1. This work was funded by NSF grant ATM-0340602, NOAA grant NA17RJ1228, and by a one-year AMS Graduate Fellowship. For further information, please visit http://tornado.atmos.colostate.edu/name/ or email progers@atmos.colostate.edu. An Observational Analysis of The 13 July Gulf Surge Event During The 2004 North American Monsoon Experiment Peter J. Rogers, Richard H. Johnson, Paul E. Ciesielski, and Brian D. McNoldy Department of Atmospheric Science, Colorado State University, Fort Collins, CO INTRODUCTION OBSERVATIONAL RESULTS AND DISCUSSION DYNAMICAL MECHANISMS DATA AND METHODS CONCLUSIONS G ulf surges are northward propagating disturbances along the Gulf of California (GOC) that advect large quantities of cool, moisture-laden air from the GOC or eastern tropical Pacific Ocean into the low deserts of the southwest United States and northwest Mexico during the North American Monsoon (NAM). Previous attempts to understand the underlying dynamical mechanisms for these surges have been hampered by a lack of data in this region. High temporal and spatial observational measurements collected during the 2004 North American Monsoon Experiment field campaign (1 July – 15 August) are used to describe the structure and probable dynamical mechanisms of two gulf surge events, one of which is discussed in this presentation. Of greatest interest was the deployment of three National Center for Atmospheric Research (NCAR) Integrated Sounding Systems (ISSs) along the east coast of the GOC, in addition to a relatively dense network of rawinsonde sites across the NAM geographical domain. Synoptic circulation patterns (subtropical ridges, tropical cyclones, mid-tropospheric easterly waves, and inverted troughs) can enhance the development of NAM-related convection. Preceded by strong anomalous warming, the gulf surge exhibited cooling, moistening, and increased south-southeasterly flow from the surface up to approximately 800 hPa. Surge passage (in the early morning) was accompanied by a rapid surface pressure rise. Surface winds shifted to the southeast and strengthened slightly. Gulf surge was amplified by the nocturnal northern GOC LLJ. Surge propagation speed along the northern GOC was approximately 20.0 m s -1. Observations suggest a nonlinear Kelvin wave-like disturbance with atmospheric bore or solitary wave-like characteristics along its leading edge. Figure 3: Infrared GOES-10 Satellite imagery at (a) 0146 UTC 13 July and (b) 0800 UTC 13 July. Colored contours represent cloud top temperatures (°C). Figure 2: T2A gridded dataset 700 (a) and 200 hPa (b) heights (10 m intervals) (black solid), temperatures (2°C intervals) (red dotted), mixing ratios (g kg -1 ) (colored contours), and wind vectors (m s -1 ) at 1200 UTC from t = -2 d (11 July) to t = +2 d (15 July) where t = 0 (13 July) corresponds to day of gulf surge onset at Puerto Peñasco. White regions represent areas below the surface. Figure 1: 2004 NAME rawinsonde sites. Site color code refers to number of launches per day during the extended observing period (EOP) and intensive observing periods (IOP). Red arrows denote ISS sites. Colored contours represent altitude above sea level (m). (a) (b)(a)(b) Figure 4: Puerto Peñasco (a), Bahia Kino (b), and Los Mochis (c) wind profiler and surface data from 0000 UTC 11 July – 0000 UTC 16 July. Wind speed (colored contours) (m s -1 ) is plotted every half hour and wind barbs every two hours. One full barb equals 5 m s -1. Surface temperature (red) (°C), dewpoint temperature (blue) (°C), and pressure (green) (hPa) are averaged over and plotted every half hour. Figure 5: Half-hour interpolated rawinsonde potential temperature and moisture flux (meridional wind times mixing ratio) anomalies at Yuma (a), Puerto Peñasco (b), Bahia Kino (c), Empalme (d), Los Mochis (e), and Mazatlan (f) from 0000 UTC 10 July to 0000 UTC 17 July. Anomalies are calculated using 5 July-16 August means. Colored contours represent potential temperature (K). Solid (dotted) thin black curves represent positive (negative) moisture flux (m g s -1 kg -1 ). Thick solid black curve equals zero, with each successive contour at (  ) 25 m g s -1 kg -1 intervals. Figure 7: Puerto Peñasco rawinsonde temperature (red) (°C), dewpoint temperature (green) (°C), and wind (1 full barb = 5 m s -1 ) profiles at 0000, 0600, 1200, and 1800 UTC 13 July. Black dotted curve represents reference dry adiabat. Figure 6: Surface pressure (black) (hPa) and temperature (red) (°C) anomalies at Puerto Peñasco (a) and Bahia Kino (b) for 13 July. Anomalies are calculated using 0000 UTC 7 July – 2359 UTC 14 August means. Panels (c) and (d) are three hour subsets surrounding the time of gulf surge onset at Bahia Kino and Puerto Peñasco, respectively as noted by the dotted vertical curves in panels (a) and (b). T he NCAR ISS sites were located at Puerto Peñasco, Bahia Kino, and Los Mochis (Fig. 1), and consisted of a Vaisala GPS sounding system, an enhanced surface observing station, a 915 MHz Doppler clear-air wind profiling radar, and a radio