3 Bechtel Systems & Infrastructure Inc., San Francisco, U.S.A, anfindik@bechtel.com

Abstract: This paper presents a 3D hydrodynamic model established for the Maracaibo system, including Lake Maracaibo, Maracaibo Strait, Tablazo Bay and the Gulf of Venezuela. The model represents in detail the important features of the complicated system: the tide; wind set-up; the strongly stratified conditions in the Tablazo Bay, especially in the dredged narrow navigational channel connecting the Lake and the Gulf; and the weakly stratified Lake with a counter-clockwise circulation. The modeling periods using this system model made up a total of 16 months. Furthermore, a 10-year Lake simulation was performed to analyze the effect of interannual variations in the run-off to the Lake. A description of the processes governing the salt intrusions is given based on the simulations.

 

Keywords: 3D modeling, salt intrusions, semi-enclosed seas, estuarine circulation, Maracaibo, stratification

A 3D hydrodynamic and eutrophication model has been set-up for the Maracaibo system to analyze the present conditions of the system and to perform scenario assessments of the effects of potential future changes to the system. The model tool is part of an integrated study on environmental remediation options for the Lake (Findikakis et al., 2001). At present Lake Maracaibo suffers from anoxic conditions in the hypolimnion and high productivity in the surface waters. Furthermore, point source discharges, especially in the Strait, contaminate the water with BOD and coliforms.

The environmental quality of the Lake depends on a large number of physical, chemical, and biological processes in Lake Maracaibo itself and the adjacent bodies of water, i.e. Maracaibo Strait, Tablazo Bay and the Gulf of Venezuela (Herman de Bautista, 1997). To understand the present conditions and to predict changes resulting from specific future actions it is necessary to consider the Lake-Strait-Bay-Gulf system in its entirety. The major challenge in modeling this system is finding the proper way to deal with the broad range of spatial and temporal scales of its most significant processes. Due to the different time and space scales of the processes affecting the environmental quality in the Lake and the computational requirements for their proper simulation, a combination of linked models was used to meet the objectives of the study.

This paper deals with the results of the hydrodynamic modeling, which also act as a basis for the eutrophication modeling (Hansen et al., 2000b).

The objective of the hydrodynamic modeling was to establish a 3D-modeling tool to improve the understanding the governing hydrodynamic features of the system and to act as a basis for the subsequent eutrophication modeling. The steps taken to achieve this goal included:

• Setting up a hydrodynamic Regional Model and two reduced model set-ups, each of which covering only a part of the Regional Model domain.
• Calibrating the Regional Model against field data from 1997-1998, especially a dedicated field monitoring program in November 1998 (Horn et al., 2001a).
• Validating the model against monitoring data from 1999, especially a dedicated field program in March-April 1999 (Horn et al., 2001a).
• Running the Regional Model for a 12-month period December 1997 – November 1998 and a Lake Subset Model for a constructed 10-year period with changing hydrologic input to the Lake.

Mike 3 is a software package developed at DHI for unsteady 3D Newtonian fluids (Rasmussen et al, 1990). The model applies a Cartesian grid representation of the modeling domain. The modeling system includes a basic full 3D hydrodynamic (HD) module, a turbulence module and add-on modules for, for example, eutrophication assessments. The turbulence module offers different approaches, one of which, the mixed Smagorinsky (in the horizontal) and k-ε (in the vertical) closure scheme, was used for this study.

The challenge in establishing a Regional Model was the large range of spatial scales, which had to be taken into account. The horizontal dimensions of the Maracaibo system are 300km north-south and 250 km east-west. The maximum depth is over 60 m. Furthermore, the important navigational channel dredged from the Gulf through Tablazo Bay and into the Strait has a width of only about 250 m and a depth of about 13.6 m. To cope with this range in scales a fully dynamic nesting facility was applied with four embedded sub-domain. The domain covering the Gulf and the Lake Maracaibo had a horizontal grid resolution of 6750 m. Embedded into this domain was a 2250 m resolution domain as a transition to a 750 m resolution domain of the Strait. Finally, a 250 m resolution domain covered Tablazo Bay including the main part of the navigation channel, see Figure 1. The vertical grid spacing was 1.5 m, with the exception of the top layer, which extended from the actual water surface elevation (varying, but typically –0.5 m to +0.5 m) down to an elevation of –2.25 m. Furthermore, the orientation of the horizontal grid of the model was chosen to match the orientation of the main part of the navigational channel in order to represent the channel well.Three model setups were applied for the modeling:

