Richard L. Thompson
Storm Prediction Center, Norman, OK
The return flow cycle and airmass modification over the Gulf of Mexico have received considerable attention over the past several decades (e.g., Henry 1979; Lewis et al. 1989; Crisp and Lewis 1992; Lewis and Crisp 1992; Merrill 1992). From these studies, a conceptual model of the return flow cycle has been developed. The model consists of: 1) rapid airmass modification following cold frontal passage, 2) the surface pressure ridge settles over the Gulf and low-level winds and the rate of modification diminish, 3) then the modified airmass returns to the southern Plains in response to pressure falls in the lee of the Rocky Mountains.
In recent years, the focus has been on operational model forecasts of boundary layer moisture (Janish and Lyons 1992, Janish 1993) in return flow situations. Such studies have identified operational model biases, and suggested means to correct for these biases. Another approach to return flow forecasts involves application of "equilibrium" thermodynamic concepts to return flows consisting of maritime tropical air (Thompson et al. 1994). Following Merrill (1987) and Betts and Ridgway (1989), this equilibrium between between the boundary layer and the underlying sea surface can be approximated by an air temperature one degree less than the sea surface temperature (SST) and a relative humidity of 80%. Thus, given SST, equilibrium dewpoint temperatures can be calculated for a return flow event. Field projects such as GUFMEX (Lewis et al. 1989) have provided detailed datasets to aid the study of return flow, but much of this information is not routinely available to National Weather Service (NWS) forecasters. The purpose of this work is to demonstrate the utility of relatively limited buoy observations from the western Gulf of Mexico during cool season return flow cycles.
Surface buoy observations from the NOAA buoy 42002 were the primary data source for this study. Buoy 42002 is located in the western Gulf of Mexico at 25.9 N latitude and 93.6 W longitude, roughly 300 km south of Beaumont, TX and 300 km east of Brownsville, TX. This location was considered ideal for investigating the open water characteristics of surface return flow affecting the southern Plains and lower Mississippi Valley. This buoy is located south of the shallow continental shelf waters. The shelf waters experience relatively large seasonal variations in temperature and are not considered relevant when evaluating air-sea equilibrium during cool season return flow episodes (Thompson et al. 1994).
Data were archived via the Automated Field Operations System (AFOS) at the Houston area National Weather Service (NWS) office. Four observations, at six hour intervals, were stored daily for the period from 0000 UTC 12 November 1993 to 1800 UTC 16 April 1994. On several occasions the data were not available from AFOS, thus National Meteorological Center (NMC) surface analyses on microfilm served as a surrogate. The few remaining unavailable observations (mainly dewpoint temperature) were omitted from the time series and subsequent analyses.
The specific data collection period of 12 November 1993 to 16 April 1994 was chosen for both logistical and meteorological reasons. November is generally considered the start of the cool season along the Gulf coast, when the frequency of surface cold frontal passages and alternating periods of offshore and onshore flow increases substantially (Henry 1979, Lewis and Crisp 1992). Also, the means to archive the buoy data on-station were not operational until 12 November 1993. The end time for the analyses corresponded to the diminished frequency of frontal intrusions and subsequent air mass modification over the Gulf of Mexico, and the common occurrence of "super-equilibrium" conditions (see Fig. 2 and discussion in section 3).
3.1 Seasonal Time Series
Time series of SST, temperature, and dewpoint temperature at NOAA buoy 42002 were constructed for the 1993-94 cool season. The data displayed in Fig. 2 are observations recorded every six hours at the synoptic observation times (00, 06, 12, 18 UTC), displayed by month. Fig. 2 shows the numerous surface frontal intrusions into the western Gulf of Mexico during the period of study. The frontal intrusions were characterized by relative maxima in the temperature and dewpoint temperature traces, followed by dramatic decreases in the temperature and dewpoint temperature often over a period of twelve hours or less.
A total of twenty-seven frontal passages and return flow episodes were observed at buoy 42002. Frontal passages were defined by wind shifts, pressure rises, and temperature/dewpoint temperature drops. In this investigation, return flow was said to begin at the buoy when surface winds veered (typically) from northeast to a consistent easterly direction, similar to the beginning of the onshore flow phase discussed by Crisp and Lewis (1992). An east wind at buoy 42002 and an east to northeast wind at buoy 42001 in the central Gulf, implied surface trajectories crossing much of the open Gulf, as opposed to short over-water trajectories from the north central Gulf coast across the shelf waters with north to northeast flow at buoy 42002.
