Application Of Allied Elastic Impedance Inversion In The Study Of Ocean Characteristics

Seismic oceanography has become a new technique using multichannel seismic(MCS) method to research physical oceanography. With the advantages in lateralresolution, this technique has been applied to the study of marine fronts, internalwaves, mesoscale eddies and other marine phenomenon. Early researches weremainly focused on describing water mass boundaries directly using seismic reflectionmethod, which lack the quantitative analysis. In order to describe oceancharacteristics more finely, seismic inversion technique has been applied to this field.Full waveform inversion and acoustic impedance inversion have made someachievements in the study of ocean characteristics. But the accuracy of acousticimpedance inversion accuracy is not high, and full waveform inversion is currentlylimited to one dimension, also they are both not very good at extracting ocean density.Therefore, it seemed quite significant to propose a comprehensive and preciseinversion method of ocean parameters.The quantitative relationship between multichannel seismic records and physicaloceanographic parameters are the basis for precise inversion. Based on the equation ofstate for seawater and Wilson formula, this paper analyses the relationships amongocean parameters, and attains contributions of these parameters respectively to thereflectivity by means of Zoeppritz equation. These works build the complicatedrelationship between seismic reflection and physical properties of seawater, whichprovides a theoretical guidance for marine inversion. Acoustic impedance is thebridge of marine parameters and seismic reflection. So this paper derives the equationof elastic impedance in seawater and names it Allied Elastic Impedance (AEI). But itsdimensionality varies with incidence angle, so AEI is normalized to acquire a betterinversion result.As for the difficulty in extracting ocean density, this paper proposes a method toextract two allied elastic impedance volumes using two angle gathers, and then calculates velocity and density, which makes full use of AVO. Considering that oceanparameters change slightly, the inverted profile is smoothed to get precise velocityand density, which makes this method more adaptable and provides available data forcalculating temperature and salinity. According to the conclusion that temperature andsalinity is sensitive to velocity and density respectively, this paper proposes a methodto fitting temperature and salinity from velocity and density by means of linear andneural network respectively. This method breaks through the limitation thattemperature and salinity can only be calculated by velocity, and also increases theaccuracy of salinity inversion.The feasibility and good noise immunity of allied elastic impedance inversionare proved by theoretical model. In real data application, it can also get preciseprofiles of velocity, density, temperature and salinity. All the results show that thismethod has superiority in the study of ocean characteristics and has potential practicalvalue.