Stream sediments have intrinsically the capability to describe the geochemistry of the sample's upstream basin, averaging each elemental contribution of the different outcropping lithotypes. Finer fraction (<150 um) is used as indicator of landscape status respect to pollutant, because of high reactivity with surficial water, groundwater and biota in a long-term environmental monitoring. Furthermore a sampling point on a stream network shows a nearly stable composition under the assumption of watershed time/space stationarity. Aside it is useful to underline that hot topic in environmental geochemistry is the definition of wide-area elemental background concentration value in geological matrixes. Moreover it is often mandatory for regulatory activities to define local elemental reference values over a discrete area to be compared to point concentration. This need is strictly related to a change of support passing from point measurement to evaluation over a representative area unit. Ultimately those reference values in whatever manner are defined over heterogenous geological settings can reach limit of applicability or worst overused if not defined in a probabilistic frame. The heterogeneity of the Ligurian Alps and Appennines invokes a change of strategy for what concerns the description of the spatial distributed geochemistry capturing the local variability of the regolith composition. With the aim to develop a flexible and highly adaptive tool based on geostatistical approach, the Geochemical Numerical Model of Liguria (GNM-L) of stream sediments is created. That digital product is the result of a joint project of the Regione Liguria and DISTAV to generate a tool able to fix the geochemistry of the environment based on stream sediments and to become a reference layer for future environmental intervention. To manage lithotypes heterogeneities computations are led on a-priori defined irregularly shaped discrete mosaic of the Liguria with the development of a geostatistical knowledge of regional variability. The database used to compute the spatial patterns of 32 elements by means of 8 statistical moments and probabilistic cut points of distribution functions is composed by 1830 sampling stations along the stream network. Each of them are sampled with a quasi-random sampling schema based on a subset of the grid GRID generated as a reference layer during the creation of the Archivio Geochimico Nazionale in the frame of the ANPA-CNR-UNIGE joint project. Through the application of a stochastic method based on gaussian sequential simulations over a multi resolution grid computation each chemical element is deeply investigated as a random function on a final support of 200m x 200m. The enhancement of the information density on each single element of the GNM allows the end-user to understand more profoundly the environment with the help of a sufficiently flexible tool that more adherently can represent the point uncertain reality.
The Geochemical Numerical Model of Liguria: a stochastic tool to evaluate the uncertainty of elemental concentration estimates.
Vetuschi Zuccolini;
2021-01-01
Abstract
Stream sediments have intrinsically the capability to describe the geochemistry of the sample's upstream basin, averaging each elemental contribution of the different outcropping lithotypes. Finer fraction (<150 um) is used as indicator of landscape status respect to pollutant, because of high reactivity with surficial water, groundwater and biota in a long-term environmental monitoring. Furthermore a sampling point on a stream network shows a nearly stable composition under the assumption of watershed time/space stationarity. Aside it is useful to underline that hot topic in environmental geochemistry is the definition of wide-area elemental background concentration value in geological matrixes. Moreover it is often mandatory for regulatory activities to define local elemental reference values over a discrete area to be compared to point concentration. This need is strictly related to a change of support passing from point measurement to evaluation over a representative area unit. Ultimately those reference values in whatever manner are defined over heterogenous geological settings can reach limit of applicability or worst overused if not defined in a probabilistic frame. The heterogeneity of the Ligurian Alps and Appennines invokes a change of strategy for what concerns the description of the spatial distributed geochemistry capturing the local variability of the regolith composition. With the aim to develop a flexible and highly adaptive tool based on geostatistical approach, the Geochemical Numerical Model of Liguria (GNM-L) of stream sediments is created. That digital product is the result of a joint project of the Regione Liguria and DISTAV to generate a tool able to fix the geochemistry of the environment based on stream sediments and to become a reference layer for future environmental intervention. To manage lithotypes heterogeneities computations are led on a-priori defined irregularly shaped discrete mosaic of the Liguria with the development of a geostatistical knowledge of regional variability. The database used to compute the spatial patterns of 32 elements by means of 8 statistical moments and probabilistic cut points of distribution functions is composed by 1830 sampling stations along the stream network. Each of them are sampled with a quasi-random sampling schema based on a subset of the grid GRID generated as a reference layer during the creation of the Archivio Geochimico Nazionale in the frame of the ANPA-CNR-UNIGE joint project. Through the application of a stochastic method based on gaussian sequential simulations over a multi resolution grid computation each chemical element is deeply investigated as a random function on a final support of 200m x 200m. The enhancement of the information density on each single element of the GNM allows the end-user to understand more profoundly the environment with the help of a sufficiently flexible tool that more adherently can represent the point uncertain reality.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.