A research project was carried out by the C.N.R. to develop an integrated geological-geotechnical model of the subsoil of Rome. Data of more than 6000 boreholes were archived in a GIS and used to develop the geological model; the results presented in this work mainly focused on the upper Pleistocene-Holocene alluvial deposits. Information of more than 2000 boreholes penetrating the alluvial deposits was encoded and elaborated using geostatistics to model the sedimentary bodies. Spatial variability of the physical and mechanical properties was also investigated to develop the geotechnical model. Multiple linear regression, kriging, and cokriging were applied to estimate the drained friction angle φ’; cross-validation demonstrates the cokriging with the PCA factors as auxiliary variables being the most suitable method. In progress work on cokriging of φ’ using granulometries as auxiliary variables demonstrates this approach to be viable for future applications
Geotechnical modelling of the subsoil of Rome (Italy) by means of multivariate geostatistics / D., FOLLE D; Raspa, Giuseppe; M., Mancini; M., Moscatelli; A., Patera; F., Stigliano; R., Vallone; G. P., Cavinato; G., Cavarretta; Milli, Salvatore; F., Garbin; S., STORONI RIDOLFI. - ELETTRONICO. - (2006), pp. S13-18-S13-18. (Intervento presentato al convegno IAMG XIth International Congress on "Quantitative Geology from multiple sources" tenutosi a Liège (belgium) nel 3-8 september 2006).
Geotechnical modelling of the subsoil of Rome (Italy) by means of multivariate geostatistics
RASPA, Giuseppe;MILLI, Salvatore;
2006
Abstract
A research project was carried out by the C.N.R. to develop an integrated geological-geotechnical model of the subsoil of Rome. Data of more than 6000 boreholes were archived in a GIS and used to develop the geological model; the results presented in this work mainly focused on the upper Pleistocene-Holocene alluvial deposits. Information of more than 2000 boreholes penetrating the alluvial deposits was encoded and elaborated using geostatistics to model the sedimentary bodies. Spatial variability of the physical and mechanical properties was also investigated to develop the geotechnical model. Multiple linear regression, kriging, and cokriging were applied to estimate the drained friction angle φ’; cross-validation demonstrates the cokriging with the PCA factors as auxiliary variables being the most suitable method. In progress work on cokriging of φ’ using granulometries as auxiliary variables demonstrates this approach to be viable for future applicationsI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
