Analysis of Ocean Current Observations in the Northern

Veracruz Coral Reef System, Mexico: 2007–12

Authors: Allende-Arandía, María E., Zavala-Hidalgo, Jorge, Romero-

Centeno, Rosario, Mateos-Jasso, Adriana, Vargas-Hernández, Juan
M., et al.
Source: Journal of Coastal Research, 32(1) : 46-55
Published By: Coastal Education and Research Foundation
URL: https://doi.org/10.2112/JCOASTRES-D-14-00148.1

BioOne Complete (complete.BioOne.org) is a full-text database of 200 subscribed and open-access titles
in the biological, ecological, and environmental sciences published by nonprofit societies, associations,
museums, institutions, and presses.

Your use of this PDF, the BioOne Complete website, and all posted and associated content indicates your
acceptance of BioOne’s Terms of Use, available at www.bioone.org/terms-of-use.

Usage of BioOne Complete content is strictly limited to personal, educational, and non - commercial use.
Commercial inquiries or rights and permissions requests should be directed to the individual publisher as
copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit
publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to
critical research.

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 Journal of Coastal Research 32 1 46–55 Coconut Creek, Florida January 2016

Analysis of Ocean Current Observations in the Northern

Veracruz Coral Reef System, Mexico: 2007–12
Marı́a E. Allende-Arandı́a†, Jorge Zavala-Hidalgo†*, Rosario Romero-Centeno†,
Adriana Mateos-Jasso†, Juan M. Vargas-Hernández‡, and Luis Zamudio§
† ‡
Centro de Ciencias de la Atmósfera Instituto de Ciencias Marinas y Pesquerı́as
Universidad Nacional Autónoma de México Universidad Veracruzana
Coyoacán, DF 04510, México Boca del Rı́o, Veracruz 94290, México
§
Center for Ocean-Atmospheric Prediction Studies
Florida State University
Tallahassee, FL 32306, U.S.A.

ABSTRACT
Allende-Arandı́a, M.E.; Zavala-Hidalgo, J.; Romero-Centeno, R.; Mateos-Jasso, A.; Vargas-Hernández, J.M., and
Zamudio, L., 2016. Analysis of ocean current observations in the northern Veracruz Coral Reef System, Mexico: 2007–12.
Journal of Coastal Research, 32(1), 46–55. Coconut Creek (Florida), ISSN 0749-0208.

The Veracruz Coral Reef System (VCRS) is a marine protected area located on the continental shelf of Veracruz, in the
SW Gulf of Mexico, that is formed by two reef conglomerates: one to the north, in front of the Port of Veracruz, and one to
the south, in front of the Antón Lizardo tip. To characterize the currents and their variability for the northern reefs of the
VCRS, ocean currents, meteorological parameters, and sea-level observations for the period 2007–12 were analyzed. The
temporal variability of the currents was dominated by 2- to 10-day lasting events associated with the atmospheric
synoptic conditions through the forcing of the along-coast wind-stress component over the Tamaulipas-Veracruz shelf.
During autumn–winter, strong episodic southeastward currents prevailed, reaching more than 1 ms1, and they were
associated with the intrusion of atmospheric high-pressure systems into the Gulf of Mexico. However, some episodes of
strong southeastward currents observed in spring–summer were forced by low-pressure systems (tropical storms). From
May to August, weak northwestward currents were predominant, with speeds lower than 0.40 ms1. Annually,
southeastward (northwestward) currents were observed 45–60% (40–55%) of the time, with an average of 55% (45%).
Tidal currents are one order of magnitude weaker than the wind-driven currents. A multiple linear regression model for
the along-shelf currents, which is based on the wind stress, explains most of the observed variance and performs well in
reproducing the autumn–winter strong southeastward events but underestimates those observed in summer.

ADDITIONAL INDEX WORDS: Shelf currents, local and remote forcing, western Gulf of Mexico, coral reefs.

INTRODUCTION suggest that the circulation in the VCRS has a seasonality,

The Veracruz Coral Reef System (VCRS) is a 400-km2 with average southeastward currents from September to
marine-protected area, rich in flora and fauna, located over March and northwestward currents from May to August, while
the continental shelf of the SW Gulf of Mexico (GoM) in the April is a transition month (Morey, Zavala-Hidalgo, and
central part of the state of Veracruz. The VCRS comprises O’Brien, 2005; Zavala-Hidalgo, Morey, and O’Brien, 2003).
islands, islets, and two groups of reefs: one formed by 11 coral This seasonality has also been observed in sea surface
reefs located to the north, in front of the Port of Veracruz and temperature satellite images by means of a persistent summer
Boca del Rı́o City, and the other formed by 12 coral reefs located upwelling occurring along the Tamaulipas-Veracruz shelf
to the south, in front of the Antón Lizardo municipality (Figure (Zavala-Hidalgo et al., 2006). A few studies on the VCRS
1). The length of the coral reefs varies from approximately 0.3 circulation have focused on time scales different from the
km to 3.2 km, and their areas range from 1 km2 to 19 km2 (Lara seasonal scale. Salas-Pérez and Granados-Barba (2008), based
on a bibliographic review of the hydrodynamics in SW GoM,
et al., 1993; Salas-Pérez and Granados-Barba, 2008). The
discussed the local and basin-wide scale forcing and suggested
northern group of reefs, which is the study area of this
that the VCRS may be forced by tides, winds, and eddies
research, is located between the coast and the 40-m isobath; it
located along the slope because the shelf is relatively narrow
is characterized by a complex ocean floor topography consisting
there. Based on high-frequency current measurements along
of shallow sand banks, reefs, and islands, forming a series of
the shelf break of the western GoM, Dubranna et al. (2011)
channels with varying sizes and depths (Figure 1).
showed the presence of coastal-trapped waves generated in the
Previous studies, based on numerical simulations forced with northern shelf of the state of Tamaulipas. They also showed
monthly climatological winds that include the entire GoM, that in addition to the along-coast wind-stress forcing, eddies in
the Campeche Bay affect the external shelf circulation, a
DOI: 10.2112/JCOASTRES-D-14-00148.1 received 1 August 2014; feature that was also observed by Salas-Pérez et al. (2012).
accepted in revision 19 February 2015; corrected proofs received
7 April 2015; published pre-print online 30 April 2015.
It is particularly important to better understand the
*Corresponding author: jzavala@atmosfera.unam.mx circulation in the VCRS because, regarding its area coverage,
Ó
Coastal Education and Research Foundation, Inc. 2016 it ranks second among the coral reef areas in the Mexican

