University of Gondar Hydraulic and Water Resources Engineering
University of Gondar Hydraulic and Water Resources Engineering
By Alemu Tezera
1
University of Gondar
Hydraulic and Water Resources Engineering
Reach scale: a selected river reach with uniform characteristics: sinuosity, slope,
width…
MISSOURI RIVER
Temporal scales
Short-term variations: waves
Daily variation: tide, water input from rainfall
Seasonal variations: high and low flow stages
Periodical variations: el niño
Long-term: climatic change, sea level rise
Depend on the amounts of sediment to be moved to produce the changes and on the flow
strength
Example assuming the same discharge:
Ripple forming: fast (minutes)
Dune forming: rather fast (days)
Point bar forming: medium (months to years)
Pool forming: medium (months to years)
Planimetric changes: slow (tens of years)
Valley formation: very slow (hundred of years)
For Irrigation:
To design stable channels
To design sediment traps
For River Restoration:
To understand present and past river conditions
To model future effects of restoration
To select proper restoration design
For RB Management:
To understand where the sediment come from
To put in place the best soil conservation practice
Hydrological modelling
A model is a simplified, schematic representations of the real world, a representations that
retains enough aspects of the original system to make it useful to the modeller.
(Chapra&Canale,2006)
“It is a system which will convert a given input (geometry, boundary conditions, force, etc.)
into an output (flow rates, levels, pressures, etc.) to be used in civil engineering design and
operation" (Novak);
“Model is a simplified representation of a natural system“ (Resfgaard);
In general, models are mathematical representations of physical phenomena in forms of
equations.
It is not reasonable to expect that a "simplified representation" converts input into exact
output.
Errors are inevitable.
All models are an approximation to reality, i.e. the models try to reflect as much as possible
behaviors of the real-world, but an imperfect representation is always expected. A model is
a simplified representation of the natural complex system.
Surface water mathematical models have been increasingly applied to water management.
Several numerical model types are available in order to simulate different systems and
emission sources [Chapra, 1997]. Those models vary from empirical models, simplified
models and detailed physically-based models.
Models - How ?
Important first steps
– Define the question or problem to be addressed with the model.
– Determine appropriate spatial and temporal scales.
– Identify important system/ecosystem components and processes that must be considered
to answer the questions.
Some specific questions to ask
– On a temporal scale
Are the steady-state conditions adequate or we need to predict changes over time?
If time is important, do we need to look at
Short-term changes (e.g., daily, seasonal) or
Long-term changes (e.g., trends over years)?
River Basin Development By Alemu Tezera 27
University of Gondar
Hydraulic and Water Resources Engineering
On a spatial scale
Is our question best addressed
On a regional scale (e.g., comparing streams in a region) or
By modeling specific processes.
Models are commonly classified into three types based on their reflections of reality.
Detailed physically-based models: Ideally, the model is constructed based on the physical
processes. This type of models is called a white-box model since the linking between model
input and output is transparently described by a large number of mathematical equations
representing the physical processes. It is expected that most model parameters of the white-
box model can be measured. An example for this type is detailed hydrodynamic models.
Conceptual models: When a macroscopic description of the processes is used, the model
becomes a grey-box model because the physical processes are less transparently described.
In such models, the physical processes are represented as they are observed and an
abstraction is made over several microscopic processes of observations so that they are
lumped together in a few processes. Because of this, model parameters in the conceptual
models require calibration. Model parameters of this type cannot often be directly measured
but require estimation. Thus, a physical interpretation is extremely important to calibrate the
model parameters in a physically-based way.
Empirical models: this type of model does not have an internal description. It is also called
a black-box model. A black-box model does not reflect any specific physical structure of a
system. It is built and calibrated according to measurements for model input and output.
The empirical model may be conditional on the period used for calibration. A serious
drawback of black-box models is that it is difficult to provide a physical interpretation of
estimated parameters. Representative for this type are regression models (generalized linear
models) and transfer function models.
TopoDrive
This two-dimensional model is designed to simulate two ground-water process. To simulate
topography-driven flow, the user may specify the shape of the water table, which bounds
the top of the vertical flow section.
A topography-driven flow system is one in which ground water flows from higher-elevation
recharge areas, where hydraulic head is higher, to lower-elevation discharge areas, where
hydraulic head is lower. This type of flow system is commonly encountered in ground-
water basins.
The main factors that control ground-water flow are basin geometry, shape of the water
table, and the distribution of hydraulic properties.
For a two-dimensional vertical section, the boundaries of the flow domain (fig. 1) are as
follows :
The top boundary (AB) is the water table, which is assumed to lie close to land surface.
The two vertical boundaries (BC and AD) and the bottom boundary (DC) are no flow
boundaries.
Steady-state flow of ground water in the vertical section (fig. 1) is governed by the
equation.
where h is hydraulic head, and Kxx and Kzz are the principal values of the hydraulic
conductivity tensor.
Step 3: Mesh
This step specifies the finite element mesh.
Step 4: Properties
This step assigns hydraulic properties (hydraulic conductivity and porosity) to model
elements.
Step 5: Head
This step computes hydraulic head in the model domain.
Assignment-2
Make an aquifer with a length of 5000 m and a height of 140 m (vertical exaggeration 18).
Draw a water table which varies from 80 m (left side) to 140 m (right side ). The number of
columns and rows should be 40 and 20 respectively.
1. Put the K of the aquifer equal to 10-4 m/s (sand). Draw 20 equipotential lines. How is
the distribution of equipotential lines? Where is the recharge area and discharge area?
2. Change value of K to 10-1 m/s (gravel). How does this influence the equipotential lines
and flow paths?
140m
80m
5000m
River Basin Development By Alemu Tezera 44
University of Gondar
Hydraulic and Water Resources Engineering
3. Draw a zone B as indicated in the figure below. Make K of layer B equal to 10-7m/s
(clay) and K of layer A and C equal to 10-4 m/s (sand). Explain the shape of the
equipotential lines and flow paths.
A
140m
80m B
C
5000m
4. Consider there are contaminated fluid particles on the right side of the aquifer as shown
in the figure. A well is located on the left of the aquifer to supply drinking water. Your task
is (i) to determine whether the contaminant migrates towards the drinking water well or not
for case of sandy aquifer, clay aquifer and gravel aquifer and (ii) if yes, how long it will
take the first particle to reach to the well. (iii) if no, why not?
140m
80m
5000m
River Basin Development By Alemu Tezera 46
University of Gondar
Hydraulic and Water Resources Engineering
Assignment-3
You will be given one assignment on HEC-HMS. You will develop one surface water
hydrological model.
Thank you!