Factor analytical approaches for evaluating groundwater trace element chemistry data

摘要: Abstract The multivariate statistical techniques principal component analysis (PCA), Q-mode factor (QFA), and correspondence (CA) were applied to a dataset containing trace element concentrations in groundwater samples collected from number of wells located downgradient the potential nuclear waste repository at Yucca Mountain, Nevada. PCA results reflect similarities elements water resulting different geochemical processes. QFA compositions, whereas CA reflects that are dominant waters relative all other included dataset. These differences mainly due ways which data preprocessed by each three methods. highly concentrated, thus possibly more mature (i.e. older), groundwaters separated dilute using 1 (PC 1). PC 2, as well dimension results, describe chemistry aquifer materials through they have flowed. Groundwaters thought be representative those flowing an composed dominantly volcanic rocks characterized elevated Li, Be, Ge, Rb, Cs, Ba, associated with dominated carbonate exhibit greater Ti, Ni, Sr, Rh, Bi. 3, lesser extent 2 show strong monotonic relationship percentage As(III) suggesting these reflect, qualitative sense, oxidizing/reducing conditions within groundwater. relatively reducing Mn, Co, oxidizing V, Cr, Ga, As, W, U.