1. Introduction
a. Background
The North American monsoon is characterized by rainfall events related to synoptic and subsynoptic atmospheric systems during summer (mid-June to September) over the southwestern United States and northwestern Mexico. Accurate forecasting of rainfall and wind fields on diurnal-to-intraseasonal time scales inside the North American monsoon domain requires correct simulations not only on continental-to-synoptic scales, but also of associated mesoscale atmospheric circulation features (Gutzler et al. 2005; Higgins and Gochis 2006). This is especially challenging in the “core monsoon” region (Fig. 1) that includes southern Arizona and New Mexico, and the coastal plains and western flank of the Sierra Madre Occidental in Sonora and Sinaloa, where the simulation of the mean monsoon and its variability is difficult because of relatively few observations and difficulties of numerical models in resolving the complex orography and coastal geometry. The special geographical configuration (Fig. 1), which includes the northwest–southeast-oriented Sierra Madre Occidental, the mountains along the Baja California peninsula, and the narrow channel of warm sea surface temperatures in the Gulf of California (hereafter “gulf” for brevity)—creates mesoscale low-level flow features and complicated rainfall patterns that prevent a simple interpretation of the North American monsoon (hereafter “monsoon”) in its entirety.
The core monsoon region, particularly within the gulf basin, often experiences an atmospheric phenomenon known as the “gulf surge” or “moisture surge.” Moisture surges (hereafter “surges” for brevity) are characterized by variations in the low-level flow within the gulf, often spanning 2–3 days, with a pronounced southeasterly wind increase, a temperature drop, and moisture and sea level pressure rises. These surges represent an important component of the transient variability of the monsoon atmospheric circulation and rainfall (Hales 1972; Brenner 1974; Adams and Comrie 1997; Higgins et al. 2004; Gochis et al. 2004), while the transient flow is as important as the time-mean flow in transporting moisture into the monsoon core region (Berbery 2001). These considerations imply that a better understanding of monsoon rainfall variability and its accurate simulation require improved documentation and knowledge of surges, especially their synoptic forcing, dynamical mechanism(s), diurnal variability, and the possible interaction of these processes.
b. Review of previous research
For the southwestern United States and northwestern Mexico, the strong dependence of intraseasonal rainfall variability on surge events was initially described by Hales (1972) and Brenner (1974) and subsequently confirmed by many other studies (e.g., Reyes et al. 1994; Stensrud et al. 1997; Fuller and Stensrud 2000; Berg et al. 2000; Douglas and Leal 2003; Higgins et al. 2004; Adams and Stensrud 2007). “Major” surges (i.e., long lived, spanning 2–3 days) and rainfall variability in the monsoon core region have been related to westward-moving tropical disturbances such as tropical easterly waves (Stensrud et al. 1997; Fuller and Stensrud 2000; Adams and Stensrud 2007), tropical cyclones (Douglas and Leal 2003; Higgins and Shi 2005), mid- to upper-tropospheric inverted troughs, and other cyclonic disturbances that can originate over the eastern coast of Mexico and Gulf of Mexico. On the other hand, surge events have been associated with rainfall patterns over the monsoon core complex terrain. For example, surge composites based on rainfall observations evolution show an enhanced northward rainfall pattern confined between the gulf coastal plains and the Sierra Madre Occidental rim (Gochis et al. 2004). Less intense surges may originate within the gulf, even in the northern gulf and over the coastal plains of Sonora (Mexico), as convective outflow from mesoscale convective systems (MCSs; Stensrud et al. 1997; Douglas and Leal 2003). The outflows are channeled northward along the gulf and often are capable of producing short-lived (6–24 h) surge-related signals or “minor” surges (Hales 1972). A compositing study by Higgins and Shi (2005) revealed that nearly half of major surge events are associated with the westward passage of tropical cyclones and their enhanced convection to the south of Baja California.
c. NAME and present study
The North American Monsoon Experiment (NAME) was a major field campaign carried out from June to September 2004 to improve prediction of warm season rainfall over the southwestern United States and northwestern Mexico on diurnal-to-intraseasonal time scales and different spatial scales (Higgins et al. 2006). It substantially enhanced the observations made in the monsoon core region (Fig. 1) during the experiment period, especially over northwestern Mexico, which lacks dense routine meteorological observations. NAME was the latest in a number of field campaigns [e.g., the multiyear SouthWest Area Monsoon Project (SWAMP-90; Douglas 1995) and (SWAMP-95; Douglas et al. 1998)] that have improved the sampling of monsoon mesoscale phenomena and their links to large-scale monsoonal patterns, and have helped to evaluate numerical simulations of those phenomena.
The NAME observations offer an opportunity to diagnose a surge life cycle in considerable detail. This paper presents the results of an observational analysis of a major 3-day surge event (12–15 July) during NAME that was associated with Tropical Storm Blas. Our study, emphasizing aircraft measurements, complements that of Rogers and Johnson (2007), who investigated the dynamical structure of this 12–15 July surge event, using the enhanced tropospheric sounding network operated during NAME. The present emphasis is on describing the surge genesis and low-level flow associated with the surge with the aid of special NAME aircraft measurements. Other observations are also used in our study to provide the spatial and time continuity needed to interpret the aircraft observations.
