• Bretherton, F. P., R. E. Davis, and C. B. Fandry, 1976: A technique for objective analysis and design of oceanographic experiments applied to MODE-73. Deep-Sea Res.,23, 559–582.

  • Carter, E. F., and A. R. Robinson, 1987: Analysis models for the estimation of oceanic fields. J. Atmos. Oceanic Technol.,4, 49–74.

  • Cho, K., 1996: Three dimensional structure and variability of low-frequency currents on the Texas–Louisiana shelf based on moored current meter data. Ph.D. dissertation, Texas A&M University, 121 pp.

  • Cochrane, J. D., and F. J. Kelly, 1986: Low-frequency circulation on the Texas–Louisiana continental shelf. J. Geophys. Res.,91, 10 645–10 659.

  • Denman, K. L., and H. J. Freeland, 1985: Correlation scales, objective mapping and a statistical test of geostrophy over the continental shelf. J. Mar. Res.,43, 517–539.

  • Gandin, L. S., 1963: Objective Analysis of Meteorological Fields (in Russian). Gidrometeorizdar, 238 pp.

  • Gill, A. E., and E. H. Schumann, 1974: The generation of long shelf waves by wind. J. Phys. Oceanogr.,4, 83–90.

  • Hastenrath, S., 1968: A contribution to the wind conditions over the Caribbean Sea and Gulf of Mexico. Tellus,1, 163–178.

  • Hsu, S. A., 1993: The Gulf of Mexico—A breeding ground for winter storms. Mar. Wea. Log,37, 4–11.

  • ——, 1994: On the incorporation of wave characteristics into drag coefficient formulation at sea. Preprints, Second Int. Conf. on Air–Sea Interaction and on Meteorology and Oceanography of the Coastal Zone. Lisbon, Portugal, Amer. Meteor. Soc., 237–238.

  • Koch, S. E., M. Desjardins, and P. J. Kocin, 1983: An interactive Barnes objective map and analysis scheme for use with satellite and conventional data. J. Climate Appl. Meteor.,22, 1487–1503.

  • Li, Y., W. D. Nowlin Jr., and R. O. Reid, 1997: Mean hydrographic fields and their interannual variability over the Texas–Louisiana continental shelf in spring, summer, and fall. J. Geophys. Res.,102, 1027–1049.

  • Mariano, A. J., and O. B. Brown, 1992: Efficient objective analysis of dynamically heterogeneous and nonstationary fields via the parameter matrix. Deep-Sea Res.,39, 1255–1271.

  • McWilliams, J. C., W. B. Owens, and B. L. Hua, 1986: An objective analysis of the POLYMODE local dynamics experiment. Part I:General formalism and statistical model selection. J. Phys. Oceanogr.,16, 483–504.

  • Miller, P. A., and S. G. Benjamin, 1992: A system for the hourly assimilation of surface observations in mountainous and flat terrain. Mon. Wea. Rev.,120, 2342–2359.

  • NOAA, 1993: Marine weather review: North Atlantic weather January, February and March, 1993. Mar. Wea. Log,37, 69–72.

  • Nowlin, W. D., Jr., and C. A. Parker, 1974: Effects of a cold-air outbreak on shelf waters of the Gulf of Mexico. J. Phys. Oceanogr.,4, 467–486.

  • NWS, 1994: National Weather Service Observing Handbook 7. 451 pp. [Available from NWS, 1325 East–West Highway, Silver Spring, MD 20910.].

  • Perrie, W., and B. Toulany, 1989: Correlations of sea level pressure fields for objective analysis. Mon. Wea. Rev.,117, 1965–1974.

  • Pond, S., and G. L. Pickard, 1978: Introductory Dynamic Oceanography. Pergamon, 241 pp.

  • Rienecker, M. M., C. N. K. Mooers, and A. R. Robinson, 1987: Dynamical interpolation and forecast of the evolution of mesoscale features off northern California. J. Phys. Oceanogr.,17, 1189–1213.

  • Smith, S. D., 1988: Coefficients for sea surface wind stress, heat flux, and wind profiles as a function of wind speed and temperature. J. Geophys. Res.,93, 15 467–15 472.

  • Smith, W. H. F., and P. Wessel, 1990: Gridding with continuous curvature splines in tension. Geophysics,55, 293–305.

  • Sturges, W., 1993: The annual cycle of the western boundary current in the Gulf of Mexico. J. Geophys. Res.,98, 18 053–18 068.

  • Thiebaux, H. J., and M. A. Pedder, 1987: Spatial Objective Analysis with Applications in Atmospheric Science. Academic Press, 299 pp.

  • Velasco, G. G., and C. D. Winant, 1996: Seasonal patterns of wind stress and wind stress curl over the Gulf of Mexico, 1990–1993. J. Geophys. Res.,101, 18 127–18 140.

  • Wang, W., 1996: Analyses of surface meteorological fields over the northwestern Gulf of Mexico and wind effects on the circulation over the LATEX shelf. Ph.D. dissertation, Texas A&M University, 122 pp.

  • ——, M. K. Howard, W. D. Nowlin Jr., and R. O. Reid, 1996: LATEX shelf data report, meteorology, April 1992 through December 1994. Tech. Rep. 96-2-T, 499 pp. [Available from Texas A&M University, Department of Oceanography, College Station, TX 77843-3146.].

  • ——, W. D. Nowlin Jr., and R. O. Reid, 1998: A comparison among LATEX, NCEP, and ERS-1 scatterometer winds over the northwestern Gulf of Mexico. J. Atmos. Oceanic Technol.,15, 1204–1214.

  • Wessel, P., and W. H. F. Smith, 1991: Free software helps map and display data. Eos, Trans. Amer. Geophys. Union,72, 441, 445–446.

  • White, W. B., and R. L. Bernstein, 1979: Design of an oceanographic network in the mid latitude North Pacific. J. Phys. Oceanogr.,9, 592–606.

  • Winant, C. D., and C. E. Dorman, 1997: Seasonal patterns of surface wind stress and heat flux over the Southern California Bight. J. Geophys. Res.,102, 5641–5653.

  • View in gallery

    Locations of sites at which the meteorological variables used in this study were measured.

  • View in gallery

    Vector mean 10-m winds and variance ellipses, mean field of wind stress curl, surface air temperature, and mean SST.

  • View in gallery

    Seasonal mean wind and wind variance ellipses, constructed from hourly gridded winds for the period April 1992–November 1994.

