The Number I Had to Throw Away

If you live here, you have felt the ground move. Probably more than once.

There is that jolt at 3 in the morning. You lie still and ask yourself: was that a big one? Then you check your phone to see if anyone else felt it.

I went to look at the records. What I found first was not about earthquakes at all. It was about counting.

The Number I Had To Throw Away

I pulled every Philippine earthquake from the USGS catalogue since 2000. That is 16,489 of them.

That is a nice big number. I nearly put it at the top of the page. I am glad I did not.

Here is why. Seismology has one rule it is very sure of. Small quakes are common. Big ones are rare. For every magnitude 5, you should get about ten magnitude 4s. And about a hundred magnitude 3s. Small ones should always outnumber big ones, by a lot.

So I counted mine by size, smallest first. I got 6. Then 115. Then 2,880. Then 9,818.

Look at that again. It goes up. That is backwards. It is like counting ants and mice in your kitchen and finding one ant and nine hundred mice.

The ground is not skipping the small ones. The machines are missing them.

Most of the sensors in this catalogue sit in other countries. A small quake in Surigao is a whisper by the time it gets to them. They only hear the loud ones. So the small numbers do not measure the earth. They measure the microphones.

That changes what I am allowed to say. Below about magnitude 4.5, I have to stay quiet. Above it, the counting works. So every number on the big page is magnitude 4.5 and up. There are 9,203 of those. A smaller number, and the only honest one.

The December Puzzle

Then I asked a question people always ask. Is there a season for earthquakes?

I split the quakes by month. December stood out at once. It held 13 out of every 100 quakes. If months were even, each would hold about 8. December looked special.

I almost wrote that down. Something felt off, so I went and looked at the actual days.

They were all bunched up. Nearly all of them came right after the 2nd of December, 2023.

That day there was a magnitude 7.6 quake off Mindanao. When a big one hits, the rock around it keeps settling for weeks. Those little settling shakes are called aftershocks. They are one event finishing, not many new ones starting.

That single quake was followed by 510 more in thirty days. A normal thirty days that year would have had about 66.

So I counted again with those thirty days taken out. December dropped to 8.9 out of 100. Right where an even month should be.

There is no earthquake season. There was one big earthquake, and I nearly turned it into a season.

What The Data Does Say

Some things held up fine.

The south shakes the most. Below 10 degrees north, which is Mindanao and the seas around it, sits 58 out of every 100 quakes. The north, where most people live, gets 26.

Most quakes are shallow. About 74 out of 100 start less than 70 km down. Shallow is the kind you feel. A shallow medium quake will rattle you harder than a deep big one.

Although even that one comes with a catch. When the machines cannot work out how deep a quake was, they write down 10 km and move on. Almost a quarter of my quakes have exactly 10 km. So "74 out of 100 are shallow" is the most it could be, not a firm count.

I also tested an old idea: that bigger quakes start deeper. They do not. I measured it and got 0.064, where 0 means no link at all. The averages look like a link, but only because a handful of huge deep ones drag the average up. The middle value barely moves.

What I Took Off The Page

The old version of this page had charts of volcanoes, tsunami warnings, and damage in pesos.

I could not find those numbers in any file I trust. So I took them off. Not fixed them. Took them off.

A wrong number that looks tidy is worse than a blank space. The blank space at least tells you to go and look somewhere else.

The Part I Am Still Unsure About

Bigger quakes, magnitude 6 and up, are loud enough that the machines have always caught them. So I can compare those across time fairly.

In the 2000s we had about 4.7 a year. In the 2020s, about 8.6.

That looks like a real rise. It might be. But seven years is a short stretch, and earthquakes come in clumps by nature. A run of busy years proves nothing on its own.

So I am putting the number up and not calling it a trend. Ask me again in twenty years.