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 Analysis of Current Observations in Veracruz, Mexico 47

Figure 1. Location of the Veracruz Coral Reef System in the SW Gulf of Mexico. The black square indicates the location of the tide gauge at Veracruz city, and the
circles indicate the location of the current meters at the northern group of reefs. Isobaths every 10 m are shown, and reefs are shaded.

shelves, surpassed only by the Mesoamerican Barrier. The tidal current components are computed, and the relation of the
VCRS is a region of great scientific interest for its high currents with the local and remote winds is revised.
ecological diversity and its connectivity with other coral reef
areas of the southern and western GoM (Sanvicente-Añorve et METHODS
al., 2014), besides its importance for the touristic and economic A combination of in situ observations and numerical
sectors of the country; however, the VCRS is under strong simulations is used to describe and analyze the currents and
pressure from the activities of the city and the Port of Veracruz their variability in the NVCRS. Measurements of ocean
and the discharge of several rivers. The city discharges polluted currents, sea level, and atmospheric variables were used
waters into the area, there is a project to extend the Port, and together with reanalysis data from the North American
the discharging rivers carry an excess of sediments from the Regional Reanalysis (NARR, 2014) and numerical simulations
deforested areas upstream. To better manage and preserve the from the Weather Research and Forecasting (WRF, 2014)
VCRS, improving the knowledge of the circulation in the area is model.
an urgent need. Results from this study may also be useful to Current Meter Data
validate and tune numerical models that will be implemented Ocean currents were measured with a 600-kHz Nortek
for the VCRS. Aquadopp Acoustic Doppler Current Profiler (ADCP) that was
The purpose of this study is to describe and analyze the deployed at a site located between the Anegada de Adentro and
variability of the intensity and direction of the observed Isla Verde reefs at a depth of approximately 30 m and 8 km
currents in the northern VCRS (NVCRS) and explain the offshore (Figure 1). Initially, ADCP measurements were also
origin of that variability at different temporal scales, including taken between La Gallega and Galleguilla reefs at a depth of 15
diurnal, synoptic (2–10 days), intraseasonal, seasonal, and m and 1.2 km offshore, a site that is located about 8 km from the
interannual scales. To do this, six years of ocean current other ADCP. Simultaneous measurements at both locations
observations from a current profiler are analyzed. Also, the were obtained during two relatively short periods, September–

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 48 Allende-Arandı́a et al.

ms1, 0.100 ms1, 0.050 ms1, and 0.059 ms1, respectively.

Therefore, most of the analysis was performed using the
rotated axes along the direction of maximum variance.

Atmospheric Data
The NARR (2014) 10-m wind data for the period 2007–12
were used to calculate the along-coast wind-stress component
(ACWSC). These reanalysis data have a spatial resolution of
~32 km and a 3-hour temporal resolution. The wind-stress
components, sx and sy, were parameterized as follows:

sx ¼ CD qūu;

sy ¼ CD qūv;

