation pattern that is closely related to synoptic drivers (Barlow et al. 1998; Mo and Berbery 2004).

In combination with the plan views of θe and moisture flux, the vertical cross section addresses the long-standing moisture source debate for the North American monsoon (Figs. 8, 11). Midlevel moisture, enhanced by deep easterly flow in wet surges, approaches the SMO two days before surge arrival in Yuma. Moisture convergence over the SMO is likely responsible for the favorable environment for enhanced organized convection along the SMO’s western slopes that ultimately generates the dynamical environment and air masses for surges. This pattern is not evident in dry surges, likely because of weaker easterly flow over the region. The amount of moisture in this zonally advected presurge air mass likely depends on GoM SSTs, convective processes over the GoM and the interior of Mexico, and evaporation and evapotranspiration over northeastern and north-central Mexico. However, the majority of low- and midlevel moisture within surge air masses originates from advection and surface evaporation over the eastern Pacific and the GoC and from land–atmosphere fluxes resulting from recent enhanced precipitation over western coastal Mexico (Dominguez et al. 2008). Details regarding the mesoscale flows and sources of moisture for surge-initiating convection near the SMO cannot be adequately resolved by NARR and need further investigation.

Mo and Berbery (2004) suggested that the NAM moisture source splits geographically—Arizona precipitation originates over the GoC and New Mexico precipitation comes from the GoM. Contrary to Mo and Berbery (2004), the analysis herein suggests that the moisture source could be more dependent on time than location. The GoM contributes significant moisture to convective precipitation early in wet surges in New Mexico and into Arizona’s Mogollon Rim, as the eastern side of the TEW advects moisture up the Rio Grande valley (Ladwig and Stensrud 2009). This revives the idea that moisture sources from east of the SMO may play at least a secondary role in controlling surge frequency and precipitation west of the SMO and in many parts of the southwestern United States through moisture transports. While Pacific basin and local moisture sources appear most important for surge-related rainfall throughout most of the western NAM region and southwestern United States, GoM SSTs and land surface memory over Mexico may also influence NAM precipitation variability on meteorological and climatological scales. These results come with the caveat that the NARR tends to overestimate the up-gulf moisture flux related to the GoC LLJ (Mo et al. 2005b), so the east-to-west moisture contribution to surge-related convection may be relatively more important than the NARR depicts, warranting further observational and modeling studies of moisture flows in the NAM.

8. Conclusions

This study uses an improved GoC surge identification technique based on H04 to investigate the spatiotemporal patterns of atmospheric flow features, moisture transports, and precipitation associated with GoC surges. Using hourly surface observations from Yuma MCAS in Arizona, the updated method can find a surge when the background dewpoint is already relatively high. Many of the surges identified herein match those identified by H04, and composites of surge characteristics also appear similar (H04; Figs. 1, 2).

Our results show that surges are indeed related to important departures from the regional and large-scale climatological flow, moisture transport, and thermodynamic state (Barlow et al. 1998). The main findings include:

  • Updated identification method provides a more quantitative and automated way to identify surges, accounts for the rapid rise in moisture, and utilizes the increase in moisture as well as the decrease in temperature to catch surges when presurge moisture is relatively high (such as late in the monsoon).

  • Wet surges are 26% more common than dry surges (similar to H04).

  • More northerly track of transient cyclonic flow anomalies, such as TEWs in wet surges, is key to enhanced moisture flux around the surge air mass.

  • θe/dewpoint is higher in Arizona and New Mexico before a wet surge than before a dry surge, but the local increase in moisture at surge arrival at Yuma is approximately the same.

  • Teleconnections to the tropics, the Great Plains, and southeastern United States confirms that large-scale patterns influence presurge air masses, the surge-generation process, and precipitation production in surges.

  • Surge moisture source may depend more on time than location, with the GoM contributing a large fraction of moisture early in the surge to convective systems that generate surges as well as moisture preconditioning in the southwestern United States, while local and Pacific sources contribute more after surge arrival at Yuma.

