Conduct effluent plume delineation: chapter 5

Receiving Environment

• Rivers
• Small Lakes and Impoundments
• Large Lakes
• Estuaries and Fjords
• Marine
• Climatic Conditions

Specific factors that should be considered in each type of receiving environment are outlined in Table 5.1, and described in this section. The influences of climatic conditions such as ice and wind on plume behaviour are described at the end of this section. This section also provides specific guidance for conducting plume delineations to supplement guidance provided in Section 3.

Table 5.1: Summary of factors to be considered for each type of receiving environment

Rivers

In rivers, currents are typically unidirectional, and flows and water levels are seasonally variable. Considerations for determining plume behaviour include the characteristics of the discharge point, shoreline and bottom attachment and rising or sinking of the plume due to density effects caused by thermal or chemical factors. The river may consist of different zones that should be delineated, including fast and slow flowing, with various depositional and erosional zones influencing any suspended solids transport from the effluent. A discussion on the temporal changes (seasonal and long term) of the river regime on the reaches affected by the plume should be provided.

Plume delineation studies should be conducted during a period that approaches the annual low river flow, which typically occurs in late summer. This will leave the extreme high and medium flows to be predicted using numerical models. Low river flows typically correspond with low overall dilution potential and reduced rates of turbulent mixing in the system. As a result, spatial extent of plumes tends to be larger during periods of low flow.

If a dye tracer is used, it should be added using a continuous flow rate injection system. The number of field tracer measurements to take will be site-specific. It is recommended that concentrations be measured at various distances downstream and that the spatial extent of the plume be determined. Slug tests (i.e., batch release of dye, as opposed to continuous flow rate tracer injection) may be a also a suitable technique for monitoring mixing in some rivers, but this generally requires a very good understanding of the plume behaviour and may require repeated slugs.

Small Lakes and Impoundments

Guidance for rivers (above) will apply if there are large effluent volume discharges, because there will typically be an easily discernable and measurable flow through the system. However, this guidance will not apply to some small lakes and impoundments where there are residual currents, or where the long- term drift is masked or even reversed by short term effects induced by factors such as wind shear, lake overturning, development of thermocline, or freshets, separately or in combination. It is important to assess the magnitude and significance of each of these comparatively short-term phenomena. This is most easily done using numerical modeling, provided there are adequate field data available for model calibration.

The residual flow, or even transitory currents, may be below the threshold of regular recording current meters. In these cases, a mass balance of the flows entering the body of water and the flows leaving will provide an estimate of the retention time in the system. This retention time may be reduced further by assessment of special circumstances or geometry. However, where currents are very low (e.g., a few cm/min), only field tracer studies can provide the confidence that an effluent will be carried away and that re-circulation does not take place. Re-circulation takes place where eddies from the plume are brought back to the mixing system and used instead of clean new receiving water. After some time this may significantly reduce the dilution of the effluent and significantly extend the 1% concentration boundary.

The residual flow through the area should be estimated as well as the currents that are induced by wind at various strengths and from various directions. An onshore wind will typically trap water against the shore and cause the surface water layer to thicken. Conversely, an offshore wind will cause deeper water to be brought inshore and upwell to replace the water being carried offshore by the wind.

The relative density of effluent and receiving water may be important to determine, because if a thermocline is present, most water movement in the lake will take place above it. If the plume stays above the thermocline, vertical mixing will be restricted and dilution will result more from horizontal mixing. Should the initial dilution be such that the plume is entrapped below the thermocline, mixing will likely be slower and movement could almost approach the volume being displaced by the effluent discharge.

Tracer studies should be designed to fulfill the general requirements and to consider seasonal variations. Prevailing currents will influence the duration and timing of each study. The recommended method for the delineation of the effluent plume for a lake discharge is the continuous flow rate injection system using a dye tracer. This is particularly true if there is any likelihood of eddies and re-circulation or entrapment back into the plume. The slug test can be used to determine dispersion characteristics, but does not give as good a visual picture, nor the same guidance regarding re- circulation.

To measure water currents, drogues should be released and tracked concurrent with dye tracing. For buoyant surface plumes, drogue vanes should be set with the upper edge at or just below the surface (within top 50 cm). For submerged plumes, drogues should be weighted and sails set so that they travel at the same initial depth of the plume, which can be determined on site from the initial tracer monitoring. If clusters of drogues are released, dispersion for a batch release may be obtained from the paths of the individual drogues by determining the variance of the individual drogues about the centroid.