• The Regional Model. This set-up was used to study the details in time and space of the exchange flow in the Bay and Strait driven by tidal, wind and freshwater discharge to the Lake, and the mixing conditions in the Lake. This model set-up was used for simulation periods of up to one year.
• A reduced Regional Model. This set-up only extended up to the southern part of the Gulf, and was used for detailed calibration of the Strait and Bay hydrodynamic settings.
• A Lake Subset Model. This model is a simple subset of the Regional Model, covering only the Lake. The Lake Subset Model was used to perform 10-year simulations. The Lake Subset Model simulations used as boundary conditions flows and salt fluxes into the Lake obtained from the MIKE 3 Regional Model simulations.

The models' bathymetry settings were based on digitized sea charts and a special bathymetric survey of the inlets from the Gulf to Tablazo Bay, conducted as part of the study.

The Regional model ran with 30 seconds time steps, however the update of the AD was only each 90 seconds. The Lake Subset Model ran with ? hour time steps.

The forces driving the hydrodynamic Regional Models include:

• Time series of water level variation along the open Gulf boundary between Punta Perret and Amuay. Prior to December 1998 these levels were generated based on tide constituents, whereas gauging data have been collected afterward this date.
• Information on typical open boundary salt and temperature conditions from literature.
• Time series of measured wind, air pressure and temperature distribution over the model area, mainly provided by the European Center for Midrange Weather Forecast (ECMWF).
• Time series of river discharges into the model domain, developed by CGR Ingeneria, based on river flow data and estimated flows were data were not available.
• Time series of precipitation distribution to the Lake. Climatological data applied.
• Information on typical cloudiness for the area.
• Information on typical humidity for the area.

The initial distribution of salt and temperature within the model domain was established based on hydrographic surveys in the area performed by the Instituto Para el Control y la Conservacion de la Cuenca del Lago de Maracaibo (ICLAM).

Figure 2 shows the hydrologic components of the simulation periods, with long-term averages of run-off into the Lake at 1627 m³/s, precipitation on the Lake at 549 m³/s and evaporation at 641 m³/s. Thus, the surplus freshwater in the Lake averages 1535 m³/s. The freshwater surplus is largest in May and in October-November and at minimum in January-March. The actual figures from the simulation periods are in good agreement with the statistical figures.

The reduced Regional Model applied data from the Malecon water level gauge, located near the inlet to Tablazo Bay, at the open boundary in the Gulf.

Two dedicated monitoring programs were conducted as part of the study: the wet season program October-November 1998 and the dry season program March-April 1999, see Horn et al. (2001b).

In addition, data from ICLAM’s quarterly water quality surveys from 1997-99, covering about 24 stations in the Lake, Strait and Bay, and from water level gauges maintained by the Instituto Nacional de Canalizaciones (INC) were used for the calibration and validation of the model set-ups.

The dedicated monitoring program October-November 1998 was used mainly to calibrate the barotropic and baroclinic features in the Bay and Strait. Figure 3 shows the agreement for the water level at the Maracaibo gauge in the middle of the Strait and discharge comparisons from the main inlet to the Bay. The results are from re-run with the full Regional Model. The tidal amplitude is reduced to 0.15m from about 0.5 m at Malecon.

The monitoring program included three ADCP recordings from the Lake. Figure 4 shows the comparison of the modeling results with data from the station in the NE part of the Lake. It should be noted that the wind forcing was only available as 4 hourly wind fields. Therefore, higher frequency fluctuations in the current can not be represented in the model. The current speeds in the calibration period match the general information on the counter-clockwise surface current in the Lake of a speed of about 0.2 m/s and reduced speeds at lower depths.