Though observations are limited across much of the NAM region, this study highlights the large range in the scale of processes important for the generation of monsoon precipitation events, including those related to surges. Synoptic-scale processes, like TEWs, seem to be particularly important not only for generating convection and the cold pool for the surge but also for postsurge moisture propagation up the GoC. Also, local moisture sources from evaporation, evapotranspiration, and precipitation recycling (Luo et al. 2007; Dominguez et al. 2008), as well as land surface moisture in eastern and central Mexico via moisture transport across the continent, may affect surge generation and characteristics. Given all these scale-dependent processes, limited observations, and necessary-but-insufficient conditions for surge-related precipitation in the NAM, additional field studies in and surrounding the complex terrain in Mexico and the United States and high-resolution regional modeling studies are needed to better understand these processes and improve predictions of heavy rainfall events in the NAM on weather and climate time scales.

Acknowledgments

This work was funded by NOAA Climate Prediction Program for the Americas (CPPA) Grant NA07OAR4310214 (program managers Jin Huang and Annarita Mariotti). Thanks to three anonymous reviewers for their help in improving the manuscript. Thanks to Drs. Alex Ruane and David Gochis for the stimulating science discussions.

REFERENCES

REFERENCES
Adams
,
D. K.
, and
A. C.
Comrie
,
1997
:
The North American monsoon
.
Bull. Amer. Meteor. Soc.
,
78
,
2197
2213
.
Adams
,
J. L.
, and
D. J.
Stensrud
,
2007
:
Impact of tropical easterly waves on the North American monsoon
.
J. Climate
,
20
,
1219
1238
.
Anderson
,
B. T.
,
J. O.
Roads
, and
S.-C.
Chen
,
2000a
:
Large-scale forcing of summertime monsoon surges over the Gulf of California and the southwestern United States
.
J. Geophys. Res.
,
105
(
D19
),
24 455
24 467
.
Anderson
,
B. T.
,
J. O.
Roads
,
S.-C.
Chen
, and
H.-M. H.
Juang
,
2000b
:
Regional simulation of the low-level monsoon winds over the Gulf of California and southwestern United States
.
J. Geophys. Res.
,
105
(
D14
),
17 955
17 969
.
Anderson
,
B. T.
,
J. O.
Roads
,
S.-C.
Chen
, and
H.-M. H.
Juang
,
2001
:
Model dynamics of summertime low-level jets over northwestern Mexico
.
J. Geophys. Res.
,
106
(
D4
),
3401
3413
.
Barlow
,
M.
,
S.
Nigam
, and
E. H.
Berbery
,
1998
:
Evolution of the North American monsoon system
.
J. Climate
,
11
,
2238
2257
.
Becker
,
E. J.
, and
E. H.
Berbery
,
2008
:
The diurnal cycle of precipitation over the North American monsoon region during the NAME 2004 field campaign
.
J. Climate
,
21
,
771
787
.
Berbery
,
E. H.
,
2001
:
Mesoscale moisture analysis of the North American monsoon
.
J. Climate
,
14
,
121
137
.
Berbery
,
E. H.
, and
M. S.
Fox-Rabinovitz
,
2003
:
Multiscale diagnosis of the North American monsoon system using a variable-resolution GCM
.
J. Climate
,
16
,
1929
1947
.
Bieda
,
S. W.
,
C. L.
Castro
,
S. L.
Mullen
,
A. C.
Comrie
, and
E.
Pytlak
,
2009
:
The relationship of transient upper-level troughs to variability of the North American monsoon system
.
J. Climate
,
22
,
4213
4227
.
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
.
Bordoni
,
S.
,
P.
Ciesielski
,
R.
Johnson
,
B.
McNoldy
, and
B.
Stevens
,
2004
:
The low-level circulation of the North American monsoon as revealed by QuikSCAT
.
