Iranian Journal of Applied Ecology (Mar 2014)
Application of Multivariate Statistical Techniques to Assess Seasonal Variation in Water Quality Parameters in Gorganrood Watershed, Iran
Abstract
Nonpoint source (NPS) pollution is a major surface water contaminant commonly caused by agricultural runoff. The purpose of this study was to assess seasonal variation in water quality parameters in Gorganrood watershed (Golestan Province, Iran). It also tried to clarify the effects of agricultural practices and NPS pollution on them. Water quality parameters including potassium, sodium, pH, water flow rate, total dissolved solids (TDS), electrical conductivity (EC), hardness, sulfate, bicarbonate, chlorine, magnesium, and calcium ions during 1966-2010 were evaluated using multivariate statistical techniques. Multivariate analysis of variance (MANOVA) was implemented to determine the significance of differences between mean seasonal values. Discriminant analysis (DA) was also carried out to identify correlations between seasons and the water quality parameters. Parameters of water quality index were measured through principal component analysis (PCA) and factor analysis (FA). Based on the results of statistical tests, climate (freezing, weathering and rainfall) and human activities such as agriculture had crucial effects on water quality. The most important parameters in differentiation between seasons in descending order were potassium, pH, carbonic acid, calcium, and magnesium. According to load factor analysis, chlorine, calcium, and potassium were the most important parameters in spring and summer, indicating the application of fertilizers (especially potassium chloride fertilizer) and existence of NPS pollution during these seasons. In the next stage, the months during which crops had excessive water requirements were detected using CROPWAT software. Almost all water requirements of the area’s major crops, i.e. cotton, rice, soya, wheat, and oat, happen in the late spring until mid/late summer. According to our findings, agricultural practices had a great impact on water pollution. Results of analysis with CROPWAT software also confirmed this conclusion.
