LECTURE 1

MODELLING AND MATHEMATICAL MODELS

Lecturer: Prof. Nguyen Ky Phung
MSc. Dang Thi Thanh Le
MSc. Tran Thi Kim
 COUSE OUTLINE

1. What is model?

2. What is environmental modelling?

3. Role of environment modelling

4. Types of models

5. Mathematical models?

2
 1. WHAT IS MODEL?

A Model is a simplification of reality that is constructed

to gain insights into select attributes of a physical,
biological, economic, or social system. A formal
representation of the behaviors of system processes,
often in mathematical or statistical terms. The basis
can also be physical or conceptual (Environment
Protection Agency, 2009)

3
2. WHAT IS ENVIRONMENTAL MODELLING ?

Environmental modelling involves the application of

multidisciplinary knowledge to explain, explore and
predict the Earth’s response to environmental change,
both natural and human-induced. (DBW,2018)

4
 3. ROLE OF ENVIRONMENT MODELLING

• Improved understanding of environmental systems.

• Developing scientific understanding - through
quantitative expression of current knowledge of a
system (as well as displaying what we know, this may
also show up what we do not know);
• Test the effect of changes in a system;
• Aid decision making, including (i) tactical decisions by
managers; (ii) strategic decisions by planners.

5
4. TYPE OF MODELS

 Physical modeling
 Empirical models
 Mathematical models

6
 4. TYPE OF MODELS

 Physical modeling
• Physical modeling is a way of modeling and simulating systems that
consist of real physical components. A physical model is a smaller or
larger physical copy of an object.
• Spatial analysis and similarity theories are used in this process to
ensure that the model results can be extrapolated to the real system
with high accuracy.
• Physical modeling is the main approach of scientists in developing
basic theories of the natural sciences.

7
Water dam model
 4. TYPE OF MODELS

 Empirical models
Empirical models describe observed behaviour between variables on
the basis of observations alone and say nothing of process. They are
usually the simplest mathematical function, which adequately fits the
observed relationship between variables. No physical laws or
assumptions about the relationships between variables are required.
Empirical models have high predictive power but low explanatory
depth, they are thus rather specific to the conditions under which data
were collected and cannot be generalized easily for application to other
conditions

8
 4. TYPE OF MODELS

 Mathematical models
• Mathematical model is a representation of real world problem in
mathematical form with some simplified assumptions which helps to
understand in fundamental and quantitative way
• Mathematical models are much more common and represent states and
rates of change according to formally expressed mathematical rules.
Mathematical models can range from simple equations to complex
software codes applying many equations and rules over time and space
discretization. One can further define mathematical models into different
types but most models are actually mixtures of many types or are
transitional between types.

Dam model is described by a

mathematical model
9
 Real Systems

Mathematic Models

Probabilistic
Deterministic Models
models
Continuous Discrete

Static Dynamic Statistical

N=1 N>1 Centralized Distributed Markov

Partial Monte Carlo

System of Ordinary Partial
Algebraic
linear differential differential differential
Equations
equations equation equation equation

Linear Non-Linear

Analytics Numerical
Classification models 10
 WHAT IS
MATHEMATICAL MODELLING?

11
 Definition & Meaning

 System and boundary

• A system is a set of one or more related objects, which can be a
physical entity with specific properties or characteristics. The
system is isolated from its surroundings by boundaries, which can
be physical or virtual
 Open and Closed, flow/non-flow systems
• A closed system is a system that is completely isolated from its
environment.
• An open system is a system that has flows of information,
energy, and/or matter between the system and its environment,
and which adapts to the exchange.
• When the flow of matter does not cross the boundary (but energy
can), the system is called a nonflow system. If the material flow
can cross the boundary, the system is called a flow system.
12
 Definition & Meaning

 Variable, parameter
• A variable is a value that changes freely in time and space (a
compartment or flow) and a state variable is one which represents
a state (compartment). A constant is an entity that does not vary
with the system under study, for example, acceleration due to
gravity is a constant in most Earth-based environmental models
(but not in geophysics models looking at gravitational anomalies,
for example).
• A parameter is a value which is constant in the case concerned but
may vary from case to case where a case can represent a different
model run or different grid cells or objects within the same model.

13
APPROACHES TO MATHEMATICAL MODEL BUILDING

Identify problem

System analysis

Choosing mathematic equation

Solving equation

Step to building mathematical models

14
 Get started

Other models
Objectives

Pre-existing
Describe the system and identify: the system,
Knowledge Data
its boundaries, and its surroundings

Make an assumption Physical Model

Establish variables and parameters

Principles,
theories,
theorems, laws Building model Experimental
relationship
Equation Minimize math formulas
Theory

STEP TO BUILDING
Equation Analytic mathematical models MATHEMMATICAL
Software Preliminary application MODELS
Calibration Training data

No
Accept?

Yes

Validation Verify data

Yes
NO
Accept?

15
Application
 APPROACHES TO MATHEMATICAL MODEL BUILDING

 CALIBRATION AND VALIDATION

Establising Model

Calibraition Independent data 1

Validation Independent data 2

Application

• Calibration is the iterative process of comparing the model with

real system, revising the model if necessary, comparing again, until a
model is accepted (validated)
• Validation is a process of comparing the model and its behavior to the
real system and its behavior
• Sensitivity analysis is the process of defining how changes in model
input parameters affect the magnitude of changes in model output..
16
 APPROACHES TO MATHEMATICAL MODEL BUILDING

 Evaluation of simulation results

There are two methods to evaluate model performance
• Graphical method (qualitative)
• Statistical methods (quantitative)
 The calibration model does not represent accurately possibly due
to the factors multiply as below:
• The model is used incorrectly or the model setting is incorrect
• The model is not suitable for this application
• Lack of data to describe the real world
• Measurement data is not reliable

17
 MATHEMATICAL MODEL BUILDING

 Evaluation of simulation results

• Graphical Method

B. Scatter graph

A. Bar graph

C. Line graph
18
APPROACHES TO MATHEMATICAL MODEL BUILDING

 Evaluation of simulation results (cont.)

• Statistical method
Percent bias – PBIAS : measures the average tendency of the
simulated values to be larger or smaller than their observed ones.

PBIAS  0: indicating accurate model simulation.

: observed value
: Simulate value
n: total observed value / Simulate
19
 APPROACHES TO MATHEMATICAL MODEL BUILDING

 Evaluation of simulation results (cont.)

• Statistical method
Correlation coefficient formulas are used to find how strong a
relationship is between observed and simulate data.

: average value of the series of observed data

: average value of the series of simulated data

20
 APPROACHES TO MATHEMATICAL MODEL BUILDING

• Statistical method
Nash – Sutcliffe ( NSE ): is a normalized statistic that determines the relative
magnitude of the residual variance compared to the measured data variance (Nash
and Sutcliffe, 1970). Nash-Sutcliffe efficiency indicates how well the plot of observed
versus simulated data fits the 1:1 line. NSE = 1, corresponds to a perfect match of the
model to the observed data. NSE = 0, indicates that the model predictions are as
accurate as the mean of the observed data, Inf < NSE < 0, indicates that the observed
mean is a better predictor than the model.

PBIAS
Properties NSE, R2
Flow Water - quality
Very Good 0.75  1.00 <  10 % < 25 %
Good 0.65 0.75  10 %   15 %  25 %   40 %
0.500.65  15 %   25 %  40 %   70 %
Unsatisfactory <0.50 >  25 % >  70 %
(Moriasi et al., 2007) 21
THE END

22