**Effluent Mixing Modeling for Liquefied Natural Gas Outfalls in a Coastal Ecosystem**

#### **Mustafa Samad and Karim El-Kheiashy**

**Abstract:** Liquid Natural Gas (LNG) processing facilities typically are located on ocean shores for easy transport of LNG by marine vessels. These plants use large quantities of water for various process streams. The combined wastewater effluents from the LNG plants are discharged to the coastal and marine environments typically through submarine outfalls. Proper disposal of effluents from an LNG plant is essential to retain local and regional environmental values and to ensure regulatory and permit compliance for industrial effluents. Typical outfall designs involve multi-port diffuser systems where the design forms a part of the overall environmental impact assessment for the plant. The design approach needs to ensure that both near-field plume dispersion and far-field effluent circulation meets the specified mixing zone criteria. This paper describes typical wastewater process streams from an LNG plant and presents a diffuser system design case study (for an undisclosed project location) in a meso-tidal coast to meet the effluent mixing zone criteria. The outfall is located in a coastal and marine ecosystem where the large tidal range and persistent surface wind govern conditions for the diffuser design. Physical environmental attributes and permit compliance criteria are discussed in a generic format. The paper describes the design approach, conceptualization of numerical model schemes for near- and far-field effluent mixing zones, and the selected diffuser design.

Reprinted from *J. Mar. Sci. Eng.* Cite as: Samad, M.; El-Kheiashy, K. Effluent Mixing Modeling for Liquefied Natural Gas Outfalls in a Coastal Ecosystem. *J. Mar. Sci. Eng.* **2014**, *2*, 493-505.

#### **1. Introduction**

Total global demand for liquefied natural gas (LNG), which is one of the cleanest fossil fuels, is estimated to have grown by approximately seven percent per year since 2000 [1,2]. Driven by national environmental preference for lower carbon fuels, economic impacts of carbon emission costs and low shale gas prices, desire to diversify energy supply sources, geopolitics, and heightened popular opposition to post-Fukushima nuclear energy, the LNG production capacity is set to experience unprecedented growth by 2018 [2,3]. The number of new construction and operation of LNG plants also heightened environmental awareness in the plant permitting process. The LNG is primarily composed of methane, CH4, which is converted to a liquid form for adequate storage and transport. In a typical LNG process the natural gas is first extracted from a deep on- or offshore gas exploration site, pre-treated and transported to an onshore or near shore processing plant where it is purified by removing condensates such as water, oil, mud, and other gases. An LNG process train would also typically be designed to remove trace amounts of mercury from the gas stream. The gas is then cooled down in stages until it is liquefied at approximately í260 °F at atmospheric pressure [4].

The LNG processing plants use large quantities of water for various process and wastewater streams. Plant wastewater is treated to meet required regulatory standards before wastewater effluents are discharged to the coastal and marine environments. Onshore LNG plants also support desalination of seawater to meet the large water demand of the LNG process and potable use during construction and operation of the LNG plant. Proper disposal of effluents from an LNG facility is essential to retain regional environmental values and ensure regulatory and permit compliance. The combined wastewater effluents are discharged through appropriately designed outfalls that commonly adopt a multi-port diffuser system in sufficiently deep waters. The diffuser system design accounts for the mixing characteristics in the near field and far field regions and generally forms a part of the plant overall environmental impact assessment.

In this paper, a diffuser system design case study is presented to meet the regulatory effluent mixing zone criteria. Although the project location is not specified, it is located on a meso-tidal coast with frequent storm surge impacts. The outfall is located in a coastal and marine ecosystem where large tidal range and persistent surface wind govern conditions for the diffuser design. Physical environmental attributes and permit compliance criteria for the case study are discussed in a generic format. The paper describes major LNG process wastewater streams, the diffuser system design approach, conceptualization of numerical model schemes for near- and far-field effluent mixing zones, and the selected diffuser design with particular emphasis on the modeling of the near-field mixing process.

#### **2. LNG Process and Outfall Wastewater Streams**

Onshore LNG plants support desalination of seawater to meet the large water demand of the LNG process and potable use during construction and operation of the LNG plant. Treated effluents from LNG plants may include several wastewater streams, for example, desalination plant brines, treated wastewater from LNG process streams, and treated sanitary wastewater.

Figure 1 shows a typical schematic description of the LNG plant wastewater process streams including outfall to natural environment. The figure shows one desalination (seawater reverse osmosis, SWRO) plant to supply SWRO product water to the LNG plant and SWRO filtration water and brines as wastewater to the equalizer tank. Treated wastewater from two sanitary wastewater streams, one from the LNG train areas and other from site facilities, are also pumped to the same equalizer tank. The final treated wastewaters from these streams are equalized in the combined equalizer tank before discharging to the outfall. The projected combined effluent flow rate from this system is estimated as approximately 750 m<sup>3</sup> /h.