Unlike the river receiving water where the plume configuration can be predicted using simple numerical models, plume behaviour in lakes is not as easily predicted. If data on water currents in the vicinity of the discharge are lacking, current meters (see Section 3.8.2) may be needed to obtain information for use in the numerical model. Climatic data may be obtained from the local airport or other local weather data source. The statistical relationship between the dilution and travel time of the effluent should be combined with the statistical characteristics of the water currents to develop spatial dilution zones around the discharge. This may be depicted as a graph showing the frequency of the dilution factor as a function of dilution and travel time.

The dilution zones should show the probability of the effluent plume being present at any receiving water location and the mean and standard deviation of the effluent concentration at any location in the receiving water. The dilution zones do not show the configuration of the effluent plume under specific current speed and direction conditions.

Large Lakes

Water currents in large lakes are often wind driven, seasonally variable, and generally less predictable that those encountered in fluvial and tidal areas. Thermal and density stratification is important to understand because the effluent may be more dense than lake water due to chemical factors, or less dense than lake water due to thermal factors. Wind and wave advection, seiches, shore-line and bottom attachment are all important. Many of the factors considered for small lakes and impoundments are also relevant to large lakes.

In large lakes, it is useful to have information on the variability in long-term water movement and the resulting residual movement. This information may be obtained using moored current meters with a low current threshold. Temperature recorders on the current meters may be used to determine whether internal waves or seiches occur. On shore anemometers may be used to obtain concurrent wind data if wind data are otherwise unavailable.

Initial dispersion of the surface plume into a relatively quiescent lake will be significantly influenced by the volume of the upwelling plume. Numerical modeling is the strongest tool to delineate the plume to this stage. Subsequent dispersion using numerical modeling will require estimates of horizontal and vertical dispersion coefficients, as well information about water currents obtained from current meters. The dispersion coefficients are best determined by either a slug release of dye or by continuous release and subsequent monitoring of a tracer in the effluent plume. If wind is present, an estimate of induced wind drift current can be determined using wind drift forecasting curves (see Section 5.6.2). Winds can also induce tilting of the water surface and of the thermocline, which may result in seiche motions that oscillate within the basin. See Section 5.6.2 for more discussion on wind-induced seiches.

Two-dimensional numerical modeling may be adequate, however in situations where there is stratification, three-dimensional modeling may be necessary.

Estuaries and Fjords

Estuaries and fjords are complicated environments in which to conduct plume studies. An understanding of the nature of the receiving waters is very important for planning the field study. This section will first provide a description of hydrographic considerations in estuaries and fjords, followed by guidance for conducting field studies.

• Hydrographic Considerations
• Conducting the Field Work

Hydrographic Considerations

The most significant influences on effluent dispersion in estuaries and fjords are tidal flows and density differences between freshwater and saltwater. Tides cause water to flow into an estuary on the rising tide and flow out again during the falling tide, causing mixing to take place. Effluents are typically similar in density to fresh water and will therefore tend to follow the fresh water in their mixing pattern. Freshwater has a specific gravity of about 1.00. Effluent and freshwater will typically rise and flow along the top of saltwater, which has a specific gravity of about 1.026.

Figure 5.1 depicts a "salt wedge estuary" in which a large river flows into an estuary with small tides and creates a sharp interface between fresh and salt water. Figure 5.1 (a) shows the fresh water layer thinning as the estuary widens and extends seaward. If the freshwater velocities get high, there will be shear created at the interface and salt water drawn into the upper layer. Figure 5.1 (b) shows how the turbulence of the tide coming in creates shear at the interface, creating waves and causing mixing to occur.

Figure 5.1: Generalized Profile of an Estuary, Showing Circulation During Ebb Tide (a) and Flood Tide (b).

The amount of shear and mixing that occurs between the freshwater and saltwater lays depends on the relative volume and velocity of freshwater (confined by the estuary sides) and of saltwater (confined more by depth). Figure 5.2 depicts stratification in a salt wedge estuary under different levels of tidal energy. A highly stratified estuary (a) can occur if tidal energy is relatively low and the gradual rise of the tide causes some mixing. As the tidal energy ("tidal prism") increases relative to freshwater flow, more mixing takes place to produce a partially mixed (b) and fully mixed (c) estuary.

Figure 5.2: Profiles of typical estuarine water masses, showing an highly stratified estuary (a), a partially mixed estuary (b), and a fully mixed estuary (c).

