地下水
含水层
火山
地质学
热液循环
氡
地球化学
空间分布
环境科学
地球科学
水文学(农业)
地震学
物理
遥感
岩土工程
量子力学
作者
Pooria Ebrahimi,Annalise Guarino,Vincenzo Allocca,Antonello Cutolo,Domenico Cicchella,Stefano Albanese
标识
DOI:10.1016/j.apgeochem.2023.105607
摘要
Campi Flegrei is one of the most active volcanic areas in the world and assessing the potential proxies of volcanic-related phenomena is critical. Therefore, the spatial distribution of radon and carbon dioxide in groundwater and the statistical relationships between the dissolved gases and other variables deserve further attention. Compositional data analysis (CoDA) was proposed at the end of the last century and further developed in the last decades for reliable data mining, but its potential has not been fully explored for characterization of the groundwater aquifers affected by hydrothermal activity. Based on a prospecting campaign mainly aimed at the determination of both radon and carbon dioxide in Campi Flegrei groundwater, this article explores the spatial patterns of these gases in the local aquifer system and uses a CoDA approach to extract the relevant information and to determine the meaningful geochemical associations. The results show that the spatial distribution of both dissolved gases corresponds to the hydrothermal system. The logratio transformed CO2 (aq) distinguishes bicarbonate-rich groundwater better than the raw values. Principal component analysis reveals two associations: A1) Ca2+, Mg2+, K+, SO42−, HCO3− + CO2 and pH; and A2) Na+, Cl−, As, B, Li, Rn, TDS and T. It highlights that the groundwater composition is generally influenced by two main factors: (1) meteoric water, which is modified by CO2‒rich magmatic gases in some cases; and (2) hydrothermal fluid and/or seawater. The results are in agreement with the literature and application of CoDA is recommended in future investigations because the study area is highly populated and considering the compositional nature of geochemical data might help mitigate the volcanic hazard at Campi Flegrei.
科研通智能强力驱动
Strongly Powered by AbleSci AI