2. Data and methodology
a. NAME observations
The NAME measurements included special networks of pilot balloon and rawinsonde stations wind profilers, a research aircraft, two research ships, special rain gauges, radars, and additional oceanographic measurements (Higgins et al. 2006). During the main NAME observing period (1 July–14 August 2004), intensive observing periods (IOPs) were called by NAME scientists to focus on key synoptic and mesoscale features, including the monsoon onset, surges, the low-level jet (LLJ), tropical waves, and MCSs. Every IOP involved an increased frequency of rawinsonde soundings, from 1–2 day−1 during routine operations to 4–6 day−1 during IOPs depending on the station and IOP objective, and also included special aircraft missions (about 8 hours per flight). The present study focuses on the second IOP that sampled the surge of 12–15 July. Figure 2 shows the locations of the observational platforms used during this IOP.
An array of surface meteorological stations along the Mexican Pacific coast and along the gulf coastal plain (Fig. 2) were used to describe the surge movement (section 4). These stations are close to the coast, away from large topographic features (except south of the gulf entrance), and fall along a northwest–southeast transect parallel to the gulf axis (line A–B in Fig. 2). South of the gulf entrance, where the Sierra Madre Occidental approaches the coast, the airflow was found to be mainly associated with local orographic effects along the contorted coastline. Because of the large diurnal cycle in surface observations, a 24-h running mean was applied to the surface temperature, dewpoint, sea level pressure, and surface winds to better depict the synoptic-scale signal associated with the surge. Finally, anomalies of these surface values, as well as of upper air winds, were obtained by subtracting the average values for the 10-day period from 7 to 18 July from each time series. This period fully encompasses the entire surge event; the anomaly fields aid in depicting the synoptic structure of the surge.
During NAME, the National Oceanic and Atmospheric Administration (NOAA) WP-3D aircraft was used to measure the three-dimensional structure of the low-level flow during normal climatological monsoon conditions over the gulf region and adjacent oceanic areas (Mejia and Douglas 2005). A major objective for the lower gulf observational activities was to determine the mesoscale structure and development associated with surge genesis. Flights over the upper gulf better resolved the horizontal gradients associated with the LLJ than did the NAME sounding network. Flight level data with a 1-s time resolution were obtained during a series of horizontal zigzags (while porpoising from 160 m above sea level to 4600 m) from Mazatlan to the midgulf (Fig. 2), from which 10-s averages were used in the subjective aircraft analyses presented in section 5.
Across- and along-gulf winds are calculated after rotation (35° counterclockwise) of the geographic coordinate system that aligned the north–south axis with the gulf. In this rotated coordinate system, southeasterly (northwesterly) flow indicates a positive (negative) wind component υ along the gulf axis. Likewise, onshore (offshore) gulf coastal plain flow has a positive (negative) wind component u perpendicular to the gulf axis.
b. Other data sources
The Geostationary Operational Environmental Satellite-12 (GOES-12) infrared satellite imagery at 30-min and 4-km grid spacing were obtained from the Comprehensive Large Array-data Stewardship System (CLASS) data server (available online at http://www.class.ncdc.noaa.gov/) to sample MCSs during the surge lifetime. Here, we define a MCS as a persistent cloud cluster structure that stands out in the IR satellite imagery as lower brightness temperature features (light colors), even after applying a 6-h running mean (which removes scattered, less organized convective clouds) to 30-min imagery.
3. Synoptic and mesoscale environments
Nine IOPs were undertaken during NAME to observe different features related to the moisture flux field over and adjacent to the NAME domain (Fig. 1). During IOP-2, a well-defined tropical easterly wave propagated westward across Central Mexico on 11 July, after which it moved over the eastern Pacific (Higgins et al. 2006; Rogers and Johnson 2007). At 1500 UTC 12 July, the system was declared a tropical depression by the National Hurricane Center (at 14.8°N, 105.8°W, Fig. 2), after which it became Tropical Storm Blas from 0000 UTC 13 July (16.4°N, 107.9°W) to 1200 UTC 14 July (22.10°N, 116.6°W), when it was downgraded to a tropical depression (Figs. 2 and 4).
Figure 3 shows the evolution of the along-gulf wind anomalies for upper-air stations aligned north–south along the gulf coastal plain. In general, all coastal stations show a single distinctive signal in the wind and (not shown) temperature anomaly fields that lasted 3–4 days. In particular, the Guaymas surge signal (Fig. 3c) resembles the Guaymas radiosonde-based composites from Douglas and Leal (2003, their Fig. 8). This surge-related signal extended from early 12 July to early 15 July at Los Mochis (extreme south), and from early 13 July to late 15 July at Puerto Peñasco (far north), where it was strongest and especially well defined. Except for Los Mochis, the southeasterly wind anomalies are stronger below the Sierra Madre Occidental average height (2500 m or ~750 hPa; Fig. 1), suggesting that the surge progressed northward confined by the topography, such as with a coastal-trapped disturbance (Mass and Albright 1987; Nuss et al. 2000). Moderate-to-weak southeasterly wind anomalies also occurred between 700 and 500 hPa (Fig. 3), associated with a midtropospheric trough that was a northeastward extension of Tropical Storm Blas. Climatological studies for Guaymas (midgulf, Fig. 2) have shown that the strongest wind anomaly during surge events is at ~950 hPa (Douglas and Leal 2003). For this particular surge event, the strongest southeasterly winds (18 m s−1, not shown) and largest southeasterly wind anomalies (9 m s−1) observed at Puerto Peñasco are associated with the LLJ observed over the far northern gulf (Rogers and Johnson 2007).
