Reading path: Ocean depth and big questions →
Questions to keep separate
| Question or feature | Explanation | What it does not establish |
|---|---|---|
| Tsunami generation | Large water displacement can follow undersea earthquakes, landslides or volcanic activity. [1] | A reported animal behavior does not identify its cause. |
| Offshore and coastal waves | Tsunamis may be inconspicuous in deep water and build in height in shallower water; speed depends on depth. [1] | Deep-water behavior does not establish a universal animal response. |
| Animal prediction | USGS reports no established consistent, reliable animal precursor behavior. [2] | Detecting a P wave before an S wave is reaction to an earthquake already underway, not days-ahead prediction. |
| Warning evidence | NOAA uses DART and coastal water-level observations in warning operations. [1] | A viral clip without an original dated observation record cannot authenticate a particular animal sighting. |
What makes a tsunami, and why it changes near shore
A tsunami begins when a large displacement moves water, often during an undersea earthquake. Landslides and volcanic activity can also cause such displacement. In deep water, a tsunami may be inconspicuous at the surface; as it reaches shallower coastal water, it can build in height. Wave speed depends on water depth. [1]
That sequence matters when assessing a claim about a deep-sea animal “sensing a tsunami.” The tsunami is a moving water disturbance, while an earthquake is a separate ground-motion process that may generate it. Evidence that an animal behaves unusually does not, by itself, identify which process it sensed or establish that it predicted either one. [1] [2]
A seconds-long warning is not a forecast
USGS says consistent, reliable animal behavior that predicts earthquakes has not been established, and a reliable mechanism remains elusive. An animal could detect a P wave, which arrives before the stronger S wave people may feel. That would be a response to an earthquake already underway, potentially seconds earlier, not a forecast days or weeks ahead. [2]
A photograph of an unusual animal before an earthquake is not enough to show predictive value. To evaluate the claim, you need a baseline: how often the animal appears when no earthquake follows, and how often earthquakes occur without the animal appearing. You also need the location and time window specified clearly enough to test, rather than chosen after the event. [2]
A claim-checking exercise
Suppose a post says: “An unusual fish appeared on Monday, then an earthquake happened Friday.” First verify the original observation’s date and location; a repost date is not necessarily the sighting date. Then set in advance the geographic radius, earthquake magnitude threshold and time window that would count as a match. Compare sightings without qualifying earthquakes and earthquakes without sightings. A memorable pair of events is not that test. [2]
An unattributed viral clip cannot establish a documented animal warning event. Without the original image or record, its observation date, location and context remain unknown. Verification of a particular sighting would still leave a second question: did that behavior predict an earthquake with useful accuracy? The general prediction claim lacks consistent, reliable support; circulating an image without its observation record supplies neither a verified event nor a successful prediction. [2]
Use monitored warnings and natural signs
NOAA warning operations use DART observations and coastal water-level measurements alongside other information. Those measurements help assess tsunami conditions; animal sightings are not a substitute for official warnings. Deep-water waves can be hard to notice, so absence of an obvious offshore surface signal does not establish that no tsunami is present. [1]
For coastal safety, follow official tsunami warnings and natural warning signs rather than waiting for an animal report. The cited tsunami guidance describes wave behavior and monitoring, but it does not prove that every deep-sea animal is unaffected by earthquakes or tsunamis. It also supplies no authenticated case here from which to infer a particular animal response. [1]
Sources & further reading
Have a correction or a source to suggest? Send a note →