Skip to content

Intermittency

Core Idea

The occurrence of irregular bursts or sporadic intensifications within an otherwise continuous or background process, leading to spikes of activity or output at unpredictable intervals.

How would you explain it like I'm…

Quiet Then Burst

Think about a faucet that mostly just drips, drip, drip, drip — but every so often, it suddenly shoots out a big splash of water, then goes back to dripping. Most of the water comes from those rare big splashes, not from all the little drips. When something acts quiet for a long time and then has a big burst, that pattern is called intermittency.

Rare Big Bursts

Intermittency is when something is usually pretty calm but every once in a while bursts out really big, then goes calm again. Think of an old volcano: quiet for a long time, then a huge eruption, then quiet again. If you add up all the energy, most of it comes from the rare eruptions, not from the calm times. Lots of things in nature work this way — wind gusts, earthquakes, even some computer network traffic — quiet stretches broken by surprise spikes.

Intermittency (Bursty Behavior)

Intermittency is the pattern where a signal is mostly quiet or mildly fluctuating but is punctuated by sporadic, high-amplitude bursts at irregular intervals. The key fact is that the statistics are dominated by those rare bursts rather than by the background: the quiet stretches add little to the totals, while a few big events can account for most of the variance and most of the cumulative impact. You see this in turbulence, earthquakes, solar flares, neuron firing, and financial markets. Because rare extremes carry the action, averages calculated from quiet periods badly underestimate what the system is really doing — which is why intermittent processes need different statistical tools than smoothly varying ones.

 

Intermittency is the pattern in which an otherwise quiescent or mildly fluctuating signal exhibits sporadic, high-amplitude bursts at irregular intervals. The essential commitment is that the signal's statistics are dominated by the burst distribution, not by the background: quiet periods contribute little to totals while rare events can carry most of the variance, higher moments, and cumulative impact. Characterizing an intermittent process requires four things: (1) the signal or process exhibiting bursts; (2) the statistical signatures (heavy-tailed distributions — distributions where extreme values are far more common than a Gaussian predicts; flatness > 3, meaning the fourth moment exceeds what a normal distribution would give; burst clustering; multifractal spectra); (3) the underlying mechanism (turbulent cascade, self-organized criticality — where the system tunes itself to a critical state — regime switching, threshold phenomena); and (4) the time and amplitude scales over which the intermittent structure lives. The classic taxonomy of Types I, II, and III came from Pomeau and Manneville's 1980 analysis of route-to-chaos transitions in dissipative dynamical systems.

Broad Use

  • Meteorology: Gusty winds or sudden bursts of precipitation during storms.

  • Data Traffic: Network traffic experiencing intermittent surges of requests.

  • Finance: Volatility "spikes" interrupting relatively calm trading periods.

  • Neural Activity: Neurons firing in short bursts separated by periods of relative calm.

Clarity

Pinpoints where a system does not change smoothly over time but instead exhibits significant, irregular peaks of energy or activity.

Manages Complexity

Focuses attention on brief, high-impact events rather than trying to model behavior as a continuous, uniform process.

Abstract Reasoning

Encourages understanding that steady averages can mask extreme local fluctuations, prompting analysis of burst patterns and their triggers.

Knowledge Transfer

Helps researchers in diverse fields—from meteorology to network engineering—recognize and adapt to sudden, powerful surges in system behavior.

Example

Turbulent Airflow: Small eddies can intermittently break off into more intense bursts of turbulence, impacting aircraft stability.

Relationships to Other Primes

One-hop neighborhood: parents above, mutual partners to the right, children below.Intermittencycomposition: RhythmRhythmsubsumption: Heavy-Tailed DistributionsHeavy-TailedDistributions

Parents (2) — more general patterns this builds on

  • Intermittency is a kind of Heavy-Tailed Distributions — Intermittency is a specialization of heavy-tailed distributions in which rare high-amplitude bursts dominate an otherwise quiescent signal.
  • Intermittency presupposes Rhythm — Intermittency presupposes rhythm because identifying bursts as exceptional requires a baseline expectation of patterned recurrence against which they register as exceptional.

Path to root: IntermittencyHeavy-Tailed Distributions

Not to Be Confused With

  • Intermittency is not Oscillation because intermittency is the bursting between quiet and active phases, whereas oscillation is sustained repetitive cycling; oscillation is regular and continuous, intermittency is irregular and episodic with quiet periods interrupting activity.
  • Intermittency is not Noise because intermittency is a structured alternation between distinct dynamical regimes, whereas noise is random fluctuation without coherent structure; intermittency has a detectable pattern (intervals of different behavior), noise has no such organization.
  • Intermittency is not Chaos because intermittency can occur in a structured sequence of bursts and quiescence, whereas chaos is continuous sensitive dependence within a bounded attractor; intermittent behavior can be chaotic during active phases, but the overall pattern is episodic rather than pervasive.