Signal extraction problems in seismology

The earth's surface is under continuous influence of a variety of natural f orces such as the effect of past earthquakes, wave, wind, tide, air pressur e, precipitation and a variety of human induced sources. Since it is almost impossible to describe the response to these noise inputs precisely, for a utomatic processing of seismic data, proper statistical modeling is necessa ry. In this paper, we describe four specific examples of time series modeli ng for signal extraction problems related to seismology, Namely, we conside r 1) the estimation of the arrival time of a seismic signal, 2) the extract ion of small seismic signal from noisy data, 3) the detection of the coseis mic effect in groundwater level data contaminated by various effects from a ir pressure etc., and 4) the estimation of changing spectral characteristic of seismic record.