Lec 3: Modelling

Conceptual modelling: (Summarise ideas)

Uses easy flow charts, to see visual effects

Limitations:

  • Doesn’t inlude alot of detail

  • Unconvincing

  • Hand wavy

Positives:

  • Summarise idead easily

  • Show how everything is connected easily

  • Can communicate across to non-experts

Emperical modelling (Statistical models)

(Ultimately fitting a line or curve to data)

(Autocorrelation is common in coastal data)

Autocorrelation is an correlation between a signal and a lagged version of the same signal for example the average daily temperature, if its warm today, can be warm tomorrow but will change as time goes on.

Linear

  • Regression models

  • Generalised LS

  • Exponential, logarithmic and other forms.

Semi-emperical models

  • Surface and bed friction

  • Wave breaking type

Coastal emperical models:

  1. SBEACH

Storm-induced/Change of beach state

Predicts over the short time of the storm, where sand will erode and creep.

Limitation:

  • Assumes no long shore sediement transport, so it can work well over the time of the storm, but instances where more long term information is needed not so much.

  1. GENESIS

Generalised model for shore line change over a longer time factoring things in like long shore transport.

General limitations:

  • Model can assume underlying normal (Distribution)probabablity (Bell-shaped etc) and sometimes is not the case.

  • Using this model can limit the ability to comapare data to another beach, as this is likely to be site specific.

Physical models: (Used for sediment transport problems)

Flumes: Artificial rivers

Types of flumes:

  • Annular flume

-Rotates flow in one direction continuously in a donut shaped circulatory system, with a lid going in the opposite direction to avoid secondary flow.

  • -The lid needs to be in the opposite direction, because if it was all in the smae direction the flow in the surface water will be caused to bulge slightly to the right, by putting a lid on in the opposite direction it cancels out this secondary flow effect.

  • Recirculating flume (Experiment erosion threshholds)

-Fluid and sediment move in a continuous loop

- Working section (Used to imitate different speeds in flow, and thresh holds for different materials)

-Can explore roughness of materials

  • 1D wave flume

-Expenive, Big (Minimise boundary effects to avoid friction from walls) and timely.

-Simulate large full-sized waves

-Can add vegetation, people, structures etc.

  • 2D basin models (Designing an testing marine infrstracture)

-Ports, vessels, wind turbines

-Tauranaga harbour was a physical model, including tides, waves, sediment transport, resulting in dredging in channels.

Advantages of physical models:

  • Dont need to fully understand processes

  • Can provide new insight

  • Can test scenarios under controlled conditions

Disadvantages of physical models:

  • Scale issues (e.g coal dust vs sand)

  • Difficult to reproduce realistic waves:

-Random waves cant be produced by most wave generators

-Wave shape, particularly for shaoling waves, is hard to generate

Analytical models:

  • Simple cases that can be solved analytically (Maths, calculaus, geometry)

-Deep vs shallow water waves

-Circular islands

-Infinitely deep ocean

Simultaneously solve:

  • Velocity

  • Temperature, salinity

  • Pressure

Given an equation, making decisions on which terms are important to keep in the equation, and re-arranging the equation to see specifically what you want on paper.

Example:

Geostrophic flow

Numerical models:

  • Breaking down of a complicated problem into simple shapes (Grid cells)

-Boxes or triangles

-Converts differential equations into little algebra problems

  • Result from each grid cell, is used to estimate the next grid cell

  • Once a grid cell is estimated for a given step, an process begins again using new result as starting point.