Reality Matters

Governors and other state and local leaders have been telling us that the decision of when and how to reopen society from the Covid-19 lockdown will be guided by science. But what does it mean to use science for decisions?

Some people think of “science” as a tool that can somehow measure how far we are from reopening like a measuring tape can tell us how much our waist has grown from quarantine baking. You should, instead, think of science as process that helps us determine the truth.

Scientists begin from the assumption that there is an objective reality, independent of our hopes and desires. We try to understand the nature of that reality through observations and experiments.  We get ideas and make models from those observations and from those models, make predictions of what will happen in the future. We test the predictions against reality to see if we are right.

And we know we are sometimes wrong because the easiest person to fool is oneself.  So we use peer review to find the mistakes. We take our research, our intellectual offspring, the prized result of our hard work, and put it in the hands of colleagues, or even a competitor, and ask them to be brutally honest, to tear it apart, find the flaws and tell us what we did wrong.  The process doesn’t always work – people make mistakes or actively commit fraud.  But the system is set up to make it difficult to keep fraud going. Because reality matters.

This novel corona virus doesn’t care what we believe about it. Our need to go back to work, and our desperation about the economy won’t change anything about how the virus infects new people. It also doesn’t put out a big sign saying “It’s safe now.” Rather, scientists are creating a model for how many people will become infected and how many will die as we do or don’t change our behavior. A good model is the best representation of that reality. We cannot wait until we have all the data to be sure our model is right, but we have to carefully monitor what happens and adjust the model as new data comes in.

Reality is real – and it doesn’t care what we believe. People’s lives depend on how well we understand, and act on, the truth. And you can’t hope, believe or wish something into becoming true. Reality matters.

The Ring of Fire and other earthquake myths

Let’s talk earthquake triggering. Every time a notable earthquake occurs, I get the same questions:

  • How does this affect the Ring of Fire?
  • How does this earthquake affect California?
  • Does this mean a big earthquake will happen soon here?

All of these reflect the human need to make patterns, especially when faced with danger. To understand if a pattern is real, we need to use statistics to tell us if a pattern is repeatable or just coincidence. Many scientists have conducted these statistical studies and we know which ones are real.

One earthquake does make other earthquakes more likely. The slip in the quake changes the state of stress around the fault on which the first quake occurs. The likelihood of triggering another earthquake dies off with distance from the fault and the time since the quake. Mostly they are smaller and we call them aftershocks. About 5% of the time the aftershock is bigger than the first earthquake and we change the names and call the first one a foreshock.

Where the triggered earthquakes occur is a bit more complicated. To understand the results, you need to remember that earthquakes don’t happen at epicenters – they happen over a fault surface and the bigger the surface, the bigger the earthquake. From the magnitude, you can guess the length of the fault that produced the earthquake, as shown in this table.

So each earthquake has a fault length, the length of the fault that moves in that quake.  Most aftershocks triggered by an earthquake will be very near its piece of fault. We use the word aftershock to described triggered earthquakes that fall within one fault length of the mainshock’s fault.  For instance, a M7 earthquake will have a fault length of about 50 km. So any earthquake triggered within 50 km of any point on the mainshock’s 50-km-long fault will be called an aftershock.

Within the first week or two after a quake, we also sometimes see triggered earthquakes farther away and we use the term triggered earthquake to describe them.  These might extend for 3 to 4 fault lengths.  So a M7 might trigger earthquakes as far away as 150 km and a M8 might trigger earthquakes out to 800 km.

Beyond 4 fault lengths, the statistics clearly show that the rate of earthquakes doesn’t change.  Mexican earthquakes have never caused a change in the rate of earthquakes in California. New Zealand earthquakes don’t trigger earthquakes in Japan. Or California. Or anywhere else.  This doesn’t mean we can’t have an earthquake in California, or New Zealand or Alaska. There are M3 earthquakes several times a week in California and a magnitude 2.5 somewhere in the world every minute. But statistics of the earthquake catalogs for the last hundred years clearly show that beyond a few fault lengths the rate of earthquakes is unaffected.

So back to those questions. I’ll answer them now and you can insert whichever earthquake has just happened:

How does this [insert quake here] affect the Ring of Fire?

It doesn’t. In fact, the Ring of Fire is a literary device, not a scientific concept. When we first started exploring the world and recording earthquakes, we saw that both volcanoes and earthquakes were more common around the Pacific Ocean, and the Ring of Fire was coined to describe that.  But the plate tectonics revolution in the 1960s explained why those volcanoes and earthquakes are there – they lie around the boundaries of the tectonic plates, and there are many plates around the Pacific Ocean. Now we know that earthquakes in southern California are occurring in the Pacific plate, while those in Mexico are in the Cocos or North American plate, the ones in Chile or in the Nazca plate and New Zealand is in the Australian plate. The plate motions do affect each other – on the time scale of tens of millions of years. One the time scale of one earthquake, statistics show us there is no relation.

How does this [insert quake here] affect California?

It doesn’t if it is more than a few hundred miles away.

Does [insert quake here] mean a big earthquake will happen soon here?

An earthquake somewhere else does not make a California quake more or less likely.


So when you want to make a pattern out of a group of earthquakes, remember that earthquakes happen all the time, and we need statistics to tell us if our pattern is just coincidence. Just because we want a pattern doesn’t make the pattern real.