Coastal Engineering Challenges Motril, 2005

1
Coastal Engineering Challenges (Motril, 2005)

• J. W. Kamphuis
• Queens University
• Kingston, ON, Canada
• K7L 3N6
• kamphuis_at_civil.queensu.ca

2
Challenges forCoastal Morphology and Evolution
(I cheat a little)
3
Approach
I trust others will adequately address science
requirements. This presentation addresses very
practical needs for engineering studies and
designs.
4
Summary

• Sediment Transport (as example)
• Data Aquisition
• Uncertainties
• Integration
• Co-operation and Communication

5

• Sediment Transport (as example)

• Bulk
• Direct measurements

6
Model Results Kamphuis (1991)
7
Model and Field Results Kamphuis (1991)
Larger Uncertainty !
Verification !
8
Filling the Gap VHP Model Tests Added
9
Newer Field Data
Newer, more modern results !
Same scatter !
10

• Sediment Transport

• Bulk
• Direct measurements
• Inferred

11
Sediment Transport at Keta, Ghana
More-or-Less Unidirectional Transport
12
Comparison of Sediment Transport Rates at
Bovbjerg, Denmark
Low with respect to DHI model
13
Comparison of Sediment Transport Rates at
Bovbjerg, Denmark
14
Sediment Transport at Adra, Spain
Much Room for Interpretation
15
Sediment Transport at Richards Bay, South Africa
Qnet Pavg ?
16
So?

• Simplest Calculation Bulk Sediment Transport
  is
• Crude
• Not good
• Does not look like it will improve substantially
  in the near future.

17

• Sediment Transport

• Bulk
• Direct measurements
• Inferred
• Detailed

18
Comparison of Detailed Expressions
gt 2 orders !
19
Is this OK ?

• No
• Credibility
• So how can we improve on this very basic
  computation (and other unknowns)?

20

• Obtain more data
• Reduce uncertainties in the data (better data)
• Write better models for detailed sediment
  transport
• Generalities ?

21

• Sediment Transport
• Data Aquisition

22

• Field measurements are the truth, but
• Test conditions cannot be designed
• Boundary conditions cannot be manipulated
• Long-term steady conditions impossible
• Tests cannot be repeated
• Close observation of details is difficult
• Bias toward low energy and good weather
• Expensive (therefore, few and short)

23

• Improve field measurement by
• Reduction of mobilization and other costs
• Making experiments more transportable
• Use new or different technologies
• Develop some common standards
• And

24

• Improve dataset through the use of physical
  models in addition to field data,
• The shortcomings of field measurements (expense,
  poor observation, long-term steady conditions,
  etc.) are strengths of physical modelling
• Perfect match?

25

• The model studies may (will?) need to be Very
  Large Physical Model (VLPM) studies,
• VLPM studies have all the strengths of physical
  modelling, but..
• Who will build the facility?
• Is it economically justified?
• Commercially not viable.

26

• Also VLPM experiments are like field
  experiments
• Extensive planning and long lead times
• Collective exhaustion of the researchers
  recovery takes years?
• Therefore extensive downtime of facility
• Someone needs to finance
• Experiments
• Facilities
• Downtime

27

• And
• We have not mentioned a 3-D VLPM
• Yet, we cannot do without the VLPM simulation of
  prototype conditions.
• Of course there is more we can get from VLPMs
  than data.

28
Challenges

• Good estimates of sediment transport rate (and
  other basics)
• Support and awakening to VLPM simulation of
  prototype conditions to add to data bank.
• 3-D VPLM

29

• Sediment Transport
• Data Aquisition
• Uncertainties

30

• Examples of uncertainties are
• Natural (climate seismic, geo, etc.)
• Operational (construction, deterioration,
  maintenance, etc.)
• Process (from beginning to end)
• Data
• Model

31

• Process uncertainties are
• Uncertainties in the end result
• Cumulative
• More sophisticated study ? more uncertainty
• Cause projects to be challenged

32

• Data uncertainties are
• Measurement errors
• Inadequate sampling (space, time, frequency,
  duration)
• Handling, transcription, etc.
• Solutions are obvious
• Discussed earlier as Data Aquisition

33

• Model uncertainties are due to
• Formulation (improving)
• Parameters (up to 2 orders and not improving)
• Numerical methods (improving)
• (So Not just the data!)