where CD is the drag coefficient, q is air density (1.23 kg m3),

and ū ¼ (u, v) is the horizontal wind vector. CD is variable and
depends on the wind speed; for low wind speeds its value is
around 1.1 3 103, while for wind speeds greater than 6 ms1, a
Figure 2. Scatter diagram of the meridional vs. zonal components of the linear relationship of the form CD ¼ (0.61 þ 0.063ū) 3 103 is
currents (ms1) (dots) and the average velocity (arrow) for 2007–12. Rotated
established (Gill, 1982).
axes along the direction of maximum variance (called the principal direction)
are also plotted. The ACWSC was computed using basic geometry by
obtaining the inclination angle of a smoothed coast (a ¼
143.58) and then the unit vector in that direction, which was
October 2006 and December 2006–January 2007, and the used to project the wind stress. The obtained series of the
correlations between the low-pass filtered series of the ACWSC were filtered using the same criteria as for the ocean-
vertically averaged along-coast current component were current data.
computed. The correlation values for the two mentioned Numerical simulations with the WRF model were used in the
periods were very high, 0.92 and 0.96, respectively, and the
analysis of the influence of tropical cyclones in the area. These
amplitude of the variations was also very similar. Therefore,
simulations are from the operational numerical weather
the measurements at the Anegada de Adentro-Isla Verde site
forecast of the Centro de Ciencias de la Atmósfera of the
were considered representative of the currents in the NVCRS
Universidad Nacional Autónoma de México (CCA-UNAM,
area.
2015).
The ADCP sampling frequency was 30 minutes, with 2-m-
Wind, relative humidity, air temperature, and atmospheric
wide cells and a blanking distance of 0.50 m (the distance above
pressure data from a meteorological station located on the
the instrument without observations). Current measurements
Veracruz city shoreline (19811 0 31 00 N, 9687 0 24 00 W) were also
were collected along 1837 days from 2007 to 2012, which
used for the analysis. These data, available for a short period in
represent ~84% of the total number of days in that six-year
March 2010, had a temporal resolution of 10 minutes and were
period. Of these days, 136 (37%) correspond to 2007, 301 (82%)
provided by the Servicio Mareográfico Nacional (SMN, Mexi-
to 2008, 304 (84%) to 2009, 365 (100%) to each of the years 2010
can Sea Level Service) of UNAM (SMN-UNAM, 2015). The
and 2011, and 366 (100%) to 2012.
temperature and relative humidity sensor is a Rotronic HC2-S
Ocean-current data were filtered to eliminate tidal and
with ranges from 40 to 608C and from 0 to 100%, respectively.
inertial frequencies by applying a Lanczos low-pass filter with a
The pressure sensor is a Setra 278 with a range from 800 to
36-hour cut-off frequency. The direction of maximum variance
1100 mb, and the wind sensor is a Gill Windobserver 70 with a
of the currents (hereafter called principal direction [PD]) was
range from 0 to 70 ms1. A low-pass filter was applied to the
determined from the filtered data following Emery and
meteorological data using the same criteria as for the current
Thompson (2001). In the principal axes system, the variations
data.
are maximized in one component (the PD) and minimized in the
orthogonal component (the secondary direction [SD]). For the Sea-Level Data
purpose of describing the dynamics, the PD can be considered Sea-level data for Veracruz were also provided by the SMN-
as the along-coast direction because both are practically UNAM for 2007–12. Float (OTT Thalimedes) and radar (OTT
coincident. The scatter diagram of the vertically averaged Kalesto) sea-level sensors are located at the same site as the
currents (zonal vs. meridional components) for the whole period meteorological station, and data were sampled every 10
shows that most of them were oriented in a nearly NW-SE minutes.
direction (Figure 2). Henceforth, northwestward and south-
eastward directions are considered parallel to the PD. RESULTS
Annually, the PD describes 96.8%, 97.2%, 96.6%, 96.1%, In general, the variability of the currents was characterized
95.2%, and 94.1% of the observed variance in years 2007 to by episodes that lasted from 2 to 15 days and with an
2012, respectively; furthermore, the mean speeds of the alternating direction along the PD (either southeastward or
southeastward currents were 0.146 ms1, 0.082 ms1, 0.053 northwestward). The duration of the episodes had a peak in the

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 Analysis of Current Observations in Veracruz, Mexico 49

Figure 4. (a) Magnitude of the vertically averaged currents along the

principal direction in the NVCRS during March 2010. Meteorological
variables measured during the same period in Veracruz city: (b) along-coast
Figure 3. Currents in the water column along the principal direction in the wind-stress component, (c) relative humidity, (d) air temperature, and (e)
NVCRS 2007–12. Positive (negative) values indicate southeastward (north- sea-level pressure. Shading indicates the occurrence of norther events. The
westward) direction. black line represents the low-pass filtered series.

interval from 2 to 4 days, which includes 38% of the cases. temperature and relative humidity dropped, and the atmo-
Episodes lasting up to 10 days include 90% of the cases, and spheric pressure increased.
those lasting up to 15 days include 97.5% of the cases. The All intense ocean currents, considered as those episodes
event that lasted longest was 30 days. when currents along the PD attained a magnitude equal or
The vertical structure of the currents along the PD for the greater than 0.7 ms1, were caused by northers and lasted from
period 2007–12 is shown in Figure 3. During most of the 1 to 4 days. In the 6 years of observations, 22 of these events
sampling period, the currents in the water column showed a occurred (recalling that there are some gaps in the time series),
weak vertical shear, with the same direction and similar with an average of 3.7 per year but with a large interannual
magnitude from surface to bottom. When currents were variability (Figure 5). These events occurred during approxi-
mately 1% of the period of observation.
intense, they had the same direction in the entire column,
Winds from the NARR for a site near the location of the
but, in general, they were more intense in the upper layers.
current meter were used to analyze the wind-stress contribu-
Therefore, most of the analysis is carried out using the
tion to the ocean currents. The time series of the local ACWSC
vertically averaged currents.
over the 25-m isobath (estimated from the NARR wind data)
Ocean Current Response to Synoptic-Scale Winds and the currents along the PD are compared in Figure 5, and
It has been reported that prevailing winds in the western their monthly and annual correlations are shown in Table 1.
GoM are easterlies/northeasterlies in autumn–winter and Strong wind events caused by the northers in autumn–winter
easterlies/southeasterlies in spring–summer (Gutiérrez de had associated intense southeastward currents in the NVCRS,
Velasco and Winant, 1996). During autumn–winter, frontal and the correlation values between the ACWSC and the
currents along the PD were high.
systems cross the GoM causing a drop in temperature, an
During the summer months, when winds are relatively
increase in atmospheric pressure, and strong northerlies
weak, the correlations reduce, and, in some cases, they are
(DiMego, Bosart, and Endersen, 1976; Schultz et al., 1997).
not significant (Table 1). The importance of remote forcing
The cold fronts collide with moist air masses, generating
was estimated from the correlations between the local
cloudiness and heavy rainfall. The winds generated by these
currents and the ACWSC over the Tamaulipas-Veracruz
systems, so-called northers, which occur mainly from Septem- (TAVE) shelf (from the Mexico-U.S. border to the VCRS) and
ber to April, can be very intense, reaching speeds above 20 over the Texas-Louisiana (LATEX) shelf (Table 1). These
ms1. computations were performed looking for the possible
The response of the currents and weather conditions to influence of coastal-trapped waves on the NVCRS currents.
northers in the NVCRS is illustrated, as an example, for a The correlations of the currents in the NVCRS with the
period during March 2010 (Figure 4). The ACWSC strongly ACWSC over the TAVE shelf were similar to those obtained
increased at the arrival of the norther, and, a short time later, when the local ACWSC was considered and were smaller
the currents strengthened, peaking later and lasting longer when the ACWSC over the LATEX shelf was considered.
than the strong winds. During the norther events, the Results show that correlations for individual months were

