Seasonal evaluation of hydro-geochemical parameters using correlation and regression analysis

Correlations and multiple linear regressions were used to develop models relating well water chemical quality parameters to a set of independent chemical variables in post- and pre-monsoon seasons in the upper Gunja-naeru River basin, Cuddapah district, Andhra Pradesh, South India. The correlation between the specific electrical conductance (SEC) and other parameters except potassium (K) is significantly positive, whereas Ca2+ + Mg2+/Na+ + K+ is significantly negative for both post- and pre-monsoon seasons. In predicting SEC for both post- and pre-monsoon, the independent variables, viz. SHCO3-, SO42- and Cl- in the model had a significant effect (from It' test for partial regression coefficient at the 5% level of probability). The multiple R-2 values 0.982 and 0.997 indicate that 98.2 and 99.7% of variability in the observed SEC could be ascribed to the combined effect of Na+, Cl-, Ca2+ + Mg2+, HCO3-, and SO42- for post- and pre-monsoon seasons, respectively. Out of the 98.2% variability in SEC due to the com bined effect of Na+, Ca2+, HCO3-, SO42- and Cl-, 44.8% is due to HCO3-, 44.4% is due to Cl-, 9% is due to SO42- and 2% each is due to Na+ and Ca2+ + Mg2+ for post-monsoon season. Whereas in the pre-monsoon season, 99.7% variability in SEC is due to the combined of Na+, Ca2+, Mg2+, HCO3-, SO42- and Cl-; 38.8% is due to Cl-, 36.9% is due to HCO3-, 12% is due to SO42- and 2% each is due to Na+ and Ca2+ + Mg2+ for This shows that HCO3-, Cl-, SO42- and Cl-, HCO3, Na+, SO42- are the most significant independent variables in predicting SEC for post- and pre-monsoon seasons respectively.