Modelling of Dense Brine Discharges in
Oman: Recirculation and Environmental
Aspects - A Case Study
W. Verbruggen, R. Morelissen, C.M. Freixa
International Conference on Desalination,
Environment & Marine Outfall Systems 2014,
Muscat, Oman
16 april 2014
Outline
Objective: Illustrate techniques required for a comprehensive
assessment of the effluent behaviour
• Introduction
• Near-field aspects
• Visual disturbance
• Dilution
• Far-field aspects
• Inclusion of near-field information
• Selection of model scenarios
• Conclusions
16 april 2014
Introduction
Plume dispersion and recirculation assessments
Objectives of plume dispersion and recirculation studies:
Optimization of intake and outfall w.r.t.:
• Compliance with environmental criteria
• Recirculation potential
Outfall
16 april 2014
Introduction
Spatial and temporal scales
current
Far-Field
•
Near-Field
•
•
Outfall
momentum
Outfall
geometry
16 april 2014
•
•
Currents
• Wind
• Tidal
Stratification
Etc.
Near-Field Aspects
Important aspects in the design of a diffuser:
• Energy losses in the diffuser
• Uniform discharge distribution along diffuser
• Optimal spacing of ports
• Possible visual disturbance
• Optimal near-field dilution
Pressure head inside
diffuser
16 april 2014
Near-Field Aspects
Visual disturbance
XL
= 1.94
D
u0 2
( ρs − ρ0 ) gD
ρ0
XL
Typical values: u0 = 4 m/s, D = 0.25 m,
ρ0-ρs= 15 kg/m3  XL ≈ 10 m
D
Abraham, G. 1967. Jets With Negative Buoyancy in Homogeneous Fluid. Journal of Hydraulic Research, 5 (4), 235248.
16 april 2014
Near-Field Aspects
Dilution
Abessi and Roberts (2014) found that dilution = 1.1-1.2 * H/D in
case of water level impact
In our case study: dilution about 30% - 50% lower if diffuser is
constructed at too limited water depth
Abessi, O. & Robberts, P.J.W. 2014. Optimization of Desalination Diffusers Using Three-dimensional Laser-induced
Fluorescence. Final Report. Atlanta, Georgia.
16 april 2014
Far-Field Aspects
Model set-up: inclusion of near-field information
Suppose:
• Q = 10 m3/s
• ΔS = +30 ppt
• Snf = 10
Source
Sink
1. Diluted source only
•
•
•
NF-information included at coupling location
Not mass conservative
Entrainment in FF-model
2. DESA method
•
•
Mass conservative
Entrainment in FF-model
Choi, K.W. & Lee, H.W. 2007. Distributed entrainment sink approach for modelling mixing and transport
in the intermediate field. J. Hydraulic Eng. 133 (7), 804-815.
16 april 2014
3/s
3/s
Q = 10
100mm
ΔS = 30
+3 ppt
ppt
+
Q = 90 m3/s
ΔS = 0 ppt
Far-Field Aspects
Model set-up: inclusion of near-field information
Thermal plume tracer
in a stratified water body
Without DESA
With DESA
Thermal plume
Thermal plume
16 april 2014
Far-Field Aspects
Model set-up: inclusion of near-field information
Velocity near the bed
Combined plume tracer
along cross-section
Combined thermal
and brine discharge
Brine
discharge
Diffuser
combined
discharge
16 april 2014
Far-Field Aspects
Model scenarios: Scenario Based Modelling Approach
Experience
Data Analysis
Set of ambient
conditions
(normal and
adverse)
-
Test simulations
16 april 2014
Which processes are dominant?
How does the plume behave (response time,
buoyancy, etc.)?
Which conditions are governing w.r.t.:
- Recirculation
- Environmental compliance
Far-Field Aspects
Schematised:
+ understanding system
- relevant?
Wind speed [m/s]
Model scenarios: Events
Unschematised:
+ Realistic
- Governing period?
- Less understanding system
Wind direction [°N]
Time
Time
Select governing periods
Tests with schematised conditions
Select relevant schematised
conditions
16 april 2014
Far-Field Aspects
Model scenarios: Events
Relevant time scales:
• Response time of the plume
Response time
• Operational criterion (6 hours)
• Environmental criterion (day – week)
 Select conditions that are corresponding to these timescales
16 april 2014
Conclusions
In order to accurately and reliably model the dispersion and
recirculation of brine discharges, it is necessary to use
comprehensive and state-of-the-art modelling techniques:
• Analyse near-field behaviour in detail (water level impact,
dilution, etc.)
• Use DESA to include entrainment in FF-model
• Iteratively (e.g. using SBAM) determine a representative set of
ambient conditions
• Take relevant time scales into account in the selection of
modelling scenarios
16 april 2014
Thank you!
Questions?
16 april 2014