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 50 Allende-Arandı́a et al.

A response lag exists between the currents and surface-wind

stress in the NVCRS, which peaked after 7.5 hours, 15.0 hours,
and 25.0 hours with respect to the local, TAVE, and LATEX
winds, respectively. The lowest correlations were obtained for
May–August, indicating that other processes, such as off-shelf
eddies (Dubranna et al., 2011), could be important forcings
during this period of the year. Other processes such as the
along-shelf pressure gradients may be important, as has been
reported for other regions (Lentz, 2008).

Vertical Structure
There were events, mainly during spring–summer (April–
September), when currents in the upper layer were northwest-
ward, while in the lower layer they were southeastward,
although some isolated events also occurred in other months
(Figure 5). Ocean currents were weak during all of these
events, 51 in total, and they lasted as much as 12 days, with a
mean of 3.5 days and a standard deviation (STD) of 2.3 days.
These events happened, in some cases, when local wind stress
was favorable to northwestward currents.
There were also periods when the ocean currents in the upper
Figure 5. Time series of the 36-h low-pass filtered vertically averaged
layer were southeastward, while in the lower layer they were
currents along the principal direction (green) and the along-coast wind-
stress component (red) for 2007–12. Positive (negative) values indicate northwestward. These events also happened when currents
southeastward (northwestward) direction. Black dots indicate events of were weak and were observed mainly during summer–autumn
intense currents during autumn–winter. Purple (pink) shading indicates (July–November), although there were some in other months
periods when currents in the upper (lower) layer had a northwestward (March 2010, May 2008 and 2012) (Figure 5). There were 31
(southeastward) direction and in the lower layer had a southeastward
(northwestward) direction.
events identified with these characteristics, with a mean
duration of 2.7 days and a STD of 1.6 days; the largest event
lasted 9 days. During these events, winds were weak,
variable, suggesting that remote forcing has an important frequently with a southeastward along-coast component. In
contribution on the NVCRS currents, as shown by Dubranna many cases, the events occurred after a period of northwest-
et al. (2011) who found a contribution of coastal-trapped ward currents, which suggests that the upper layer responds
waves in the western GoM currents. faster to the local wind, generating the observed shear. In other

Table 1. Monthly and annual correlations between the current along the principal direction, and the along-coast wind stress component (ACWSC) measured
at the NVCRS; the mean ACWSC along the 25-m isobath over the Texas-Louisiana shelf (LATEX), and the mean ACWSC along the 25-m isobath over the
Tamaulipas-Veracruz shelf (TAVE). Only months with more than 50% of measurements were considered for the calculations (see Table 2). Values in boldface
indicate significant correlation at the 95% confidence level.

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual
2007
LATEX 0.67 — — — — — — — 0.55 0.72 0.47 0.32 0.60
TAVE 0.90 — — — — — — — 0.33 0.80 0.54 0.75 0.79
NVCRS 0.88 — — — — — — — 0.43 0.77 0.65 0.80 0.73
2008
LATEX 0.57 0.13 0.17 0.07 0.40 — 0.06 0.39 0.10 0.62 0.15 0.15 0.22
TAVE 0.76 0.60 0.71 0.71 0.58 — 0.26 0.27 0.68 0.76 0.64 0.73 0.68
NVCRS 0.71 0.77 0.83 0.86 0.54 — 0.57 0.33 0.73 0.85 0.65 0.81 0.68
2009
LATEX 0.35 0.08 0.13 0.17 0.74 — 0.77 — 0.33 0.41 0.56 0.47 0.36
TAVE 0.84 0.76 0.73 0.78 0.86 — 0.82 — 0.55 0.57 0.84 0.76 0.77
NVCRS 0.87 0.56 0.73 0.77 0.89 — 0.45 — 0.70 0.46 0.75 0.70 0.65
2010
LATEX 0.49 0.50 0.10 0.23 0.13 0.61 0.23 0.17 0.65 0.76 0.49 0.32 0.42
TAVE 0.87 0.64 0.74 0.65 0.39 0.44 0.03 0.45 0.76 0.76 0.67 0.73 0.70
NVCRS 0.86 0.70 0.84 0.73 0.28 0.61 0.43 0.25 0.86 0.78 0.79 0.71 0.70
2011
LATEX 0.46 0.62 0.22 0.43 0.47 0.48 0.22 0.02 0.25 0.09 0.30 0.34 0.31
TAVE 0.80 0.89 0.80 0.76 0.83 0.53 0.14 0.20 0.71 0.71 0.76 0.76 0.77
NVCRS 0.82 0.89 0.75 0.79 0.85 0.44 0.35 0.52 0.80 0.79 0.78 0.79 0.72
2012
LATEX 0.11 0.62 0.34 0.35 0.34 0.71 0.38 0.31 0.37 0.51 0.49 0.40 0.36
TAVE 0.71 0.76 0.75 0.59 0.48 0.28 0.34 0.45 0.34 0.66 0.60 0.79 0.62
NVCRS 0.79 0.75 0.69 0.53 0.57 0.27 0.19 0.57 0.43 0.71 0.52 0.79 0.58