  • View in gallery

    Seasonal wind steadiness (magnitude of vector mean divided by mean speed), constructed from gridded hourly values for the period April 1992–November 1994.

  • View in gallery

    Seasonal mean wind stress fields, constructed from hourly gridded winds for the period April 1992–November 1994.

  • View in gallery

    Seasonal mean wind stress curl (10−4 Pa km−1), constructed from hourly gridded fields for the period April 1992–November 1994.

  • View in gallery

    Seasonal mean sea level atmospheric pressure (hPa), constructed from gridded hourly values for the period April 1992–November 1994.

  • View in gallery

    Seasonal mean surface air temperature (°C), constructed from hourly gridded values for the period April 1992–November 1994.

  • View in gallery

    Seasonal mean SST (°C), constructed from hourly gridded values for the period April 1992–November 1994.

  • View in gallery

    Daily mean SST (solid) and surface air temperature (dashed) for 1993 from 27.5°N, 94°W.

  • View in gallery

    Fig. 11a. Monthly vector mean surface wind fields with variance ellipses for February and May, constructed from hourly gridded values.

  • View in gallery

    Fig. 11b. Monthly vector mean surface wind fields with variance ellipses for August and November, constructed from hourly gridded values.

  • View in gallery

    Surface wind fields during the passage of a cold front in November 1992.

  • View in gallery

    Surface air temperature (°C) during the passage of a cold front in November 1992.

  • View in gallery

    Sensible heat flux (W m−2) during the passage of a cold front in November 1992.

  • View in gallery

    Surface wind fields during the passage of the Storm of the Century in March 1993.

  • View in gallery

    Sensible heat flux (W m−2) fields during the passage of the Storm of the Century in March 1993.

  • View in gallery

    Near-surface (approximately 10-m depth) currents during the passage of the Storm of the Century in March 1993.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 34 34 0
PDF Downloads 25 25 1

Analyzed Surface Meteorological Fields over the Northwestern Gulf of Mexico for 1992–94: Mean, Seasonal, and Monthly Patterns

View More View Less
  • 1 Department of Oceanography, Texas A&M University, College Station, Texas
© Get Permissions
Full access

Abstract

The primary objective of this work is to formulate surface meteorological fields over the northwestern Gulf of Mexico for the period from April 1992 through November 1994 useful for the study of mesoscale processes and for model forcing of the near-coastal circulation. Observations were adjusted to standard heights, and a method of statistical interpolation was applied to time series of in situ observations to produce the required surface fields. Resulting monthly and seasonal mean fields show two principal patterns over the Texas-Louisiana shelf region—for summer and nonsummer. From June through August, surface winds are relatively constant, with alongshore wind components generally directed upcoast (from Mexico toward the Mississippi Delta). In other (nonsummer) months, surface winds are much more variable with alongshore wind components generally directed downcoast. The relatively large interannual variability is illustrated. Using these meteorological fields together with rather complete oceanographic data available from the same period, the effects of episodic atmospheric events on the circulation and properties of the Texas-Louisiana shelf may be examined. As examples, two extreme atmospheric events are characterized in terms of wind, surface air temperature, SST, and sensible heat flux fields: a cold air outbreak in November 1992 and a cyclone generated in March 1993 known as the “Storm of the Century.”

Corresponding author’s address: Wensu Wang, Department of Oceanography, Texas A&M University, College Station, TX 77843-3146.

Email: wwang@latexsun.tamu.edu

Abstract

The primary objective of this work is to formulate surface meteorological fields over the northwestern Gulf of Mexico for the period from April 1992 through November 1994 useful for the study of mesoscale processes and for model forcing of the near-coastal circulation. Observations were adjusted to standard heights, and a method of statistical interpolation was applied to time series of in situ observations to produce the required surface fields. Resulting monthly and seasonal mean fields show two principal patterns over the Texas-Louisiana shelf region—for summer and nonsummer. From June through August, surface winds are relatively constant, with alongshore wind components generally directed upcoast (from Mexico toward the Mississippi Delta). In other (nonsummer) months, surface winds are much more variable with alongshore wind components generally directed downcoast. The relatively large interannual variability is illustrated. Using these meteorological fields together with rather complete oceanographic data available from the same period, the effects of episodic atmospheric events on the circulation and properties of the Texas-Louisiana shelf may be examined. As examples, two extreme atmospheric events are characterized in terms of wind, surface air temperature, SST, and sensible heat flux fields: a cold air outbreak in November 1992 and a cyclone generated in March 1993 known as the “Storm of the Century.”

Corresponding author’s address: Wensu Wang, Department of Oceanography, Texas A&M University, College Station, TX 77843-3146.

Email: wwang@latexsun.tamu.edu

1. Introduction

Atmospheric phenomena and associated air–sea interactions play important roles in modifying the oceanic regime on the Texas-Louisiana continental shelf. As examples: winds with alongshelf components directed from the Mississippi River toward Mexico, here referred to as downcoast (upcoast) components result in downcoast (upcoast) water flow over the inner shelf (Cochrane and Kelly 1986); and extreme atmospheric phenomena, such as cold air outbreaks, significantly modify coastal circulation and oceanic property distributions by large momentum and heat fluxes (Nowlin and Parker 1974).

These oceanographic shelf processes are dominated by spatial scales on the order of tens to hundreds of kilometers and are forced by atmospheric events with scales on the order of hundreds of kilometers. Understanding the coastal ocean–atmosphere system requires observations on commensurate scales. In reporting on seasonal patterns of surface wind stress and heat flux over the southern California Bight, Winant and Dorman (1997) noted that “significant gradients in ocean–atmosphere fluxes occur on spatial scales which are similar to those which characterize the variability of ocean properties,” and certainly at scales considerably less than 1°.

The Texas-Louisiana Shelf Circulation and Transport Processes Study (LATEX-A) provided observations of surface meteorological parameters and ocean properties over the Texas-Louisiana continental shelf from April 1992 through November 1994. The surface meteorological measurements from LATEX buoys supplemented by contemporary measurements from other buoys and platforms in the northwestern Gulf of Mexico and from selected coastal airport stations in Texas and Louisiana constitute the database used in this study; they are discussed in more detail in the next section. Section 2 also describes methods used in processing and analyzing the surface meteorological fields. The resulting fields are described in section 3. A brief summary is presented in section 4.

2. Data and methodology

a. LATEX meteorological measurements

In April 1992 eight meteorological buoys were deployed over the central area of the Texas-Louisiana continental shelf (Fig. 1) as part of the LATEX program. Buoys were maintained at a subset of the eight locations through November 1994; four buoys were removed during summers to avoid possible hurricane damage.