The frontal passages were well-correlated (negatively) with the pressure tendency (correlation coefficient of -0.67 for the entire dataset). Trends in surface pressure agreed well with trends in dewpoint temperature, with rising (falling) pressure accompanied by falling (rising) dewpoint temperature (Fig. 3). Mean statistics for the 12 November 1993 to 16 April 1994 dataset are summarized in Table 1.
It should be noted from the time series in Fig. 1 that the SST reached a seasonal minumum in the period from late February to mid-March, consistent with observations during 1995 and other previous years.
3.2 Air-sea Equilibrium at Buoy 42002
An objective of this study was to examine the air-sea equilibrium timescale from the perspective of an operational forecaster. Air-sea equilibrium has shown promise as a return flow forecast tool (Thompson et al. 1994; Merrill 1987 personal communication), and real-time experiences of forecasters at the Storm Prediction Center (SPC) in Kansas City, MO support its utility.
Air-sea equilibrium can be approximated by an air temperature one degree (C) less than the underlying SST, with a relative humidity of 80%. Therefore, given SST, equilibrium dewpoint temperatures can be calculated (see Table 2). Careful examination of the 1993-94 cool season time series yields several interesting observations concerning air-sea equilibrium over the Gulf.
First, air-sea equilibrium is not common at buoy 42002. Some of the return flow episodes fail to reach equilibrium with respect to SST at the buoy, while most eventually exceed equilibrium for the given SST at buoy 42002 (see Fig. 4).
The episodes that fail to reach equilibrium are characterized by synoptic scale flow regimes that produce return flow (Rfd) durations comparable to or less than the offshore flow duration (Ofd) (i.e. ratio Rfd/OFd near or < unity). These return flow events fail to reach equilibrium due to relatively short surface trajectories (say < 18-24 hours) over SSTs at or greater than observations at buoy 42002.
The long-duration return flow cases often exhibited surface conditions that exceeded equilibrium at the buoy. An explanation for this is surface air being transported from a region of warmer SSTs (such as the Caribbean Sea), where air-sea equilibrium supports greater temperatures and dewpoint temperatures than the SST at buoy 42002.
From Figure 4, a majority of return flow dewpoint temperatures exceeded equilibrium by 2-4 C for the SST at buoy 42002. This suggests that the source regions for the cool season return flows with the greatest dewpoints may be either the Bay of Campeche or Gulf of Tehuantepec, the Loop Current and Florida Straits, or the western Caribbean Sea - all regions of SST that support dewpoint temperatures >20 C during the cool season. Such "super-equilibrium" return flow conditions may be linked to cool season cases of strong/severe convection in the southern/central United States, but this hypothesis awaits further research.
From the perspective of an operational forecaster, important questions to answer are: 1) What is the SST in the source region for a return flow event?, and 2) Will surface air reach equilibrium during the return flow? To answer these questions, the forecaster needs to estimate surface trajectories, and the time spent over relatively warm SSTs. In addition, if a forecaster considers particular dewpoint temperatures "significant", then he/she can estimate surface trajectories that may result in dewpoints at or above the desired value(s) by determining the appropriate equilibrium dewpoints for the various return flow source regions (refer to Table 2).
Important considerations in return flow forecasting are the Ofd and subsequent Rfd, and whether or not the return flow air will be modifying polar air or maritime tropical near equilibrium. The longer the return flow, the greater the likelihood that the return flow air will be maritime tropical in nature (i.e. ratio Rfd/OFd >2). An exception to this is the scenario where a cold front stalls over the Gulf, and air on the warm side of the frontal zone has already achieved equilibrium. In such cases, a rapid return of maritime tropical air to the coast is likely.
Equilibrium thermodynamic principles can be applied successfully to forecasts of deep convection. From operational experience, the majority of southern/central United States outbreaks of strong/severe thunderstorms during the cool season are associated with return flows having greater than equilibrium dewpoint temperatures for much of the Gulf (dewpoints >21 C). Examination of several years of cool season SST data reveals that the area of the Gulf near buoy 42002 typically supports equilibrium dewpoint temperatures ranging from 17-19 C during the late winter and early spring months (late January through mid/late March). Thus, dew point observations at or greater than 21 C (~69 F) suggest region(s) of warmer SSTs (Bay of Campeche, Loop Current, Caribbean Sea) are the likely source for those return flows associated with the strongest severe thunderstorm episodes.
Equilibrium dewpoint temperatures, in conjunction with the techniques of Johnson and Graschel (1991), may also prove useful in the forecasting of sea fog over the relatively cool continental shelf waters of the northern Gulf during the cool season. Surface trajectories and SST can be combined for estimates of equilibrium dewpoint temperatures. These forecast values should be compared to available buoy and platform observations to determine the likelihood of sea fog development.