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 Analysis of Current Observations in Veracruz, Mexico 51

Seasonal Variability
The ocean currents in the NVCRS have a strong seasonal
modulation, with events of intense currents from September to
April and weak to moderate currents from May to August (see
Figures 3 and 5). During summer, the currents persisted for
longer periods with the same direction. Some events of
relatively strong southeastward currents during June–July
2010 are subsequently discussed.
The annual frequency distributions of currents along the PD
show that southeastward currents attained speeds above 0.50
ms1 around 5% of the time, reaching occasionally more than
1.0 ms1 (Figure 6). Northwestward currents were weaker,
mainly with speeds below 0.20 ms1 and maximum intensities
of about 0.40 ms1. The frequency distribution of currents along
the PD for the whole period of observations shows a maximum
in the 0.15 to 0.00 ms1 interval and a positive skewness
toward values in the southeastward direction (Figure 6).
The average monthly percentages of occurrence of south-
eastward currents (Table 2), computed with respect to the
corresponding number of days with measurements for 2007 to
2012, show that they were more frequently observed from
September to February and less often from March to August,
except in June when they occurred more than 50% of the time.
The frequency of occurrence of southeastward currents has a
peak in September and the northwestward currents in April
Figure 6. Annual frequency distributions of the ocean current magnitude
(ms1) along the principal direction from 2007–12 (upper panels) and the
(Table 2).
frequency distribution for the whole period of observations (bottom panel).
Positive (negative) values indicate southeastward (northwestward) currents.
Interannual Variability
There is a strong interannual variability in the NVCRS,
which is evident in the different percentages of occurrence of
cases, the events occurred in the middle of a period with southeastward currents from one year to the next (Table 2).
southeastward favorable winds, which may be caused by When a 36-hour low-pass filtering is applied to the currents
upwelling coastal-trapped waves that drive the subsurface along the PD, the percentage of occurrence changes by less
local currents in the opposite direction to the weak local winds. than 2% as a consequence of the relatively low contribution of

Table 2. Monthly and annual number of days with ocean current measurements and percentage of occurrence of southeastward currents (the component along
the direction of maximum variance) computed with respect to the corresponding number of days with measurements for 2007–12. The global monthly and
annual percentages are also included.

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual
2007
NDM 27 — — — — — — 2 30 20 30 27 136
PO 70.0 — — — — — — 68.2 69.9 82.1 67.2 41.4 65.4
2008
NDM 31 18 30 30 15 9 31 30 27 31 18 31 301
PO 62.1 58.3 50.5 34.8 82.3 78.8 44.0 41.9 78.6 96.2 25.9 62.9 53.5
2009
NDM 31 10 27 30 19 3 31 31 30 31 30 31 304
PO 51.6 30.7 35.1 24.4 12.8 0 7.7 62.2 82.4 42.6 78.6 79.9 48.0
2010
NDM 31 28 31 30 31 30 31 31 30 31 30 31 365
PO 58.0 95.8 53.6 49.4 45.8 59.3 56.8 48.3 79.6 58.3 58.2 56.1 59.6
2011
NDM 31 28 31 30 31 30 31 31 30 31 30 31 365
PO 43.2 44.2 29.2 23.5 31 47.4 39.1 52.7 42.4 66.4 59.8 58.5 44.8
2012
NDM 31 29 31 30 31 30 31 31 30 31 30 31 366
PO 61.2 73.1 39.5 46.7 64.4 68.1 70.6 32.5 75.1 42.7 51.4 43.7 55.6
Global
NDM 182 113 150 150 127 102 155 156 177 175 168 182 1837
PO 57.4 65.5 41.7 35.7 38.5 53.2 43.6 47.7 71.2 63.6 59.1 57.4 53.4

NDM ¼ Number of days with measurements.

PO ¼ Percentage of occurrence.

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 52 Allende-Arandı́a et al.