Each buoy had sensors measuring hourly SST and the following surface meteorological properties: wind speed and direction, air temperature, and sea level barometric pressure. Wind speed and direction were measured at 3.6 m above sea level, air temperature at 3 m, pressure at sea level, and SST at 1 m below sea level. The ranges, accuracies, and resolutions of variables measured on the LATEX meteorological buoys are listed in Table 1. Wind speed and air temperature were adjusted to 10 m above sea level based on marine boundary layer theory (following Smith 1988) before further analysis.

b. Other meteorological measurements

Other meteorological measurements were collected from six National Data Buoy Center (NDBC) sites, nine Coastal-Marine Automated Network (C-MAN) sites, and five National Weather Service (NWS) airport weather stations located on land near the coast. All locations are shown in Fig. 1. The variables measured were the same as measured on LATEX buoys at the same temporal intervals (see Table 2). However, the measurement heights for winds and air temperatures varied among the NDBC and C-MAN buoys and platforms (Table 3). The measured winds and air temperatures were adjusted to 10-m above sea level using the same method used to adjust winds and air temperatures from LATEX buoys, thus avoiding bias created by different methods. Reported atmospheric pressures used had been adjusted to sea level using the method described in NWS (1994).

At the airport weather stations, the same surface properties except SST were measured hourly. Wind speed and direction were measured at 10 m above local ground level and air temperature at approximately 1.5 m above ground level. Reported surface pressures had been adjusted to sea level using the method described in NWS (1994).

c. Methodology

For many uses the variables of interest are needed at regularly spaced grids. The average spacing of observations (distance between each observation site and its nearest neighbor) is 0.86° lat. According to classical sampling theory, typical bounds on the ratio between the grid spacing and the data spacing lie in the range 0.3–0.5 (Koch et al. 1983). For a ratio of 0.5, the grid spacing would be 0.43°. A value of 0.5° was used. During summers, four of the LATEX meteorological buoys were removed; during those periods the average spacing of observations increased to 1.05°.

The spacing of observation sites is considerably coarser offshore of the 200-m isobath than on the shelf. This likely reduces the relative confidence of the analysis in the off-shelf area. However, it is expected that the principal uses of these analyzed fields will be in the study of phenomena and processes over the shelf.

Several objective analysis methods are in use to estimate values at grid points using weighted averages of nearby observed data. Statistical interpolation (also called optimal interpolation) was used in this study. Statistical interpolation was introduced as optimal estimation in meteorological data analysis (Gandin 1963). It is now widely used in meteorology and oceanography for both analysis and observation array design (Bretherton et al. 1976; White and Bernstein 1979; McWilliams et al. 1986; Carter and Robinson 1987; Rienecker et al. 1987; Thiebaux and Pedder 1987; Mariano and Brown 1992; Miller and Benjamin 1992). In the SI method, weights are chosen based on the spatial correlation of the observed data to minimize the error variance of gridded values.

The statistical interpolation method requires knowledge of the covariance structure of variables. The real covariances depend on time and location and are difficult to estimate well directly from existing data. To avoid this difficulty, covariances are usually modeled in terms of mathematical forms that involve parameters whose values can be estimated from a given data sample. Examples of models are second-order and third-order autoregressive models or Gaussian distribution models (Denman and Freeland 1985; Perrie and Toulany 1989). The following parametric form (Rienecker et al. 1987) was applied to obtain correlations in this study:
i1520-0493-126-11-2864-e1
where r is the separation between observation pairs, δ is the Kronecker delta (δ = 1 if r = 0; δ = 0 otherwise), and a, b, and E2 are the parameters to be determined. The length a represents the zero-crossing scale; b is the Gaussian decay scale. For surface meteorological variables, a and b are approximately the same; here we take b as equal to a. Here E2 is the noise variance of observations due to submesoscale process as well as instrumental noise. The assumption used to obtain this covariance function is that the spatial covariance is isotropic and homogeneous.

Whether the covariances were isotropic was examined using data from 10 months distributed throughout the seasons and years sampled: June 1992, November 1992, January 1993, March 1993, June 1993, October 1993, January 1994, April 1994, July 1994, and October 1994. Pairs of observing sites located approximately perpendicular to the isobaths were used to estimate cross-shelf covariances; pairs located approximately parallel to the isobaths were used to estimate along-shelf covariances. Scales were estimated by zero crossings of the autocorrelations. The mean scales with standard deviations are shown in Table 4. For surface barometric pressure and eastward and northward wind components, the cross-shelf and along-shelf covariances are not statistically different, with spatial scales (a) near 350 km in both cross-shelf and along-shelf directions. This implies that covariances of surface pressure and two wind components may be considered isotropic. However, for SST and air temperature, the along-shelf spatial scales were considerably longer than the cross-shelf scales (Table 4). This is expected because the temperature distributions have pronounced cross-shelf gradients. Nevertheless, for simplicity the spatial scales used were obtained by assuming isotropic spatial variations in all variables.

To test homogeneity, zero crossings from correlations were calculated for each variable using November 1992 data for the inshore region (measurements over water with depths less than 50 m) and compared with zero crossings from the offshore region. No allowance was made for direction of separation of two data sites; time series from each location were correlated with those from every other location in the subdomain. No significant difference in scales as a function of region was noted for these variables.

The covariances between any pair of hourly observations were calculated within 50-km bins. The monthly mean covariances were estimated by averaging hourly values for each month. Then the spatial scale (a) and the noise variance (E2) in Eq. (1) were fitted to the normalized average covariances squared; that is, correlated, by the method of least squares. The noise variance E2 was determined by subtracting from unity the value of the analytical function at r = 0. From Table 4 we see that the spatial scales for surface pressure and wind components are quite similar. Therefore, an identical covariance function (with the same a and E2 values) was used to represent the covariances for the two wind components and surface pressure by simultaneously fitting covariance data for the three variables. The parameters in Eq. (1) for SST and air temperature were estimated from separate covariance datasets.

For each month from April 1992 through November 1994, the monthly mean covariance data were fitted by the covariance model represented in Eq. (1). For surface pressure and the 10-m wind components, the spatial scale (a) oscillated around 370 km with a standard deviation of 32 km, and the noise variance (E2) oscillated around 0.4 with a standard deviation of 0.1. The averaged spatial scale over fall season (September–November) was 356 km, smaller than in other seasons, possibly because of frequent frontal passages. The scales of surface air temperature and SST had relatively large variations. The mean value for air temperature was approximately 320 km, with a standard deviation of 63 km, and the mean noise variance was about 0.4; for SST the mean spatial scale was about 310 km, with a noise variance somewhat less than 0.4.

