Assessment of the uncertainty in spatial-correlation models for earthquake ground motion due to station layout and derivation method

Abstract The evaluation of the aggregate risks to spatially distributed infrastructures and portfolios buildings requires quantification estimated shaking over a region. To characterize spatial dependency ground motion intensity measures (e.g. peak acceleration), common geostatistical tool is semivariogram. Over past decades, different fitting approaches have been proposed in geostatistics literature fit semivariograms thus correlation structure. A theoretically optimal approach has not yet identified, as it depends on number observations configuration layout. In this article, we investigate estimation methods based likelihood function, which, contrast classical least-squares methods, straightforwardly define without needing further steps, such computing experimental Our outcomes suggest that maximum-likelihood may outperform methods. Indeed, former provides estimates, do depend bin size, unlike ordinary weighted regressions. addition, lead lower percentage errors dispersion, independently both stations their layout well underlying Finally, propose some guidelines account for uncertainty within seismic hazard risk assessments. consideration dispersion regional assessments could more realistic estimations corresponding losses.