Figure 7. Winds at 10 m (barbs in km h1) and sea-level pressure contours (every 2 hPa) from the Weather Research and Forecasting model simulations for (a) 25
June 2010 (1100 local time), (b) 7 July 2010 (1500 local time), and (c) 18 July 2010 (0900 local time). Color contours indicate wind intensity according to the color
bar scale (where C1 to C5 stand for hurricane categories).

the higher frequencies (daily or higher). Southeastward meteorological events occurred during summer 2010, which
currents occurred from 44.8% in 2011 to 59.6% in 2010 (2007 generated intense southeastward currents that reached up to
is not considered because of the large amount of missing data), 0.50 ms1. To analyze these wind events, numerical simula-
with an average of 53.4% for the whole period, noting that there tions using the WRF model were examined. The intense
were about 83% of measurements in 2008 and 2009 (Table 2). northerly winds were better reproduced by the WRF simula-
The variability of the occurrence of southeastward currents tion than by the NARR data, probably because of its higher
is larger when individual months are compared. For instance, resolution and better representation of the topographic
in October 2008, these currents were observed 96.2% of the features of the Sierra Madre mountain range, inshore of the
time; however, in 2009 and 2012, they occurred around 43% of VCRS. The first event was a tropical cyclone that became
the time (Table 2). Also, February is a month with a large hurricane Alex and lasted from 25 June to 1 July (NASA,
interannual variability. 2010a). On the 25 June, relatively strong northerly winds along
Ocean currents in the NVCRS also showed large variations the coast of Veracruz developed as a consequence of the
in the summer months. In July 2012, for instance, the atmospheric cyclonic circulation associated with the storm,
percentage of occurrence of southeastward currents was as which was centered on 168N, 848W and extended from the
high as 70.6%, while in 2009, it was only 7.7% (Table 2). In western Caribbean to the SW GoM (Figure 7a). The atmo-
autumn–winter, dominated by northers and strong currents, spheric low-pressure system became a tropical storm on 26
as well as in summer, characterized by weaker winds, currents June, reaching the Yucatan Peninsula and the GoM by the 28
showed large variability. June. On 30 June, the storm strengthened into hurricane Alex,
Intraseasonal Variability moving toward northern Tamaulipas, where it made landfall.
During summer, there were episodes of relatively strong It was observed that the strongest currents in the NVCRS
southeastward currents that lasted for several days. Three developed on the 25 and 26 June, before the system reached

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 Analysis of Current Observations in Veracruz, Mexico 53

Table 3. Diurnal and semidiurnal tidal constituents of sea level (SL) and can be associated with the sea breeze, although the energy in
currents (CU) along the principal direction. The standard deviations this direction is much smaller. For this computation, the data
(STDs) were computed from the annual estimations for the years 2007–12.
without filtering was used (not shown).
SL (3102 m) CU (3102 ms1)
Period (h) 6 STD 6 STD DISCUSSION
Diurnal Nearly half of the time, the currents in the NVCRS flow in
K1 23.93 16.00 6 0.83 0.66 6 0.48 each of the directions along the PD (southeastward or
O1 25.82 16.51 6 1.26 0.37 6 0.20 northwestward), but the mean currents are southeastward.
P1 24.07 4.73 6 0.38 0.66 6 0.40 The duration of episodes with intense southeastward
Semidiurnal
M2 12.42 8.64 6 0.14 0.14 6 0.12
currents during autumn–winter was different from year to
S2 12.00 2.17 6 0.07 0.33 6 0.22 year (see Figures 3 and 5) because of the different duration of
N2 12.66 2.18 6 0.11 0.11 6 0.08 the northers that penetrated into the GoM. The interannual
variability of intense southeastward currents during summer
hurricane strength, as a consequence of the large atmospheric is associated with the number of tropical storms and easterly
cyclonic circulation that dominated wind patterns in the SW waves whose trajectories affect the VCRS. The mean current,
GoM. although it is a limited indicator, corroborates an important
The second event of strong southeastward currents was interannual variability because it was 0.10 ms1 in 2010 and
observed during 7–13 July 2010, caused by a low-pressure around 0.05 ms1 in 2011. Therefore, meteorological events
system that crossed the Yucatan Peninsula and became modify the winds in the VCRS, particularly along the coast of
Tropical Depression No. 2 (Figure 7b) (NASA, 2010b). The Veracruz, generating persistent northerlies/northeasterlies
system crossed the GoM following a SE-NW path, offshore of and causing changes in the direction and intensity of the
Veracruz city, and generated northeasterlies along the western currents (see Figures 5 and 7).
GoM that caused intense southeastward currents in the The seasonality of the low-frequency currents reported by
NVCRS for several days (see Figure 5). Zavala-Hidalgo, Morey, and O’Brien (2003) was corroborated,
Another low-pressure system over the western Caribbean but a large interannual variability exists; the frequency of
and the Yucatan Peninsula, with an associated extended southeastward currents varies from 35.7% (April) to 71.2%
cyclonic circulation, was observed on 17 July. At the same (September), with a large observed interannual variability (see
time, a low pressure over the Gulf of Tehuantepec favored the Table 2). Results from this study show that there is a high
development of northerlies over the VCRS (Figure 7c). The probability of occurrence of currents in both directions in any
situation was similar to that observed before the development time of the year, which is important for the advection of low-
of hurricane Alex, but this system did not reach hurricane or salinity and low-temperature waters from the NW shelves of
tropical storm category; however, it generated northerly winds the GoM into the VCRS, favoring the presence of lower surface
and southeastward currents in the NVCRS during several temperatures and temperature inversions (Mateos-Jasso et al.,
days, although they were weaker than those associated with 2012), as well as population connectivity, larval transport, and
previous events. In June 2012, a low-pressure system over the pollution dispersion (James, 2002; Pineda, Hare, and Sponau-
Bay of Campeche also caused southeastward currents with gle, 2007; Sanvicente-Añorve et al., 2014). These results may be
similar patterns to those previously described. important to develop better strategies to reduce the environ-
These results show that during the tropical storm season, mental impact of the new Port, which is planned to be located in
easterly waves (atmospheric systems coming from the tropical the northern area of the NVCRS.
Eastern Atlantic) and tropical cyclones strongly affect the
A Model for the Currents in the NVCRS
circulation in the area for several days while they approach and
The high correlation between the measured currents along
cross the southern and western GoM. Usually, these events
the PD and the local and remote ACWS suggested a model for
occur one after another, following similar trajectories and
the currents in the NVCRS, considering stresses with a lag
causing a considerable impact on the NVCRS circulation,
with respect to the currents. The multiple linear regression
regardless of their intensity.
model, following Dzwonkowski and Park (2010), considers the
Tidal Currents and Diurnal Frequency local, TAVE, and LATEX along-shelf stresses, with lags
Tidal components were computed using the tide gauge data obtained from the lag correlations and running means of the
for sea level and currents measured with the current meter. stresses for a 36-hour window.
The most critical contributions to the tidal components of the The coefficients were computed for the 2012 data and then
sea level were the diurnal components, K1 and O1, while the applied to the 2007–12 series. Annual correlations between
semidiurnal component, M2, had around half of the amplitude modeled and observed currents were from 0.75 to 0.85, with an
of K1 and O1 (Table 3). The tidal contribution to the currents is average of 0.79. The modeled currents reproduce the strong
around 1 3 102 ms1, which is one or two orders of magnitude autumn–winter events, in most cases with a similar intensity,
smaller than the instantaneous currents. and the northwestward currents between norther events, but
A harmonic analysis was applied to the currents (following the summer events are underestimated (Figure 8). The simple
Emery and Thompson [2001]), which shows that along the PD model does not account for the contributions of the along-shelf
there is more energy in the subtidal periods (2–10 days), while pressure gradient and off-shelf influence of eddies over the
along the SD there is a peak in the diurnal frequency, which slope, which may be important in those events.