The variation of monthly mean covariances is large. We might expect even larger variations over shorter timescales. The covariance function really depends on the state of the atmosphere, which varies temporally and spatially. For example, during a frontal passage spatial scales are reduced relative to calm conditions. This influence on spatial scales will likewise influence the gridded values and estimated errors and is of concern. However, we have opted to use monthly mean parameter values instead of synoptic values in order to increase statistical confidence.

To illustrate the effect on the gridded values of using different spatial scales, we carried out analyses with a range of spatial scales from 200 to 800 km. As expected, the larger the correlation scale, the smoother the field. As an example, for a scale of 200 km, the sea level atmospheric pressure field at 0600 UTC 4 November 1992, had a minimum valued contour of 1008 hPa, whereas the minimum values for spatial scales of 400 and 800 km were 1008.5 hPa and 1009 hPa, respectively. In summary, the interpolation method proved not very sensitive to the scale so long as it was within a factor of 2 of the optimal value.

3. The surface meteorological fields

In addition to hourly fields of 10-m wind, SST, sea level atmospheric pressure, and air temperature, fields of wind stress and wind stress curl also were produced using the gridded winds. Hourly wind stress was estimated by applying a bulk aerodynamic formula to hourly gridded 10-m winds using Hsu’s (1994) drag coefficient Cd that was estimated for the outer Texas-Louisiana shelf:
3CdU10
where U10 is the 10-m wind speed. Hourly values of wind stress curl were calculated at grid points using centered differences of hourly wind stress (consistent with the Stokes theorem for a quadrangle domain). Finally, for the period April 1992 to November 1994, hourly values of all variables (wind, wind stress, wind stress curl, surface atmospheric pressure, surface air temperature, and SST) were averaged at each grid point to form 32-month, seasonal, and monthly means. Discussed briefly here are the 32-month and seasonal means, for more details and a description of monthly mean fields, see Wang et al. (1996) or Wang (1996).

a. Mean fields for April 1992–November 1994

Shown in Fig. 2a are the 32-month vector mean winds with their variance ellipses. Ellipses are plotted only at every other grid point to avoid cluttering. The mean wind vectors are generally easterly with stronger winds and larger southerly components over the western portion. This can be explained by the fact that the study area is located in the trade wind zone, and the winds are blocked somewhat in the west by land forms and thus channeled to flow more southerly. The mean speeds range between less than 1 m s−1 over the northeastern region to 3 m s−1 in the west. Mean wind speeds decrease toward land over most of the region, particularly over the eastern study region, probably due to greater friction over land.

The fluctuations relative to the mean are much larger than the mean speeds—two to three times as large over the eastern shelf where ratios between minor and major axes are close to unity. Over the western shelf the fluctuations prefer a north–south orientation, possibly due to frontal incursions.

The mean wind stress curl field is shown in Fig. 2b. The mean curl is negative (anticyclonic) over the region, with largest values located over the central region between 93° and 95°W, where absolute values reach 1 × 10−4 Pa km−1. The minimum absolute values of 0.2 × 10−4 Pa km−1 are found in the southwestern corner of the region. For comparison, an order of magnitude for surface wind stress over the North Atlantic near 35°N might be 104 Pa km−1 (Pond and Pickard 1978). Gradients are largest over the south Texas shelf.

The isotherms of mean surface air temperature (Fig. 2c) have a basic zonal orientation, as expected. The orientation of isotherms reflects continental influence by curving to match the orientation of the coast off Texas and Louisiana. Averaged air temperatures are lowest (21.0°C) near shore off Mississippi and gradually increase westward and offshore to maxima near 24.8°C. Gradients are larger near shore and over the shelf than over the open sea.

SST (Fig. 2d) has a pattern similar to that for air temperature, but has a smaller range (2.2°C) than air temperature (3.0°C). SST is always higher than air temperature—about 1°C over open sea and 2°C near the coast.

b. Seasonal patterns

1) Wind

Seasonally averaged 10-m LATEX winds are shown in Fig. 3. In spring (March–May), wind speeds range from 1 to 4 m s−1; with strongest winds in the west and weakest winds in the northeastern corner. East of 92°W, winds are easterly; moving west of 92°W they become southeasterly. The variance ellipses indicate northwest–southeast fluctuations dominate in this season. Nearly all fluctuations are larger than means, especially in the east where winds are weak. The large fluctuations are due to frequent frontal passages during this season. According to A. Jochens (1996, personal communication), 35 fronts passed over this shelf during the three spring seasons included in the study period.

The summer wind field is characterized by the southeast or south winds and the smallest fluctuations of any season. Speeds range from 1 m s−1 in the northeast to 6 m s−1 in the west. The strong summer winds are due to the core of the easterlies lying over the western part of this region in that season (Hastenrath 1968). East of 92°W the major axes of variance ellipses are comparable to the mean wind speeds, but west of 92°W the mean wind speeds far exceed the major variance axes. Variances show no strong preferred direction. The small fluctuations are attributed to the fact that frontal passages occur infrequently in summer; according to Jochens, only six fronts occurred during these three summer seasons.

In contrast to summer and spring, when winds over the eastern study area are significantly weaker than those to the west, the wind speeds in fall are nearly uniform spatially, with values near 3 m s−1. Directions are to the southwest over the eastern area and to the west over the western area. The magnitudes of fluctuations are generally the same as those during spring. Approximately 40 fronts passed over the area in fall during the observing period. We can see from the principal axes of the variance ellipses, especially over the western shelf, that the winds shifted frequently from southerly to northerly.

The mean winter winds are out of the east or northeast with stronger winds (near 4 m s−1) over the eastern region. Winter fluctuations are the greatest among the four seasons. In the west, the speeds are smaller than those in summer, but instantaneous winds in winter and summer are comparable in magnitude. The winds in winter (December–February) are generally much more variable in direction than winds in summer; this is also evidenced by the variance ellipses for winter. Consequently, the averaged wind vectors are less in winter than in summer for this area. The seasonal wind steadiness (the magnitude of vector mean divided by the mean speed) is shown in Fig. 4. This presentation gives an alternative quantitative comparison of season-to-season variability. The relatively steady direction of summer winds compared to other seasons, especially winter (where values greater than 0.5 are found only in two small subregions), is very clear. Spring and fall fields are comparable in steadiness.