Tidal frequency and magnitude are considered when planning a plume delineation study and in subsequent numerical modeling. Tides in Canadian estuaries typically rise and fall with approximately 12.42 hours between high waters. The vertical range between high and low water, even within one coastal system, can range is from a few centimetres to several metres. Also, there is a beat every 7, 14 and 29 days indicating the "spring and neap" cycles where the forces of the sun and moon first act together and then oppose. In some areas of Canada every second high or low water differs considerably from its predecessor. These differences are generally referred to as semidiurnal and diurnal responses of the particular water mass to the differing pulls of sun and moon.

Other considerations for conducting plume delineation studies in estuaries include the following:

• duration of the tidal cycle: it is not uncommon for the duration of the falling tide to exceed 7 hours and the time of the rising tide to be significantly less than 6 hours; and
• slack tide: in some estuaries, a slack period can occur in vertical and horizontal movement around high and low water; this can amount to a few minutes or in exceptional cases almost an hour (in this case generally low water only); lack of significant horizontal movement for a period can lead to considerable pooling of effluent in the area of the discharge, leading to possible expansion of the boundary of the zone of >10% effluent concentration.

A special case occurs in fjords where the depth of the estuary is well over ten times the thickness of the tidal prism. Here the longitudinal profile will look like a salt wedge estuary or a well-stratified estuary with the fresh water lying in a layer at the surface and very significant water depth below. The tide can rise and fall, causing negligible mixing, except possibly at the entrance where there is typically a sharp sill. In some fiords there may be a 2 to 10 m thick layer of fresh water that extends the full length of the estuary with an interface of less than a meter thick. In longer fjords extending over several tens of kilometres, diffusion and other influences such as wind and internal waves will cause this interfacial layer to thicken and the upper layer to become brackish.

Estuaries may demonstrate two or more estuarine types along their length or during seasonal variation in river flow or in the spring to neap tidal cycle. In partially mixed estuaries or well stratified estuaries, there is often little seaward movement along the bed during the ebbing tide creating an increase in stratification during this period. On the flood tide, the stronger currents run deep and conditions closer to well mixed are more likely to occur. In Canada, because of the Coriolis force and the difference in density between salt and fresh water, ebb currents tend to run stronger and salinity tends to be lower towards the right hand bank (looking seaward) of an estuary when looking seaward. Similarly, flood tide currents are stronger and salinities higher towards the left when looking seaward. This can be very marked in wide estuaries. Where an estuary has a high fresh water discharge or a narrow exit to the open coastal waters, a fresh or brackish water plume can extend well out to sea particularly during the ebb tide. Depending on the dynamics at such an entrance a portion of this plume may re-enter the estuary with the flood tide.

An estuary can change from one stratification type to another, either with distance or with time. During spring run-off, freshwater may dominate the upper reaches of the estuary and create a higher degree of stratification downstream. During lower flow periods, the upper reaches of the estuary will return to partially mixed conditions and the lower salinity water will penetrate much further upstream. Internal waves (or seiches) may form along the fresh or brackish water/salt water boundary in fjords and estuaries as a result of wind action, and may temporarily affect plume dispersion. These seiches are discussed in more detail in Section 5.6.2.

During slack tide, a significant pool of effluent can gather in the vicinity of the discharge pipe and at a concentration close to that delivered at the top of the rising plume. Once the horizontal tidal currents begin to move again, this large effluent pool is the leading edge of the effluent plume and may take some time to dissipate. As the tidal currents strengthen, the effluent leaving the discharge area will behave more like a standard streaming pool. If the effluent pool gathered during the low water slack, the rising tide will typically come underneath the plume as denser water and carry the effluent pool upstream. The somewhat diluted effluent in the pool may pass down through the discharge area on the falling tide and provide already polluted water to the discharging effluent. If the effluent pool gathered during the high water slack, it will tend to stream off downstream in a very thin layer due to density differences. When the tidal currents strengthen, this high water pool of effluent will tend to disperse more quickly than the low water effluent pool, but it may substantially extend the limit of the plume at the surface, particularly if it is shore-attached.

Plunging plumes may occur in estuaries and fjords. Effluents with a fairly high initial dilution on the rising jet can pick up salinity from the lower layers such that the plume attains a density greater than that of the surface layer. If the rising momentum is sufficient, the plume may break the surface and then slowly plunge or sink down to a water layer consistent with its density and travel and mix with this lower layer. This is depicted in Figure 2.1 (i).

In other situations, particularly where the surface layer is thick or very distinct, the jet will rise but be trapped at the interface between the upper and lower layer. This is likely to be the case in fjords if the initial discharge is located in the deep saline layer. Effluent mixing into the upper layer is more likely during an ebbing tide. During a rising tide, the strength of the current in a shallow estuary is close to the sea floor and may cause significant mixing with, and entrainment into, the lower layers.

