Environmental Modeling

1
Environmental Modeling

• Steven I. Gordon
• Ohio Supercomputer Center
• sgordon_at_osc.edu
• June, 2004

2
Environmental Models Offer Many Options

• Many models
• Atmospheric processes
• Hydrologic processes
• Ecological systems
• Natural hazards
• Many interactions
• Many scales
• Local habitats
• Regional mesoscale
• Global

3
Problems in Instruction

• Modeling complex, dynamic systems
• Changes occur both spatially and temporally
• Quality of data to confirm model validity often
  questionable
• High degrees of uncertainty
• Many different processes cross disciplinary
  boundaries
• Challenge for students with varying background
• Challenge for faculty trying to apply to
  instruction

4
Mixed Approaches

• Models based on physical theory
• Fluid dynamics
• Mechanics
• Biochemistry
• Models based on statistical and empirical
  estimates
• Used to simplify the complex dynamic systems
• Based on abstractions that do not always apply

5
Many Places Many Parameters

• Requirements for data describing initial
  conditions at each place in the model
• Amount of data required dependent on model scale
• Data acquisition difficult
• Increasing availability of spatial data from
  public sources
• Most models embed many parameter choices
• Values found under different circumstances
• Calculated based on different principles
• Choices can make model use decisions dizzying

6
Basic Model Components

• State variables describing status as different
  places at time zero
• Flow over time and space of matter, energy,
  organisms
• Transformation of physical, chemical, or
  biological characteristics over time

7
Alternative Representations

• What governs the movement from one place to
  another?
• How does movement vary with changes in
  environmental conditions? How is this change
  represented (steady steady, stochastically,
  statistically)?
• How will space be represented implicitly, one,
  two or three dimensions?

8
First Example Dissolved Oxygen in a Stream

• Measure of health ability to support diverse
  ecosystem
• Basic relationship
• Inversely related to temperature
• Range between 0 and 14 ppm (mg/l)

9
Conceptual model

• Organic waste (BOD) decomposed by bacteria that
  use oxygen
• DOf(1/BOD)
• Two processes
• Deoxygenation
• Reaeration

10
Graphical Representation of Point Discharge and DO
11
Basic Equations

• Where D dissolved oxygen deficit over time
• L concentration of organic matter requiring
  decomposition
• k1 coefficient of deoxygenation
• k2 coefficient of reaeration

12
Stella Model Example
13
Excel Engineering Version

• Qual2K
• Based on EPA code for DO called Qual2E
• http//www.epa.gov/athens/wwqtsc/html/qual2k.html
• Example run

14
Complexity of the Model

• Choose which aspects to focus on
• Leave the rest as a black box
• Create an exercise that focuses on variables of
  interest
• E.G. BOD load sensitivity to reaeration
  parameter and temperature

15
Simple Point Source/Receptor Model
16
Simple Point Source/Receptor Model
17
Simple Point Source/Receptor Model
18
Gaussian Plume Model

• Dispersion in downwind direction proportional to
  wind speed (x)
• Dispersion in y and z direction normally
  distributed along the plume center line
• Mean concentration and dispersion vary with
  stability class in known empirical quantity

19
Equation
20
Where

• C (x,y,z) concentration of pollutant in 3
  dimensions given steady state emission
• X horizontal distance from source in direction
  of wind vector and along plume centerline
• Y horizontal distance perpendicular to and
  measured from the plume center line
• Z vertical distance from ground to plume center
  line
• Q mass rate of production of pollutant over time

21
Where

• U mean wind speed in the x direction
• H effective height of plume

22
Equation
Emission dispersed as statistical dispersion in 3
directions
Dispersion in cross-wind and vertical dimension
23
Empirically Solve for Coefficients
24
Solving the Equation

• Probability distribution of different wind speed,
  direction, stability class occurrence
• Solve the model for each condition
• Weight the answer by the frequency of each
  condition

25
Stability Wind Rose
26
Excel Version of the Model
27
The Climatological Dispersion Model
28
Alternative Approaches

• Find a simple version of a model to run in Stella
  or a spreadsheet
• Have students add to the simple model by taking
  advantage of the documentation/discussion in more
  complex models
• Run a more complex model but vary only a few
  variables most relevant to the class topics

29
Create and Test a Set of Exercises

• Regardless of approach need to carefully
  prepare instructions that include
• Readings on the model basis
• Step-by-step instructions
• Realistic scenarios
• Clear list of expected exercise outcomes
• Opportunities for feedback

30
Sources of Information

• See attached sheet with web links to a variety of
  modeling and data sites