Delta Atlas - The Basin

A resilient thing is not one that never gets hit. It is one that returns to itself after the hit, and holds a wide enough margin that the hit does not throw it somewhere it cannot come back from. This lesson makes both of those measurable, with a bowl, a ball, and a shock you control. Runs entirely on your device; stores nothing.

What this lesson is for. To give one honest, shared picture behind every tool on this site: a system at rest sits in a basin, a shock knocks it out, and resilience is two separate quantities the basin makes visible: how fast it returns, and how hard a shock it can take before it never returns. Once you can see the basin, the audits that follow stop being jargon.
Two resiliences, not one

Ecologist C. S. Holling drew the line in 1973 and it still holds. Engineering resilience is the speed of return to normal after a disturbance: the time to zero. Ecological resilience is the size of disturbance a system can absorb before it flips into a different state entirely: the width of the basin. They are different, and a system can be strong in one and fatally weak in the other. A steep narrow bowl returns fast and tips easily. A broad shallow bowl is almost impossible to tip but recovers slowly. Most real fragility is a system optimized for return speed that quietly lost its width.

The instrument
Kick the ball and watch the time to zero.
Read what the modes are telling you

One rest state. Slide the width. A narrow bowl snaps back fast but a big kick clears it. A wide bowl is slow but nearly impossible to eject from. That trade is the whole subject: you cannot maximize both return speed and shock tolerance, and a plan that only measures one has not measured resilience.

Two rest states. There are now two places the ball can settle, with a ridge between them. The ridge is the separatrix. A small kick is absorbed; a kick past the ridge drops the ball into the other basin, and no ordinary nudge brings it back. The second basin is stable too. This is the hard truth behind a health crisis, a bankruptcy, a broken trust: the bad place is not chaos, it is another basin with real walls, which is exactly why "just climb out" is bad physics.

Losing stability. Watch the return time as the basin flattens. It gets longer, and longer, before anything visibly breaks. That lengthening is critical slowing down, and it is a measured early-warning signal in ecosystems, heart rhythms, and markets (Scheffer and colleagues, 2009). A system taking longer each time to recover from the same size shock is a system whose walls are closing in. The slowing is the smoke detector.

Design rules that survive the mapping

1. Measure width, not just speed. Ask of any plan not only "how fast does it recover" but "how big a shock before it cannot." A backup with no stated restore time, or a control with no failure margin, is a speed claim pretending to be a resilience claim.

2. Do not optimize to the razor. The deepest narrowest minimum is the most efficient and the most fragile state you can build. Leave slack on purpose. The slack is what absorption spends.

3. Track the trend, not the moment. One recovery tells you little. Recovery times getting longer across repeated shocks tell you the basin is eroding while there is still time to act.

Fidelity, marked honestly. For physical and ecological systems the basin, the separatrix, and critical slowing are measured law with the citations below. For a household, a business, a court, or a life, the mapping is an organizing analogy: the same shape of tradeoff, not the same equations. A marriage is not a particle in a potential well. What transfers is the discipline of asking, for anything you care about keeping: how fast does it recover, how hard a hit can it take, and is its recovery getting slower. This page checks structure, not correctness, and it runs entirely in your browser.
This lesson's wordThe established termWhat it names
rest statestable equilibrium / attractorwhere a system settles when nothing pushes it
basinbasin of attractionthe set of states that return to that rest state
time to zeroreturn / settling timeengineering resilience
basin widthecological resiliencethe disturbance absorbed before a flip
the ridgeseparatrixthe boundary past which return fails
slowing recoverycritical slowing downan early warning that resilience is eroding
Sources. C.S. Holling, "Resilience and stability of ecological systems," Annual Review of Ecology and Systematics 4 (1973) · S.L. Pimm, "The complexity and stability of ecosystems," Nature 307 (1984) · M. Scheffer et al., "Early-warning signals for critical transitions," Nature 461 (2009) · A.M. Lyapunov (1892) on stability via a descending function · S. Strogatz, Nonlinear Dynamics and Chaos, on basins, bifurcation, and synchronization.
The simulation is a damped particle in a potential well, integrated in your browser; it is illustrative of the shapes above, not a model of any specific system. Lesson v0.1 (2026-07-11), part of the Delta Atlas explorable series. CC BY 4.0. Back to the atlas · Previous lesson: the Cadence Dial.