Submerged plumes may resurface when the plume encounters an influx of differing density water coming into the receiving water. In this case, surface water will initially plunge, but will resurface down drift. This scenario may occur if there is an influx of less dense water downstream of the discharge, such as from a tributary or another major effluent discharge.

Conducting the Field Work

Two boats are recommended for tracer studies in estuary environments. The recommended technique for dye tracer measurement is use of a fluorescence sensor or sampler intake off the bow of the boat, as described in Section 3.3.3. Salinity and temperature verticals should be taken along the centre line of the estuary and in the vicinity of the discharge. The sampling time should be noted relative to time of high or low water, and sampling should always proceed in a counter current direction relative to the receiving water movement.

It is important to consider the differences between successive high waters and between spring and neap tidal ranges and reconcile them with establishing the representative density regime in the estuary and also the objectives of the plume delineation study. The vertical range relative to the water depth at low tide gives some indication of potential turbulence. Another method, but requiring more work, is to estimate the water volume passing various sections during flood and ebb and hence to obtain the average velocities. This is particularly valuable where there are large inter-tidal volumes or additional river flows added at particular points along the estuary.

If using a dye tracer, injection should begin about half an hour before the turn of the tide so that the study may begin around low water. For most estuary studies, the plumes can be delineated in the field in the same way as a river plume. When sampling, the boat should always work against the tidal currents to avoid just drifting with the receiving water and possibly sampling the same water mass. If a separate boat is doing the drogue work or when being done on a separate day, it is useful to check salinity and temperature at the drogue depth each time the position is confirmed. Care must be taken when the tide reverses and the partly diluted effluent comes back over the discharge point as there will probably be a new thin poorly diluted plume on top of a thicker well mixed plume. When the difference in flood and ebb tide surface excursions is small (e.g., < 20% of the excursion) there may be a build up of effluent from more than one tide. Generally the numerical model can provide the most effective predictions of this phenomenon.

For plunging plumes, drogues can be set to follow the plume at depth. However, caution should be taken to ensure that the vertical dimension of vanes is not too large, because the upper and lower edges may get caught in two different water flows. If the depth of the plume is known from initial field sampling, the plume may be tracked as outlined in Section 3.9, except that instead of air there is a layer of water above the plume. This is one case where a fluorescence sensor or sampler intake may be towed astern, because it is likely to be well below the influence of mixing from the boat hull and propulsion unit. This technique requires more boat operator skill to maintain constant speed either with a water speed gauge or a revolution counter. A depressor will hold the fluorometer head or intake down, but a depth gauge (either pressure or upward looking sonar) is essential to record depth.

It is much more challenging to follow a plume that rises to the surface and then plunges and also where there is no well marked interface. Initial field measurements of salinity taken to characterize the environment will demonstrate how the naturally occurring surface layer behaves relative to the lower layer down drift.

Numerical modeling may need to be carried out over a number of tidal cycles to show short-term and long term effluent dilution fields. Three-dimensional modeling may be required for stratified receiving waters.

Marine

Thermal and salinity stratification are important features to assess for coastal marine environments because the effluent is usually less dense than sea water due to chemical and thermal factors. In very calm weather, the plume may be contained within a few centimetres of the water surface. Significant influences in this environment include wind and wave advection, shore-line attachment, bottom attachment, and tidal activity (often with rotary currents). Oceanic receiving waters are distinguished from estuaries in that the main circulation is not dominated by fresh water in the vicinity of the discharge. The effluent plume is diluted and transported by currents. Tidal activity, while present, does not always provide the main contribution to net effluent movement, but does affect effluent dispersion. A continuous dye release of one tide cycle is generally sufficient, but the duration of the numerical modeling program may be over several tidal cycles.

The residual current patterns may be dominated by a coastal circulation pattern or by local wind and wave influences. It is useful to have local water current data, such as may be obtained from positioning coastal current meters near the discharge. Likewise, local salinity, water temperature and wind records are also useful.

Prior to any dye release, the tidal heights and times at the site should be correlated to the nearest recording tide gauge maintained by the Canadian Hydrographic Service. This is of value for the dye study and also in the use of current and other longer term records. Spatial salinity and temperature profiling should also be conducted at this time.

The initial concept of effluent dispersion should be useful for planning the dye release. Field delineation of a surface plume may follow the standard practice outlined in Section 3.9, although judgement will be required on the spacing of the transects, particularly if the rotary current is marked and the plume begins to come back over the discharge area. Tracking of submerged plumes should be conducted as described in Section 5.4 above. It is usually beneficial to use larger scale drogues with radar reflectors and GPS receivers in coastal areas, especially if wave action is present.

