Application of Multivariate Statistical Modelling in Temporal Patterns of Water Chemistry in Haraz River (Mazandaran Province)

Principal component analysis (PCA) was used to extract the factors associated with the physico-chemical variables in the Haraz River during four seasons in 2004-05. Using the results which were analyzed using PCA, a data matrix was produced. From the annually correlation matrix, seven principal components (PC) were extracted which explain 81.49% of the total variance of the raw data. PC1 (21.31% of the variance) is associated with the nitrogen compounds in terms of nitrate, DIN and DON and also CFU. PC2 (14.49% of the variance) is characterized by TA, TH and EC (physical parameters). PC3 (11.16% of the variance) is mainly contributed by the phosphorous compounds (DIP and DOP) and TSS. PC4 which explains 10.44% of the variance is associated with temperature and BOD5, while, PC5, PC6 and PC7 explain 10.39%, 7.25% and 6.89% of total variance are contributed by NH4+, DOP and pH and DO respectively. This study highlights the advantage of combining simple but powerful statistics with water quality monitoring.

Application of Multivariate Statistical Modelling in Temporal Patterns of Water Chemistry in Haraz River (Mazandaran Province) Keywords:

Multivariate Statistical , Physico-chemical , Haraz River , Mazandaran , Iran