Stephen Elston | Harvard University Extension School (original) (raw)
Related Authors
Graduate Center of the City University of New York
Uploads
Papers by Stephen Elston
Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r... more Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r&ust estimamrs when stacking seismic signals will minimize the effect of largeamplitude noise 0n the result. The paf0rmance of the mean, median, Huber and rcdcscwding M+stimetol are investigated, ualng simulati0ns, 10 determine which estimators performed best for a variety of noise dktributi0ns. St&s made withthtHubats~~~thecharactaofthoamadewithalinear estimamr, but have significantly ess random noise. Stack made with median and redtsccnding Mestimatcrs have poor rekcmr character. Velrzity panels made with either the median 0r mdescending Meetimat0r am affected the lean by large-amplitude noise.
Seg Technical Program Expanded Abstracts, 1999
Seg Technical Program Expanded Abstracts, 1999
Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r... more Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r&ust estimamrs when stacking seismic signals will minimize the effect of largeamplitude noise 0n the result. The paf0rmance of the mean, median, Huber and rcdcscwding M+stimetol are investigated, ualng simulati0ns, 10 determine which estimators performed best for a variety of noise dktributi0ns. St&s made withthtHubats~~~thecharactaofthoamadewithalinear estimamr, but have significantly ess random noise. Stack made with median and redtsccnding Mestimatcrs have poor rekcmr character. Velrzity panels made with either the median 0r mdescending Meetimat0r am affected the lean by large-amplitude noise.
Geochimica et Cosmochimica Acta, 1992
surface reaction model for the kinetics of quartz dissolution is derived from an analysis of low-... more surface reaction model for the kinetics of quartz dissolution is derived from an analysis of low-temperature dissolution rate data published by seven experimenters. The model correlates seventynine quartz dissolution rate measurements with calculated surface species distributions to predict reaction kinetics. Using the triple layer electrostatic surface model, we compute the distributions of predominant acid-base surface species for the solution composition associated with each reported rate measurement. Reactivities of the modeled species are then obtained by fitting the calculated site distributions and the experimental rate data to a simple rate equation. This results in an expression valid for 25"C, pH 2-l 3 and O-O.5 molal sodium chloride given by rate = 10-'3'o(8*ioH)* + 10-'0'8(B~i&um)' + 10-9~2(B~si~~m)2r where rate has units of mol me2 s-' , 0-iOH is the fraction of =SiOH surface complexes and * indicates that the reaction order dependence on this species is ill-determined. The Oeiau,,, term gives the sum of =SiO-Na+ and =SiO-fractions. This correlation removes much of the apparent scatter that is observed between experimenters and demonstrates that most measurements follow a consistent trend when a sodium complex, =SiO-Na+, is included in the model. The model predicts that reaction rates are increased by a factor of 12 at pH 8 with the addition of 0.2 molal sodium to the reacting solutions. While this approach is an indirect measure of the reactivity of bonds at mineral surfaces, when compared to previous models, it offers a better understanding of rate-controlling processes because it specifically includes the calculated distributions of surface reactants. This analysis, in conjunction with evidence from the literature for cation-specific influences on silicon dioxide reactivity, supports a dissolution model incorporating cation interactions.
Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r... more Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r&ust estimamrs when stacking seismic signals will minimize the effect of largeamplitude noise 0n the result. The paf0rmance of the mean, median, Huber and rcdcscwding M+stimetol are investigated, ualng simulati0ns, 10 determine which estimators performed best for a variety of noise dktributi0ns. St&s made withthtHubats~~~thecharactaofthoamadewithalinear estimamr, but have significantly ess random noise. Stack made with median and redtsccnding Mestimatcrs have poor rekcmr character. Velrzity panels made with either the median 0r mdescending Meetimat0r am affected the lean by large-amplitude noise.
Seg Technical Program Expanded Abstracts, 1999
Seg Technical Program Expanded Abstracts, 1999
Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r... more Seismic data is often contaminated by largeamplitude. nonGaussian distributed noise. Ihe USC of r&ust estimamrs when stacking seismic signals will minimize the effect of largeamplitude noise 0n the result. The paf0rmance of the mean, median, Huber and rcdcscwding M+stimetol are investigated, ualng simulati0ns, 10 determine which estimators performed best for a variety of noise dktributi0ns. St&s made withthtHubats~~~thecharactaofthoamadewithalinear estimamr, but have significantly ess random noise. Stack made with median and redtsccnding Mestimatcrs have poor rekcmr character. Velrzity panels made with either the median 0r mdescending Meetimat0r am affected the lean by large-amplitude noise.
Geochimica et Cosmochimica Acta, 1992
surface reaction model for the kinetics of quartz dissolution is derived from an analysis of low-... more surface reaction model for the kinetics of quartz dissolution is derived from an analysis of low-temperature dissolution rate data published by seven experimenters. The model correlates seventynine quartz dissolution rate measurements with calculated surface species distributions to predict reaction kinetics. Using the triple layer electrostatic surface model, we compute the distributions of predominant acid-base surface species for the solution composition associated with each reported rate measurement. Reactivities of the modeled species are then obtained by fitting the calculated site distributions and the experimental rate data to a simple rate equation. This results in an expression valid for 25"C, pH 2-l 3 and O-O.5 molal sodium chloride given by rate = 10-'3'o(8*ioH)* + 10-'0'8(B~i&um)' + 10-9~2(B~si~~m)2r where rate has units of mol me2 s-' , 0-iOH is the fraction of =SiOH surface complexes and * indicates that the reaction order dependence on this species is ill-determined. The Oeiau,,, term gives the sum of =SiO-Na+ and =SiO-fractions. This correlation removes much of the apparent scatter that is observed between experimenters and demonstrates that most measurements follow a consistent trend when a sodium complex, =SiO-Na+, is included in the model. The model predicts that reaction rates are increased by a factor of 12 at pH 8 with the addition of 0.2 molal sodium to the reacting solutions. While this approach is an indirect measure of the reactivity of bonds at mineral surfaces, when compared to previous models, it offers a better understanding of rate-controlling processes because it specifically includes the calculated distributions of surface reactants. This analysis, in conjunction with evidence from the literature for cation-specific influences on silicon dioxide reactivity, supports a dissolution model incorporating cation interactions.