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 54 Allende-Arandı́a et al.

with the northers, but underestimated the events observed in

summer. There are not observations at different sites in the
cross-shelf direction; therefore, the cross-shelf circulation was
not analyzed in this study.

ACKNOWLEDGMENTS
The authors would like to thank the valuable comments
and suggestions from three anonymous reviewers that helped
to greatly improve this manuscript. Marı́a Elena Osorio-Tai
helped in the elaboration of the figures, and Octavio Gómez
Ramos helped with the numerical simulations outputs. The
Servicio Mareográfico Nacional from the Universidad Nacio-
nal Autónoma de México provided the sea-level data. This
work was funded by projects CNA-CONACYT 2004-02-24,
SEP-CONACYT 133299, PAPIIT 120408-3, and PAPIIT
IN116111.

LITERATURE CITED
CCA (Centro de Ciencias de la Atmósfera) of UNAM (Universidad
Nacional Autónoma de México), 2015. http://www.atmosfera.unam.
mx.
DiMego, G.J.; Bosart, L.F., and Endersen, G.W., 1976. An examina-
Figure 8. Time series of the vertically averaged currents along the principal tion of the frequency and mean conditions surrounding frontal
direction (black) compared with the modeled currents (gray) for 2007–12. incursions into the Gulf of Mexico and Caribbean Sea. Monthly
Weather Review, 104(6), 709–718.
Dubranna, J.; Pérez-Brunius, P.; López, M., and Candela, J., 2011.
CONCLUSIONS Circulation over the continental shelf of the western and
This study analyzes the ocean currents in the NVCRS, their southwestern Gulf of Mexico. Journal of Geophysical Research,
variability in different time scales and forcing mechanisms, 116(C8). doi:10.1029/2011JC007007
Dzwonkowski, B. and Park, K., 2010. Influence of wind stress and
using several years of measurements. Results show that discharge on the mean and seasonal currents on the Alabama shelf
currents in the NVCRS have high energy in the atmospheric of the northeastern Gulf of Mexico. Journal of Geophysical
synoptic time scale (2–10 days), with currents changing in Research, 115(C12). doi:10.1029/2010JC006449
intensity and direction in that frequency. Events of strong Emery, W.J. and Thompson, R.E., 2001. Data Analysis Methods in
Physical Oceanography. Amsterdam: Elsevier Science, 638p.
currents that last a few days occur in the area, and they are Gill, A.E., 1982. Atmosphere-Ocean Dynamics. Volume 30. San Diego,
mainly observed from September to March. These intense California: Academic Press, Inc., 662p.
currents have a southeastward direction, reaching speeds Gutiérrez de Velasco, G. and Winant, C.D., 1996. Seasonal patterns of
higher than 0.5 ms1 and, in a few cases, above 1 ms1; they are wind stress and wind stress curl over the Gulf of Mexico. Journal of
Geophysical Research, 101(C8), 18127–18140.
caused by the northers that penetrate into the GoM in James, I.D., 2002. Modelling pollution dispersion, the ecosystem and
autumn–winter. On the other hand, northwestward currents water quality in coastal waters: A review. Environmental Model-
occurred more frequently from April to August, and, on ling & Software, 17(4), 363–385.
average, they were weaker than southeastward currents, Lara, M.; Padilla, C.; Garcı́a, C., and Espejel, J.J., 1993. Coral reef of
Veracruz Mexico I. Zonation and community. Proceedings of the 7th
showing magnitudes lower than 0.40 ms1 most of the time.
International Coral Reef Symposium, Vol. 1, Mangilao, Guam:
The annual averages of the along-shelf currents showed that University of Guam Press, pp. 535–544.
from 45% to 60% of the time, the currents in the NVCRS had a Lentz, S.J., 2008. Observations and a model of the mean circulation
southeastward direction, but an important interannual vari- over the Middle Atlantic bight continental shelf. Journal of
Physical Oceanography, 38(6), 1203–1221. doi:10.1175/
ability occurred that became larger when individual months