34
Primary Chain of Impacts
Fish
Dredge Plume
Birds
Less Exposed Lee Shore
Mammals
Fishing
Widening of Littoral Zone
Construction
Safety
Recreation
Decrease in Nearshore Depth
Ports
Navigation
New Sea-Land Interface
Cables, Pipe
35
One Secondary Chain of Impacts Dredge Plume
Dredging Operation
Silt Content
Foam Formation
Phytoplankton
Recreation
Benthic Community
ZooPlankton
Fish
Fishing
Birds
Mammals
36
Another Secondary Chain of Impacts New Sea-Land
Interface
37
Growth of Uncertainty
38
Contemporary Decision Making Process
Stakeholders from many backgrounds, each with
personal and organisational biases and
perception of risks, who do not understand how to
deal with uncertainties
Knowledge (Theory Experience)and Data
Models, Design, Trial and Error Best Engineering
Estimates, Includes Subjective Probabilities (Engi
neering Judgment)
Decision-making authority
(GAMSI) Decision
39
Suggestions to reduce Process Uncertainty
40
Suggestion

• Improve the numerical models (now the most
  crucial tool in analysis and design!)
• Formulation, parameters, numerical
• Research on interactions (e.g. flow and)
• Cohesive materials (erosion)
• Pollutants (water quality)
• Vegetation (fish habitat)
• Better calibration, verification data

41
Suggestion

• (Much) more (good) research on biology,
  fisheries, etc.
• (Much) more communication with biologists,
  chemists, etc.
• (Much) more interdisciplinary research and
  exchange of technologies

42
Suggestion
Sediment and Morphology
New (Aggregate) Models
43
Suggestion

• Engineering means solutions to real-life problems
• Statistics and mathematics walked away from
  reality (pure vs ?? objective vs subjective)
• But pure and objective alone cannot provide the
  complete answers for real problems.
• Therefore judgment (subjective) needs to be
• Cultivated and used
• Explained to decision makers and stakeholders
• Developed in our young people

44
Suggestion

• Significance needs to be carefully defined and
  determined (large uncertainty in an insignificant
  aspect is OK)

45

• Sediment Transport
• Data Aquisition
• Uncertainties
• Integration

46
Integration

• Clearly the heart of any solution today is the
  numerical model.
• But it needs to be carefully integrated with the
  other tools and information.

47
Integration
Output
48
Composite Model
Integration

• Processes are modelled
• physically and/or numerically
• models must be trustworthy representations
• Process model results forms the bricks
• simple, inexpensive, easy to understand tests
• generic tests and results
• repeatable results
• repeated tests

49
Integration
Composite Model

• Computational module forms the mortar
• not necessarily a numerical model (spreadsheet)
• calibration (inexpensive)
• what if? (inexpensive)
• Not necessarily same provider for various parts
• Bank of process model bricks (I wish)

50
Integration
Composite Model
Output
Model Results
B.C.
51

• Sediment Transport
• Data Aquisition
• Uncertainties
• Integration
• Co-operation and Communication

52
Co-operation and Communication

• Anti-fragmentation
• Tsunami is not an Indonesian Problem
• Katrina is not an American problem
• Global Climate change
• Coastal erosion is not only a management problem
• Coastal is more than rocks and sand includes
  plants, animals and man
• Research is not only university based

53
Co-operation and Communication

• Co-operation between
• Coastal scientists and engineers
• Coastal managers with scientists and engineers
• Coastal scientists and engineers with related
  coastal disciplines
• Universities with consultants and contractors
• Sharing of technology (as EU)

54
Co-operation and Communication

• World Centres of Excellence
• Distinguished through topics and disciplines,
  rather than geography.
• Examples
• Fluid flow
• Sediment transport and morphology
• Mathematical and numerical methods (basic tools
  development)
• Physical modelling
• Coastal biology

55
Co-operation and Communication

• Technology Platforms
• C of Es set up and manage
• Information is sold by shareholders
• Information is sold to shareholders
• Only contributors benefit

56
Co-operation and Communication

• Who will fund this?
• No-one will leap forward
• We must plant the seed
• We must start co-operating more
• We must make an effort to invite strangers (other
  disciplines, nationalities) into our shop
• We must communicate to funding agencies that we
  can no longer work on a disciplinary or national
  basis.

57
Co-operation and Communication

• I see this happening !
• But we need more
• And we need it soon

58
Summary

• Sediment Transport (as example)
• Data Acquisition
• Uncertainties
• Integration
• Co-operation and Communication

59
Summary

• Sediment Transport (as example of 2, 3, 4, 5)
• Data Acquisition high cost
• Uncertainties medium cost
• Integration low cost
• Co-operation and Communication medium cost

60
The paper is posted onwww.civil.queensu.cagoogle
kamphuis canada
Thank You