The 200-hPa velocity potential Madden–Julian oscillation (MJO) composite index from the NOAA/Climate Prediction Center indicates that the 12–15 July surge coincided with an active MJO event (Johnson et al. 2007). Higgins and Shi (2001) and Lorenz and Hartmann (2006) have suggested that some enhanced rainfall events in the NAME core region tend to occur during the MJO active phase. Possible surge associations with this low-frequency tropical MJO disturbance mode are intriguing but not within the scope in this paper.
The 12–15 July surge was associated with the development of two MCSs over northwestern Mexico. A 6-h mean infrared imagery composite (Fig. 4) shows that the first MCS (MCS1) emerged over the Nayarit/Sinaloa coast 350 km to the south of the gulf entrance at 0600 UTC 12 July and moved offshore and dissipated by 1800 UTC. A second MCS (MCS2) developed over the central gulf coastal plain after 0000 UTC 13 July, dissipating by 1200 UTC.
4. Surface analysis for the 12–15 July moisture surge
The routine and special surface observations during NAME are useful in describing the movement of the surge up the gulf. This is important for helping to place into context the research aircraft observations described in the next section.
Surface observations from Yuma, Arizona, clearly reflect the 12–15 July surge (Fig. 5a), but they also reveal that the diurnal variation of surface quantities often is as large as the variability induced by the surge. Therefore, as explained in section 2a, the diurnal cycle and 7–18 July mean were removed from each station’s data to facilitate interstation comparisons. Figure 5b illustrates the effect of applying this filtering procedure to the sea level pressure (SLP), temperature (T), dewpoint temperature (Td), and across- and along-gulf wind components (u, υ). The resulting anomalies for these parameters henceforth are indicated by primes. At Yuma, the 3–4-day changes observed in SLP′ (~7 hPa), Td′ (~9 K), T ′ (~4 K), and u′ and υ′ (from northwesterly to southeasterly) indicate the persistence and magnitude of this surge. However, observations from surface stations farther south along the gulf (e.g., Los Mochis, Figs. 5c,d) can differ from the surge characteristics illustrated by Yuma. Interestingly, while the amplitude of SLP′, u′, and υ′ at Los Mochis are as large as those at Yuma, the amplitude of Td′ (~1.5 K) and T ′ (~2 K) are relatively small.
Figure 6 shows the progression of surge-related surface anomalies for stations along the Mexican Pacific coast and gulf coastal plain. The presurge conditions include negative SLP′ (~−2 hPa) along the southern gulf and southwestern Mexico that became slightly larger (~−3 hPa) over the northern gulf and southwestern United States (Fig. 6a). To the south of the gulf, the Td′ decreases little—by about 1 K over the surge period. Over the gulf coastal plain and northern gulf, a relatively warm (T ′ > 0) and dry (Td′ < 0) surface environment prevailed before the 13 July surge onset, associated with weak northwesterly wind anomalies (υ′ < 0; Figs. 6b–d). This presurge pattern along the gulf agrees well with presurge conditions found in the compositing studies of Douglas and Leal (2003, their Fig. 8) and Higgins and Shi (2005, their Fig. 3).
The most important feature in Fig. 6 is the rather smooth up-gulf progression of the surge, compared to the sudden changes usually observed in gravity current or convective outflows (Koch et al. 1991; Haertel et al. 2001). A number of additional important aspects of Fig. 6 now are described in more detail.
a. SLP′
This time sequence shows a relatively up-gulf progression of positive SLP′ with its magnitude increasing northward (Fig. 6a). Over the northern gulf and southern Arizona, SLP′ increases by nearly 7.5 hPa (from −3 hPa before surge onset to +4.5 hPa around 0000 UTC 15 July) compared with a 5-hPa increase at the gulf entrance. The SLP′ increase at the gulf entrance started at ~1000 UTC 12 July and reached the northern end of the gulf at ~0200 UTC 13 July, which implies an up-gulf velocity of 18 m s−1. Rogers and Johnson (2007) inferred the speed of the surge from high time-resolution wind profiler data to be 17 m s−1 between Los Mochis and Kino Bay, and 22 m s−1 from Kino Bay to Puerto Peñasco (see Fig. 2 for locations). By way of comparison, the motion of Tropical Storm Blas averaged only 10 m s−1 toward the northwest.
