The swine flu has been spreading around the world, and there has been much discussion about whether the threat was exaggerated.

The recurring problem is making predictions, whether it is of earthquake, flood, volcanic eruption, mudslide, tornado, hurricane, tsunami or pandemic.

We have many more data points than ever before to help us recognize and understand all kinds of natural events, whether physical or biological. Prediction comes down to making educated guesses about whether something imminent will be realized.

The denser data allows us to see things that we could not see otherwise.

Hurricanes are probably the best-known example.

It used to be that hurricanes would sweep in and catch everyone off guard. Now we can track them by satellite and see them developing days ahead of time far out in the ocean and watch their tracks.

Even so, we cannot predict when, where or even whether they will strike land or what their intensity will be. Pandemics are similar in this respect. The progress of the disease can be tracked, and there is no way to predict with certainty its path or its severity.

It might appear to be a good thing to predict disasters with 100 percent accuracy. Suppose we could say with certainty that 100,000 people on Oahu would die from the flu or that we knew that a magnitude-9 earthquake would hit San Francisco on a specific day.

There could be a mass exodus even if no one could say exactly who would become infected or die, or the exact details of the damage to come.

What effect would this have on business and commerce? What would happen to the value of property? Who would buy property or move into the area knowing that a disaster would happen?

How ironic it would be if people did evacuate or migrate to avoid an earthquake only to be stricken by a flood or some other of nature’s many potentially destructive events!

We have to realize that there is no place that is completely safe. You can move from California to escape earthquakes, but where do you go?

Making predictions is a lose-lose proposition. Perfectly reliable predictions might not be desirable for the above reasons. But the imperfect predictions we can make now carry the risk of the cry-wolf effect. Every warning for an event that does not materialize causes a ripple effect such that subsequent warnings are greeted with increasing skepticism.

Experts are expected to know things that are unknowable. The H1N1 flu virus has not been as virulent or as deadly as feared (not yet, anyway), and there is a good chance that people are less likely to respond with the same level of interest and concern to a future pandemic warning.

We must keep in perspective that nature has no wrath. Natural events are disasters only to the extent of their impact on us. Nature does what it does, and we are its slaves and not its master.

Richard Brill is a professor of science at Honolulu Community College. E-mail questions and comments to .(JavaScript must be enabled to view this email address).