AI summary by TwinMind
The Butterfly Effect is Much Worse Than We Thought
Sabine Hossenfelder
7m 24s
Butterfly Effect Background
- Traditional butterfly effect: small disturbances in one place can grow to large disturbances elsewhere
- Edward Lorenz 1969 paper introduced concept using simplified equations (Lorenz model)
- Lorenz model became most famous example of chaos theory
- Solutions very sensitive to exact starting conditions
- Tiny differences lead to completely uncorrelated outcomes
- Solutions plotted in 3D space converge to butterfly-shaped attractor
- Original paper discussed seagulls, not butterflies
- Lorenz model doesn't use Earth surface coordinates, can't tell actual butterfly effects in China
Weather Chaos Complexity
- Weather isn't always chaotic - sometimes chaotic, sometimes not
- Weather forecast accuracy varies: 2 weeks in some places, can't predict next morning elsewhere
- Scientists confirmed perturbations from ~100km scales influence much larger areas, potentially entire planet
- Energy consideration: changes requiring little energy can move huge amounts of energy long-term
Navier-Stokes Equation Challenge
- Atmosphere governed by Navier-Stokes equation
- Nonlinear equation notoriously hard to solve
- Gives rise to turbulence and chaos
- Clay Institute offers $1 million prize for solution
- Scientists use computer simulations with finite-size grids as approximations
New Research Findings
- Recent paper published in Physical Review Letters (PRL)
- Computer simulations show molecular motion sufficient to trigger turbulence globally
- "Inevitably present molecular noise is sufficient to trigger spontaneous stochasticity"
- Even molecules can cause butterfly effect, not just butterflies, elephants, or jumbo jets
Implications for Scientific Modeling
- Affects understanding of multiple systems:
- Climate modeling
- Nuclear fusion
- Galaxy formation
- Climate models: even projected 1km horizontal resolution in next decade won't eliminate need for stochastic models
- Impossible to capture all noise sources - would need to go down to molecular size
- More powerful computers with smaller grids won't solve the fundamental problem
Broader Scientific Context
- Potential relevance to Jonathan Oppenheim's post-quantum gravity work with stochastic noise
- Galaxy formation modeling likely affected