b. T′
At the gulf entrance, cold anomalies are observed just before the surge formation (at 1000 UTC 12 July), probably associated with convective outflows from previous storms as indicated by satellite IR images (Fig. 4). Marked cooling is observed along the entire gulf, with changes of 4–5 K, associated with enhanced up-gulf winds, shown in Fig. 6e.
c. Td′
In contrast with the changes in T ′ and SLP′, the central and southern gulf experienced relatively small changes in Td during the surge passage. However, the northern gulf and surrounding low desert experienced impressive 24-h Td′ increases of ~10 K.
d. u′ and υ′
The surge progression is more evident in υ′ than u′ (Figs. 6d,e). The υ′ increase coincides with SLP′ increases and T ′ decreases. Over the lower gulf, the southeasterly wind increase is coincident with the organized convection offshore of Puerto Vallarta between 0600–1000 UTC 12 July (Fig. 4). Over the northern gulf, the largest υ′ increases are observed at Puerto Peñasco between 0800 UTC 13 July and 0000 UTC 14 July. Although the changes in cross-gulf wind, u′, are smaller than those along-gulf (Fig. 6d), they clearly show the surge’s synoptic variation, with a weak easterly wind anomaly after surge passage. Westerly anomalies are evident both before and well-after the surge passage, coinciding with anomalous down-gulf flow.
5. Aircraft observations of the moisture surge leading edge
Aircraft data analyses (Figs. 7, 8, and 10) provide an offshore “snapshot” perspective of the surge formation. They complement, with much greater spatial detail and less effect from local surface forcing, the previous analyses based on data from surface stations and other platforms such as wind profilers (Rogers and Johnson 2007). Figure 7 shows that relatively cool, moist air was associated with stronger southeasterly 950-hPa winds over the gulf entrance. However, the vertical structure was complex and included two features associated with the surge leading edge (Fig. 8).
The most apparent surge feature consists of strong potential temperature and wind speed gradients near the gulf entrance between the surface and the 302.5-K isentrope (represented by the line S1 in Figs. 7 and 8) that resemble a convective outflow (Simpson 1987; Haertel et al. 2001). The second feature is a more gradual and deeper (750–950 hPa) potential temperature and mixing ratio perturbation that decreases downwind of the 900-hPa S1 feature (S2 line in Fig. 8). This structure also is manifest in uplifted isentropes within and over the marine boundary layer stable layer, a wind-direction shift from easterly on the warm side of S2 to southeasterly on the cool side, and a subsequent increase in magnitude of the southeasterly flow behind S2.
a. Structure 1 at gulf entrance
Some convective outflows have been characterized as gravity currents (Klemp et al. 1994; Haertel et al. 2001), driven by density gradients and involve advection of dense, colder air behind the propagating boundary. Convective storms observed between 0600 and 1800 UTC 12 July offshore of Nayarit (Figs. 1 and 4) were associated with S1, and the movement of radar echoes evident from the NAME radar network permitted us to track the speed of this convective outflow (Fig. 9). Additionally, the radar-observed rainfall echoes at 1415 UTC 12 July are collocated with S1. Animation of the high-reflectivity features on the radar mosaic using the maximum (15 min) radar time resolution (not shown), together with the assumption that convection occurred near the leading edge of S1, gave a speed of the convective outflow of ~14 m s−1. Consistent with this result, sustained winds of 18–20 m s−1 were recorded at the S1 front when the observed mean background flow ahead of S1 was 5–6 m s−1 (at 500 m), which also suggests a speed of 13–14 m s−1.
There are several problems with assuming that S1 was the main factor producing the surge signal along the gulf (Figs. 7 and 10). The filtered surge signal in Fig. 6 shows that the wind, temperature, and SLP vary gradually, and that the speed of the surge was somewhat faster (18 m s−1, section 4) than would be expected from a convective outflow (~13–14 m s−1, above). In addition, Zehnder (2004) noted that the horizontal extent of a convective outflow is constrained by geostrophic adjustment with a time scale of 1/fo (~4 h in this case), where fo is the Coriolis parameter. The S1 already shows weakening of its wind and temperature features by the time of the aircraft’s return legs 6.5 h later (Fig. 10). Furthermore, the expected hydrostatic pressure perturbation associated with this type of disturbance can be estimated as 
b. Structure 2 at gulf entrance
The elevated second feature (S2) extended about 300 km downwind (at 900 hPa) of S1. Ahead of S2, the winds were easterly-northeasterly and nearly calm (1–2 m s−1), whereas beneath S2 the flow was southeasterly with a modest speed increase (Fig. 8). At 850 hPa, ahead of (behind) this feature the mixing ratio was 10 g kg−1 (13 g kg−1) and potential temperature was 311 K (307 K). The vertical structure and evolution of S2 are documented further using Los Mochis rawinsonde soundings and aircraft flight level soundings (at locations in Figs. 7a,b), respectively (Fig. 10). The Los Mochis soundings (Fig. 10a) show low-level warming between 1100 and 1800 UTC 12 July, likely due to surface heating and mixing, while the aircraft soundings over the gulf at ~1500 UTC 12 July (Fig. 10b) show unperturbed conditions below the marine boundary layer capping stable layer. Figure 10 also shows that the perturbed flow, which experiences abrupt cooling, extends from the top of the marine boundary layer to ~2500 m MSL (750 hPa), at least over the gulf.