2) Wind stress

Seasonal wind stresses without variance ellipses are shown in Fig. 5. The magnitudes of stresses in spring are small, between 0.01 and 0.03 Pa, with generally easterly directions. In summer, wind stresses are southerly or southeasterly. The maximum values near 0.06 Pa are found over the western shelf. The magnitudes of fall wind stresses are spatially rather uniform (about 0.03 Pa), from east or southeast. Winter wind stresses are somewhat larger than in fall and southeasterly, with maximum values near 0.05 Pa in the southern part of the region. However, the magnitudes of mean stresses in winter are smaller than in summer; refer to seasonal steadiness of the wind fields shown in Fig. 4.

3) Wind stress curl

Figure 6 shows seasonal wind stress curl fields. In spring an anticyclonic pattern with low values of −1 × 10−4 Pa km−1 is centered off the shelf near 26°N, 94°W. Values vary by about −0.2 × 10−4 Pa km−1 near the south Texas coast. By summer, the minimum has shifted somewhat northward and is located over the shelf edge;central values have intensified to −1.4 × 10−4 Pa km−1. In the southwestern and northeastern corners, the curl magnitudes are only about 0.2–0.4 × 10−4 Pa km−1. Fall wind stress curl has a center located somewhat farther southwest and central values of about 1.2 × 10−4 Pa km−1; gradients are weaker than for summer. In winter the center is diffuse, with the lowest values near 1 × 10−4 Pa km−1 located near the coast off the Texas-Louisiana border. East of 92°W the curl magnitude increases to the east, reaching values of −0.8 × 10−4 Pa km−1 in the southeastern corner of the study region. Our curl values are in reasonable agreement with the seasonal fields shown for the Gulf of Mexico by Velasco and Winant (1996), although there is more detail and extreme values are greater in our fields because they are based on a much more closely spaced dataset.

We remind the reader that, while Ekman pumping by wind stress curl is a primary driving mechanism for large-scale circulation in the upper layers of the stratified, deep ocean, the vector product of the wind stress and the gradient of water depth is the dominant driving term over the continental shelf and slope (Gill and Schumann 1974). For downwelling favorable winds over the continental shelf, this near-coastal pumping effect tends to produce cyclonic circulation over the shelf, with strongest downcoast flow near shore. Over the deeper northwest Gulf of Mexico the wind field tends to have negative curl that favors an anticyclonic circulation regime with an annual cycle (Sturges 1993). However, the strongest currents over the continental slope and deeper waters in the western Gulf of Mexico are associated with the energetic anticyclonic eddies whose origin is the Loop Current in the eastern Gulf of Mexico.

4) Sea level atmospheric pressure

The annual cycle of seasonal mean sea level atmospheric pressure fields is shown in Fig. 7. Isobars are oriented more north–south in summer than in other months, corresponding to upcoast wind components. In fall and winter the isobars are oriented more northwest–southeast. Pressures increase from northeast or east to southwest or west for all seasons. Naturally, values decrease from winter to spring and summer. Spring and summer pressure gradients are more intense in the west than in the east, corresponding with stronger winds in the west than in the east for these two seasons (see Fig. 3). Gradients in winter are somewhat stronger than in fall, resulting in larger wind speeds in winter than in fall.

Comparing patterns of seasonal surface winds (Fig. 3) with those for sea level atmospheric pressure (Fig. 7), the wind fields are seen to be approximately along the isobars in summer and to have the largest cross-isobaric components in winter. This is consistent with the concept that the surface winds are more nearly geostrophic in summer when wind speeds generally are small and frictional effects least, whereas the opposite situation occurs in winter.

5) Surface air temperature

Spring surface air temperatures (Fig. 8) vary from 20.8°C near the northern coast to 22.4°C over the open Gulf. Isotherms are oriented generally east–west except south and east of the Mississippi Delta. In summer, air temperatures are almost uniform with values near 28.2°C. West of the Mississippi Delta, isotherms generally follow isobaths, showing the influence of cold air outbreaks over the south Texas shelf. The range is from 22.4° to 25.6°C. South and east of the Delta the continental effect is even stronger. With the strongest cross-shelf gradients among the four seasons, the winter pattern shows more influence of cooling, with warm air temperatures maintained by the effect of the open Gulf waters between 93° and 94°W.

6) SST

SST fields (Fig. 9) are very similar to surface air temperature fields for all seasons. Spring SST is about 0.8°C higher than surface air temperature. Summer SST shows more variation than surface air temperature, increasing slightly from west to east. The cause may be the upcoast currents over the inner shelf during that season (Cochrane and Kelly 1986; Li et al. 1997). In summer, winds directed upcoast drive upcoast currents and force upwelling near the south Texas coast. These upcoast currents carry cooler upwelled water. Summer SST is still higher (by approximately 1°C) than air temperature over the entire region. In fall seasonal air–sea temperature differences reach maximum values—greater than 3°C near the coast and 2°C over the open Gulf. Fall and winter SST isotherms closely follow those of surface air temperature, with largest cross-shelf gradients in winter.

Figure 10 shows time series of daily average SST and surface air temperature for 1993 at the location 27.5°N, 94°W, near the center of the study region. Examining many such plots, this is typical. SST is seen to be warmer than surface air temperature except in December 1993, reflecting the warm winter of 1993/94. The SST is seen to vary quite smoothly, especially in contrast with the surface air temperature, which has variations of as much as 12°C, in response to frontal passages.

c. Annual cycle of monthly mean winds

Shown in Fig. 11 are monthly mean vector winds with fluctuation ellipses constructed from hourly values of the LATEX analyzed fields for the months of February, May, August, and November during the observing period April 1992 through November 1994. There is a regular annual cycle in the monthly fields with two dominant modes—one for summer and one for nonsummer (discussed further in Wang et al. 1996).

Summer (nominally June–August) winds differ from those in other seasons in that they have an upcoast (in the direction from Brownsville toward the Mississippi Delta) component, at least over the inner shelf, and fluctuations are small. Winds in fall, winter, and spring have downcoast components and greater variability. Near the Mexican border (26°N), the winds begin in April to change from the nonsummer to the summer pattern, as evidenced by upcoast components. This shift moves northward and eastward along the coast as a function of time. The upcoast wind component increases until July and then decreases. Toward the end of August or early September the winds shift back to a downcoast component. A very similar annual cycle occurs in the alongcoast currents on the inner shelf in response to the winds (Cochrane and Kelly 1986; Li et al. 1997).

