Earthquake prediction by animals: Evolution and sensory perception

Authors
Citation
Jl. Kirschvink, Earthquake prediction by animals: Evolution and sensory perception, B SEIS S AM, 90(2), 2000, pp. 312-323
Citations number
90
Categorie Soggetti
Earth Sciences
Journal title
BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA
ISSN journal
00371106 → ACNP
Volume
90
Issue
2
Year of publication
2000
Pages
312 - 323
Database
ISI
SICI code
0037-1106(200004)90:2<312:EPBAEA>2.0.ZU;2-S
Abstract
Animals living within seismically active regions are subjected episodically to intense ground shaking that can kill individuals through burrow collaps e, egg destruction, and tsunami action. Although anecdotal and retrospectiv e reports of animal behavior suggest that although many organisms may be ab le to detect an impending seismic event, no plausible scenario has been pre sented yet through which accounts for the evolution of such behaviors. The evolutionary mechanism of exaptation can do this in a two-step process. The first step is to evolve a vibration-triggered early warning response which would act in the shore time interval between the arrival of P and S waves. Anecdotal evidence suggests this response already exists. Then if precurso ry stimuli also exist, similar evolutionary processes can Link an animal's perception of these stimuli to its P-wave triggered response, yielding an e arthquake predictive behavior. A population-genetic model indicates that su ch a seismic-escape response system can be maintained against random mutati ons as a result, of episodic selection that operates with time scales compa rable to that of strong seismic events. Hence, additional understanding of possible earthquake precursors that are presently outside the realm of seis mology might be gleaned from the study of animal behavior, sensory physiolo gy, and genetics. A brief review of possible seismic precursors suggests th at tilt, hygroreception (humidity), electric, and magnetic sensory systems in animals could be linked into a seismic escape behavioral system. Several testable predictions of this analysis are discussed, and it is recommended that additional magnetic, electrical, tilt, and hygro-sensors be incorpora ted into dense monitoring networks in seismically active regions.