Figure 11 shows streamline analyses for different pressure levels using wind soundings from different platforms (Fig. 2) during the 12 July flight. Both S1 and S2 are superposed in Fig. 10 at the same location shown in Fig. 8. In these streamline analyses, S2 appears as an enhanced southeasterly surgelike feature over the midgulf and is confined to below 700 hPa. In the cross-gulf direction, at 850 hPa, S2 is less defined farther offshore from the Mexican mainland and its length scale (~400–500 km), measured normal to the Sierra Madre Occidental, is comparable to that predicted by the Rossby radius of deformation (~300 km). Figures 10b,c also show that the effect of S2 was felt over the western side of the southern Baja California Peninsula. Additionally, positive sea level pressure anomalies over this region were also associated with the surge passage (Rogers and Johnson 2007). These features highlight the importance of understanding the role of the elevated terrain and different multiscale atmospheric disturbances along the Baja California peninsula on the surge evolution (e.g., its diurnal circulations, relatively strong vertical turbulent transport, and channeling and low-level horizontal confinement of the surge). Although the observations are not definitive, the synoptic-scale component of the trough extending northeastward from Tropical Storm Blas (Fig. 11f) may be associated with S2 as the trough interacts with topography.
6. The moisture surge: Mature phase
The IOP-2 involved two consecutive research flights. The flight on 13 July sampled a mature surge, showing characteristics previously identified as characteristic of such events. In this section we focus on characteristics of the mature surge that were uniquely sampled by the flight on 13 July. Coupled with the observations on the previous day, these provide a unique description of changes over the gulf associated with this strong surge.
The 13 July flight extended to the northern extreme of the gulf and included segments over the surrounding coastal regions near midgulf and at the northern end of the gulf (Fig. 12d). The flight segments were intended to cut the low-level flow at multiple locations to provide cross-section information along the gulf to help identify along-gulf changes of the surge. The overland segments were carried out to determine how the low-level flow changed with distance inland. The midgulf flight segment trending east–west was duplicated on the return, to help identify any time changes during the flight. This segment was common to five other NAME flights.
a. Horizontal structure of the flow on 13 July
The lowest-level flows on 13 July (Fig. 12a) were clearly up-gulf over the mid- and northern gulf, with the flow shifting to easterly and even northeasterly over the gulf entrance region. This probably reflects the influence of Tropical Storm Blas (Figs. 2 and 13 at 1200–1800 UTC 13 July); the fact that flow over most of the gulf is away from Blas indicates that the direct influence of Blas is limited. Figure 12a also reveals that the coastal terrain in the western gulf blocks some of the offshore low-level flow over the gulf’s eastern coast. On the other hand, strongest winds are over the northernmost gulf, but the up-gulf components increase over the entire gulf. The strongest winds are closer to the Baja California Peninsula over the central gulf, but in the northern gulf they are strongest over the middle, eastern gulf (Fig. 13). This is most apparent from vertical cross sections (Fig. 14). Strongest winds are near 975 hPa, though some variability is seen between individual aircraft soundings and the vertical shear is quite weak east of the jet core. This may be due to daytime heating of the overland boundary layer since the aircraft traversed this region about 4 h after sunrise.
More apparent from Fig. 14c is the marked decrease in the capping stable layer as the aircraft flew westward. The weak stability overland confirms the likely reason for the weak vertical shear below 850 hPa, with strongest stability over the western gulf. The origin of the very warm air over the western gulf between 950 and 850 hPa is unclear, but it appears from Fig. 12b to be related to cross-peninsula advection from the southwest—from air moving around the southern end of high terrain in northern Baja California. This warm air over the extreme northwestern gulf appears in the North American Regional Reanalysis (NARR) mean fields for July–August 2004 (not shown) and also in mean 4-km WRF runs for the NAME period, so the feature may be more than a transient.
The impact of surges on subsequent rainfall over the region of the northern gulf and low deserts of northwestern Mexico and the southwestern United States is likely to be associated with enhancement of moisture fluxes during the surge. The aircraft data allow for description of both the horizontal and vertical variations in the moisture flux; Fig. 15 shows three cross-gulf slices of along-gulf kinematic moisture flux. Two features are apparent: 1) the up-gulf flux is much larger over the northern gulf and 2) substantial values of moisture flux are only found in the boundary layer. The latter is a direct result of both the shallow moist layer (approximately a few hundred meters) and shallow low-level jet (level of maximum winds within 500 m of surface). With winds and moisture rapidly decreasing above the over-gulf stable layer, the moisture flux values decrease to 10%–20% their near-surface values by 850 hPa. The challenge in accurately estimating this flux for routine forecasting or climate monitoring applications should be apparent.
