Earth has a centre of mass, but it is not a point permanently fixed inside the rock. Water, air and snow move across the planet through the seasons. Because they have mass, those migrations move the balance point of the complete Earth system by several millimetres relative to the solid surface.
A study published in Geophysical Journal International has produced two new estimates of that annual movement by combining satellite laser ranging, GPS measurements and the tracked orbits of low-Earth satellites. This is one study, not settled consensus. Its main contribution is a cleaner way to separate movement of the mass centre from deformation of the ground stations used to measure it.
The distinction is important. Earth is not being knocked off course by the monsoon, and its core is not sliding back and forth. Geocentre motion is the changing offset between two reference points: the centre of mass of the whole Earth system and the geometric centre of the solid planet.
Earth has several useful centres
For an everyday globe, “the centre” seems unambiguous. At geodetic precision, the choice depends on what is included. The total-system centre of mass counts the solid Earth, oceans, atmosphere, ice, soil moisture, rivers and groundwater. It is the balance point about which satellites orbit.
The centre of figure describes the geometry of the solid surface. A third quantity, the centre of a geodetic network, is obtained from the positions of tracking stations. That network centre has often served as a practical approximation to the centre of figure, even though the stations are neither perfectly distributed nor perfectly still.
It is equally valid to say that the mass centre moves relative to the crust or that the crust’s geometric centre moves relative to the mass centre. The offset is relative. It is not a claim that Earth as a whole wanders by millimetres across its solar orbit.
Nor is this polar motion, the movement of Earth’s rotation axis relative to the crust. Mass redistribution can influence both phenomena, but the quantities describe different pieces of the planet’s dynamics.
Satellites follow the mass
An orbit responds to gravity, so an artificial satellite naturally moves around Earth’s total centre of mass. Geodesists exploit that fact by tracking satellites from the ground and solving backwards for the reference point their orbits reveal.
The two LAGEOS satellites are particularly useful. Each is a dense, passive sphere covered with retroreflectors. Ground stations fire laser pulses upwards and time their return, building an extremely precise record of the satellite’s changing distance. SpaceDaily has previously examined how little else LAGEOS-1 needs to do its job.
Traditional satellite laser ranging has a geographical weakness. Roughly half of its observing stations are in Europe and North America, according to the study. Taking the average coordinates of an uneven network does not reproduce the centre of the solid surface exactly.
The new analysis therefore draws on more than one observing system. One solution uses conventional satellite laser ranging. A second integrates laser ranging with GPS tracking and the orbits of low-Earth satellites. GPS stations have a broader global distribution, while the satellites supply additional information about the total centre of mass.
The measuring points move too
Uneven coverage is only part of the problem. Rain, snow, ocean water and air pressure do more than redistribute mass horizontally. Their changing weight pushes down on the crust. The solid Earth bends elastically beneath a load and rebounds as that load moves or disappears.
A ground receiver can therefore shift because the ground beneath it is responding to the same seasonal water and atmosphere that the network is trying to locate. If that elastic movement is left in the station coordinates, part of the loading signal can be mistaken for motion of the reference frame.
Donald Argus of NASA’s Jet Propulsion Laboratory and his co-authors approached the problem by starting with models of seasonal mass in the atmosphere, oceans and water stored on continents. They calculated the elastic displacement expected at each geodetic site, then subtracted it from the site’s measured position.
The mean of the adjusted positions supplies an estimate of the solid Earth’s centre of figure. In principle, the method does not require stations to be distributed almost uniformly over the globe. If the elastic displacement at a site is known accurately, even that site’s position contains information about the centre of figure.
That does not make the result assumption-free. The correction depends on models of where water and air are located and how the solid Earth responds to their weight. The study is best read as a new way to make several observing systems agree, not as the final possible description of geocentre motion.
Snow and the Amazon lead the cycle
Seasonal snow over Eurasia and North America reaches its modelled maximum around March. In NASA’s account of the analysis, that northern load moves Earth’s centre of mass by about three millimetres towards the North Pole.
The Amazon basin follows on another axis. At its April maximum, the team’s continental-water model places about 2,400 gigatonnes of additional water across the region. That is associated with a shift of about 2.2 millimetres towards South America.