Even though there is an annual cycle, comparison of monthly wind fields in Fig. 11 for different years shows the degree of interannual variability that might be expected. This gives rise to differing lengths of the summer, versus nonsummer, “season.” For example, the mean winds of August in 1992 and 1994 had predominantly downcoast components over much of the inner shelf, in contrast to those for August 1993; thus, the summer season lasted longer in 1993 than in 1992 or 1994. Another striking example of an anomalous monthly mean is illustrated by May 1994. In that year May winds were unusually light and steady over the eastern part of the region for a spring month.

This interannual variability—especially the differing lengths of the summer upcoast wind forcing—is mimicked in the alongshelf currents. Cho (1996) produced monthly mean maps of streamfunctions for this shelf based on 32 months of moored current meter data. He also represented the patterns of variability in these fields by decomposing them into empirical orthogonal functions. The first mode, containing 81% of the total mean square variability, showed a pattern of up- or downshelf flow over the inner shelf, with flow in the opposite direction over the outer shelf. The temporal variability of this pattern is shown by Cho et al. (1997, submitted to J. Geophys. Res.) to correspond very closely with a time series of the alongshelf components of our monthly mean LATEX wind stress averaged at six locations along the 20-m isobath.

d. Examples of extreme atmospheric events

Several extreme meteorological events occurred over the Texas-Louisiana shelf during the study period, including cold air outbreaks, cyclogenesis, and a hurricane passage. Here we describe briefly two examples to illustrate the potential usefulness of the LATEX meteorological fields in analyses of such phenomena and their effects on the shelf waters. We describe first a cold air outbreak which passed over the shelf from 3 November 2100 UTC to 5 November 0300 UTC 1992.

Before the cold air outbreak, winds were southeasterly with speeds from 3 to 8 m s−1 (Fig. 12a). Six hours later, a cold front oriented northeast–southwest between 92° and 96°W had moved into the region (Fig. 12b). Ahead of the front, winds remained southeasterly. Behind the front, winds were northerly; speeds were as great as 13 m s−1 in the southwest corner of the region. Another 6 h later (Fig. 12c) the front was located between 90° and 94°W. Behind the front northerly winds had increased to about 13–14 m s−1. From 4 November 1800 UTC to 5 November 0000 UTC the front was stalled over the eastern edge of the Texas-Louisiana shelf (Figs. 12d and 12e). One reviewer has suggested that part of the reason why this storm stalled over the outer shelf might be due to the deeper water, and consequently higher thermal inertia, of the offshore waters. By 5 November 0600 UTC (Fig. 12f), the front had cleared the shelf and was over the open Gulf south of the Mississippi Delta. Winds were northerly or northwesterly over the entire shelf and had begun to weaken.

Now we examine changes in surface air temperature associated with the frontal passage. Before the front at 4 November 0000 UTC (Fig. 13a), surface air temperature was generally uniform, with a range of about 2°C. As the front moved into the shelf area (Fig. 13b), temperatures behind the front decreased rapidly. After the front cleared the region, isotherms were generally zonal with a gradient of about 13°C from north to south across the region. The final state (Fig. 13f) showed temperatures ranging from 9° to 14°C over the shelf. This corresponds to decreases of 10° to 16° relative to prefront conditions.

The ocean releases considerable heat to the atmosphere during such cold air outbreaks. Figure 14 shows sensible heat flux fields during the cold air outbreak. At 0000 UTC on 4 November, sensible heat flux was near zero because there was little difference between SST and surface air temperatures. As the front moved across the region, isopleths of sensible heat exchange generally followed the orientation of the front. Maximum values increased to about 160 W m−2 over the south Texas shelf (Figs. 17d and 17e) and then began to decrease as the front cleared the shelf. This fast moving front reduced SST by about 2°C over the inner shelf.

The second example presented is the large cyclone, called the “Storm of the Century,” which formed in this region during March 1993 (Hsu 1993; NOAA 1993). Before the storm, winds were easterly or southeasterly except in the northeastern corner of the study region where winds were northeasterly (not shown here). At 1200 UTC on 12 March the storm center was just inshore of the coast near 26.5°N as shown in Fig. 15a. At that time, winds were stronger in the northern than the southern region of the study area. Maximum speeds exceeded 12 m s−1 at 28.8°N, 95.5°W, to the northeast of the storm center. One hour later the center had moved east to about 26.5°N, 95.7°W and the pattern of an overwater cyclone was clear (Fig. 15b). Wind speeds ranged from 1.6 m s−1 at the storm center to 12.5 m s−1 northeast of the center. During the following 4 h, the storm moved rapidly to the east. It then stalled for several hours. Then the storm again moved rapidly eastward and cleared the study area by 0000 UTC on March 13 as seen in Fig. 15i. Strong southeasterly winds with maximum speeds greater than 22 m s−1 were seen west of 92°W after the storm passed.

Using our surface fields, Fig. 16 was prepared showing sensible heat flux fields at times corresponding to the wind fields of Fig. 15. The increase from moderate values to very large heat fluxes is seen to progress as the front passed rapidly through the region. At 1200 UTC on 12 March 1993, the fluxes were relatively small:ranging from 80 W m−2 near the coast and about 93°–95°W to zero over the open southern boundary of the region. During the next 3 h, the maximum heat flux increased to 140 W m−2 near the Texas coast. The area of maximum heat flux was associated with strong winds located on the northeastern side of the storm center at and before 1500 UTC. In the next several hours the center of the storm moved eastward with strong winds to its northwest, so the area of maximum flux moved somewhat southwest along the Texas coast, with largest values greater than 200 W m−2. By 0000 UTC on 13 March, the storm had cleared the shelf area, leaving maximum fluxes over 200 W m−2 over the western study area. Comparing this figure with Fig. 14, the sensible heat flux associated with this cyclone is seen to be considerably larger than for the cold air outbreak.

Shown in Fig. 17 are analyzed fields of near-surface currents (nominally at 10-m depths) at times corresponding to the surface wind fields shown in Fig. 15 for the passage of the Storm of the Century. These currents were produced by objective analysis of hourly currents measured at 31 LATEX current meters distributed over the shelf. The objective analysis method used to interpolate the currents to a 15-min grid is based on an extension of the minimum curvature method of gridding described by Smith and Wessel (1990), contained in the Generic Mapping Tools software package (Wessel and Smith 1991). The striking response of the currents to the wind field is evident in the strong downcoast flow over the inner shelf and the eastward progression of the center of the cyclonic shelf circulation in response to the eastward progression of the atmospheric storm. The near-surface currents generated by this storm exceeded 100 cm s−1 in many locations. One would not be able to relate wind and current fields so closely using ordinary analysis products of numerical weather prediction procedures for this area.

