Common use of Simulation of Salinity Clause in Contracts

Simulation of Salinity. In order to analyze the horizontal salinity variability during the simulation period, a series of data, sampled at 5-day intervals was extracted (Figures 2-13a through 2-13c). All data periods selected were near the slack before flood in order to remain consistent relative to tidal phase. On day 5, since the stage at the moment of the snapshot was near the beginning of flood for the mouth of Barataria Pass, maximum plume excursion from the mouth of the passes was observed. Inside the bay, lower salinities extended to the mouth of the basin. It can be seen easily that the high salinities are coming from the open boundary. It is notable that there are large gradients in the coastal waters due to the time and space variations of salinity imposed along the open boundary. On day 10, the situation is similar to that of day 5 except for the higher salinity values coming from the open boundary. By day 25, very high values of salinity are coming from the open boundary and penetrating far into the basin. There are big salinity gradients in the northeast region of the Gulf of Mexico in the model domain. On day 30, after several days of high salinity values at the open boundary, the high gradients increase at the coast. In shallow, bar-built estuaries, it is often assumed that the flow field is vertically homogeneous (▇▇▇▇▇▇▇, 1996). When the estuary is shallow, the velocity shear on the bottom may be large enough to mix the water column completely and make the estuary vertically homogeneous (▇▇▇▇, 1973). The concern over the use of a two-dimensional model for this modeling study is in its application to deep channels, waterways, or passes, such as the Barataria Pass. As mentioned earlier, the Barataria Basin has five passes that connect to the Gulf of Mexico. Except for Barataria Pass, all of the other passes are very shallow. Barataria Pass, however, is wide (about 1 km) and deep (deeper than 12 m in the middle of the pass) perhaps allowing some stratification to be present, i. e., salt wedge might exist along the bottom during flood tide. This phenomenon can not be accounted for in the model. Figure 2-14 shows observed and simulated salinity at eight stations. It is apparent that at all the stations simulated salinity can capture low-frequency salinity variability but higher-frequency signal is missed. This is due to the fact that the salinity initial condition was based on salinity observations at the eight stations, and it completely missed any smaller horizontal-scale variability that might have existed at that time. Subsequent tidal advection of the smaller-scale salinity variability cannot be simulated by the model as long as the salinity initial condition does not contain the signal. As far as the low frequency variability is concerned, the model appears to capture large portion of the signal at all the stations. After Day 16, the observed salinity at S1 shows significant increase apparently due to an intrusion of saltier water from the Gulf of Mexico. This salty intrusion propagates upstream reaching S2 after Day 25, and S5 soon after. Presently, the model is not capable of capturing the observed increases in salinity at S1 due to the present model open boundary set-up where its corresponding salty water has to propagate from the open boundary to S1. At station S5, the instrument error resulted in the loss of usable data at Figure 2-13a. Simulated horizontal salinity (ppt) distribution at Day 5 and Day 10 during the simulation period 7/71999 – 8/5/1999. All time slices selected were near the slack before flood. Figure 2-13b. Simulated horizontal salinity (ppt) distribution at Day 15 and Day 20 during the simulation period 7/7/1999 – 8/5/1999. All time slices selected were near the slack before flood. Figure 2-13c. Simulated horizontal salinity (ppt) distribution at Day 25 and Day 30 during the simulation period 7/7/1999 – 8/5/1999. All time slices selected were near the slack before flood. Figure 2-14. Observed (blue) and simulated salinity (red) at eight stations (listed in Figure 2-1). 56 the beginning of the record until Day 5. At stations further upstream, the overall results look better than those downstream due to the relatively short simulation period that did not allow upstream propagation of the salty intrusion coming from the Gulf of Mexico. One way to improve simulation is to assimilate observed salinity at the mouth of Barataria Pass, thus directly forcing the salinity at the mouth of the bay rather than at the offshore open boundary. Availability of better salinity initial condition with much higher spatial resolution can certainly contribute to improving model simulation. In summary, it is rather encouraging that a 30-day model simulation based on the salinity initial condition estimated from only eight stations can capture bulk of low-frequency signal. In order to describe tidal mixing of salinity and tidal plume characteristics, a series of 2- hour interval salinity snapshots over a tidal cycle were extracted and zoomed in on at the mouths of the basin (Figures 2-15a and 2-15b). The first snapshot, begins at hour 500, shows slack before flood. Outflow plumes are easily seen at the three passes. During this stage, the plumes reach maximum size. The largest plume, 3 km in diameter, extends from Barataria Pass. There are sharp salinity gradients along the frontal zone of the plume. A mushroom-like plume 2 km in diameter formed at the mouth of Quatre Bayou Pass. Although the size of the plume is slightly smaller than that from Barataria Pass, the salinity gradient is greater than that at Barataria Pass due to the lower salinity inside the bay. Although the flood has already begun, the scales of plume do not change significantly for the next several hours, while their shapes do change. However, inflow plumes are noticeable inside the bay by hour 506. At hour 508, the plumes from all three passes were clearly retracting into the basin under incoming flood tide and forming well-defined inflow plumes inside the bay. It should be noted that high salinity values are detectable along the left side of Barataria Pass during the flood, probably due to the alongshore salinity gradient immediately offshore. Unlike the outflow plumes, the inflow plumes spread much wider. By hour 512, most of the outflow plumes disappeared. At hour 518, inflow plumes reached their maximum length of 11 km. Some of this water remained in the bay over the entire cycle. After hour 520, the inflow plume began to move back out through the passes. In summary, the high resolution of the model allows detailed simulation of exchange and mixing processes near the main passes. It appears that mixing and exchange processes at the passes are very complex partly due to the highly complex morphological characteristics of the bay, and a very high model grid resolution is required in order to correctly simulate mixing and exchange processes at those passes.