By Jayson Chaw. Tue Jul 28.
From atomic diagrams to climate simulations, scientific models are simplifications by design. Here is why that is a strength, not a flaw.
The little diagram of an atom, with a nucleus in the middle and electrons circling it like tiny planets, is one of the most recognisable images in science. It is also, strictly speaking, wrong: electrons do not orbit in neat circular paths. This does not make the diagram useless. It makes it a model, a deliberately simplified representation of something too complex, too small, or too abstract to observe or explain directly in full detail.
What a model is actually for
A scientific model is a tool for thinking, not a claim of perfect accuracy. Models exist to make complicated systems easier to reason about, to predict, and to communicate, by stripping away detail that is not relevant to the question being asked. A map is a useful comparison: a map of a train network is enormously simplified compared with reality, distorting distances and ignoring most physical features, yet it is far more useful for navigating the system than a perfectly accurate satellite image would be.
Models change as understanding improves
The atomic model is a good example of how models evolve. Early models pictured the atom as a solid, uniform sphere. Later evidence led to a model with a central nucleus surrounded by electrons in fixed orbits, followed by an even more refined model describing electrons as existing in probability clouds rather than fixed paths. Each version was not simply wrong and then corrected; each was a model useful for its time, replaced by a more accurate one as new evidence demanded it.
A model can be useful and incomplete at the same time
This is often summarised by the well-known statement, commonly attributed to the statistician George Box, that all models are wrong but some are useful. The simplified orbital diagram of the atom is technically inaccurate, yet it remains genuinely useful for explaining chemical bonding at an introductory level. Rejecting a model because it is not perfectly accurate misses the point of what a model is trying to do in the first place.
Models across different sciences
The particle model of matter, used to explain states of matter and changes between them
Food webs and energy pyramids, used to represent feeding relationships in an ecosystem
Genetic diagrams and Punnett squares, used to predict the likelihood of inherited characteristics
Circuit diagrams, used to represent how components in an electrical system are connected
Climate models, used to represent how energy moves through the atmosphere and oceans over time
…