Geophys. Res. Lett.
,
31
,
L10109
,
doi:10.1029/2004GL020009
.
Bosilovich
,
M. G.
,
2003
:
Numerical simulation of the large-scale North American monsoon water sources
.
J. Geophys. Res.
,
108,
8614
,
doi:10.1029/2002JD003095
.
Brenner
,
I.
,
1974
:
A surge of maritime tropical air—Gulf of California to the southwestern United States
.
Mon. Wea. Rev.
,
102
,
375
389
.
Diem
,
J. E.
,
2005
:
Northward extension of intense monsoonal activity into the southwestern United States
.
Geophys. Res. Lett.
,
32
,
L14702
,
doi:10.1029/2005GL022873
.
Diem
,
J. E.
, and
D. P.
Brown
,
2009
:
Relationships among monsoon-season circulation patterns, gulf surges, and rainfall within the lower Colorado River basin, USA
.
Theor. Appl. Climatol.
,
97
,
373
383
.
Dixon
,
P. G.
,
2005
:
Using sounding data to detect gulf surges during the North American monsoon
.
Mon. Wea. Rev.
,
133
,
3047
3052
.
Dominguez
,
F.
,
P.
Kumar
, and
E. R.
Vivoni
,
2008
:
Precipitation recycling variability and ecoclimatological stability—A study using NARR data. Part II: North American monsoon region
.
J. Climate
,
21
,
5187
5203
.
Douglas
,
A. V.
, and
P. J.
Englehart
,
2007
:
A climatological perspective of transient synoptic features during NAME 2004
.
J. Climate
,
20
,
1947
1954
.
Douglas
,
M. W.
,
1995
:
The summertime low-level jet over the Gulf of California
.
Mon. Wea. Rev.
,
123
,
2334
2347
.
Douglas
,
M. W.
, and
J. C.
Leal
,
2003
:
Summertime surges over the Gulf of California: Aspects of their climatology, mean structure, and evolution from radiosonde, NCEP reanalysis, and rainfall data
.
Wea. Forecasting
,
18
,
55
74
.
Douglas
,
M. W.
,
R. A.
Maddox
,
K.
Howard
, and
S.
Reyes
,
1993
:
The Mexican monsoon
.
J. Climate
,
6
,
1665
1677
.
Dunn
,
L.
, and
J.
Horel
,
1994a
:
Prediction of central Arizona convection. Part I: Evaluation of the NGM and eta model precipitation forecasts
.
Wea. Forecasting
,
9
,
495
507
.
Dunn
,
L.
, and
J.
Horel
,
1994b
:
Prediction of central Arizona convection. Part II: Further examination of the eta model forecasts
.
Wea. Forecasting
,
9
,
508
521
.
Finch
,
Z. O.
, and
R. H.
Johnson
,
2010
:
Observational analysis of an upper-level inverted trough during the 2004 North American Monsoon Experiment
.
Mon. Wea. Rev.
,
138
,
3540
3555
.
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
.
Gochis
,
D. J.
,
C. J.
Watts
,
J.
Garatuza-Payan
, and
J.
Cesar-Rodriguez
,
2007
:
Spatial and temporal patterns of precipitation intensity as observed by the NAME Event Rain Gauge Network from 2002 to 2004
.
J. Climate
,
20
,
1734
1750
.
Hales
,
J. E.
,
1972
:
Surges of maritime tropical air northward over the Gulf of California
.
Mon. Wea. Rev.
,
100
,
298
306
.
Higgins
,
R. W.
,
Y.
Yao
, and
X. L.
Wang
,
1997
:
Influence of the North American monsoon system on the U.S. summer precipitation regime
.
J. Climate
,
10
,
2600
2622
.
Higgins
,
R. W.
,
K. C.
Mo
, and
Y.
Yao
,
1998
:
Interannual variability of the U.S. summer precipitation regime with emphasis on the southwestern monsoon
.
J. Climate
,
11
,
2582
2606
.
Higgins
,
R. W.
,
W.
Shi
, and
C.
Hain
,
2004
:
Relationships between Gulf of California moisture surges and precipitation in the southwestern United States
.
J. Climate
,
17
,
2983
2997
.
Huffman
,
G. J.
, and
Coauthors
,
2007
:
The TRMM Multisatellite Precipitation Analysis (TMPA): Quasi-global, multiyear, combined-sensor precipitation estimates at fine scales
.
J. Hydrometeor.
,
8
,
38
55
.
Johnson
,
