Evaluating the potential of bank infiltration as a source of water supply in Jenderam Hilir, Selangor, Malaysia

This study was inspired by the Klang Valley water crisis, for which Bank Infiltration (BI) was considered as a potential solution. Higher incidences of pollution to the rivers in Malaysia has lead to the decrease in resources for drinking water. Many developing countries like Malaysia, are faced with a big challenge to provide safe drinking water to the ever-increasing population. Currently, the treatment of river water uses chemicals to reduce several contaminants due to pollution. The chemical contents could give long term effects to the health of the consumers. However, there is hope in a more effective, but low-cost technology, known as Bank Infiltration (BI) or Riverbank Filtration (RBF). BI is a natural process using natural soil (aquifer) to treat river water and at the same time utilizes groundwater. Moreover, due to its ability to remove even the most persistent contaminants and microbes, BI can enhance treatment steps in a water treatment facilities, especially in a conventional water treatment. BI is a method that has long been known but it is still new and has not been practiced in Malaysia. Nevertheless, BI is site specific and requires thorough site investigations to assess its feasibility based on the local site characteristics. The study area was located in the southwest state of Selangor within the Langat Basin which covered an area of 10 km2. Besides, this study was specifically conducted to explore the possibility of using the BI systems to source the polluted surface water with groundwater. Three major factors were considered for evaluation: (i) an investigation on the contribution of surface water through BI, (ii) the input of local groundwater, and (iii) the water quality characteristics of water supply. In the earlier part of the investigations, the geophysical method was employed to define the subsurface geology and hydrogeology of the area. Isotope techniques were performed to identify the source of groundwater recharge and interaction between the surface water and the groundwater. A total of 25 monitoring wells and 2 test wells (DW1 and DW2) were constructed to circumscribe the BI system. The physicochemical and microbiological parameters of the local surface water bodies and groundwater were analysed before and during the water abstraction. The abstraction of water revealed a 5–98% decrease in turbidity, as well as HCO3-, SO4-, NO3-, Al, As and Ca concentrations reduction compared to the Langat River water. However, the water samples from test wells during pumping showed high concentrations of Fe2+ and Mn2+. In addition, the amounts of E. coli, total coliform, and Giardia were significantly reduced (99.9%).The pumping test results indicated that the two test wells (DW1 and DW2) were able to sustain yields of 15.9 and 128 m3/hr, respectively. The BI method looks closely at surface water and groundwater interaction. The water quality interaction was assessed through multivariate statistical analyses based on analytical quantitative data. Multivariate statistical analyses were used, including discriminant analysis (DA) and principal component analysis (PCA), based on 36 water quality parameters from the rivers, lakes, and groundwater sites at Jenderam Hilir, which were collected from 2009 to 2011 (56 observations). The DA identified six significant parameters (pH, NO2, NO3, F, Fe, and Mn) from 36 variables to distinguish between the river, lake, and groundwater groups (classification accuracy = 98%). The PCA confirmed and identified 10 possible causes of variation in the groundwater quality with an eigenvalue greater than 1, which explained 82.931% of the total variance in the water quality data set: hydrochemistry; redox conditions; groundwater flow; surface runoff; groundwater contact with rock or weathered rock (Kenny Hills Formation); river water and groundwater interaction, discharge of untreated sewage and industrial wastes into the river; interaction between the river, lake, and groundwater; heavy metal pollution from past mining activities; and industrial pollution, such as dye or paint operations. Hence, in order to evaluate the effects of groundwater pumping and BI operation on the operations of wells, as well as to determine the effects of pumping rate on flow paths, travel time, size of pumping, and to capture zone delineation and groundwater mixing in a pumping well, numerical modelling simulation using MODFLOW and MODPATH were used. The simulation involved trying different scenarios by changing the variables to perform infiltration safely and to achieve the ideal pumping rate. The results indicated that the migration of river water into the aquifer has been generally slow and depended on the pumping rate and the distance from the pumping well to the river. Most water arrived at the well by the end of the pumping period of 1 to 5 days at 3072 m3/day for test wells of DW1 and DW2, and during the simultaneous pumping for DW2 and PW1 for a well located 40 m and 20 m, respectively, from the river. During the 9.7-day pumping period, 33% of the water pumped from the DW1 well was river water, and 38% percent of the water pumped from DW2 throughout 4.6 days was river water. The models provided necessary information to water operators in the design, and the construction of pumping and sampling schedules of the BI practices. The detailed field investigation programmes obtained from this study can be used for preliminary BI system design and pilot construction by stakeholders, water operator or related agencies in Malaysia, as the study provided good methods and investigation (geophysical, isotope, grain size analysis, colloidal borescope systems and water quality analysis). The statistical and numerical methods to study the BI may also be used in other areas of similar hydrological characteristics and climate conditions.

Preview PDF FPAS 2014 3 IR.pdf Download (1MB) | Preview

Actions (login required)

View Item