Earth is quietly dusted with thousands of tonnes of space material every year, most of it as grains smaller than sand — and some of it can be sifted from the grit of ordinary roof gutters, tiny meteorites hiding in plain sight in cities.

Extraterrestrial dust falls on every part of the Earth, including the parts covered in cities. Most of it arrives as grains a few hundredths to a few tenths of a millimetre across, finer than a grain of sand or close to it, and it lands without anyone noticing. Some of that material can, with effort, be recovered from the grit that collects in ordinary roof gutters. The effort is the part usually left out.

How much arrives, and what it is

The dust is made of micrometeorites: tiny fragments of asteroids and comets that survive the fall through the atmosphere and reach the ground. They are the small end of a size range that runs up to full meteorites, and by mass they are the dominant form of extraterrestrial material arriving on Earth.

How much arrives is a genuinely difficult thing to measure, and estimates have varied. The most carefully controlled figure comes from a 2021 study in Earth and Planetary Science Letters by Julien Rojas, Jean Duprat and colleagues, based on roughly two decades of collecting from ultra-clean snow at Dome C in central Antarctica. They put the flux of micrometeorites reaching the Earth’s surface at about 5,200 tonnes a year, with a stated uncertainty of around 1,500 tonnes. The total mass of dust entering the atmosphere before any of it burns up is higher, estimated in the same work at around 15,000 tonnes a year.

For comparison, the material arriving as larger meteorites amounts to less than ten tonnes a year. The dust does the bulk of the delivering.

Most micrometeorites do not arrive intact. The heat of atmospheric entry melts them, and they reach the ground as small solidified droplets known as cosmic spherules. Unmelted micrometeorites, which preserve more of their original mineralogy, are the rarer and more scientifically valuable kind.

Why cities were considered hopeless

For most of the time micrometeorites have been studied, they were collected from places with as little human material as possible: Antarctic ice, deep-sea sediments, remote deserts. Antarctica in particular offers clean snow and a low background of terrestrial dust.

Cities were considered the opposite of that. The assumption was that urban environments were too contaminated, too full of industrial particles, for any genuine micrometeorite to be picked out. This was not idle caution. Magnetic spherules had been gathered from urban areas as far back as the 1940s, and studies in the 1950s concluded that those particular spherules were artificial. The lesson taken from that was that the city was a dead end.

The urban collection

That assumption was overturned, slowly, by a Norwegian musician. Jon Larsen, a guitarist with no formal background in the field, became interested in cosmic dust around 2009 after a tiny dark grain landed on a table where he was eating breakfast. He began collecting roof debris and sorting it, an effort he called Project Stardust.

The work was eventually taken up with professional researchers. In 2017, Matthew Genge of Imperial College London, with Larsen and two colleagues, published a paper in the journal Geology reporting around 500 micrometeorites recovered from urban rooftops, mostly in Norway. Because they had fallen recently, within roughly the previous six years, they were the youngest large micrometeorites collected up to that point.

The part the headlines skip

The story is often reduced to a simple instruction: run a magnet through your gutter and you will pull out meteorites. That oversells it considerably.

Almost everything magnetic in gutter grit is of human origin. Cities produce a steady supply of small metallic and glassy spheres: spray from welding, debris from grinding wheels and high-speed drills, residue from combustion. To a casual eye, and even to a magnet, these are indistinguishable from cosmic spherules. They are the imposters, and the title of Larsen’s first book, which pairs micrometeorites with their terrestrial imposters, makes the point plainly.

Telling the two apart is not a matter of looking harder. It requires laboratory analysis. In the 2017 study, particles were examined under a scanning electron microscope and measured by electron microprobe. A genuine cosmic spherule carries a specific chemical signature: the relative abundances of elements such as aluminium, calcium, titanium, magnesium, silicon and iron match those of chondritic meteorite material, and certain volatile elements are depleted in a way consistent with heating during atmospheric entry. That signature, not the magnetism and not the shape, is what confirms an extraterrestrial origin.

The genuine contribution of the urban work was not the discovery that space dust lands on roofs. That was always likely. It was the development of a reliable way to separate the real particles from a far larger population of convincing fakes.

Why the collection is worth having

Urban micrometeorites earn their place for a specific reason. Antarctic and deep-sea collections are difficult to reach and contain particles of varied and often considerable age. Rooftop particles are recent and their fall can be roughly dated, which makes them a clean modern reference sample.

That has a use beyond novelty. The 2017 paper noted that the proportions of different spherule types in the modern urban sample did not match the average seen in older deposits, which the authors read as a sign that the dust flux has not been constant through the recent geological past. A present-day sample gives one fixed point against which the older record can be compared.

The dust was always falling on cities. What changed was not the rain but the ability to find a few real grains of it in the noise of everything else a city sheds onto its own roofs.