The validation period includes the dry season December 1998 – April 1999. The salt intrusions into the Lake typically occur during this season. However, in 1998-89 the available forcing data did not include the conditions that would create any significant intrusions in the model, with the only exception of a minor event in late January. The result became a slow vertical homogenization of the Lake profile up to the minor intrusion (Figure 5), after which the Lake almost mixed completely in the model. The monitoring data shows almost the same development up to the end of February, but in March 1999 the salinity at the bottom level increased a little again, possibly as a result of a minor intrusion not reflected in the available forcing data.

The profile comparison shown in Figure 5, just before the Lake became almost uniform vertically, displays the important column features of a uniform salinity in the surface layer extending to a depth of more than 10 m, followed by an almost constant salinity gradient down to 25 m. The bottom layer with about 2 psu of further elevated salinity is not represented in the model. The general underestimation of about 1 psu is probably due to a vertical displacement of the doming structure in the model at this time, as the surface salinity a month earlier and a month later are in good agreement.

The full year simulation December 1997 – November 1998 provides a good basis for analyzing the actual hydrodynamic development in the Maracaibo system. From Figure 6a it is seen that the intrusion events mainly occur in the dry season. The largest appear at the end of February, when the mean inflow to the Lake became about 6000 m³/s of water with an average salinity of 11 psu (peak salinity 18 psu), passing through the northern Lake boundary. In the wet season the saline wedge is arrested in the northern end of the Strait. Occasionally, fresh water overflow from the Lake enhances and pushes the saline wedge as far as the inlet from the Gulf.

The doming structure in the Lake was reproduced well by the model (Figure 7). Part of the time a secondary clockwise surface circulation develops in the southernmost part of the Lake, probably due to the freshwater input from the major rivers to the southwest.

The intensive intrusions of saline water plunge to the center of the Lake and replace the former bottom water at the center station, see Figure 6b. Figure 6b is from a 10-year simulation using the Lake Subset Model, developed to illustrate the effects of the variability in the freshwater input to the Lake. The forcing to this model includes, first the actual period December 1997 – November 1998, followed by historical records from the period December 1975 - November 1984. The forcing data at the open boundary at the exit from the Lake is taken from the Regional Model 1-year simulation, applying monthly subsets matching the run-off variations. The model reflects the variations in the run-off; 1977 had a low run-off and 1981 had a high run-off.

The calibration, validation and 1-year simulations, the simulated results for which can be compared to monitored values, constitute the basis for a joint evaluation of the capability of the modeling system to describe the governing processes and responses of the Lake Maracaibo system to external forcings.

The tidal elevations in the system and the flow rates at the mouth of the Bay are in agreement with the field measurements, confirming that the barotropic features of the system are described satisfactorily. The major problem regarding the water level variations is that no actual boundary data exist for the entire simulation period. For part of this period the water levels used to define the boundary condition to the model were constructed from the characteristics of the tidal constituents in this area. These water levels do not include the effect of other factors potentially affecting the water levels, such as the regional wind setup. Furthermore, the applied forcing wind fields have a 0.5 degree horizontal resolution (about 60 km), which is rather coarse for the present system with a large gradient in speed and direction between the northern and southern parts of the model domain.

The important saline intrusions are reproduced acceptably, with the exception of some minor events, probably due to less appropriate water level boundary data for the Gulf. The circulation in the Lake is reproduced well and is in agreement with the field measurements collected as part of the study and general information from past studies. In general, the salinity stratification in the Lake with intensification after an intrusion event and only a slow reduction in the stratification periods with no significant intrusions or mixing events agrees well with the monitoring data. Under certain conditions the Lake becomes almost fully mixed, as observed in February 1999. The simulated temperature stratification reflects the measurements, although the Lake in general becomes about 1-1.5oC too cold during the 1-year simulation (in the validation simulation with an adjusted air temperature forcing the match is fine).

