How often are meteorites seen to fall?

Over the sixty years from 1950 to 2009, NASA’s Meteorite Landings dataset (records to 2013) gains 338 observed falls, about 5.6 a year. That figure counts falls that were seen, collected and formally recorded, not every rock that reaches the ground. It is the honest yardstick the record offers, and it has changed a great deal over two centuries.

Five and a half a year since 1950

The table splits the record into half-centuries. The rate climbs through the 1800s, peaks in the first half of the twentieth century and then settles a little lower. The last row stops in 2013, where the dataset ends, so its rate rests on only fourteen years.

Seen decade by decade the shape is clearer. The 1930s hold the most falls, 88, and the busiest single year in the whole record is 1933, with 17 falls; 1949 follows with 12. The record cannot tell whether a busy year had more fireballs or simply more stones that were picked up and reported.

Since 1950 the decades are remarkably even: the quietest full decade, the 1980s, has 54 falls and the busiest, the 1950s, has 58. That stability is what makes the long-run average of about five and a half a year a fair summary of the modern record, as long as it is read as a count of recorded falls.

The busiest years

Eight years in the record have ten or more falls. The table names the heaviest fall of each year and counts the countries involved, which shows that a busy year is spread across the map rather than concentrated in one place. In 1933, for example, the 17 falls came down in 10 countries; the most in one country was 5, in the United States.

None of these years stands out in the decade chart as much as the 1930s as a whole, which collected strong years one after another.

A slow start

Only 35 falls in the record are dated before 1800. The oldest is Nogata in Japan, dated 860. After a gap of more than five centuries come Elbogen (1399) in the Czech Republic, then Rivolta de Bassi (1490) in Italy and Ensisheim (1491) in France. Before 1700 the record holds 13 falls in all. A fall from that period survives only if a specimen was kept and its arrival written down, which explains why the early centuries look so empty.

Where the falls were recorded changed

In the 1800s, 51 % of the recorded falls came down in Europe. From 1950 onward Europe’s share is 18 %. Africa shows the opposite movement: 13 falls in the whole nineteenth century against 78 since 1950. The table gives the counts by continent.

The table shows where falls were witnessed and collected. It cannot show where meteorites land, because a fall that nobody saw or picked up leaves no entry.

Years without a recorded fall

Since 1803 almost every year has at least one fall in the record. The exceptions are 1816, 1832, 1888 and 2005. A year without a fall in the table means no witnessed and recovered meteorite from that year reached the compiled list, not that nothing fell.

The short 2010s column

The record holds 12 falls dated 2010 or later: Huaxi (2010), Košice (2010), Lorton (2010), Mifflin (2010), Varre-Sai (2010), Boumdeid (2011) (2011), Sołtmany (2011), Thika (2011), Tissint (2011), Battle Mountain (2012), Sutter's Mill (2012) and Chelyabinsk (2013). Falls classified after the dataset was compiled are missing, so the decade looks thinner than it was.

Six countries, four periods

The countries with the most falls did not build their records at the same time. France collected most of its falls early: 31 of its 58 are dated before 1850 and only 5 since 1950. China is the mirror image, with 45 of its 54 falls dated 1950 or later. The United States and Russia peak in the first half of the twentieth century, and India in the second half of the nineteenth.

Even in the busiest countries, falls are spaced out. Since 1900 the longest run of years without a recorded fall is 6 years in the United States, 8 in India and 24 in France. In China the longest gap since 1900 is 12 years.

So a national count is a history of observation as much as a count of rocks. A country whose record grew late, like China, can look sparse on a nineteenth-century map and crowded on a modern one. The decade pages show these shifts one decade at a time, with a world map of every fall dated to that decade.

Using the rate

A rate of about 5.6 recorded falls a year spread across the whole planet makes any single place unlikely to see one. To see what the record holds near you, try the What fell near tool, or open a decade from the chart above. The method behind every count is on Reading the record.