Aim of the present work is represented by the estimation of dust dispersion in quarries that is considered one of the most pollutant factors in such activities. The subject is developed according with a geostatistical approach that is proposed as an alternative way of analysis to be added to classical models in order to predict dust concentration in quarrying activities. Each single value of a regionalized variable (dust concentration), and above all its spatial variability structure, is assumed to be described efficiently in a stochastic framework. Data values are the results of a miscellaneous of complex factors and can be reasonably viewed as possible outcomes of a random process; in general, such randomness can be poorly described in a deterministic framework. According with this foreword, a field study is proposed taking into account a validation process. In this phase sampled values represented the input data grid used to obtain a continuous map of PM 10 concentration in the quarry itself. Two different methods have been tested: as for the monovariate approach, after a preliminary analysis based on variogram computation, the Kriging method is developed whereas in the multivariate one, the estimation is realized according with the Co-Kriging method taking into account two variables (PM10 and PTS). The final part consists in comparing both Kriging and Cokriging results in order to test their quality and determining the minimum number of sampled values needed to estimate PM 10 concentration without losing significant spatial information.
ALFARO DEGAN, G., Lippiello, D., Pinzari, M. (2006). Geostatistics and airborne dust: an Italian limestone quarry. In Mine planning and equipment selection 2006 (pp.923-928). TORINO : M. Cardu, R. Ciccu, E. Lovera, E. Michelotti (eds).
Geostatistics and airborne dust: an Italian limestone quarry
PINZARI, Mario
2006-01-01
Abstract
Aim of the present work is represented by the estimation of dust dispersion in quarries that is considered one of the most pollutant factors in such activities. The subject is developed according with a geostatistical approach that is proposed as an alternative way of analysis to be added to classical models in order to predict dust concentration in quarrying activities. Each single value of a regionalized variable (dust concentration), and above all its spatial variability structure, is assumed to be described efficiently in a stochastic framework. Data values are the results of a miscellaneous of complex factors and can be reasonably viewed as possible outcomes of a random process; in general, such randomness can be poorly described in a deterministic framework. According with this foreword, a field study is proposed taking into account a validation process. In this phase sampled values represented the input data grid used to obtain a continuous map of PM 10 concentration in the quarry itself. Two different methods have been tested: as for the monovariate approach, after a preliminary analysis based on variogram computation, the Kriging method is developed whereas in the multivariate one, the estimation is realized according with the Co-Kriging method taking into account two variables (PM10 and PTS). The final part consists in comparing both Kriging and Cokriging results in order to test their quality and determining the minimum number of sampled values needed to estimate PM 10 concentration without losing significant spatial information.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