7. Summary
This study has provided an observational description of a Gulf of California moisture surge associated with a synoptic-scale tropical disturbance (Tropical Storm Blas), from a synthesis of enhanced surface, upper-air, aircraft, radar, and satellite observations obtained during NAME. The analyses focused on the aircraft-observed structure of the surge on 12–13 July, including the apparent surge genesis on 12 July and its subsequent establishment over the central and northern gulf on 13 July. This surge represents a clear example of an event associated with a tropical disturbance over the eastern Pacific, which ultimately modulated the rainfall over northern Mexico and the southwestern United States. The general characteristics of this surge resemble the mean surge patterns described in other studies, such as Douglas and Leal (2003).
Composites of observed surface perturbations along the gulf suggested that the surge major signal moved up the gulf at 18 m s−1, which compares well with estimates obtained from wind profiler observations (Rogers and Johnson 2007) of 17 m s−1 between Los Mochis and Kino Bay and 22 m s−1 between Kino Bay and Puerto Peñasco. However, the above analyses of aircraft observations (integrated with radar and upper-air soundings) over the gulf and adjacent Pacific Ocean were critical for describing and interpreting two mesoscale features associated with the surge’s leading edge during its early stages. The first mesoscale feature consisted of a convective outflow that moved at 12–14 m s−1 and was associated with MCS outflows over the southern gulf. It was established that this feature was incapable of producing the major surge signal. The second mesoscale feature consisted of a disturbance within and above the marine inversion, and downwind of the aforementioned convective outflow, the timing and vertical structure of which resembled key aspects of the surge signature.
The aircraft observations on 13 July were of value in showing the increase in winds northward along the gulf and the position of strongest winds at the head of the gulf and over the ocean. The vertical profiling of the aircraft yielded detailed cross sections at multiple locations, showing that the up-gulf moisture flux was very shallow, essentially below 850 hPa, and largest at the extreme northern end of the gulf. Over the southern gulf the flux was actually down-gulf, associated with the proximity of Tropical Storm Blas.
The observations and analysis presented here are part of a larger research effort into the role of mesoscale convective systems in the genesis of surges and their modulation by synoptic-scale forcing. Some of these results will be reported separately. The present results, when combined with other studies of the NAME surge events, add to the basic understanding of surge structure that is needed to evaluate the fidelity of high-resolution mesoscale models now being used to simulate the North American monsoon environment.
Acknowledgments
This research was sponsored by the NOAA’s Climate Program Office. Many Mexican and U.S. organizations and individuals contributed to the NAME data collection effort. Useful discussions with Joseph Zehnder of Creighton University are greatly appreciated. Most NAME datasets used in this study were provided by NCAR/EOL under sponsorship of the National Science Foundation. We thank David Gochis, R. H. Johnson, and two anonymous reviewers for their thorough reviews and highly appreciate the comments and suggestions, which significantly contributed to improving the quality of this paper.
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.
Berbery, E. H. , 2001: Mesoscale moisture analysis of the North American Monsoon. J. Climate, 14 , 121–137.
Berg, W. K. , D. M. Anderson , and J. J. Bates , 2000: Satellite observations of a Pacific moisture surge associated with flooding in Las Vegas. Geophys. Res. Lett., 27 , 2553–2556.
Brenner, I. S. , 1974: A surge of maritime tropical air—Gulf of California to the southwestern United States. Mon. Wea. Rev., 102 , 375–389.
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. , A. Valdez-Manzanilla , and R. G. Cueto , 1998: Diurnal variation and horizontal extent of the low-level jet over the northern Gulf of California. Mon. Wea. Rev., 126 , 2017–2025.
Fuller, R. D. , and D. J. Stensrud , 2000: The relationship between easterly waves and surges over the Gulf of California during the North American Monsoon. Mon. Wea. Rev., 128 , 2983–2989.
Gochis, D. J. , A. Jimenez , C. J. Watts , W. J. Shuttleworth , and J. Garatuza-Payan , 2004: Analysis of 2002 and 2003 warm-season precipitation from the North American Monsoon Experiment (NAME) Event Rain Gauge Network (NERN). Mon. Wea. Rev., 132 , 2938–2953.
Gutzler, D. S. , and Coauthors , 2005: The North American Monsoon model assessment project: Integrating numerical modeling into a field-based process study. Bull. Amer. Meteor. Soc., 86 , 1423–1429.
Haertel, P. T. , R. H. Johnson , and S. N. Tulich , 2001: Some simple simulations of thunderstorm outflows. J. Atmos. Sci., 58 , 504–516.
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.
Higgins, R. W. , and W. Shi , 2005: Relationships between Gulf of California moisture surges and tropical cyclones in the eastern Pacific basin. J. Climate, 18 , 4601–4620.
Higgins, R. W. , and D. J. Gochis , 2006: Multi-scale interactions during the North American Monsoon. Preprints, Symp. on Connections between Mesoscale Processes and Climate Variability, San Antonio, TX, Amer. Meteor. Soc., 1.2.
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.
Higgins, R. W. , and Coauthors , 2006: The NAME 2004 field campaign and modeling strategy. Bull. Amer. Meteor. Soc., 87 , 79–94.
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.
