Water samples were collected randomly from some parts of the Phulbari Upazilla under the district of Dinajpur. The study was conducted during the period February, March and April, 2006. From each sampling site, sample was collected in each month. The time schedule for sampling was chosen as the groundwater levels lower down due to heavy abstraction of groundwater and low/no precipitation during this period. Within the study area, 40 sites were selected for collecting representative water samples of which 35 shallow tubewells, 3 deep tubewells and 2 hand tubewells. Details of the sampling sites for different sources of waters have been shown in Figure 1 and Table 1. Water samples were collected following the techniques outlined by Hunt and Wilson (1986) and APHA (1998). The water samples were collected in 2 liter plastic bottles. These bottles were cleaned with dilute hydrochloric acid (1:1) and then washed with tap water followed by distilled water. Before sampling, containers were again rinsed 3 to 4 times with water to be sampled. The water carried to the laboratory of the department of Agricultural Chemistry and Biochemistry, Hajee Mohammad Danesh Science and Technology University, Dinajpur, Bangladesh for testing. The samples were analyzed as quickly as possible on arrival at the laboratory.
In order to assess the suitability for irrigation, drinking and industrial uses, the important chemical constituents of surface water like pH, (EC), (TDS), Ca2+, Mg2+, Na+ , K+ , Zn2+ , Cu2+ , Mn2+, Fe3+ , PO4 3- , As3+ , CO3 2- , HCO3 - , SO4 2- , NO3 - and Cl- were considered. All bottles and glassware used were cleaned with P-free detergent (Extrans), rinsed three times with ultra-pure water (Milli-Q), soaked in 10% (v/v) HCl for at least 24 h, and finally rinsed three times with ultra-pure water. The pH and electrical conductivity (EC) were determined electrometrically (APHA, 1998). TDS was estimated after Chopra and Kanwar (1980). Ca2+ and Mg2+ were analyzed by complexometric titration (Page et al., 1982) whereas K+ and Na+ were estimated by flame emission spectrophotometer (Ghosh et al., 1983). SO4 2- was determined turbidimetrically (Wolf, 1982) while CO3 2- and HCO3 -, were analyzed titrimetrically (Chopra and Kanwar, 1980; Ghosh et al., 1983). Chloride was estimated by argentometric titration (APHA, 1998; Ghosh et al., 1983) and PO4 3- and NO3 - were determined colorimetrically (APHA, 1998). Arsenic was determined by atomic absorption spectrophotometer equipped with hydride generator (APHA, 1998). Zn2+, Cu2+, Mn2+ and Fe3+ were analyzed by atomic absorption spectrophotometer (APHA, 1998) in the laboratory of Soil Chemistry Division, Bangladesh Agricultural Research Institute (BARI), Gazipur, Bangladesh. A Perkin–Elmer Analyst 100 atomic absorption spectrophotometer equipped with a FIAS-100-flow injection hydride generation system was used for all arsenic determination. All samples were pre-reduced with concentrated HCl (1 ml), 5% KI and ascorbic acid mixture prior to hydride generation.