Isobars, and what they are for
Pressure drawn as lines turns a page of measurements into something a forecaster can read at a glance — and the distance between the lines says how hard the wind will blow.
Pressure drawn as lines turns a page of measurements into something a forecaster can read at a glance — and the distance between the lines says how hard the wind will blow.
Reading pressure as a picture
Every surface weather observation includes a barometric pressure reading — corrected to sea level so that a station on a mountain and one at the coast can be compared on equal terms. Feed enough of those readings to a chart, and you have a table of numbers. It tells you a great deal and shows you almost nothing. The isobar is the fix: a line drawn through every point where pressure is equal, at regular intervals — typically every four hectopascals on a synoptic chart. Join the four-hectopascal steps and the numbers dissolve into a landscape.

The landscape has obvious features. Closed loops mark centres of high and low pressure; the tightest loops sit over the most extreme values. Between the loops, isobars run in curves that bow toward one pole or another, tracing ridges of high pressure and troughs of low. None of that requires a legend. A person who has never seen a synoptic chart before will instinctively pick out the lows.
What is less obvious, and more useful, is that the spacing of the isobars is the wind — or rather, the pressure gradient that drives it. Where isobars crowd together, pressure is falling steeply over a short horizontal distance, and air accelerates hard across that gradient. Where they spread apart, the gradient is slack and the wind is light. Forecasters learn to read isobar spacing the way a musician reads note spacing: the density carries the tempo.
Gradient, geostrophic, and where they diverge
The formal name for what isobar spacing encodes is the pressure gradient force. In the free atmosphere — away from surface friction — that force is almost exactly balanced by the Coriolis effect: air moving toward low pressure is deflected sideways until it flows roughly parallel to the isobars, not across them. This equilibrium is called geostrophic flow ↗, and it lets a trained eye translate isobar spacing into an approximate wind speed with a geostrophic wind scale printed on the chart margin.

The approximation holds well over open ocean and at altitude. It fails close to the surface, where friction slows the wind and allows it to cut across isobars at an angle — toward low pressure, into the system. That inflow feeds the ascent that produces cloud and rain. It is why a surface low and a weather event are so reliably the same thing: the geometry of the isobars, and the failure of perfect geostrophic balance near the ground, bring air together and push it upward.
This relationship — pressure gradient to wind, wind to convergence, convergence to ascent — was one of the central insights that Vilhelm Bjerknes formalised in the early twentieth century when he argued that weather forecasting could be made into a rational scientific problem ↗. The Bergen school that followed built synoptic analysis on exactly this logic: draw the isobars, read the winds, locate the fronts where air masses collide. The isobar was not decoration; it was the primary diagnostic.
The limits of a line
An isobar only exists where there are observations to anchor it. In the early era of surface charts, that meant coasts and ships and a sparse interior. The lines between stations were interpolated, and forecasters knew — or learned to know — that the chart was a best estimate, not a measurement. That interpolation problem has not gone away. Even now, the isobars on an analysis chart are as much the output of a numerical model's data assimilation step as they are a direct summary of what was observed. The model fills the gaps; the isobar inherits the model's assumptions.
How the spacing works
- Pressure gradientthe rate at which pressure changes over horizontal distance; steep gradient → strong wind
- Geostrophic windthe equilibrium flow parallel to isobars, where pressure gradient force and Coriolis balance; used to estimate wind speed from isobar spacing
- Hectopascal (hPa)the unit of atmospheric pressure on modern charts; 1 hPa = 1 millibar, a legacy name still in common use
- Four-hPa intervalthe standard spacing between isobars on a synoptic surface chart; tighter intervals would clutter dense lows, wider ones would lose detail
None of which makes the isobar less useful. It compresses complex information into something a human visual system can parse in seconds — which is exactly what it was designed to do. The chart on your weather app, stripped to its simplest form, is still this: a field of pressures, a set of contours, and the wind implied by the space between them.
Elsewhere in The room
Richardson's hall is now a room of screens. Everything in this sectionfollows from that change, from the first machine forecast to the way a model is run today.
- ENIAC, and the first machine forecastLongIn 1950 a machine reproduced in about a day the kind of calculation Richardson had done by hand in six weeks, and the result was good enough to prove the method rather than the weather.
- How a model is actually runLongRunning a numerical weather model is not pressing a button and waiting.
- Somebody still draws the isobarsMediumThe synoptic chart looks like it predates the computer.
- The People on ShiftMediumA forecast office runs continuously, and the handover between shifts is where judgement is actually transferred from one person to the next.