Environmental Modelling Flashcards
What is the difference between a static and a dynamic model?
A static model represents a moment in time, whereas a dynamic model represents changes over time.
Which type of models requires more scientific knowledge of the system: explanatory or process-based models?
Explanatory or process-based models are the same thing and it requires more scientific knowledge than a descriptive model.
Give three reasons why you would use a model in Environmental Systems Analysis.
To quantify the cause and effect; Identify and understand hidden feedback loops; Support communication to stakeholders and decision making
How could you use the results of sensitivity analysis in follow-up research?
To determine which variable/ parameter needs to be investigated with more care/ in more detail. To collect more data on that sensitive variable/parameter.
How could you integrate the ESA tools Scenario Analysis and Environmental modelling?
The inputs of environmental modelling could be used to compare different scenarios.
Is this a positive or a negative feedback loop? Panic among traders leads a stock to crash, this will lead to the massive sale of shares, causing the prices to fall even further.
Positive
Can agriculture and human population be considered to be in a positive feedback loop?
Depends, because more factors are at play and should be taken into account.
What different types of models are there?
Descriptive and explanatory; static and dynamic.
What are the uses of models?
- Help describe and represent the dynamic behaviour of a system, e.g. shifts in potential vegetation as a response to shifts in climate.
- Help quantify cause and effect, e.g. to find if building a new dam will negatively affect the endangered native species found at the proposed dam site.
- Help investigate interactions, eg. the effects of clouds on the climate.
- Help test the degree of completeness of understanding or process (are we able to model a system).
- Make a tool to experiment, eg to perform experiments with the system without altering and thus potentially harming the system.
- Help investigate alternative future states under alternative management interventions.
What are the limitations of models?
- “Garbage in - garbage out” principle holds for every model.
- Users of a model need to understand the assumptions
- Model outcomes are to a certain extent always uncertain.
- Applying a model to another place or system for which it has been developed should be done with care
- Models cannot be used for specific prognosis (i.e. predictions), but can be used for scenario analysis
What are the important steps in model development and application?
- Identify key elements of the system. Make sure the elements are quantifying e.g. food demand [kg/capita]. Avoid vague concepts such as tastes, style etc.
- Identify driving forces
- Identify interactions which should be included in the system
- Identify data that is needed
- Define conceptual (or causal) model: defining relationships in the system ( including feedbacks). Cause - effect diagrams provide first overview.
- Create flow diagram (more refined). Flow diagrams can be used to derive mathematical equations. State (t+delta t) = state (t) + delta t * rate (t)
- Model evaluation:
- Does the model perform as we expect?
- Calibration: minimize differences between model output and measured values by adjusting parameters within reasonable limits.
- Sensitivity analysis: which factors ( input and parameters) determine the output?
- Validation: does the model mimic the system dynamics from independent datasets? ( standard approach, can also be done for historic conditions)
- Uncertainty analysis ( based on knowledge from previous steps assess): which parts of the model structure is most uncertain and which parameters?
- Model and data comparisons ( benchmarking, peer-review)
- Model experimentation and scenarios
What is integrated assessment models ( IAMs)?
- IAMs integrate knowledge of different scientific disciplines and generate information for policy making. IAMs need to be sufficiently detailed to address the problems, yet simple enough to be applicable in assessments.
What is the structure of IAMs?
- The major building blocks are well-established models from specific disciplines, e.g. climate change models, demographic models etc.
- Often simplified versions of these sub-models have to be developed
What are the strengths of IAMs?
- Consistent integration between human and earth systems
- Ability to investigate policy strategies in terms of trade-offs and synergies
- Ability to discover feedbacks between different domains
- Large spatial coverages and long time scales
What are the disadvantages of IAMs?
- Necessary trade-off between specialisation (depth of coverage) and integration ( breadth of coverage)
- Problems of combining qualitative and quantitative research
- High complexity, may be not user friendly
- May be high demand for computer power
- Not appropriate for short-term forecasts
- Uncertainty analysis very difficult because of the amount of various components