Klemp, J. B. , R. Rotunno , and W. C. Skamarock , 1994: On the dynamics of gravity currents in a channel. J. Fluid Mech., 269 , 169–198.
Koch, S. E. , P. B. Dorian , R. Ferrare , S. Melfi , W. C. Skillman , and D. Whiteman , 1991: Structure of an internal bore and dissipating gravity current as revealed by Raman lidar. Mon. Wea. Rev., 119 , 857–887.
Lorenz, D. J. , and D. L. Hartmann , 2006: The effect of the MJO on the North American Monsoon. J. Climate, 19 , 333–343.
Mass, C. F. , and M. D. Albright , 1987: Coastal southerlies and alongshore surges of the west coast of North America: Evidence of mesoscale topographically trapped response to synoptic forcing. Mon. Wea. Rev., 115 , 1707–1738.
Mejia, J. F. , and M. W. Douglas , 2005: Mean structure and variability of the low-level jet across the central Gulf of California from NOAA WP-3D flight level observations during the North American Monsoon Experiment. Preprints, Sixth Conf. on Coastal Atmospheric and Oceanic Prediction and Processes (6COASTAL), San Diego, CA, Amer. Meteor. Soc., J5.8.
Nuss, W. A. , and Coauthors , 2000: Coastally trapped wind reversals: Progress toward understanding. Bull. Amer. Meteor. Soc., 81 , 719–743.
Reyes, S. , M. W. Douglas , and R. A. Maddox , 1994: El Monzón del suroeste de Norteamérica (TRAVASON/SWAMP). Atmósfera, 7 , 117–137.
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.
Simpson, J. E. , 1987: Gravity Currents: In the Environment and the Laboratory. John Wiley & Sons, 244 pp.
Stensrud, D. J. , R. L. Gall , and M. K. Nordquist , 1997: Surges over the Gulf of California during the Mexico monsoon. Mon. Wea. Rev., 125 , 417–437.
Zehnder, J. A. , 2004: Dynamic mechanisms of the gulf surge. J. Geophys. Res., 109 , D10107. doi:10.1029/2004JD004616.
North American monsoon core domain and topography and locations mentioned in text. Darker colors represent higher elevations. Sierra Madre Occidental crest is ~3000 m MSL.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Distribution of key instrument platforms (identified in legend) during NAME IOP 2 (12–15 Jul 2004). The broken line is the alongshore transect A–B used for some analyses (Fig. 6), and the solid line provides an example of a typical WP-3D flight track. Also shown is the location of Tropical Storm Blas, which was associated with the major surge event.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Along-gulf time–height wind anomalies (m s−1) for 10–18 Jul 2004 using rawinsonde data from (a) Puerto Peñasco, (b) Kino Bay, (c) Guaymas, and (d) Los Mochis. Stations are oriented north–south (see Fig. 2). Diurnal variability was removed by obtaining 24-h running means from 4- and 6-hourly rawinsonde data. See text for details.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Time sequence of 6-h mean infrared satellite imagery (GOES-12) from 0000 UTC 12 Jul to 1800 UTC 14 Jul 2004. The tropical depression (“L”) that later became Tropical Storm Blas moved west-northwest immediately to the south of the gulf. Note the organized convection (light colors) developing along the gulf during the night and early morning (0000–1200 UTC) of each day, especially the MCSs (discussed at the end of section 3) that developed between 0600 and 1400 UTC 12 July (MCS1) and 0600 UTC 13 July (MCS2). The geographical axes are rotated clockwise (35°) for display purposes.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
(a) Surface station meteogram and (b) anomaly values for Yuma for 11–17 Jul 2004, with hourly sampling frequency. (c),(d) As in (a),(b), but for Los Mochis with 10-min sampling frequency. Anomalies are calculated with respect to mean quantities for 10–17 Jul 2004 after removing diurnal cycles using 24-h running means (see section 2a). Black circles indicate sea level pressure (left ordinate), and orange/blue gives surface temperature/dewpoint temperature (right ordinate). Wind barbs are plotted in the standard coordinate system every 3 h to avoid cluttering, where half a barb indicates a 2.5 m s−1 wind speed, a full barb indicates 5 m s−1, and staffs show actual/anomaly wind directions. Wind speed anomalies are amplified by a factor of 5. Pairs of solid green vertical lines enclose the NOAA WP-3D missions for time reference. Broken vertical lines indicate 0000 UTC for each day.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Time (10–17 Jul 2004) vs south–north gulf alongshore distance (transect A–B, Fig. 2) for surface anomalies of (a) sea level pressure (SLP′ in hPa), (b) temperature (T ′ in K), (c) dewpoint temperature (Td′ in K), (d) cross-gulf surface wind (u′ in m s−1), and (e) along-gulf surface wind (υ′ in m s−1). Shaded (dashed) contours indicate positive (negative) perturbations. Heavy dotted line shows surge formation based on its pressure signal as it progresses northward along the gulf coastal plains. Vertical solid line shows position of gulf entrance. Vertical dashed lines show the position of Puerto Vallarta and Puerto Peñasco (Fig. 2).
