Page 50 - {PDF Title}
P. 50

Adhikari and Neupane







































                                         Figure 1. Map of the Bagmati River and sampling sites

                were  considered  presumptive  positive  for  coliform   3. Results and discussion
                bacteria. The MacConkey agar medium was used for
                the  confirmed  test.  Water  samples  from  the  positive   3.1. Physiochemical and bacteriological
                confirmed  test  were  incubated  on  nutrient  agar  and   characterization of Bagmati River water
                lactose broth at 37°C for 24 h. The quantitative analysis   The physiochemical and bacteriological measurements
                                                                    are essential to determine the river water quality.
                was acquired using the membrane filter method, in which   Physicochemical parameters, such as color, turbidity,
                100 mL of the samples were placed on the MacConkey   conductivity, total hardness, pH, iron level, and chromium
                agar media by passing through 0.45 µm filter paper. The   level, and biological parameters, such as MPN and SPC
                samples were then incubated for 24 h, and bacteria were   were measured accordingly (Table  1).  The color of
                counted after incubation.  The isolation and counting   Bagmati River water at the Pashupati site (B-1) was brown,
                                      24
                of the viable microorganisms in the water sample were   while that of the Balkhu site (B-2) in Kathmandu Valley
                determined using the SPC technique. Briefly, 9 mL of   was black (Figure 2). Turbidity indicates the presence of
                sterile distilled water was added to separate test tubes   suspended solids, including both organic and inorganic
                and 1 mL of sample water was added to the first test tube;   substances, in  the  water.  The  turbidity was about 35.4
                from the first test tube, 1 mL of water was transferred to   nephelometric turbidity units (NTU) at B-1 and 26.4 at
                the second test tube. The serial dilution was up to 10    B-2. The lower turbidity in the B-2 site validates the color
                                                                -9
                mL. Thereafter, 0.1 mL of the 10  dilution series was   of the water sample.  The observed turbidity exceeded
                                              -9
                transferred to the center of the surface of the nutrient   the World Health Organization’s (WHO) recommended
                agar plates; hence, the colony-forming unit (CFU) limit   value.  The turbidity and color of the water suggested that
                                                                         26
                was 250 × 10  CFU/100 mL. The sample was evenly     before the confluence of the tributaries, the river water
                            -12
                spread using a sterile  L-shaped glass spreader and   consisted of soil particles, which  increased turbidity.
                incubated at 37ºC for 24 h. The CFU was calculated and   However, after mixing with the tributaries, the soil
                multiplied by the calculated dilution factor to determine   particles disappeared, resulting in a transparent but black
                the number of colonies in CFU/100 mL of the original   color. The black color is attributed to the mixing of sewer
                sample. 25                                          from either local sources or tributaries. 4,27



                Volume 22 Issue 1 (2025)                        44                                 doi: 10.36922/ajwep.8434
   45   46   47   48   49   50   51   52   53   54   55