4. Summary

Hourly surface meteorological values were produced on a 0.5° × 0.5° lat/long grid. Hourly wind stress and wind stress curl were then estimated at grid points from hourly gridded winds. The spatial resolution and temporal intervals used here are thus appropriate for the study of mesoscale processes (100 km, 6 h) in the marine boundary layer and their interaction with the ocean.

Monthly and seasonal, as well as 32-month mean, meteorological fields for the LATEX period were obtained by averaging hourly values at each grid point. Here, seasonal and mean patterns of these fields and monthly mean wind fields were presented and discussed. More detailed descriptions are presented in the LATEX meteorological data report (Wang et al. 1996).

The mean wind vectors were generally easterly because the study area is located in the trade wind zone. Mean speeds were less than 1 m s−1 south of the Mississippi Delta, increasing to more than 3 m s−1 off the south Texas coast. Mean winds over the open Gulf were nearly uniform with speeds about 2 m s−1.

The seasonal surface meteorological patterns showed the lower atmosphere over the LATEX shelf region in summer to be different than in other seasons. In summer, the atmosphere is relatively stable; only six fronts passed through this region in summer during the observing period from April 1992 to November 1994. Summer winds were southeasterly with very strong southerly components (about 6 m s−1) over the western area. Alongshore wind components were upcoast driving upcoast currents over the inner shelf. Air temperatures were the highest (28.2°C) in the annual cycle and were spatially quite uniform. Air–sea temperature differences were small (only about 0.5°C).

In other seasons, the lower atmosphere was more variable due to more frequent frontal passages. In spring, winds were southeasterly or easterly; in fall and winter, winds were northeasterly or easterly. The fluctuations for other seasons were larger than for summer, especially in winter. Surface air temperature as well as SST in these three seasons showed that isotherms basically followed isobaths, with values increasing toward the open Gulf. Cross-shelf gradients for both air temperature and SST were the largest in winter. Fall air–sea temperature differences reached maximum values of about 4°C near the coast. Seasonal values of SST are always higher than for surface air temperature, with differences smallest in spring (0.5°C) and largest in fall (3°C).

Two examples of extreme atmospheric events were presented to illustrate that these LATEX analyzed fields can be used to describe in detail such events. The example of a cold air outbreak also illustrates the very profound effects of such events on the shelf waters.

A number of studies are now under way using these meteorological fields to better understand atmospheric effects on the circulation of the Texas-Louisiana continental shelf. These include study of the response time of the reversal of alongcoast ocean currents over the inner shelf to reversals in the alongcoast component of the surface wind stress, a complete study of the effects of the Storm of the Century on shelf-wide circulation regime, model simulation of wind-driven circulation over the inner shelf (Current and Reid 1998, manuscript submitted to J. Geophys. Res.), nonlinear wave–wave interactions on the Texas-Louisiana shelf during Hurricane Andrew, comparison of WAM Cycle 4 model results with wave observations in the northwestern Gulf of Mexico, and comparisons among LATEX, NCEP (National Centers for Environmental Prediction), and European Remote Sensing Satellite-1 scatterometer winds over the northwestern Gulf of Mexico (Wang et al. 1998).

Acknowledgments

This study was funded by the Minerals Management Service under OCS Contract 14-35-0001-30509. Additional funding has been provided by Texas A&M University, the Texas Engineering Experiment Station, and Texas Institute of Oceanography. The authors thank Matthew Howard, who assisted in obtaining the datasets.

REFERENCES

  • Bretherton, F. P., R. E. Davis, and C. B. Fandry, 1976: A technique for objective analysis and design of oceanographic experiments applied to MODE-73. Deep-Sea Res.,23, 559–582.

  • Carter, E. F., and A. R. Robinson, 1987: Analysis models for the estimation of oceanic fields. J. Atmos. Oceanic Technol.,4, 49–74.

  • Cho, K., 1996: Three dimensional structure and variability of low-frequency currents on the Texas–Louisiana shelf based on moored current meter data. Ph.D. dissertation, Texas A&M University, 121 pp.

  • Cochrane, J. D., and F. J. Kelly, 1986: Low-frequency circulation on the Texas–Louisiana continental shelf. J. Geophys. Res.,91, 10 645–10 659.

  • Denman, K. L., and H. J. Freeland, 1985: Correlation scales, objective mapping and a statistical test of geostrophy over the continental shelf. J. Mar. Res.,43, 517–539.

  • Gandin, L. S., 1963: Objective Analysis of Meteorological Fields (in Russian). Gidrometeorizdar, 238 pp.

  • Gill, A. E., and E. H. Schumann, 1974: The generation of long shelf waves by wind. J. Phys. Oceanogr.,4, 83–90.

  • Hastenrath, S., 1968: A contribution to the wind conditions over the Caribbean Sea and Gulf of Mexico. Tellus,1, 163–178.

  • Hsu, S. A., 1993: The Gulf of Mexico—A breeding ground for winter storms. Mar. Wea. Log,37, 4–11.

  • ——, 1994: On the incorporation of wave characteristics into drag coefficient formulation at sea. Preprints, Second Int. Conf. on Air–Sea Interaction and on Meteorology and Oceanography of the Coastal Zone. Lisbon, Portugal, Amer. Meteor. Soc., 237–238.

  • Koch, S. E., M. Desjardins, and P. J. Kocin, 1983: An interactive Barnes objective map and analysis scheme for use with satellite and conventional data. J. Climate Appl. Meteor.,22, 1487–1503.

  • Li, Y., W. D. Nowlin Jr., and R. O. Reid, 1997: Mean hydrographic fields and their interannual variability over the Texas–Louisiana continental shelf in spring, summer, and fall. J. Geophys. Res.,102, 1027–1049.

  • Mariano, A. J., and O. B. Brown, 1992: Efficient objective analysis of dynamically heterogeneous and nonstationary fields via the parameter matrix. Deep-Sea Res.,39, 1255–1271.

  • McWilliams, J. C., W. B. Owens, and B. L. Hua, 1986: An objective analysis of the POLYMODE local dynamics experiment. Part I:General formalism and statistical model selection. J. Phys. Oceanogr.,16, 483–504.