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Temperature, mixing ratio, and wind analyses at 950 hPa for NOAA WP-3D flight on 12 Jul 2004. Isentropes are red solid lines (K) and mixing ratio is given by blue broken lines (g kg−1). Wind barbs (half barb, 2.5 m s−1; full barb, 5 m s−1) are plotted along the flight track (thin solid black line) every time the aircraft crossed the 950-hPa level (within ±3 hPa). Bold broken black line locates the vertical cross section analyzed in Fig. 8. Letters A and B give locations of soundings (flight level observations) “ahead” and “behind” in Fig. 11b. The flight started from Mazatlan at 1330 UTC, flying the across-gulf legs first, and ended after the Pacific Ocean transects at 2000 UTC (Fig. 2). Bold black solid line shows S1 (convective outflow).
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Along-gulf vertical cross section for transect shown in Fig. 7, based on NOAA WP-3D observations on 12 Jul 2004. The soundings included were made within 20 km of the transect line between 1330 and 1630 UTC. Isentropes are red solid lines (K) and mixing ratio is given by blue broken lines (g kg−1). Wind barbs (half barb, 2.5 m s−1; full barb, 5 m s−1) are colored coded to indicate speed (increases from blue to red). Bold black solid lines show S1 (convective outflow) and S2 (level of directional wind shear).
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Gridded (2 km) composite of ≥45 dBZ near-surface radar reflectivity echoes from the NAME radar network (Fig. 2). Broken circles are 200-km range limits for San Lucas, Los Mochis, and S-band dual-polarization Doppler radar (S-Pol). Evolution of radar reflectivity is shown from lighter (0700 UTC) to darker (1600 UTC) gray–black progression. Solid lines indicate the hourly position of the leading edge of the convective outflow.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
(a) Potential temperature θ profiles from Los Mochis radiosonde before (1100 UTC, broken line) and after (1800 UTC, solid line) the surge passage on 12 Jul 2004. (b) NOAA WP-3D flight level θ profiles made behind (solid line) and ahead (dashed line) of S2 in Fig. 8, obtained at ~1500–1600 UTC 12 Jul 2004 (sounding locations in Fig. 7).
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Streamline analyses using the 1330–2000 UTC 12 Jul 2004 WP-3D winds and in situ soundings (rawinsonde, wind profiler, pibal) made within ±2 h of flight time for (a) 950, (b) 900, (c) 850, (d) 800, (e) 750, and (f) 700 hPa. Flight strategy is described in caption of Fig. 7. Wind barbs follow the standard convention (half barb, 2.5 m s−1; full barb, 5 m s−1). Solid/broken streamlines show early (1300–1700 UTC)/late (1700–2000 UTC) flow structures. Broken thick lines indicate locations of S1 and S2 shown in Fig. 8. Shading indicates topography above the standard atmospheric height for the pressure level in (a)–(f).
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Streamline analyses using 1300–2100 UTC 13 July 2004 WP-3D winds and in situ soundings (rawinsonde, wind profiler, pibal) made within ±2 h of flight time frame for (a) 950, (b) 850, and (c) 750 hPa. (d) Terrain and flight track and timing is displayed. Wind barbs follow the standard convention (half barb, 2.5 m s−1; full barb, 5 m s−1). NARR winds helped provide continuity over the fringes of the flight track, mainly to the east of the SMO and over the eastern Pacific. Shading in (a) and (b) indicates topography above the standard atmospheric height for the pressure level in (a)–(d). Letters A to G indicate the locations of end points for vertical cross sections in Figs. 14 and 15.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Northern gulf 950-hPa streamline and isotach (m s−1) analyses using 1300–2100 UTC 13 Jul 2004 WP-3D winds and in situ soundings (rawinsonde, wind profiler, pibal). Wind barbs follow the standard convention (half barb, 2.5 m s−1; full barb, 5 m s−1).
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Northern gulf vertical cross section for transects (left) A–B and (right) D–C shown in Fig. 12d, based on NOAA WP-3D observations on 13 July. (a) Wind barbs (half barb, 2.5 m s−1; full barb, 5 m s−1) and isotach analysis (m s−1), (b) temperature and isotherm analysis (°C), and (c) potential temperature and isentrope analysis (K). Observations are color coded with values increasing from blue to red. Observations are shown in 20-s averages of 1-s data along the vertical saw-tooth flight path. Terrain is displayed by thick solid line at bottom of cross sections, where the gulf surface is at 1010 hPa.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
Kinematic moisture flux (m s−1 g kg−1) vertical cross-sectional analyses for transects (top) A–B (northern gulf), (middle) E–F (central gulf), and (bottom) G–F (south-central gulf) shown in Fig. 12d, based on NOAA WP-3D observations on 13 Jul 2004. Observations are colored coded with values increasing from blue to red. Observations are shown in 20-s averages of 1-s data along the vertical sawtooth flight path. Terrain is displayed by thick solid line at bottom of the cross sections, where the gulf surface is at 1010 hPa.
Citation: Monthly Weather Review 138, 9; 10.1175/2010MWR3228.1
