Association of wastewater determinants with fish hematological and plasma biochemical responses: Multivariate analysis approach

The aim of the study was to assess value and applicability multivariate tools for hematological plasma biochemical responses fish living in treated wastewater. Physicochemical water properties heavy metals concentration spring fall were used as determinants multiple stressors. Three methods data analysis (Agglomerative Hierarchical Clustering, Factor Analysis, Principal Component Analysis) one method modeling (Partial Least Square Regression) applied. These enabled identification clustering based on observed parameters, significant variables set, correlation with samples collected different places at seasons. Prediction total leukocytes, lymphocytes, granulocytes, hematocrit, glucose, alanine aminotransferase, triglycerides cholesterol from blood (R2 > 0.9) better than variables, regardless sampling site 0.98). For hematocrit glucose (determination coefficient over 0.99), prediction successful season site. effectiveness models also evaluated using ratio standard error performance deviation (RPD), range (RER). High these found purposes (RPD 8 RER 15), including parameters regard quality.