The 10-year simulation of the response of Lake Maracaibo to interannual changes in the run off conditions is also considered reasonable. Furthermore, the Lake conditions simulated with the Regional Model suggest that the system is in a state of dynamic equilibrium, as the Lake conditions do not show any significant trends within the 10-year Lake Subset Model simulation. This is also a strong indication that the vertical diffusion in the model is calibrated appropriately. The time scale for water exchange in the Lake is about 5 years.

The analyses have shown that the important major saline intrusions occur as a consequence of a general depression of the daily averaged water level at the inlet from the Gulf, resulting in large outflow from the Lake and lowering of the water level in the Lake. The saline intrusions appear when the depression is followed by a significant increase in the water level in the Gulf due to a wind set-up, and the run off to the Lake is low and thus unable to fill up again the Lake. The model analyses suggest that the critical point for the intrusions is to reach the southern part of the Bay, where the navigational channel widens up. Once the saline water has reached this point, it eventually ends up in the Lake as a density current. Local mixing up in the surface water in the Strait can reduce the saline intrusion volume somewhat.

The ability of the established model system to describe the governing processes and responses of the Lake Maracaibo system to the external forcings was evaluated based on 16-month Regional Model simulations for the system (the calibration, validation and 1-year simulation).

The conclusion of the evaluation of the model performance is that the comprehensive aim of establishing a general 3D modeling system for Lake Maracaibo, with a high resolution in space and time for use for scenario assessment, was successfully achieved. The model provides a detailed description of the processes and responses in the Lake Maracaibo system with a high level of reliability.

The models reproduce the important features of the system, including tidal exchange, saline intrusion, stratification in the Lake due to salinity and temperature differences and doming in the Lake due to its counter-clockwise circulation. The outcome of this evaluation is crucial to the reliability of the model results and their use as a basis for eutrophication assessments and subsequently in predictive simulations under various assumptions for future conditions.

The work reported in this paper was part of the Integral Study for the Environmental Restoration of Lake Maracaibo, conducted for PDVSA by Bechtel International. Luis Delgado and Nelson Corrie managed the study for PDVSA. The authors wish to acknowledge the support of PDVSA and the constructive criticism of the initial model results by PDVSA’s Technical Committee overseeing the study. The comments by Dr. Reinaldo Garcia-Martinez of the Venezuelan Central University, are especially appreciated.

 

Findikakis, Angelos N., J?rg Imberger, Ian Sehested Hansen, Luis A. Delgado, Reinaldo García-Mantínez, and Erich Gundlach, 2001: A Study Of Environmental Remediation Options, For Lake Maracaibo, Venezuela, Proceedings XXIX IAHR Congress, Beijing, China, September 17-21, 2001.

Hansen, Ian Sehested, J?rgen Krogsgaard Jensen, and Angelos N. Findikakis, 2001b: 3 Dimensional Eutrophication Modeling of the Maracaibo System, Venezuela, Proceedings XXIX IAHR Congress, Beijing, China, September 17-21, 2001.

Herman de Bautista, Susana, 1997: Proceso de salinización en el Lago de Maracaibo, Instituto Para el Control y la Conservacion de la Cuenca del Lago de Maracaibo (ICLAM), Maracaibo.

Horn, David A., Peter Yeates, J?rg Imberger, and Angelos N. Findikakis, 2001a: One-Dimensional Water Quality Modelling of Lake Maracaibo, Proceedings XXIX IAHR Congress, Beijing, China, September 17-21, 2001.

Horn, David A., Bernard Laval, J?rg Imberger, and Angelos N. Findikakis, 2001b: Field Study of Physical Processes in Lake Maracaibo, Proceedings XXIX IAHR Congress, Beijing, China, September 17-21, 2001.

Rasmussen, E.B., H.J. Vested, P. Justesen and L.C. Ekebj?rg, (1990), SYSTEM 3: A Three Dimensional Hydrodynamic Model, Danish Hydraulic Institute.