**Figure 1.** Schematics of a typical LNG plant wastewater process streams.

#### **3. Effluent Diffuser Design Approach**

Environmental impacts from wastewater discharges are often evaluated based on the characteristics of effluent outfall plumes including mixing, dispersion and dilution, and ambient hydraulic characteristics such as currents, winds, temperature and density. Detailed evaluation of effluent outfall plumes is also important for meeting regulatory mixing zone criteria. Such effluent evaluation studies are performed employing numerical models that account for diffuser system design and resolve flow dynamics for the near field and far field regions. The near field region is characterized by small scales near the discharge location where flows are governed by diffuser designs, discharge properties and strong turbulent and jet mixing. On the other hand, the far field region is defined by large scales in the ambient receiving water where buoyant spreading motions and passive diffusion governs effluent dilution. Combination of these models is routinely used for mixing zone evaluation and wastewater disposal designs [5–8]. Several models are available for each flow types (near- and far-fields) focusing on the specific scales, resolutions and processes for each flow fields [9–12]. Plume models, which provide average plume characteristics in the near field zone, use spatial and temporal averaging of the flow field using equivalent diffuser characteristics. Consequently, the plume models simplify the processes using empirical techniques and analytical solutions based on the simplified geometry [9,10,13,14]. Hydrodynamic circulation models resolves the flow, density and temperature fields in three-dimensions solving the unsteady, baroclinic, shallow water equations [15,16].

In addition models that combine the near-field and far-field mixing processes are also proposed [17,18] to dynamically couple information exchanges between the models of varying time and space scales. Numerical fluid dynamic simulations, which are computationally very demanding, have also been performed to resolve the mixing process [19].

Present study uses a combination of near- and far-field models to evaluate outfall discharge configuration and mixing zone behavior. This paper presents results from both near- and far-field analyses with particular emphasis on the near-field mixing process. Effluent near-field mixing is evaluated by employing the Cornell Mixing Zone Expert System (CORMIX) modeling tool [9,10,13,14]. CORMIX is a steady-state mixing zone model for single or multi-port discharges particularly suitable for near-field mixing. CORMIX has the capability to analyse negatively buoyant effluents such as effluents from desalination plants that have higher density than the receiving water body. This is simulated in CORMIX such that the effluent flow from a submerged discharge port provides a velocity discontinuity between the discharged fluid and the ambient fluid causing an intense shearing action which breaks down into turbulent motion. This turbulent intensity progresses in the direction of the flow by entraining more of the ambient less turbulent fluid [9].

The far-field effluent mixing process is studied using the three-dimensional numerical model, Delft3D-FLOW [15]. The study investigates the far-field dispersion of the outfall discharge, the potentials for re-circulation of the effluent discharge as well as the dredged bathymetry impact on the extent of the mixing zone. Delft3D-FLOW is a three-dimensional hydrodynamic and transport process model, which simulates unsteady flow and transport phenomena that result from tidal and meteorological forcing. Delft3D-FLOW solves the governing flow equations for an incompressible fluid, under the shallow water assumptions and solves the equations on a rectilinear or a curvilinear boundary fitted grid system. The far-field model primarily was used to confirm the selected diffuser design from the near-field model.

### **4. Model Setup**

The shoreline and layout of the proposed outfall is shown in Figure 2. The outfall is located approximately 2.2 km from the shoreline at water depth of 5 m with respect to the lowest astronomical tide (LAT). For this analysis tidal water level variation is considered and the corresponding water depth and current velocity are used for each studied tidal conditions.

#### *4.1. Physical Environment*

Ambient conditions of the receiving water body of the coastal environment where the effluent discharges are described below.

### 4.1.1. Bathymetry

The bathymetry input comprises of bathymetry data (below water surface) and topography data (above water surface) from several data sources. A navigation channel is planned at the project site with a water depth of approximately 14.8 m relative to mean sea level (MSL). Figure 3 shows the pre-development bathymetry of the area. The diffuser system would be placed approximately at elevation 5 m below the lowest astronomical tide (LAT) (see Figure 2), which is considered as the uniform bottom elevation of the offshore-ward unbounded near-field model (CORMIX). The difference between MSL and LAT is approximately 1.5 m.

**Figure 2.** Outfall diffuser location and receiving water body.

**Figure 3.** Bathymetry near the outfall location (depths are in meter relative to MSL).
