IB ESS SL1.2 SystemsPaper 1 & 2Core skill~11 min read
Using Models in ESS
You cannot run an experiment on the atmosphere. You cannot wait fifty years to see whether a policy works. So environmental scientists build a simplified copy of the system and experiment on that instead. Every diagram in these notes has been one.
📚 What you need to know
A model is a simplified version of reality, usually used to represent a system.
Models are analysed and tested to learn how a system works and to predict how it will respond to change.
Models take many forms: a graph, a diagram, an equation, a computer simulation, even a description in words.
All models involve approximation and simplification, so all models lose some accuracy.
Strengths: they simplify complexity, allow predictions, and let you change inputs without waiting for real events.
Limitations: output quality depends on input data, uncertainty grows the further ahead you predict, and different models can disagree.
Daisyworld is a well-known model showing how life could regulate a planet’s temperature by negative feedback.
What a model actually is
Definition
A model is a simplified version of reality, built to represent a system so that it can be studied, tested and used to make predictions.
The point of a model is not to be right about everything. It is to be simple enough to work with while still behaving like the real thing in the ways that matter.
Notice the two lines in the last panel. Feed a model slightly different assumptions and it will happily give you two different futures.
The many shapes a model can take
Form
Example from this course
Diagram
A systems diagram of a tree, with storages and flows
Graph
Predator and prey populations plotted against time
Equation
A calculation of net primary productivity
Physical model
A sealed terrarium standing in for a closed system
Computer simulation
Climate models run on supercomputers; Daisyworld
Words
A written description of how a feedback loop works
Some models are very simple, like a child’s toy car. Others need the power of a supercomputer, like the climate models used to predict how our climate will change. Both are doing the same job.
Every ball-and-valley picture, every loop of boxes, every graph in the last four pages was a model. If an exam asks you to give an example of a model in ESS, you already have a dozen — just say which system it represents and what it leaves out.
Daisyworld: a model doing its job
James Lovelock and Andrew Watson built the Daisyworld computer simulation in the 1980s to test an idea: could life itself keep a planet’s temperature stable?
The imaginary planet has only two organisms: black daisies and white daisies.
They affect the planet only through albedo — how much solar radiation the surface reflects.
As solar luminosity rises, black daisies do well because they absorb more sunlight and stay warm.
More black daisies means lower albedo, so more heat is trapped and global temperature rises.
The warmer planet now suits white daisies, which spread.
More white daisies means higher albedo, so more sunlight is reflected and temperature falls.
The two populations compete and settle into a steady-state equilibrium that holds the surface temperature stable, and both species survive long term.
Run the same model with no daisies and the planet’s climate drifts to an extreme, hot or cold, and cannot support life.
That last point is the whole reason the model was built. It shows how life itself can act as a negative feedback mechanism regulating global temperature — the core of the Gaia hypothesis, tested in a place where you can safely turn the daisies off.
Strengths and limitations
Strengths
Limitations
They simplify complex systems into something you can actually think about
They can be oversimplified, and therefore inaccurate
They allow predictions about how a system will react to change
Results are only as good as the data put into them
Inputs can be changed to see the effect, without waiting for real events
Results become more uncertain the further into the future they predict
They are easier to understand than the real system
Different models can give very different outputs from the same data
Results can be shared between scientists, engineers and companies, and explained to the public
The same results can be interpreted differently by different people
They can warn us about future environmental problems in time to act
Environmental systems are so complex that no model can include every variable
🧠
Easy way to remember it
Every strength has a matching weakness. Simple is useful but simplistic is wrong. Predicts the future is useful, but the further ahead, the shakier the prediction.
Evaluating a model in an exam
🧩 A method that works every time
Say what the model represents. Which system, and which parts of it.
Give one thing it does well. Usually: it makes a complex system understandable, or it lets us predict without waiting.
Say what it leaves out. Every model simplifies — name a factor it ignores.
Comment on the data. Poor input data gives poor output, however clever the model.
Comment on time. Predictions get less reliable the further ahead they go.
Finish with a judgement. Useful for X, but should not be treated as fact. That final sentence is what “evaluate” is asking for.
Why two models disagree. Give two climate models the same data and they can still produce different outputs, because each one makes different assumptions about which processes matter and how strongly they are linked. This is not a scandal — it is why scientists run many models and look at the range of results rather than trusting a single number.
Worked examples
WE 1
Outline the value of models
Outline two strengths and two limitations of using models to study environmental systems. (4 marks)
Strength 1
Models simplify complex systems, making them easier to understand than the real thing.
Strength 2
Inputs can be changed to see the effect, so predictions can be made without waiting for real events to happen.
Limitation 1
Results depend entirely on the quality of the data put in, and become more uncertain the further ahead they predict.
Limitation 2
Different models can produce very different outputs from the same data, and results can be interpreted differently by different people.
Useful because they simplify; unreliable for the same reasongive two of each — a one-sided answer cannot reach full marks on this command term
WE 2
Explain what Daisyworld shows
Explain how the Daisyworld model demonstrates negative feedback. (4 marks)
Step 1
Rising solar luminosity favours black daisies, which absorb more sunlight, so albedo falls and the planet warms.
Step 2
The warmer planet then favours white daisies, which spread.
Step 3
White daisies raise the albedo, so more solar radiation is reflected and the planet cools again.
Step 4
Each change triggers the response that reverses it, so the populations reach a steady-state equilibrium and temperature stays stable.
Life adjusting albedo acts as a stabilising negative feedback loopmention the dead planet with no daisies, where climate becomes extreme — it shows you understand what the model is testing
WE 3
Evaluate a model
A climate model predicts global temperature in the year 2100. Evaluate the usefulness of this model. (4 marks)
Point 1: what it is good for
It simplifies an extremely complex system and allows predictions to be made without waiting for the events themselves.
Point 2: it supports action
Results can be shared with governments and the public, warning of future problems in time to reduce them.
Point 3: the weaknesses
Output quality depends on the input data, uncertainty grows the further ahead the prediction goes, and no model can include every interacting variable.
Point 4: the judgement
Useful for showing trends and comparing scenarios, but the exact figure for 2100 should be treated as an estimate, not a fact.
Good for direction of travel, weak on precise numbers“evaluate” needs a final judgement sentence — without one you are only describing
💡 Exam tips
Learn three strengths and three limitations. Most model questions are worth 3 or 4 marks.
Say that all models involve approximation and simplification. It is the safest sentence in the topic.
For “evaluate”, always end with a judgement.
Know Daisyworld well enough to describe the loop in four steps.
If you are asked for an example of a model, a systems diagram is a perfectly good answer.
Remember that uncertainty grows with time, and say so whenever a question mentions a future date.
⚠ Common mistakes
Saying models are wrong. They are simplified, which is different, and it is deliberate.
Only listing strengths. Nearly every question wants both sides.
Thinking a model must be a computer program. A labelled diagram is a model.
Calling Daisyworld positive feedback. It is the standard example of negative feedback.
Blaming disagreement between models on bad science. Different assumptions give different outputs; that is expected.
Describing instead of evaluating. No judgement sentence, no top marks.
That completes 1.2 Systems. The five notes tell one story: a system is storages and flows, what crosses the boundary decides its type, feedback holds it steady or shoves it off course, resilience decides how much shoving it can take, and models are how we study all of it without breaking anything. Up next: 1.3 Sustainability, where you put these tools to work on real decisions.
Want this explained one-to-one?
Book a free session with an experienced IB ESS tutor and get your trickiest topics made simple.