Monsoon water across Southeast Asia supplies a smaller but still measurable component later in the year. NASA gives its seasonal mass at roughly 600 gigatonnes. The exact timing depends on the model and on how the region and its storage cycle are defined, but the physical point is straightforward: water parked on land changes the planet’s balance.
From August into October, ocean water reaches its annual maximum in the model used by the researchers. The Pacific contribution dominates, moving the centre towards the South Pacific. These values are components of a three-dimensional path, not a list of distances that can simply be added into one large displacement.
The cycle is also not a perfectly fixed annual loop. Drought, floods, snow variability and large atmosphere-ocean patterns alter the distribution from one year to the next. The paper estimates a recurring seasonal signal; it does not say every year must trace an identical route.
Air completes the picture
The atmosphere contains far less mass than the oceans or solid Earth, but it moves enough weight to matter when the target precision is a millimetre. Air pressure is mass per unit area, and the high- and low-pressure systems familiar from weather maps are therefore part of the geocentre calculation.
The authors used an atmosphere model from the European Centre for Medium-Range Weather Forecasts. It placed the maximum in the atmosphere’s seasonal mass contribution in July, about three months after the continental-water maximum and three months before the ocean maximum.
The geographical pattern changes with winter. Denser air adds weight across Arabia, Asia and North Africa around the northern winter solstice, then across South America and southern Africa near the southern winter solstice. Air is the lighter part of the ledger, but leaving it out would blur the timing and direction of the total motion.
Four approaches now sit closer together
The paper compares its two orbit-based estimates with two other answers: the seasonal geocentre model built into the International Terrestrial Reference Frame 2020, and a result inferred from the shifting gravity field measured by the GRACE satellite mission.
Across those four approaches, the estimates of the annual peak-to-peak oscillation agree to within 2.5 millimetres. That is a meaningful tightening because the 2017 and 2023 international reference-frame estimates had differed by about seven millimetres, nearly as large as the seasonal movement being sought.
The new annual estimate is also about half the magnitude believed eight years ago. That change should not be read as Earth suddenly becoming more stable. It reflects a revised measurement of a small signal that has always been difficult to distinguish from the behaviour of the observing network.
GRACE offers a useful independent comparison because its paired spacecraft detect changes in Earth’s gravity field as water moves. SpaceDaily has described the extraordinarily small distance changes behind that measurement. Agreement with that mass-sensitive method makes the orbit solutions more persuasive.
Why millimetres matter
A few millimetres is less than one part in a billion of Earth’s radius. It will not be noticed by a person walking across the ground, and it does not mean an ordinary phone’s location jumps visibly with each rainy season.
The importance lies in the global reference frame beneath precision geodesy. A satellite orbit, a sea-level record and a map of changing ice or groundwater all need an origin. If that origin is misplaced, some of the error can leak into the quantity being measured.
Long-term signals make the problem particularly demanding. Sea-level rise, ice-sheet mass loss and the movement of stored water are inferred by comparing measurements made across years. A seasonal wobble in the coordinate origin must be distinguished from an enduring change in the planet.
The result also demonstrates why several techniques are needed. Laser ranging ties orbits to ground stations, GPS widens the network, low-Earth satellites add orbit information, and GRACE-type gravity measurements test whether the inferred motion follows the mass being moved. Each method carries a different set of weaknesses.
A moving balance point, not a moving world
The study does not describe a single lump of material travelling through Earth’s core. The centre of mass is a calculated balance point. It shifts because the distribution around it changes, much as the balance point of a tray changes when a glass is moved from one side to the other.
Most of the mass remains where it was, and the seasonal transfers reverse or move on. Snow melts. Rivers carry water back to the ocean. Monsoon rain drains, evaporates or enters storage. Pressure belts migrate with the Sun. The annual motion is the combined geometry of those familiar processes.
The geocentre is therefore both abstract and physically consequential. No instrument can be placed at Earth’s centre to watch it move, yet satellite trajectories and ground coordinates carry the displacement in their measurements.
The result is a useful reminder that even the origin of a planetary coordinate system is something that must be measured. Each wet season, snowfall and pressure pattern rearranges only a tiny fraction of Earth’s mass, yet modern geodesy is precise enough to watch the planet rebalance.