2007JPO3768.1
were considered. During tropical storm season, atmospheric Mateos-Jasso, A.; Zavala-Hidalgo, J.; Romero-Centeno, R., and
easterly waves and tropical cyclones cross the southern GoM, Allende-Arandı́a, M.E., 2012. Variabiliaty of the thermohaline
generating northerly winds in the NVCRS and inducing structure in the northern Veracruz Coral Reef System, Mexico.
Continental Shelf Research, 50(51), 30–40.
relatively strong and persistent southeastward currents that
Morey, S.L.; Zavala-Hidalgo, J., and O’Brien, J.J., 2005. The seasonal
may last longer than those observed in autumn–winter. variability of continental shelf circulation in the northern and
The ocean currents are mainly wind driven, and the western Gulf of Mexico from a high resolution numerical model. In:
contribution of tides is very small. The contribution of the Sturges, W. and Lugo-Fernandez, A. (eds.), Circulation in the Gulf
of Mexico, Observations and Models. Geophysical Monograph
along-shelf pressure gradient in the VCRS was not reviewed,
Series. Washington, D.C.: American Geophysical Union, pp. 203–
and eddies may have a contribution to the observed currents 218.
when they are located over the slope, being that their influence NARR (North American Regional Reanalysis). 2014. http://nomads.
is noticeable when the wind forcing is weak. ncdc.noaa.gov/.
NASA (National Aeronautics and Space Administration), 2010a.
A simple model based on the along-shelf wind stresses
Hurricane Season 2010: Hurricane Alex (Atlantic Ocean). http://
reproduced most of the variability observed in the NVCRS, in www.nasa.gov/mission_pages/hurricanes/archives/2010/h2010_
particular the strong events during autumn–winter associated Alex.html.

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)
 Analysis of Current Observations in Veracruz, Mexico 55

NASA, 2010b. Hurricane Season 2010: Tropical Depression 2 (Gulf of SMN (Servicio Mareográfico Nacional) of UNAM. http://www.
Mexico). www.nasa.gov/mission_pages/hurricanes/archives/2010/ mareografico.unam.mx/.
h2010_02.html. Schultz, D. M.; Bracken, W.E.; Bosart, L.F.; Hakim, G.J.; Bedrick,
Pineda, J.; Hare, J.A., and Sponaugle, S., 2007. Larval transport and M.A.; Dickinson, M.J., and Tyle, K.R., 1997. The 1993 superstorm
dispersal in the coastal ocean and consequences for population cold surge: Frontal structure, gap flow, and tropical impact.
connectivity. Oceanography, 20(3), 22–39. Monthly Weather Review, 125(1), 5–39.
Salas-Pérez, J.J. and Granados-Barba, A., 2008. Oceanographic WRF (Weather Research and Forecasting), 2014. http://wrf-model.
characterization of the Veracruz reefs system. Atmósfera, 21(3),
org/.
281–301.
Zavala-Hidalgo, J.; Gallegos-Garcı́a, A.; Martı́nez-López, B.; Morey,
Salas-Pérez, J.J.; Salas-Monreal, D.; Monreal-Gómez, M.A.; Riveron-
Enzastiga, M.L., and Llassat, C., 2012. Seasonal absolute acoustic S.L., and O’Brien, J.J., 2006. Seasonal upwelling on the western
intensity, atmospheric forcing and currents in a tropical coral reef and southern shelves of the Gulf of Mexico. Ocean Dynamics, 56(3–
system. Estuarine, Coastal and Shelf Science, 100, 102–112. 4), 333–338.
Sanvicente-Añorve, L.; Zavala-Hidalgo, J.; Allende-Arandı́a, M.E., Zavala-Hidalgo, J.; Morey, S.L., and O’Brien, J.J., 2003. Seasonal
and Hermoso-Salazar, M., 2014. Connectivity patterns among coral circulation on the western shelf of the Gulf of Mexico using a high
reef systems in the southern Gulf of Mexico. Marine Ecology resolution numerical model. Journal of Geophysical Research:
Progress Series, 498, 27–41. doi:10.3354/meps10631 Oceans (1978–2012), 108(C12), 3389.

Journal of Coastal Research, Vol. 32, No. 1, 2016

Downloaded From: https://bioone.org/journals/Journal-of-Coastal-Research on 18 Oct 2022

Terms of Use: https://bioone.org/terms-of-use Access provided by Universidad Nacional Autonoma de Mexico (UNAM)