Seasonal time series may also help the forecaster anticipate the potential for extended periods of deep convection. Edwards and Weiss (1996) found strong positive monthly correlations between Gulf SST anomaly tendencies and severe thunderstorm reports in the southern/central United States. Daily or weekly examinations of Gulf and Caribbean SST data (including buoy 42002) can give forecasters the ability to anticipate potentially active periods of deep convection (i.e. expect greater potential for convection when SST anomalies are increasing, all else being equal).
Buoy observations, though limited over the Gulf of Mexico, can prove valuable to diagnoses and forecasts of return flow into the southern United States. To demonstrate the utility of buoy data, observations from NOAA buoy 42002 in the western Gulf of Mexico were examined for 1993-94 cool season.
Time series of temperature, dewpoint temperature and wind revealed a total of 27 cold frontal passages during the period from 12 November 1993 through 16 April 1994. These frontal passages marked the beginning of successive return flow cycles, where the durations of "offshore" and "onshore" flow were linked to the degree of moisture return. More specifically, the return flow events were compared to so-called "equilibrium" values for the underlying SST at buoy 42002. Time series of maximum dewpoint temperature, equilibrium dewpoint, and the ratio of return flow duration to offshore flow duration for each cycle repeatedly indicated observed dewpoints in excess of those suggested by equilibrium thermodynamics. "Super-equilibrium" conditions were almost always associated with cases where return flow duration exceeded offshore flow duration (ratio Rfd/OFd >2), implying ample time for airmass modification and surface trajectories originating over warmer waters of the southern Gulf or Caribbean.
By monitoring buoy observations and SST during the cool season, operational forecasters are able to anticipate particular return flow dewpoint values with each return flow cycle and compare observations with calculated equilibrium values. The use of equilibrium concepts and examination of buoy data can also aid forecasters in judging the quality of numerical model forecasts of low level moisture return. On longer time scales, results from Edwards and Weiss (1996) illustrate the positive correlation between Gulf SST anomaly trends and number of severe thunderstorm reports in the southern United States.
In conclusion, several utilitizations of offshore buoy data have been presented. Ultimately, it is hoped that operational weather forecasts will benefit from the application of equilibrium thermodynamic principles and simple inclusion of buoy observations in the standard forecast routine.
The author is grateful to Reid Hawkins (SOO with NWS in Wilmington, NC) for his assistance in archiving the data used in this study, and to Steve Corfidi and Roger Edwards of SPC for reviewing the document. Thanks also to Bill Read (MIC) for encouraging this and other projects by entry-level meteorologists at the NWS office in Houston, TX.
Betts, A. K. , and W. Ridgway, 1989: Climatic equilibrium of the atmospheric convective boundary layer over a tropical ocean. J. Atmos. Sci., 46, 2621-2641.
Crisp, C. A., and J. M. Lewis, 1992: Return flow in the Gulf of Mexico. Part I: A classificatory approach with a global historical perspective. J. Appl. Meteor., 31, 868-881.
Edwards, R., and S. J. Weiss, 1996: Comparisons between Gulf of Mexico sea surface temperature anomalies and southern U. S. severe thunderstorm frequency in the cool season. Preprints, 18th Conf. on Severe Local Storms, San Francisco, CA.
Henry, W. K., 1979: Some aspects of the fate of cold fronts in the Gulf of Mexico. Mon. Wea. Rev., 107, 1078-1082.
Johnson, G. A., and G. Graschel, 1991: Sea fog and stratus: A major aviation hazard in the northern Gulf of Mexico. Preprints, 4th Int. Conf. On Aviation Weather Systems, 24-28, Paris, FR.
Lewis, J. M., and C. A. Crisp, 1992: Return flow in the Gulf of Mexico. Part II: Variability of return-flow thermodynamics inferred from trajectories over the Gulf. J. Appl. Meteor., 31, 882-898.
_____, C. Hayden, R. Merrill, and J. M. Schneider, 1989: GUFMEX: A study of return flow in the Gulf of Mexico. Bull. Amer. Meteor. Soc., 70, 24-29.
Merrill, R. T., 1987: Equilibrium dewpoint temperature over the Gulf of Mexico in March. Technical paper persented at GUFMEX planning meeting, National Severe Storms Laboratory, Norman, OK, 4pp.
_____, 1992: Synoptic analysis of the GUFMEX return flow event of 10-12 March 1988. J. Appl. Meteor., 31, 849-867.
Thompson, R. L., J. M. Lewis, and R. A. Maddox, 1994: Autumnal return of tropical air to the Gulf of Mexico's coastal plain. Wea. Forecasting, 9, 348-360.