  • Miller, P. A., and S. G. Benjamin, 1992: A system for the hourly assimilation of surface observations in mountainous and flat terrain. Mon. Wea. Rev.,120, 2342–2359.

  • NOAA, 1993: Marine weather review: North Atlantic weather January, February and March, 1993. Mar. Wea. Log,37, 69–72.

  • Nowlin, W. D., Jr., and C. A. Parker, 1974: Effects of a cold-air outbreak on shelf waters of the Gulf of Mexico. J. Phys. Oceanogr.,4, 467–486.

  • NWS, 1994: National Weather Service Observing Handbook 7. 451 pp. [Available from NWS, 1325 East–West Highway, Silver Spring, MD 20910.].

  • Perrie, W., and B. Toulany, 1989: Correlations of sea level pressure fields for objective analysis. Mon. Wea. Rev.,117, 1965–1974.

  • Pond, S., and G. L. Pickard, 1978: Introductory Dynamic Oceanography. Pergamon, 241 pp.

  • Rienecker, M. M., C. N. K. Mooers, and A. R. Robinson, 1987: Dynamical interpolation and forecast of the evolution of mesoscale features off northern California. J. Phys. Oceanogr.,17, 1189–1213.

  • Smith, S. D., 1988: Coefficients for sea surface wind stress, heat flux, and wind profiles as a function of wind speed and temperature. J. Geophys. Res.,93, 15 467–15 472.

  • Smith, W. H. F., and P. Wessel, 1990: Gridding with continuous curvature splines in tension. Geophysics,55, 293–305.

  • Sturges, W., 1993: The annual cycle of the western boundary current in the Gulf of Mexico. J. Geophys. Res.,98, 18 053–18 068.

  • Thiebaux, H. J., and M. A. Pedder, 1987: Spatial Objective Analysis with Applications in Atmospheric Science. Academic Press, 299 pp.

  • Velasco, G. G., and C. D. Winant, 1996: Seasonal patterns of wind stress and wind stress curl over the Gulf of Mexico, 1990–1993. J. Geophys. Res.,101, 18 127–18 140.

  • Wang, W., 1996: Analyses of surface meteorological fields over the northwestern Gulf of Mexico and wind effects on the circulation over the LATEX shelf. Ph.D. dissertation, Texas A&M University, 122 pp.

  • ——, M. K. Howard, W. D. Nowlin Jr., and R. O. Reid, 1996: LATEX shelf data report, meteorology, April 1992 through December 1994. Tech. Rep. 96-2-T, 499 pp. [Available from Texas A&M University, Department of Oceanography, College Station, TX 77843-3146.].

  • ——, W. D. Nowlin Jr., and R. O. Reid, 1998: A comparison among LATEX, NCEP, and ERS-1 scatterometer winds over the northwestern Gulf of Mexico. J. Atmos. Oceanic Technol.,15, 1204–1214.

  • Wessel, P., and W. H. F. Smith, 1991: Free software helps map and display data. Eos, Trans. Amer. Geophys. Union,72, 441, 445–446.

  • White, W. B., and R. L. Bernstein, 1979: Design of an oceanographic network in the mid latitude North Pacific. J. Phys. Oceanogr.,9, 592–606.

  • Winant, C. D., and C. E. Dorman, 1997: Seasonal patterns of surface wind stress and heat flux over the Southern California Bight. J. Geophys. Res.,102, 5641–5653.

Fig. 1.
Fig. 1.

Locations of sites at which the meteorological variables used in this study were measured.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 2.
Fig. 2.

Vector mean 10-m winds and variance ellipses, mean field of wind stress curl, surface air temperature, and mean SST.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 3.
Fig. 3.

Seasonal mean wind and wind variance ellipses, constructed from hourly gridded winds for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 4.
Fig. 4.

Seasonal wind steadiness (magnitude of vector mean divided by mean speed), constructed from gridded hourly values for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 5.
Fig. 5.

Seasonal mean wind stress fields, constructed from hourly gridded winds for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 6.
Fig. 6.

Seasonal mean wind stress curl (10−4 Pa km−1), constructed from hourly gridded fields for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 7.
Fig. 7.

Seasonal mean sea level atmospheric pressure (hPa), constructed from gridded hourly values for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 8.
Fig. 8.

Seasonal mean surface air temperature (°C), constructed from hourly gridded values for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 9.
Fig. 9.

Seasonal mean SST (°C), constructed from hourly gridded values for the period April 1992–November 1994.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 10.
Fig. 10.

Daily mean SST (solid) and surface air temperature (dashed) for 1993 from 27.5°N, 94°W.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

i1520-0493-126-11-2864-f11a

Fig. 11a. Monthly vector mean surface wind fields with variance ellipses for February and May, constructed from hourly gridded values.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

i1520-0493-126-11-2864-f11b

Fig. 11b. Monthly vector mean surface wind fields with variance ellipses for August and November, constructed from hourly gridded values.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 12.
Fig. 12.

Surface wind fields during the passage of a cold front in November 1992.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 13.
Fig. 13.

Surface air temperature (°C) during the passage of a cold front in November 1992.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 14.
Fig. 14.

Sensible heat flux (W m−2) during the passage of a cold front in November 1992.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 15.
Fig. 15.

Surface wind fields during the passage of the Storm of the Century in March 1993.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 16.
Fig. 16.

Sensible heat flux (W m−2) fields during the passage of the Storm of the Century in March 1993.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Fig. 17.
Fig. 17.

Near-surface (approximately 10-m depth) currents during the passage of the Storm of the Century in March 1993.

Citation: Monthly Weather Review 126, 11; 10.1175/1520-0493(1998)126<2864:ASMFOT>2.0.CO;2

Table 1.

Ranges, accuracies, and resolutions for variables measured on LATEX meteorological buoys.

Table 1.
Table 2.

Ranges, accuracies, and resolutions for variables used from National Data Buoy Center and C-MAN measurement sites.

Table 2.
Table 3.

Heights of wind and temperature recorders for National Data Buoy Center and C-MAN sites from which measurements were used to construct meteorological fields.

Table 3.
Table 4.

Mean and standard deviation of zero-crossing spatial scales (km) from correlation functions based on monthly averages for surface atmospheric pressure, 10-m wind components, surface air temperature, and SST. The 10 months were distributed over all seasons and the three years of observations. Alongshelf (subscript a) and cross-shelf (subscript c) were calculated separately.

Table 4.
Save