R. H.
,
P. E.
Ciesielski
,
B. D.
McNoldy
,
P. J.
Rogers
, and
R. K.
Taft
,
2007
:
Multiscale variability of the flow during the North American Monsoon Experiment
.
J. Climate
,
20
,
1628
1648
.
Kalnay
,
E.
, and
Coauthors
,
1996
:
The NCEP/NCAR 40-Year Reanalysis Project
.
Bull. Amer. Meteor. Soc.
,
77
,
437
472
.
Kelly
,
W. E.
, and
D. R.
Mock
,
1982
:
A diagnostic study of upper tropospheric cold lows over the western North Pacific
.
Mon. Wea. Rev.
,
110
,
471
480
.
Ladwig
,
W.
, and
D.
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
.
Luo
,
Y.
,
E. H.
Berbery
,
K. E.
Mitchell
, and
A. K.
Betts
,
2007
:
Relationships between land surface and near-surface atmospheric variables in the NCEP North American Regional Reanalysis
.
J. Hydrometeor.
,
8
,
1184
1203
.
McCollum
,
D.
,
R.
Maddox
, and
K.
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
.
Mo
,
K. C.
, and
E. H.
Berbery
,
2004
:
Low-level jets and the summer precipitation regimes over North America
.
J. Geophys. Res.
,
109
,
D06117
,
doi:10.1029/2003JD004106
.
Mo
,
K. C.
,
M.
Chelliah
,
M. L.
Carrera
,
R. W.
Higgins
, and
W.
Ebisuzaki
,
2005a
:
Atmospheric moisture transport over the United States and Mexico as evaluated in the NCEP regional reanalysis
.
J. Hydrometeor.
,
6
,
710
728
.
Mo
,
K. C.
,
J.-K.
Schemm
,
H.-M. H.
Juang
,
R. W.
Higgins
, and
Y.
Song
,
2005b
:
Impact of model resolution on the prediction of summer precipitation over the United States and Mexico
.
J. Climate
,
18
,
3910
3927
.
Mullen
,
S. L.
,
J. T.
Schmitz
, and
N. O.
Rennó
,
1998
:
Intraseasonal variability of the summer monsoon over southeast Arizona
.
Mon. Wea. Rev.
,
126
,
3016
3035
.
Nesbitt
,
S. W.
,
D. J.
Gochis
, and
T. J.
Lang
,
2008
:
The diurnal cycle of clouds and precipitation along the Sierra Madre Occidental observed during NAME-2004: Implications for warm season precipitation estimation in complex terrain
.
J. Hydrometeor.
,
9
,
728
743
.
Pytlak
,
E.
,
M.
Goering
, and
A.
Bennett
,
2005
:
Upper tropospheric troughs and their interaction with the North American monsoon. Preprints, 19th Conf. on Hydrology, San Diego, CA, Amer. Meteor. Soc., JP2.3. [Available online at http://ams.confex.com/ams/Annual2005/techprogram/paper_85393.htm.]
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
.
Rogers
,
P. J.
, and
R. H.
Johnson
,
2007
:
Analysis of the 13–14 July gulf surge event during the 2004 North American Monsoon Experiment
.
Mon. Wea. Rev.
,
135
,
3098
3117
.
Ruane
,
A. C.
,
2010
:
NARR’s atmospheric water cycle components. Part I: 20-year mean and annual interactions
.
J. Hydrometeor.
,
11
,
1205
1219
.
Saleeby
,
S. M.
, and
W. R.
Cotton
,
2004
:
Simulations of the North American monsoon system. Part I: Model analysis of the 1993 monsoon season
.
J. Climate
,
17
,
1997
2018
.
Serra
,
Y. L.
,
G. N.
Kiladis
, and
K. I.
Hodges
,
2010
:
Tracking and mean structure of easterly waves over the Intra-Americas Sea
.
J. Climate
,
23
,
4823
4840
.
Svoma
,
B. M.
,
2010
:
The influence of monsoonal gulf surges on precipitation and diurnal precipitation patterns in central Arizona
.
Wea. Forecasting
,
25
,
281
289
.
Wilks
,
D. S.
,
2006
:
Statistical Methods in the Atmospheric Sciences. International Geophysics Series, Vol. 91, 2nd ed. Academic Press, 627 pp
.
Yanai
,
M.
,
S.
Esbensen
, and
J. H.
Chu
,
1973
:
Determination of average bulk properties of tropical cloud clusters from large-scale heat and moisture budgets
.
J. Atmos. Sci.
,
30
,
611
627
.