acoustic sounding system (not used in this study). Rawinsonde data from these sites and numerous others (Fig. 1) were objectively analyzed onto a 1°x1° horizontal grid with 25 hPa vertical resolution two (four) times per day over the T2A (T1A) domain of 90°-120°W and 15°-40°N (100°-115°W and 22°-35°N). S trong evaporational cooling associated with convective rainfall over the SMO foothills and GOC coastal plain aid in the development of a surface mesohigh (Fig. 8a). The anticyclonic flow around this system and Coriolis deflection piles cooler air along the north-south barrier of the SMO, inducing south-southeasterly flow and a linear Kelvin wave-like disturbance that propagates north. As the surge moves north, it develops nonlinearities along its leading edge due to the greater depth of the surge compared to that of the low-level nighttime stable layer into which it is propagating (Fig. 8b). These nonlinearities are evident in the surface (Fig. 6) and boundary layer (Fig. 7) observations as characteristics similar to those of atmospheric bores or solitary waves. The GOC LLJ may also enhance gulf surge flow along the northern gulf. Figure 8: Conceptual schematic detailing the evolution of proposed gulf surge dynamic mechanism at times t 0 (a) and t 1 (b). Dotted contour represents surge onset and perpendicular yellow arrows represent gulf surge propagation speed. Colored contours represent altitude above sea level (m). Bottom panels depict gulf surge horizontal structure (O ~ hundreds of km). (a)(b)(c) (a) (b) (c)(d) (a) (b) (c) (d) (e) (f) T he synoptic environment from two days before (11 July, t = -2 d) to two days after (15 July, t = +2 d) gulf surge onset at Puerto Peñasco was influenced by several features. At 700 hPa (Fig. 2a), the subtropical high weakened and broadened as it built north and west from the LA/MS border to the southern Great Plains. Tropical Storm Blas propagated northwest from the eastern tropical Pacific Ocean to the south and west of Baja CA. At 200 hPa (Fig. 2b), a southwest- northeast aligned inverted trough was located over central Mexico and quickly weakened as it moved northwestward. T he convective environment on 13 July consisted of many thunderstorm complexes along the Sierra Madre Occidental (SMO) and GOC coastal plain (Fig. 3a) that later merged together to form a large mesoscale convective system (MCS) between Bahia Kino and Los Mochis (Fig. 3b). G ulf surge onset occurred at ~ 1000 UTC 13 July (0600 UTC 13 July) at Puerto Peñasco (Bahia Kino) as evident in the strong south-southeasterlies that deepened and strengthened with time (Figs. 4a-b, top panels). This pattern may be enhanced by the GOC low-level jet (LLJ) and lasts less than 12 hours. Maximum wind speeds approached 20.0 m s -1 about two hours after initial onset, but at a higher altitude at Bahia Kino. Surge onset at Los Mochis (Fig. 4c) was more difficult to discern, but was perhaps related to the south-southeasterly flow that began around 0000 UTC 13 July up to 2.0 km. Surge surface signals at the northern sites (Figs. 4a-b, bottom panels) included a weakening of the diurnal thermal signature, substantial pressure rises, and a wind shift to the south-southeast. The south-southeasterly flow on 14 July at each site was most likely associated with Tropical Storm Blas. P receded by anomalous warming, gulf surge passage was also accompanied by strong anomalous low-level (below 800 hPa) cooling (– 4-8°C) and moisture flux (> + 175 m g s -1 kg -1 ), most evident from Empalme northward centered about 1200 UTC 13 July (Figs. 5a-d). The impressive cooling/moistening signals on 14 July at the northern sites were most likely due to southerly advection of cool, moist air via Tropical Storm Blas from the east tropical Pacific Ocean. I nitial gulf surge onset as identified from Figs. 4a-b was accompanied by two distinct anomalous surface pressure and temperature pulses at both Puerto Peñasco (Figs. 6a and d) and Bahia Kino (Figs. 6b and c). During the two hours following surge onset, the pressure (temperature) increased approximately 1.25 hPa (1.0°C) at Puerto Peñasco and 2.25 hPa (0.7°C) at Bahia Kino. Similar signals at Los Mochis were not evident. Tracking these features between the northern sites resulted in a propagation speed of 20.5 m s -1, assuming an initial surge front perpendicular to 160°. Several hours after onset, the pressure anomaly curves remained at or above their newly achieved levels and the temperature anomaly curves decreased. P rior to gulf surge passage, (Fig. 7, 0600 UTC) the local thermodynamic environment at Puerto Peñasco consisted of a low-level thermal inversion and nighttime stable layer. Conditions above the inversion were near neutral and dry. By 1200 UTC, the well-mixed boundary layer depth substantially increased as evident in the lifted height of the thermal inversion, and was a result of surge passage. Strong cooling, moistening, and increased south- southeasterly flow also occurred both above and below the thermal inversion. SURGE PROGRESSION, ~ 20 m s -1