From the dye tracer profiles, the effluent plume envelopes will be developed and related to the measured currents, wind and waves on that day. The probability of the occurrence of these envelopes may be determined using statistical data from available data on water currents, and numerical models used to extend these envelopes over a broader time period to represent "average" conditions.

Climatic Conditions

Ice and wind may have significant effects on effluent dispersion. The following sections provide guidance on incorporating ice and wind conditions in interpreting plume behaviour. The applicability of numerical models using ice are very limited at this time.

• Ice
• Wind

Ice

Ice is a common occurrence in many Canadian water bodies during winter. Its general effect on the effluent discharge is two-fold. Ice cover shields the receiving waters from the effects of wind stress and may also alter or cause stratification in the water column and hence modify water circulation. However, in the case of land fast ice, the underside of the ice provides a solid rough boundary to flow and creates turbulence similar to flowing over the stream bottom.

Most effluents are warm and this will cause some melting and a weakened ice surface. The buoyant plume will cling to this ice surface in the same way that a dense plume will cling to the bottom substrate. As far as is known, no detailed effluent plume field studies have been made under ice conditions, but it is reasonable to assume that mixing is likely to be decreased. This can result in the extent of plume concentrations being under-estimated when based on open water considerations alone.

Land fast ice cover removes wind stress mixing from the surface waters. In the presence of fresh water discharges (e.g., river discharges), mixing will be reduced between the effluent and river water. Ridges may appear in the formation of the ice surface. Some of these ridges may be the result of early ice flows colliding, and others may be formed by thermal expansion of the ice. Ridges may form in the same position each year and may have a keel or downward projection as much as 7 times the height of the surface ridge. These ridges can interfere with the plume and may divert it in a very different direction to its normal open water mode.

Moving ice still generally shields the surface water from direct wind effects. However, wind influences may be greater than initially predicted, because wind motion will be transmitted from the below water profile of ice flows to the receiving water and/or the plume. Where moving ice comes in contact with land fast ice or an island, a significant ice ridge may form.

The standard numerical models for plume delineation are generally not appropriate for considering ice conditions since they assume a free water surface (i.e., open channel flow).

Wind

Wind acting on large bodies of water induces currents and waves in direct proportion to its strength, its duration, and fetch over the water body. Fortunately, nomographs are available to provide estimates of current speed, and wave height and period that can contribute to understanding wind effects on the dispersion of effluents. One of the most valuable sources of information is literature relating to oil spill trajectory modeling, or ocean search and rescue. For wind currents, the James wind-drift forecasting curves (James 1966) are probably the easiest to use to obtain a magnitude. When using the James curves, the direction is usually taken as being 20 degrees to the right of the wind direction, due to Coriolis forces.

In more detailed work this can be modified to accommodate variations in duration, fetch, wind speed and water depth, among other factors. In many coastal areas it is the predominant wind direction coupled with local underwater topography that provides the driving force for currents along the shore. The momentum built up can make these currents very persistent. Wind waves also transport water, particularly when they get into shallow water. These waves can run out of the wind field and continue to transport their energy over hundreds of miles.

The nomograms most commonly used for wave predictions are those developed by the U.S. Army Corps of Engineers (1984). Wind wave energy is transmitted in two ways: first, on the surface by water particles orbiting in a circular motion with a radius of half the wave height at the surface; and second, at depth with the radius diminishing to almost zero at a depth of half the wave length. Typical coastal waves have periods of 4, 6, and 8 seconds, and have wave-lengths of 25 m, 56 m and 100 m, respectively. When this motion at a depth of half wave length comes in contact with the bottom substrate, friction causes the wave to slow and the circular particle motion changes to elliptical. On a long beach, this will cause the wave crests to try to align themselves parallel to the contours and eventually the shore. This has two effects on the plume: increased mixing within the plume and currents along the shore generated by the waves.

Internal waves, or seiches, moving along the interfacial boundary between two water masses can also impact the dispersion of a plume. The generation of these waves is not well understood but they are formed as eddies by the shear forces acting between the two opposing water layers, and they may be found moving along the thermocline in lakes, the sharp fresh water/salt water boundary in fiords, or between the estuarine and marine water in large estuaries or coastal waters. These internal waves can be of large amplitude (e.g., on the order of meters) with waves of 80 m being recorded in the St Lawrence off the Saguenay. Their impact on the plume is transitory, and are mentioned here only for guidance should an anomaly occur during a dye test. Their occurrence would be picked up on the records of recording thermographs or salinometers.